Catalytic Amine Oxidation under Ambient Aerobic Conditions: Mimicry of Monoamine Oxidase B

Article abstract of DOI:10.1002/anie.201503654

The flavoenzyme monoamine oxidase (MAO) regulates mammalian behavioral patterns by modulating neurotransmitters such as adrenaline and serotonin. The mechanistic basis which underpins this enzyme is far from agreed upon. Reported herein is that the combination of a synthetic flavin and alloxan generates a catalyst system which facilitates biomimetic amine oxidation. Mechanistic and electron paramagnetic (EPR) spectroscopic data supports the conclusion that the reaction proceeds through a radical manifold. This data provides the first example of a biorelevant synthetic model for monoamine oxidase B activity.

Full text of DOI:10.1002/anie.201503654

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201503654
German Edition: DOI: 10.1002/ange.201503654
Enzymes
Catalytic Amine Oxidation under Ambient Aerobic Conditions:
Mimicry of Monoamine Oxidase B**
Alexander T. Murray, Myles J. H. Dowley, Fabienne Pradaux-Caggiano,
Amgalanbaatar Baldansuren, Alistair J. Fielding, Floriana Tuna, Christopher H. Hendon,
Aron Walsh, Guy C. Lloyd-Jones, Matthew P. John, and David R. Carbery*
Abstract: The flavoenzyme monoamine oxidase (MAO)
regulates mammalian behavioral patterns by modulating
neurotransmitters such as adrenaline and serotonin. The
mechanistic basis which underpins this enzyme is far from
agreed upon. Reported herein is that the combination of
a synthetic flavin and alloxan generates a catalyst system which
facilitates biomimetic amine oxidation. Mechanistic and elec-
tron paramagnetic (EPR) spectroscopic data supports the
conclusion that the reaction proceeds through a radical mani-
fold. This data provides the first example of a biorelevant
synthetic model for monoamine oxidase B activity.
aggression, known as the “warrior gene”, ultimately impact-
ing human evolution.[2] Inhibition of MAO has been an
important area for medicinal chemistry with MAO inhibitors
(MAOIs) acting as potent antidepressants and having poten-
tial applications as neuroprotective agents.[3] Mechanistic
studies have also helped in understanding the role of lysine-
specific demethylase 1 (LSD1), a key epigenetic modulator,
with MAOIs impacting a number of key biological process-
es.[4]
It is remarkable that no consensus has been reached with
respect to a mechanism of action, despite over 45 years of
investigation.[5] There are two isozymes of MAO: MAO-A
and MAO-B. While the flavin active sites are identical, each
form displays a different substrate and inhibitor profile, and
the mechanistic basis of this selectivity unknown.[6]
M
onoamine oxidase (MAO) is a mitochondrial flavin-
dependent oxidoreductase enzyme which oxidizes a range of
important amines to imines, for example, the neurotransmit-
ters serotonin, histamine, and noradrenaline.[1] With such an
integral role in the neurochemical network, MAO function
has been pinpointed as an underlying rationale for a range of
behavioral, evolutionary, and physiological observations. For
example, variations in the MAO A gene can lead to increased
2H primary kinetic isotope (KIE) effects have been
À
observed for the C H bond cleavage step(s) with both
MAO A and B. In principle, rate-contributing cleavage may
be envisaged as proceeding by either H+-, HÀ-, or HC-transfer
mechanisms (Scheme 1). These options have been widely
+
discussed,[5] with rate-contributing C H cleavage by H
À
transfer being the most prevalent mechanistic description.
Two mechanistic postulates have been developed to account
for the requisite increase in acidity of the relevant a-amino
[*] A. T. Murray, M. J. H. Dowley, Dr. F. Pradaux-Caggiano,
C. H. Hendon, Prof. A. Walsh, Dr. D. R. Carbery
Department of Chemistry, University of Bath
Claverton Down, Bath (UK)
À
C H bond: the formation of a covalent flavin–amine
conjugate,[7] and the formation of an aminium radical
cation[8] after single-electron transfer from amine to flavin.
