Synthesis and characterization of model compounds of the lysine tyrosyl quinone cofactor of lysyl oxidase

Article abstract of DOI:10.1021/ja0214274

4-n-Butylamino-5-ethyl-1,2-benzoquinone (1_ox) has been synthesized as a model compound for the LTQ (lysine tyrosyl quinone) cofactor of lysyl oxidase (LOX). At pH 7, 1_ox has a λ_max at 504 nm and exists as a neutral o-quino

Full text of DOI:10.1021/ja0214274

Published on Web 04/29/2003
Synthesis and Characterization of Model Compounds of the
Lysine Tyrosyl Quinone Cofactor of Lysyl Oxidase
Minae Mure, Sophie X. Wang, and Judith P. Klinman*,†
Contribution from the Department of Chemistry, UniVersity of California,
Berkeley, California 94720-1460
Received December 10, 2002; E-mail: klinman@socrates.berkeley.edu
Abstract: 4-n-Butylamino-5-ethyl-1,2-benzoquinone (1ox) has been synthesized as a model compound for
the LTQ (lysine tyrosyl quinone) cofactor of lysyl oxidase (LOX). At pH 7, 1ox has a λmax at 504 nm and
exists as a neutral o-quinone in contrast to a TPQ (2,4,5-trihydroxyphenylalanine quinone) model compound,
4, which is a resonance-stabilized monoanion. Despite these structural differences 1ox and 4 have the
same redox potential (ca. -180 mV vs SCE). The structure of the phenylhydrazine adduct of 1ox (2) is
reported, and 2D NMR spectroscopy has been used to show that the position of nucleophilic addition is at
C1. UV-vis spectroscopic pH titration of phenylhydrazine adducts of 1ox and 4, 2, and 11, respectively,
reveals a similar red shift in λmax at alkaline pH with the same pKa (∼11.8). In contrast, the red shift in λmax
at acidic pH conditions yields different pKa values (2.12 for 2 vs -0.28 for 11), providing a means to
distinguish LTQ from TPQ. Reactions between in situ generated 4-ethyl-1,2-benzoquinone and primary
amines give a mixture of products, indicating that the protein environment must play an essential role in
LTQ biogenesis by directing the nucleophilic addition of the ꢀ-amino group of a lysine residue to the C4
position of a putative dopaquinone intermediate. Characterization of a 1,6-adduct between an o-quinone
and butylamine (3-n-butylamino-5-ethyl-1,2-benzoquinone, 13) confirms the assignment of LTQ as a 1,4-
addition product.
functions such as developmental regulation,8 tumor sup-
pression,9-13 cell growth control,14,15 and cell adhesion.16,17
Introduction
Lysyl oxidase (LOX; EC 1.4.3.13) is a copper-containing
enzyme that plays an essential role in stabilizing extracellular
matrices.1-3 It catalyzes the oxidative deamination of the
ꢀ-amino group of peptidyl lysine in collagen and elastin,
generating peptidyl R-aminoadipic-δ-semialdehyde. The peptidyl
aldehyde then undergoes nonenzymatic condensation with either
a neighboring aldehyde or an unmodified lysine side chain to
form the intra- and intercovalent cross-linking of collagen and
elastin. These molecules are deposited as insoluble fibers that
provide strength and/or elasticity to connective tissues. Altered
LOX activity has been associated with many disorders such as
vascular rupture and torsion, loose skin and joints, and fibrotic
diseases.3-5 Assays in vitro have shown that LOX can also
oxidatively deaminate a variety of aliphatic monoamines and
diamines, synthetic elastin-like polypeptides, and lysine in
histone H1 and other proteins.6,7 Recently, increasing numbers
of publications have suggested that LOX may possess alternative
While the physiological role of LOX has been well studied, its
structure has not been solved, and the catalytic mechanism is
not fully characterized.
LOX has been classified as a copper amine oxidase (CAO;
EC 1.4.3.6), but it is distinguished from other CAOs by having
a monomeric structure (32 kDa), whereas the CAOs are
homodimers (75-80 kDa per monomer).18-22 Analogous to
other CAOs, LOX is believed to contain an organic cofactor
and one copper(II) at its active site.23,24 The nature of the organic
(8) Butler, E.; Handlin, J.; Benson, S. Exp. Cell Res. 1987, 173, 174-182.
(9) Kenyon, K.; Contente, S.; Trackman, P. C.; Tang, J.; Kagan, H. M.;
Friedman, R. M. Science 1991, 253, 802.
(10) Hajnal, A.; Klemenz, R.; Schafer, R. Cancer Res. 1993, 53, 4670-4675.
(11) Oberhuber, H.; Seliger, B.; Schafer, R. Mol. Carcinog. 1995, 12, 198-
204.