As both mechanisms require discrete steps prior to the rate-
E-mail: d.carbery@bath.ac.uk
Dr. A. Baldansuren, Dr. A. J. Fielding, Dr. F. Tuna
EPSRC National EPR Facility, Photon Science Institute
School of Chemistry
À
contributing C H cleavage, it is notable that no intermediates
University of Manchester, Oxford Road, Manchester (UK)
À
accumulate to observable populations. C H cleavage in the
context of a direct hydride transfer has also been suggested.[9]
However, such a synchronous event would not be consistent
with the 15N KIE measured for amine oxidation by MAO B,
Prof. G. C. Lloyd-Jones
School of Chemistry, Joseph Black Building
West Mains Road, Edinburgh EH9 3 JJ (UK)
Dr. M. P. John
GlaxoSmithKline Research and Development
Gunnels Wood Road, Stevenage (UK)
À
thus pointing to an absence of synchronicity between C H
cleavage and sp2!sp3 nitrogen atom re-hybridization.[10]
Finally, HC transfer from the substrate to the flavin has been
[**] We acknowledge funding of a studentship (A.T.M.) from GSK,
EPSRC, and the University of Bath. The Bath, Bristol, and Cardiff
Catalysis alliance is acknowledged for support (F.P.-C.). C.H.H. and
A.W. are supported by the ERC (Grant 277757) and the computing
was enabled by access to ARCHER through membership of the UK’s
HPC Materials Chemistry Consortium, which is funded by EPSRC
(Grant No. EP/L00202).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201503654.
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co.
KGaA. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly
cited.
Scheme 1. MAO-catalyzed oxidation of amines and qualitative overview
À
of possible modes of C H bond cleavage.
Angew. Chem. Int. Ed. 2015, 54, 8997 –9000
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8997

.
Angewandte
Communications
suggested.[11] This possibility was discarded on the grounds
that no hydrogen-atom abstracting moiety, which was reactive
Generally, high yields of imine products are attainable,
although substrates with a strongly electron-withdrawing
para-substituent group (entry 9) are less reactive, thus
mirroring MAO B reactivity trends.
À
enough to overcome relevant a-amino C H bond dissociation
energies, could be identified in the enzyme active site.[12]
Studies using synthetic flavins have played a crucial role in
elucidating flavoenzyme mechanisms.[13] Accordingly, insight
gained from studying model cofactors is a valid strategy to
unlocking mechanistic problems in flavoenzymology. Pio-
neering work on primary amines by various groups supported
the polar, proton-transfer mechanism, but the low turnover,
tendency of catalysts to decompose, and requirement of
heating in an enriched O2 atmosphere for several days meant
that they are perhaps of limited relevance to biological
processes.[14] We,[15] and others,[16] have previously applied
cationic flavin catalysts in biomimetic monooxygenase con-
texts, as well as donor–acceptor chemistry,[17] and now report
the oxidation of biologically pertinent amines as a vehicle to
understanding MAO mechanism.
Upon attempted in situ 1H NMR analysis, the inability to
locate the lock signal suggested paramagnetic behavior.
Accordingly, EPR studies at the X-band were initiated.
Mixing 2a and Me2S generated the flavin radical cation 2a’
(Figure 1). The structure was further confirmed by pulsed
Initial exploratory studies demonstrated catalytic aerobic
oxidation of benzylamine, with formation of the imine 4a
being consistent with oxidase rather than monooxygenase-
like reactivity (Table 1).[18] Excellent yields of 4a were
obtained if a thioether additive (Me2S) and a cocatalyst,
alloxan (3a), were used (Table 1). Initially 3a was present as
an undetected by-product from the synthesis of 2a, however,
was found to be crucial for this transformation. N,N-
dimethylalloxan (3b) was found to be inactive (entry 3)
despite possessing structural similarity to 3a. Additionally,
cobalamin synthase, BluB, has been implicated in the
cannibalization of flavin mononucleotide to form alloxan,
which acts as a crucial multifunctional redox catalyst in the
biosynthesis of vitamin B12.[19] A series of substituted benzyl-
amines, typical substrates for MAO-B, have been examined.