(12) Peyrol, S.; Raccurt, M.; Gerard, F.; Gleyzal, C.; Grimaud, J. A.; Sommer,
P. Am. J. Pathol. 1997, 150, 497-507.
(13) Ren, C. Z.; Yang, G.; Timme, T. L.; Wheeler, T. M.; Thompson, T. C.
Cancer Res. 1998, 58, 1285-1290.
(14) Csiszar, K.; Entersz, I.; Trackman, P. C.; Samid, D.; Boyd, C. D. Mol.
Biol. Rep. 1996, 23, 97-108.
† J.P.K. is also with the Department of Molecular and Cell Biology,
University of California, Berkeley, CA 94720-1460.
(15) Li, W. D.; Liu, G. M.; Chou, I. N.; Kagan, H. M. J. Cell. Biochem. 2000,
78, 550-557.
(1) Csiszar, K. Prog. Nucleic Acid Res. Mol. Biol. 2001, 70, 1-32.
(2) Rucker, R. B.; Kosonen, T.; Clegg, M. S.; Mitchell, A. E.; Rucker, B. R.;
UriuHare, J. Y.; Keen, C. L. Am. J. Clin. Nutr. 1998, 67, S996-S1002.
(3) SmithMungo, L. I.; Kagan, H. M. Matrix Biol. 1998, 16, 387-398.
(4) Kagan, H. M. Pathol. Res. Pract. 1994, 190, 910-919.
(5) Streichenberger, N.; Peyrol, S.; Philit, F.; Loire, R.; Sommer, P.; Cordier,
J. F. Virchows. Arch. 2001, 439, 78-84.
(6) Kagan, H. M.; Williams, M. A.; Williamson, P. R.; Anderson, J. M. J.
Biol. Chem. 1984, 259, 11203-11207.
(7) Ohkawa, K.; Fujii, K.; Nishida, A.; Yamauchi, T.; Ishibashi, H.; Yamamoto,
H. Biomacromolecules 2001, 2, 773-779.
(16) Ito, H.; Akiyama, H.; Iguchi, H.; Iyama, K.; Miyamoto, M.; Ohsawa, K.;
Nakamura, T. J. Biol. Chem. 2001, 276, 24023-24029.
(17) Omori, K.; Fujiseki, Y.; Suzukawa, J.; Inagaki, C. Matrix Biol. 2002, 21,
337-348.
(18) McIntire, W. S.; Hartmann, C. In Principles and applications of quino-
proteins; Davidson, V. L., Ed.; Dekker: New York, 1993; pp 97-171.
(19) Knowles, P. F.; Dooley, D. M. In Metal ions in biological systems; Sigel,
H., Sigel, A., Eds.; Dekker: New York, 1994; Vol. 30, pp 360-403.
(20) Klinman, J. P.; Mu, D. Annu. ReV. Biochem. 1994, 63, 299-344.
(21) Jalkanen, S.; Salmi, M. EMBO J. 2001, 20, 3893-3901.
(22) Mure, M.; Mills, S. A.; Klinman, J. P. Biochemistry 2002, 41, 9269-9278.
9
10.1021/ja0214274 CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 6113-6125
6113

A R T I C L E S
Mure et al.
Chart 1. Structures of Quinonoid Cofactors
cofactor has been controversial and was once thought to be
pyrroloquinoline quinone (PQQ).25 In 1996, following the
discovery of a new quinonoid cofactor, topaquinone (TPQ) of
a CAO,26 lysine tyrosylquinone (LTQ) was identified as the
cofactor of LOX.27 LTQ is a new variant of the quinonoid
cofactor and derives from cross-linking of a modified tyrosine
with a lysine residue within the same polypeptide chain. There
are two other identified cross-linked quinonoid cofactors,
namely, tryptophan tryptophylquinone (TTQ)28 and cysteine
tryptophylquinone (CTQ) in bacterial amine dehydrogenases.29-31
Among the quinonoid cofactors, TPQ has been the most
extensively studied.20,22,32 TPQ has been shown to form from a
tyrosine residue by a self-catalytic process that requires only
copper and dioxygen; a presumed intermediate is dopaquinone,
which can undergo attack by hydroxide ion to yield the mature
cofactor.32-35 A number of compounds that model TPQ have
been synthesized, and their ability to catalyze the oxidative
deamination of primary amines was examined.36-42 Further,
TPQ derivatives that mimic both the proposed reaction inter-
mediates and products of CAOs with substrate amines and
inhibitors have been prepared and characterized with regard to
their spectroscopic and redox properties.37-39 These solution
studies of TPQ model compounds and their derivatives have
played an important role in helping to understand the mechanism
of enzymatic reaction (see Chart 1).