Figure 1. EPR spectra of 2a’ and 4a’, and DFT-calculated spin den-
sities measured from solutions of 2a + Me2S (top) and 2a + Me2S +
3a + 1a (bottom).
EPR studies. In particular, the protonation state of 2a’ was
assessed by electron spin echo envelope modulation
(ESEEM), and is a rare example of an aerobically generated
flavin semiquinone, having demonstrable relevance to catal-
ysis, observed by EPR spectroscopy.[20] The use of a strong
hydrogen-bonding solvent, trifluoroethanol, may aid the
stabilization of the semiquinone formation, as discussed by
Massey and co-workers, for flavins with amino acids.[21] Upon
sequential addition of alloxan and amine, a new EPR
spectrum was observed and characterized as the radical 4a’,
and is consistent with charge-transfer-initiated hydrogen-
atom abstraction from 4a. Hybrid-DFT and post-Hartree
Fock calculations were performed on 2a’ and 4a’ and the spin
density isosurfaces are shown in Figure 1.[22] Importantly, the
theoretical calculations quantify the local spin density dis-
tribution, thus further corroborating the simulations of the
continuous-wave EPR spectra.[23]
Table 1: Flavin-organocatalyzed amine oxidation.[a]
Entry
Substrate
R1
Product
Yield [%]
1
1a
1a
1a
1b
1c
1d
1e
1 f
1g
1h
1i
1j
1k
1l
1m
1n
1o
Ph
Ph
Ph
4a
4a
–
94
49
0
2[b]
3[c]
4
4-MeC6H4
4-MeOC6H4
4-tBuC6H4
4-FC6H4
4-ClC6H4
4-CF3C6H4
3-MeC6H4
3-OMeC6H4
2-MeC6H4
2-MeOC6H4
2-ClC6H4
2-furyl
4b
4c
4d
4e
4 f
4g
4h
4i
4j
4k
4l
4m
4n
4o
99
95
92
98
77
37
91
72
96
68
66
70
87
72
5
6
7[d]
8[d]
9[d,e]
10
11
12
13
14[d]
15[d]
16
17
Kinetic studies provided additional important mechanistic
information with the transformation being first order in
benzylamine[24] and showing a KIE of kH/kD = 1.9 when using
À
PhCD2NH2 (7), thus supporting rate-contributing C H bond
cleavage (Figure 2). A range of studied para-substituted
benzylamines provided a negative Hammett correlation
(1 = À2).
2-thiophenyl
1-naphthyl
[a] Reaction conditions: 2a (2 mol%), 3a (2 mol%), Me2S (10 equiv),
5 h. [b] 2b used. [c] 3b used. [d] 18 h. [e] 2a (4 mol%), 3a (4 mol%)
used.
The observed rates of reaction were found to be
independent of the Me2S concentration. Kinetic analysis for
8998 www.angewandte.org
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 8997 –9000

Angewandte
Chemie
alloxan. Additionally, a purple by-product, consistent with the
dye murexide (6; UV/vis lmax = 521 nm; lit = 520 nm),[30] is
observed to accumulate from 3a’’, 3a, and ammonia This
observation is consistent with a two-electron over-reduction
of 3a, thus leading to catalyst deactivation and suggesting that
3a’’ is not a catalytically active species.
À
This model study supports a homolytic C H bond
cleavage mediated by a flavin semiquinone, and with a sub-
strate preference for benzylamines, it has prompted us to ask
whether any reasonable insight into the enzymatic mechanism
of MAO B can be achieved through consideration of this
currently presented model system. A linear correlation exists
between the substrate pKa value and steady-state kcat for
MAO B (Figure 3),[31] and is consistent with a neutral amine
substrate. It is significant that Hammett electronic correla-
tions for MAO B are only apparent at pH 9.0. The similarity
of the model’s KIE and Hammett profiles to the equivalent B
isozyme data, when the enzyme kinetics are measured at
pH 9.0, which is similar to this unbuffered system, is notable,
(MAO B: kH/kD = 2.25, 1 = À0.9 at pH 9.0).[32]
Figure 2. Hammett and kinetic isotope effect study carried out by
HPLC analysis of imine formation from amines against an internal
standard of naphthalene.