It has been proposed that nucleophilic addition of the ꢀ-amino
group of a lysine residue to the dopaquinone intermediate,
derived from a tyrosine residue as proposed in TPQ biogenesis,
yields LTQ.27 Analogous cross-linking of a modified tyrosine
with glutathione or histidine has been observed in insects.43-45
Further, such chemistry has been implicated in Lewy body
formation (neuronal inclusions)46 and also in degeneration of
dopaminergic neurons47 where both processes are proposed to
be related to Parkinson’s disease. To date, detailed investigations
of the mechanism of LTQ formation in LOX have been limited
by the lack of a suitable in vitro overexpression system.
To fully understand the mechanism of LTQ formation in and
subsequent to catalysis by LOX, it is important to characterize
the basic chemical and physiological properties of LTQ. We
have used the compound 4-n-butylamino-5-ethyl-1,2-benzo-
quinone (1ox) as a model of LTQ.27,48 The structure of LTQ in
LOX was first elucidated by comparison of the spectroscopic
properties between the phenylhydrazine adduct of enzyme and
the phenylhydrazine derivative of 1ox (2).27 A subsequent study
showed that resonance Raman spectra of underivatized LOX
were in excellent agreement with that of 1ox.48 Recent studies
of LTQ model compounds focused on their reactivity toward
amines and decomposition during the catalytic turnover reac-
tion.49,50 However, there has been no systematic characterization
of the spectroscopic, acid-base, and electrochemical properties
of any LTQ model compound as they relate to the cofactor in
LOX.
(23) Gacheru, S. N.; Trackman, P. C.; Shah, M. A.; Ogara, C. Y.; Spacciapoli,
P.; Greenaway, F. T.; Kagan, H. M. J. Biol. Chem. 1990, 265, 19022-
19027.
(24) Iguchi, H.; Sano, S. Connect. Tissue Res. 1985, 14, 129-139.
(25) Williamson, P. R.; Moog, R. S.; Dooley, D. M.; Kagan, H. M. J. Biol.
Chem. 1986, 261, 16302-16305.
(26) Janes, S. M.; Mu, D.; Wemmer, D.; Smith, A. J.; Kaur, S.; Maltby, D.;
Burlingame, A. L.; Klinman, J. P. Science 1990, 248, 981-987.
(27) Wang, S. X.; Mure, M.; Medzihradszky, K. F.; Burlingame, A. L.; Brown,
D. E.; Dooley, D. M.; Smith, A. J.; Kagan, H. M.; Klinman, J. P. Science
1996, 273, 1078-1084.
(28) McIntire, W. S.; Wemmer, D. E.; Chistoserdov, A.; Lidstrom, M. E. Science
1991, 252, 817-824.
(29) Datta, S.; Mori, Y.; Takagi, K.; Kawaguchi, K.; Chen, Z. W.; Okajima,
T.; Kuroda, S.; Ikeda, T.; Kano, K.; Tanizawa, K.; Mathews, F. S. Proc.
Natl. Acad. Sci. U.S.A. 2001, 98, 14268-14273.
(30) Vandenberghe, I.; Kim, J. K.; Devreese, B.; Hacisalihoglu, A.; Iwabuki,
H.; Okajima, T.; Kuroda, S.; Adachi, O.; Jongejan, J. A.; Duine, J. A.;
Tanizawa, K.; Van Beeumen, J. J. Biol. Chem. 2001, 276, 42923-42931.
(31) Satoh, A.; Kim, J. K.; Miyahara, I.; Devreese, B.; Vandenberghe, I.;
Hacisalihoglu, A.; Okajima, T.; Kuroda, S.; Adachi, O.; Duine, J. A.; Van
Beeumen, J.; Tanizawa, K.; Hirotsu, K. J. Biol. Chem. 2002, 277, 2830-
2834.
(32) Schwartz, B.; Klinman, J. P. Vitam. Horm. 2001, 61, 219-239.
(33) Klinman, J. P. J. Biol. Chem. 1996, 271, 27189-27192.
(34) Dooley, D. M. J. Biol. Inorg. Chem. 1999, 4, 1-11.
(35) Kim, M.; Okajima, T.; Kishishita, S.; Yoshimura, M.; Kawamori, A.;
Tanizawa, K.; Yamaguchi, H. Nat. Struct. Biol. 2002, 9, 591-596.
(36) Mure, M.; Klinman, J. P. J. Am. Chem. Soc. 1993, 115, 7117-7127.
(37) Mure, M.; Klinman, J. P. J. Am. Chem. Soc. 1995, 117, 8698-8706.
(38) Mure, M.; Klinman, J. P. J. Am. Chem. Soc. 1995, 117, 8707-8718.
(39) Mure, M.; Klinman, J. P. Methods Enzymol. 1995, 258, 39-52.
(40) Lee, Y.; Sayre, L. M. J. Am. Chem. Soc. 1995, 117, 11823-11828.
(41) Lee, Y. H.; Sayre, L. M. J. Am. Chem. Soc. 1995, 117, 3096-3105.
(42) Lee, Y. H.; Huang, H.; Sayre, L. M. J. Am. Chem. Soc. 1996, 118, 7241-
7242.
In this paper, we describe the syntheses of 1ox, its reduced
form 1red, and its N-alkylated derivative (3ox) and compare their
spectroscopic and redox properties with TPQ model compounds
(4, 5) and other quinones (6-9). The acid-base properties of
the phenylhydrazone (2) and the analogous 4-nitrophenylhy-
drazine derivative (10) are studied and compared with those of
the corresponding adducts of TPQ (11, 12). In relation to LTQ
(46) Montine, T. J.; Farris, D. B.; Graham, D. G. J. Neuropathol. Exp. Neurol.
1995, 54, 311-319.
(47) Hastings, T. G.; Zigmond, M. J. J. Neurochem. 1994, 63, 1126-1132.
(48) Wang, S. X.; Nakamura, N.; Mure, M.; Klinman, J. P.; SandersLoehr, J. J.
Biol. Chem. 1997, 272, 28841-28844.
(43) Leem, J. Y.; Nishimura, C.; Kurata, S.; Shimada, I.; Kobayashi, A.; Natori,
S. J. Biol. Chem. 1996, 271, 13573-13577.
(44) Xu, R. D.; Huang, X.; Hopkins, T. L.; Kramer, K. J. Insect Biochem. Mol.
Biol. 1997, 27, 101-108.
(49) Ling, K. Q.; Kim, J.; Sayre, L. M. J. Am. Chem. Soc. 2001, 123, 9606-
9611.
(45) Kramer, K. J.; Kanost, M. R.; Hopkins, T. L.; Jiang, H. B.; Zhu, Y. C.;
Xu, R. D.; Kerwin, J. L.; Turecek, F. Tetrahedron 2001, 57, 385-392.
(50) Akagawa, M.; Suyama, K. Biochem. Biophys. Res. Commun. 2001, 281,
193-199.
9
6114 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

LTQ Cofactor of Lysyl Oxidase
A R T I C L E S
Chart 2
) 20/1 v/v) to yield 10.26 g (69% yield) of 4,5-(methylenedioxy)-2-
nitrostyrene as a light yellow solid: 1H NMR (CDCl3) δ 5.404 (1H, d,
J ) 10.9 Hz, -CHdCH2 cis), 5.610 (1H, dd, J ) 0.5, 17.0 Hz, -CHd
CH2 trans and geminal), 6.098 (2H, -OCH2O-), 6.971 (1H, s, ring
proton at C6), 7.187 (1H, dd, J ) 10.9, 17.2 Hz, -CHdCH2), 7.466
(1H, s, ring proton at C3).
(B) Synthesis of 2-Ethyl-4,5-methylenedioxyaniline. The reduction
of 4,5-(methylenedioxy)-2-nitrostyrene was performed according to the
reported method51 with some modifications. NaBH4 (0.95 g, 25 mmol)
was added to a yellow solution of 4,5-(methylenedioxy)-2-nitrostyrene
(2.62 g, 13.6 mmol) in THF (50 mL). Palladium on carbon (10%, 0.4
g) was added in one portion, and the mixture was stirred at room
temperature for 30 min. At this point, the reduction was incomplete as
determined by TLC. Acidification by dropwise addition of 2 N HCl
caused bleaching of the reaction mixture, indicating the completion of
the reduction, which was confirmed by TLC. The catalyst was removed
by filtration, and the filtrate was extracted three times with ethyl acetate.
Then the pH of the aqueous layer was adjusted to >10 by addition of
10 N NaOH and then extracted five times with ethyl acetate. The
combined ethyl acetate layers were dried over MgSO4, and the solvent
was removed under reduced pressure to yield 2.13 g (95%) of 2-ethyl-
4,5-methylenedioxyaniline as a slightly discolored oil. Prolonged drying
under vacuum yielded opaque needles: 1H NMR (CDCl3) δ 1.206 (3H,
t, J ) 7.5 Hz, -CH2CH3), 2.436 (2H, q, J ) 7.5 Hz, -CH2CH3), 3.430
(2H, br s, exchangeable, -NH), 5.829 (2H, s, -OCH2O-), 6.290 (1H,
s, ring proton at C6), 6.604 (1H, s, ring proton at C3).