3a did not demonstrate a simple reaction order, with
saturation behavior observed over the concentrations exam-
ined (see the Supporting Information). The kinetic order in
2a was probed by means of a ln(kobs) versus ln([flavin]) plot,
which was linear with a slope of 0.25 and consistent with de-
aggregation of a higher order resting state, but with a mono-
meric semiquinone being catalytically active. Significantly,
Our proposal for the MAO B mechanism is informed by
the presented data, the substrate reactivity trends, and the pH
sensitivity of MAO B.[34] This mechanistic suggestion centers
upon a charge-transfer event promoted by the free-base
substrate interacting with an electron-rich phenol of Y398
near the flavin acceptor, as demonstrated by Scrutton and co-
workers.[35] This acceptor is itself activated by the H2O–K296
hydrogen-bonding motif. The neutral semiquinone thus
formed can mediate hydrogen-atom transfer from the sub-
strate, with the tyrosinyl radical cation now able to accept the
second substrate electron, in direct analogy to the role played
the less oxidizing flavin 2b also mediates this reaction (kobs2a
/
k
obs2b = 2.96) with an electrochemical reduction potential of
+ 66 mV vs. SHE, which parallels MAO-B at + 40 mV
(Table 1, entry 2)[25] Therefore, the flavin catalysts 2a,b offer
themselves as realistic mimics of MAO through the neutral
N(5)-H semiquinone.[26]
A mechanism that accounts for EPR and kinetic data is
underpinned by the realization that rate-determining C H
cleavage is mediated by 2a’ (Scheme 2). The radical cation 2a’
is formed by a proton-coupled electron
transfer from Me2S, as observed by
EPR. BnNH2 promotes the formation
of 2a’ by mediating the de-aggregation
and deprotonation of 2a’, thus gener-
ating the neutral semiquinone 2a’’, with
À
subsequent H
*
transfer, initiated by
a charge-transfer event, from 1a to 2a’’.
An a-amino radical is formed (1a’),
and it acts as a potent reductant,[27] thus
reducing alloxan and forming 1a’’.
Electron transfer from a-amino radi-
cals to vicinal dicarbonyl compounds is
regarded as one of the fastest reactions
between a radical and a neutral closed-
shell organic molecule.[28] Alloxan (3a)
reacts as an amide tautomer, thus
allowing stabilization of a developing
oxyanion character, a feature which is
impossible for the inactive 3b (Table 1,
entry 3). This captodative-stabilized
radical[29] subsequently reacts with O2,
thus generating 5. The peroxyl radical 5
oxidizes 2a’’’ to 2a’, thus forming the
hydroperoxide 5’ and completing the
catalytic cycle. Formation of stoichio-
metric DMSO is observed. Therefore
Scheme 2. Proposed reaction mechanism showing amine oxidation mediated by the key flavin
Me2S mediates the reduction of 5’ to semiquinone 2a’’.
Angew. Chem. Int. Ed. 2015, 54, 8997 –9000
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 8999

.
Angewandte
Communications
[14] a) G. A. Hamilton, L. E. Brown, J. Am. Chem. Soc. 1970, 92,
7225 – 7227; b) J. M. Kim, I. S. Cho, P. S. Mariano, J. Org. Chem.
1991, 56, 4943 – 4955; c) J. M. Kim, M. A. Bogdan, P. S. Mariano,
J. Am. Chem. Soc. 1993, 115, 10591 – 10595; d) W.-S. Li, L. M.
Sayre, Tetrahedron 2001, 57, 4523 – 4536.
[15] a) B. J. Marsh, D. R. Carbery, Tetrahedron Lett. 2010, 51, 2362 –
2365; b) A. T. Murray, P. Matton, N. W. G. Fairhurst, M. P. John,
D. R. Carbery, Org. Lett. 2012, 14, 3656 – 3659.
[16] a) Y. Imada, H. Iida, S. Ono, S.-I. Murahashi, J. Am. Chem. Soc.