(C) Synthesis of N-n-Butyl-2-ethyl-4,5-methylenedioxyaniline. 4,5-
(Methylenedioxy)-2-nitrostyrene (1.93 g, 11.7 mmol) in anhydrous THF
(120 mL) was treated with molar equivalents of n-butyllithium (1.6 M
in hexane, 7.3 mL) and n-butyl bromide (1.26 mL) while being cooled
in an ice bath under an atmosphere of dry N2. 1,4-Dioxane (10.5 mmol,
0.89 mL, dehydrated by treating with 4 Å molecular sieves previously
dried at 300 °C) was then added, and the mixture was heated to 60 °C
for 2 h. The solvent was removed by evaporation under reduced
pressure, and the crude product was purified by silica gel chromatog-
raphy (eluent: hexanes/ethyl acetate ) 20/1 v/v) to give 2.07 g (80%)
of N-n-butyl-2-ethyl-4,5-ethylenedioxyaniline as opaque needles: 1H
NMR (CDCl3) δ 0.970 [3H, t, J ) 7.4 Hz, -NH(CH2)3CH3], 1.205
(3H, t, J ) 7.5 Hz, -CH2CH3), 1.450 [2H, m, -NH(CH2)2CH2CH3],
1.628 (2H, m, -NHCH2CH2CH2CH3), 2.410 (2H, q, J ) 7.5 Hz, -CH2-
CH3), 3.086 [2H, t, J ) 7.1 Hz, -NHCH2(CH2)2CH3], 3.3 (1H, br s,
exchangeable, -NH-), 5.827 (2H, s, -OCH2O-), 6.318 (1H, s, 6-H),
6.632 (1H, s, 3-H); MS (EI) m/z 211.1 (M+); HRMS (EI) C12H19NO2
(M+ - HCl) calcd 221.14158, obsd 221.14152.
(D) Synthesis of 4-n-Butylamino-5-ethyl-1,2-dihydroxybenzene
(1red). N-n-Butyl-2-ethyl-4,5-ethylenedioxyaniline (0.22 g, 1 mmol) in
dichloromethane (25 mL) was stirred at room temperature under an
atmosphere of deoxygenated Ar for 30 min. Boron trichloride (1 M in
dichloromethane, 5 mL) was added to the mixture, which was then
stirred at room temperature under Ar for 1.5 h. The reaction was
quenched by addition of methanol (10 mL) which had been deoxy-
genated with Ar. The solvent was removed in vacuo, and the residue
was redissolved in degassed methanol (5 mL). The methanol solution
was heated under reflux for 5 min, and again the solvent was removed
in vacuo. This operation was repeated three times to give a hydrochlo-
ride salt of 4-n-butylamino-5-ethyl-1,2-dihydroxybenzene (1red) as a
slightly discolored oil which formed into opaque needles in vacuo
(quantitative yield): 1H NMR (CD3CN) δ 0.831 [3H, t, J ) 7.4 Hz,
-NH(CH2)3CH3], 1.182 (3H, t, J ) 7.4 Hz, -CH2CH3), 1.327 [2H,
sextet, J ) 7.4 Hz, -NH(CH2)2CH2CH3], 1.749 (2H, tt, J ) 7.4, 7.9
Hz, -NHCH2CH2CH2CH3), 2.249 (br s, exchangeable, NHCH2CH2-
CH2CH3), 2.636 (2H, q, J ) 7.4 Hz, -CH2CH3), 3.158 (2H, t, J ) 7.9
Hz, -NHCH2CH2CH2CH3), 6.761 (1H, s, 3-H), 7.118 (1H, br s,
exchangeable, -OH), 7.209 (1H, s, 6-H), 8.272 (1H, br s, exchangeable,
biogenesis, we report preliminary findings involving the oxida-
tion of 3-n-butylamino-5-ethylcatechol (13red) and the reaction
of 4-ethylphenol and N-methyl-n-butylamine catalyzed by a
mushroom tyrosinase. These studies provide support for the
proposed 1,4-addition of lysine to an o-quinone intermediate
during LTQ production (see Chart 2).
Experimental Section
Materials. 6-Nitropiperonal, 5-nitrovanillin, and N-methyl-n-butyl-
amine were purchased from Aldrich. 4-Ethylcatechol was purchased
from Lancaster. 2-Hydroxy-5-ethyl-1,4-benzoquinone (4), 2-hydroxy-
5-tert-butyl-1,4-benzoquinone (5), 4-tert-butyl-1,2-benzoquinone (6),
and 4-methoxy-5-tert-butyl-1,2-benzoquinone (7) were prepared ac-
cording to the methods described previously.37 Mushroom tyrosinase
(EC 1.14.18.1) was purchased from Sigma.
Methods. UV-vis absorbance data were obtained on a HP8452A
diode array spectrophotometer equipped with a thermostated cell holder
at 25 ( 0.2 °C (path length of 1 cm). Mass spectra (MS) were obtained
on a VG 70-SE or a VG ZAB 2-EQ instrument.
(A) Synthesis of 4,5-(Methylenedioxy)-2-nitrostyrene. Anhydrous
methyltriphenylphosphonium bromide (37 g, 103 mmol) was suspended
in anhydrous THF (500 mL) in a 1 L round-bottomed flask under an
atmosphere of dry N2. The flask was cooled on ice prior to addition of
an equimolar amount of sodium bis(trimethylsilyl)amide (1 M in THF,
103 mL), and the mixture was stirred at room temperature under N2
for 1 h. Then 6-nitropiperonal (15.33 g, 77 mmol) was added in one
portion under N2, and the reaction mixture was stirred at room
temperature for a further 5 h. The solvent was removed by rotary
evaporation, and the resulting residue was diluted with water and
extracted with ethyl acetate. The combined organic layer was washed
with saturated aqueous NaCl and dried over MgSO4, and the solvent
was removed under reduced pressure. The resulting crude product was
purified by silica gel chromatography (eluent: hexanes:ethyl acetate
(51) Vitale, A. A.; Chiocconi, A. A. J. Chem. Res., Suppl. 1996, 336-337.
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6115

A R T I C L E S
Mure et al.
-OH), 10.229 (2H, br s, exchangeable); 13C NMR (CD3CN) δ 12.8,
br m, -N(CH3)CH2(CH2)2CH3], 6.869 (1H, s, ring proton at C3), 7.071
(1H, s, ring proton at C6), ∼8.5 (br s, -OH), 10.10 (br s, -OH); 13C
NMR (CD3CN) δ 12.8, 15.3, 19.2, 23.2, 29.8, 47.3, 59.5, 108.2, 117.0,
128.9, 129.2, 131.7, 144.8, 146.4; MS (EI) m/z 223 (M+, 34%), 180
(M+ - C3H7+, 100%); HRMS (EI) C13H21NO2 (M+) calcd 223.15723,
obsd 223.15703.
15.3, 19.2, 23.2, 26.8, 47.3, 59.5, 108.2, 117.0, 128.9, 129.2, 144.8,
146.4; MS (EI) m/z 209 (M+ - HCl, 56%), 166 (M+ - HCl - C3H7
,
+
100%); HRMS (EI) C12H19NO2 (M+ - HCl) calcd 209.14158, obsd
209.14132.
(E) Synthesis of 4-n-Butylamino-5-ethyl-1,2-benzoquinone (1ox).
1red (22.5 mg, 0.09 mmol) was dissolved in a CH3CN-H2O solution
(1:4, 10 mL), and a 10 µL aliquot of NaOH (1 M) was added to the
mixture to facilitate the autoxidation. The reaction was monitored by
UV-vis spectroscopy (formation of a species absorbing at 504 nm),
and when the oxidation was complete, the reaction mixture was frozen
in liquid N2. The solvent was removed by lyophilization to yield 1ox as
a dark purple solid (quantitative yield): 1H NMR (CD3CN) δ 0.947
[3H, t, J ) 7.4 Hz, -NH(CH2)3CH3], 1.167 (3H, t, J ) 7.3 Hz,
-CH2CH3), 1.395 [2H, m, -NH(CH2)2CH2CH3], 1.640 (2H, m,
-NHCH2CH2CH2CH3), 2.485 (2H, dq, J ) 1.5, 7.3 Hz, -CH2CH3),
3.270 (2H, m, -NHCH2CH2CH2CH3), 5.410 (1H, s, 3-H), 6.154 (1H,
t, J ) 1.5 Hz, 6-H), 6.226 (1H, br s, -NH); MS (EI) m/z 207 (M+,
73%), 150 (M+ - C4H9+, 100%); HRMS (EI) C12H17NO2 (M+) calcd
207.12593, obsd 207.12559.
4-(N-n-butyl-N-methylamino)-5-ethyl-1,2-benzoquinone (3ox) was
generated in situ by autoxidation of N-n-butyl-N-methylamino-5-ethyl-
1,2-dihydroxybenzene as described in the synthesis of the quinone (1ox).
3ox was stable in a dilute aqueous solution but decomposed upon either
solvent removal by lyophilization or when concentrated, thereby
preventing full characterization. Attempts to extract 3ox with organic
solvents were unsuccessful due to the instability of the quinone.
(H) Synthesis of 2-N-n-Butylamino-5-tert-butyl-1,4-benzoquinone
(8) and 2-(N-n-Butyl-N-methylamino)-5-tert-butyl-1,4-benzoquinone
(9). 8 and 9 were formed from the reactions of 2-tert-butyl-1,4-
benzoquinone and n-butylamine and N-methyl-n-butylamine, respec-
tively, and then purified by silica gel chromatography. The reaction
conditions were same as described for the reaction of 2-tert-butyl-1,4-
benzoquinone and benzylamine.37 8: 1H NMR (CDCl3) δ 0.946 [3H,
t, J ) 7.3 Hz, -NH(CH2)3CH3], 1.296 [9H, s, -(CH3)3], 1.396 [2H,
m, -N-(CH2)2CH2CH3], 1.611 (2H, m, -NHCH2CH2CH2CH3), 3.062
[2H, 2t, -NHCH2(CH2)CH3], 5.382 (1H, s, ring proton at C3 + 1H, br
s, NH), 6.425 (1H, s, ring proton at C6); 13C NMR (CDCl3) δ 13.6,
20.0, 29.6, 30.1, 35.6, 41.9, 100.1, 127.2, 134.8, 145.3, 184.7, 185.6.