2003, 125, 2868 – 2869; b) Y. Imada, H. Iida, S.-I. Murahashi, T.
Naota, Angew. Chem. Int. Ed. 2005, 44, 1704 – 1706; Angew.
Chem. 2005, 117, 1732 – 1734; c) S. Chen, M. S. Hossain, F. W.
Foss, Org. Lett. 2012, 14, 2806 – 2809; d) S. Chen, F. W. Foss, Org.
Figure 3. a) MAO B steady-state kcat relationship to substrate pKa;
relevant kcat values from Ref. [31]. b) Crystal structure of MAO B with
4-nitrobenzylamine inhibitor.[33]
ˇ
ˇ
Lett. 2012, 14, 5150 – 5153; e) J. Zurek, R. Cibulka, H. Dvorµ-
kovµ, J. Svoboda, Tetrahedron Lett. 2010, 51, 1083 – 1086; f) H.
ˇ
Kotoucovµ, I. Strnadovµ„ M. Kovandovµ, J. Chudoba, H.
by alloxan in the currently discussed model. Indeed, both
components can be viewed as redox-active hydroxylated
units.
ˇ
Dvorµkovµ, R. Cibulka, Org. Biomol. Chem. 2014, 12, 2137 –
2142.
[17] a) A. Niemz, J. Imbriglio, V. M. Rotello, J. Am. Chem. Soc. 1997,
119, 887 – 892; b) J. B. Carroll, B. J. Jordan, H. Xu, B. Erdogan, L.
Lee, L. Cheng, C. Tiernan, G. Cooke, V. M. Rotello, Org. Lett.
2005, 7, 2551 – 2554; c) S. T. Caldwell, L. J. Farrugia, S. G.
Hewage, N. Kryvokhyzha, V. M. Rotello, G. Cooke, Chem.
Commun. 2009, 1350 – 1352.
[18] S. Oae, K. O. Asada, T. Yoshimura, K. Fujimori, Heterocycles
1992, 33, 189 – 194.
[19] X.-L. Wang, J.-M. Quan, J. Am. Chem. Soc. 2011, 133, 4079 –
4091.
In summary, an aerobic, catalytic oxidation of benzyl-
amines which mimics MAO B activity proceeding through
charge-transfer-initiated substrate HC abstraction has been
developed. EPR spectroscopy has revealed the operation of
an aerobically generated flavin semiquinone. KIE and
Hammett studies have demonstrated a pH-dependent kinetic
parallel to MAO B activity. This model system has opened up
an additional mechanistic model of MAO B activity, that is,
a charge-transfer event is harnessed to access a reactive
[20] T. Akiyama, F. Simeno, M. Murakami, F. Yoneda, J. Am. Chem.
Soc. 1992, 114, 6613 – 6620.
[21] D. Xu, R. M. Kohli, V. Massey, Proc. Natl. Acad. Sci. USA 1999,
96, 3556 – 3561.
À
neutral flavin semiquinone as the C H abstracting species in
MAO B.
[22] C. H. Hendon, D. R. Carbery, A. Walsh, Chem. Sci. 2014, 5,
1390 – 1395.
[23] For full discussion of EPR spectroscopy and calculations, see the
Supporting Information.
[24] The reaction surprisingly becomes apparently zero order in
amine after 1 h of reaction time, perhaps because oxygen
becomes rate contributing.
Keywords: amines · enzymes · EPR spectroscopy · oxidation ·
reaction mechanisms
How to cite: Angew. Chem. Int. Ed. 2015, 54, 8997–9000
Angew. Chem. 2015, 127, 9125–9128
[25] a) W.-S. Li, N. Zhang, L. M. Sayre, Tetrahedron 2001, 57, 4507 –
4522; b) D. P. H. M. Heuts, N. S. Scrutton, W. S. McIntire, M. W.
Fraaije, FEBS J. 2009, 276, 3405 – 3427.
[26] ES-MS monitoring did not reveal any catalyst–amine adducts
that would be expected from polar mechanisms.