9: 1H NMR (CDCl3) δ 0.936 [3H, t, J ) 7.3 Hz, -NH(CH2)3CH3],
1.277 [9H, s, -(CH3)3], 1.396 [2H, m, -N-(CH2)2CH2CH3], 1.578 [2H,
m, -N(CH3)CH2CH2CH2CH3], 2.982 [3H, s, -N(CH3)(CH2)3CH3],
3.477 [2H, 2t, -N(CH3)CH2(CH2)CH3], 5.478 (1H, s, ring proton at
C3), 6.303 (1H, s, ring proton at C6); 13C NMR (CDCl3) δ 13.8, 20.0,
29.5, 29.9, 35.0, 40.1, 53.7, 106.1, 129.6, 148.9, 156.8, 185.5, 186.5.
(I) Synthesis of a Phenylhydrazine Adduct of 1ox (2). To a heated
stirred solution of phenylhydrazine hydrochloride (70.1 mg, 0.485
mmol) in glacial acetic acid (5 mL) at 100 °C was added an acetic
acid solution (2 mL) of 1ox (18.6 mg, 0.09 mmol). The reaction mixture
was stirred at 100 °C for 30 min and then cooled to room temperature.
Water (50 mL) was added, and the mixture was neutralized by NaHCO3
and extracted with diethyl ether. The combined organic layers were
washed with water and dried with MgSO4, and the solvent was removed
in vacuo. The crude product was purified by silica gel chromatography
to yield 5.7 mg (21%) of 2 as an orange-red crystalline solid: 1H NMR
(CD3CN) δ 0.950 [3H, t, J ) 7.4 Hz, -NH(CH2)3CH3], 1.220 (3H, t,
J ) 7.5 Hz, -CH2CH3), 1.412 [2H, m, -NH(CH2)2CH2CH3], 1.630
(2H, m, -NHCH2CH2CH2CH3), 2.460 (2H, dq, J ) 1.0, 7.5 Hz, -CH2-
CH3), 3.240 (2H, m, J ) 8.1 Hz, -NHCH2CH2CH2CH3), 5.231 (1H,
br s, -NH), 5.940 (1H, s, 6-H), 7.240 (1H, s, 3-H), 7.267-7.656 (5H,
m, -C6H5), 14.775 (1H, br s, -OH); 13C NMR (CD3CN) δ 13.4, 14.1,
20.9, 23.3, 31.5, 43.6, 96.9, 120.3, 123.9, 128.4, 130.3, 132.0, 132.4,
149.8, 153.4, 162.2; MS (EI) m/z 297 (M+, 100%), 192 (M+ - C6H5-
N2+, 50%); HRMS (EI) C18H23N3O (M+) calcd 297.18411, obsd
297.18467.
(F) Synthesis of N-n-Butyl-N-methyl-2-ethyl-4,5-methylenedioxy-
aniline. Aqueous formaldehyde (37%, 2 mL) was added to an ice-
cold mixture of N-n-butyl-2-ethyl-4,5-methylenedioxyaniline (108.1 mg,
0.49 mmol) and formic acid (2 mL). The reaction mixture was stirred
overnight at 90 °C and then cooled to room temperature. Concentrated
HCl (2 mL) was added, and an extraction was performed with diethyl
ether. The aqueous layer was treated with 50% NaOH to adjust the pH
to 10 and then reextracted with diethyl ether. The combined ether layers
were washed with water and dried over Na2SO4. The solvent was
removed under reduced pressure to give N-n-butyl-N-methyl-2-ethyl-
4,5-methylenedioxyaniline as a yellow oil (55% yield): 1H NMR
(CDCl3) δ 0.888 [3H, t, J ) 7.2 Hz, -N(CH3)(CH2)3CH3], 1.167 (3H,
t, J ) 7.6 Hz, -CH2CH3), 1.312 [2H, m, -N(CH3)(CH2)2CH2CH3],
1.395 [2H, m, -N(CH3)CH2CH2CH2CH3], 2.549 [3H, s, -N(CH3)(CH2)3-
CH3], 2.657 (2H, q, J ) 7.6 Hz, -CH2CH3), 2.743 [2H, t, J ) 7.5 Hz,
-N(CH3)CH2(CH2)2CH3], 5.890 (2H, s, -OCH2O-), 6.701 (1H, s,
ring proton at C6), 6.710 (1H, s, ring proton at C3).