[27] I. R. Gould, J. R. Lenhard, A. A. Muenter, S. A. Godleski, S.
Farid, J. Am. Chem. Soc. 2000, 122, 11934 – 11943.
[28] J. C. Scaiano, J. Phys. Chem. 1981, 85, 2851 – 2855.
[29] H. G. Viehe, Z. Janousek, R. Merenyi, L. Stella, Acc. Chem. Res.
1985, 18, 148 – 154.
[30] J. Torreilles, M.-C. Guerin, M.-L. Carrie, Biochimie 1989, 71,
1231 – 1234.
[31] A. K. Tan, R. R. Ramsay, Biochemistry 1993, 32, 2137 – 2143.
[32] a) T. Jonsson, D. E. Edmondson, J. P. Klinman, Biochemistry
1994, 33, 14871 – 14878; b) R. Orru, M. Aldeco, D. E. Edmond-
son, J. Neural Transm. 2013, 120, 847 – 851.
[1] C. M. McEwen, J. Biol. Chem. 1965, 240, 2003 – 2010.
[2] R. McDermott, D. Tingley, J. Cowden, G. Frazzetto, D. D. P.
Johnson, Proc. Natl. Acad. Sci. USA 2009, 106, 2118 – 2123.
[3] M. B. H. Youdim, D. Edmondson, K. F. Tipton, Nat. Rev.
Neurosci. 2006, 7, 295 – 309.
[4] K. Helin, D. Dhanak, Nature 2013, 502, 480 – 488.
[5] D. E. Edmondson, C. Binda, J. Wang, A. K. Upadhyay, A.
Mattevi, Biochemistry 2009, 48, 4220 – 4230.
[6] L. Pisani, M. Barletta, R. Soto-Otero, O. Nicolotti, E. Mendez-
Alvarez, M. Catto, A. Introcaso, A. Stefanachi, S. Cellamare, C.
Altomare, A. Carotti, J. Med. Chem. 2013, 56, 2651 – 2664.
[7] D. E. Edmondson, C. Binda, A. Mattevi, Arch. Biochem.
Biophys. 2007, 464, 269 – 276.
[8] R. B. Silverman, Acc. Chem. Res. 1995, 28, 335 – 342.
[9] a) P. F. Fitzpatrick, Arch. Biochem. Biophys. 2010, 493, 13 – 25;
ˇ
b) R. Vianello, M. Repic, J. Mavri, Eur. J. Org. Chem. 2012,
[33] M. Li, C. Binda, A. Mattevi, D. E. Edmondson, Biochemistry
2006, 45, 4775 – 4784.
[34] a) L. De Colibus, M. Li, C. Binda, A. Lustig, D. E. Edmondson,
A. Mattevi, Proc. Natl. Acad. Sci. USA 2005, 102, 12684 – 12689;
b) C. Binda, P. Newton-Vinson, F. Hubalek, D. E. Edmondson,
A. Mattevi, Nat. Struct. Biol. 2002, 9, 22 – 26.
7057 – 7065.
[10] S. MacMillar, D. E. Edmondson, O. Matsson, J. Am. Chem. Soc.
2011, 133, 12319 – 12321.
[11] M. C. Walker, D. E. Edmondson, Biochemistry 1994, 33, 7088 –
7098.
[12] T. L. Cottrell, The Strength of Chemical Bonds, 2nd ed., Butter-
worth and co., London, 1958.
[13] a) C. Kemal, T. C. Bruice, Proc. Natl. Acad. Sci. USA 1976, 73,
995 – 999; b) C. Kemal, T. C. Bruice, J. Am. Chem. Soc. 1977, 99,
7064 – 7067; c) T. C. Bruice, J. B. Noar, S. S. Ball, U. V. Venka-
taram, J. Am. Chem. Soc. 1983, 105, 2452 – 2463.
[35] R. V. Dunn, A. W. Munro, N. J. Turner, S. E. J. Rigby, N. S.
Scrutton, ChemBioChem 2010, 11, 1228 – 1231.
Received: April 21, 2015
Published online: June 18, 2015
9000 www.angewandte.org
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 8997 –9000