(G) Synthesis of N-n-Butyl-N-methylamino-5-ethyl-1,2-dihy-
droxybenzene (3red) and 4-(N-n-Butyl-N-methylamino)-5-ethyl-1,2-
benzoquinone (3ox). A solution of N-n-butyl-N-methyl-2-ethyl-4,5-
methylenedioxyaniline (63.6 mg, 0.27 mmol) in dichloromethane was
stirred at room temperature for 30 min under Ar and treated with boron
trichloride (1 M in dichloromethane, 1.4 mL). The mixture was then
stirred under Ar at room temperature for a further 3 h. The reaction
was quenched by addition of methanol (5 mL), and the resulting solution
was heated under reflux for 5 min after which time the solvent was
removed in vacuo. The hydrochloride salt of N-n-butyl-N-methylamino-
5-ethyl-1,2-dihydroxybenzene was isolated as a yellow-brown solid in
quantitative yield. The compound is pure as indicated by 13C NMR
1
and mass spectrometry. The H NMR in CD3CN indicated that 3red is
present as a mixture of two forms. The chemical shifts and integrations
of the methylene protons of the n-butyl side chain [CH3CH2CH2CH2N-
(CH3)-], the methyl protons [CH3CH2CH2CH2N(CH3)-], and the
methylene protons of the ethyl side chain suggest that there are two
inequivalent forms of each. The resonances for the remaining protons
are equivalent for the two forms. This could result from 3red existing
as a mixture of the protonated and neutral forms in CD3CN, or the R
and S forms of 3red (where the protonated amine is the chiral center)
may form hydrogen-bonded dimers in CD3CN, thereby creating the
(J) NMR Assignment of 2. 1H and 13C NMR spectra were obtained
on a Bruker AM-500 spectrophotometer fitted with an inverse 1H-13C
probe operating at a proton frequency of 500.14 MHz for protons. The
spectra were recorded for a CD3CN solution containing 11 mg (0.037
mmol) of 2 in a 5 mm NMR tube. 1H chemical shifts were relative to
residual CH3CN set to 1.940 ppm. 13C chemical shifts were relative to
the 13C signal of CD3CN set to 118.2 ppm. The NOESY spectrum was
acquired at a mixing time of 2 s at 27 °C. The details of the NMR
assignment of 2 are available in the Supporting Information.
(K) Synthesis of a 4-Nitrophenylhydrazine Adduct of 1ox (10). A
stirred solution of 4-nitrophenylhydrazine (316.8 mg, 2.1 mmol) in
acetic acid (20 mL) was heated to 100 °C, and then 1ox (82.8 mg, 0.4
mmol) was added as an acetic acid solution. The reaction mixture was
stirred at 100 °C for 1 h and cooled to room temperature. Water (50
mL) was added, and the mixture was neutralized by addition of NaHCO3
1
diastereomers. In contrast to H NMR, the 13C NMR in this solvent
can be assigned to a single species, 3red
:
1H NMR (CD3CN) δ 0.763
[3H, t, J ) 7.2 Hz, -N(CH3)(CH2)3CH3], 1.122 (3H, t, J ) 7.6 Hz,
-CH2CH3), 1.20-1.23 [3H, m, -N(CH3)CH2CH2CH2CH3], 1.563 [1H,
m, -N(CH3)CH2CH2CH2CH3], 2.710 (1H, 2q, J ) 7.6 Hz, -CH2-
CH3), 2.821 (1H, 2q, J ) 7.6 Hz, -CH2CH3), 3.090 [1.5H, s,
-N(CH3)(CH2)3CH3], 3.102 [1.5H, s, -N(CH3)(CH2)3CH3], 3.423 [2H,
9
6116 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

LTQ Cofactor of Lysyl Oxidase
A R T I C L E S
and then extracted with diethyl ether. The combined organic layers
were washed with water and dried with MgSO4, and the solvent was
removed in vacuo. The crude product was purified by silica gel
chromatography to yield 38 mg (30%) of 10 as an orange-red crystalline
solid: 1H NMR (CDCl3) δ 0.990 [3H, t, J ) 7.3 Hz, -NH(CH2)3CH3],
1.298 (3H, t, J ) 7.3 Hz, -CH2CH3), 1.456 [2H, m, -NH(CH2)2CH2-
CH3], 1.694 (2H, m, -NHCH2CH2CH2CH3), 2.420 (2H, q, J ) 7.3
Hz, -CH2CH3), 3.263 (2H, m, -NHCH2CH2CH2CH3), 4.849 (1H, br
s, -NH), 5.788 (1H, s, ring proton at C6), 6.951 (1H, s, ring proton at
µL of 50% NaOH and 3red subjected to autoxidation in air, resulting in
a stock solution of 3ox (7.85 × 10-2 M, pH 6.46). A 100 µL aliquot of
the stock solution was diluted with 900 µL of a buffer at the desired
pH value. UV-vis spectra of the samples were taken immediately
(7 s) following dilution.
pKas of 8, 2, and 11: A 40 µL aliquot of the appropriate CH3CN
stock solution ([8] ) 2.4 mM, [2] ) 0.46 mM, [11] ) 0.37 mM) was
diluted into 960 µL of an aqueous buffer at the desired pH. UV-vis
spectra of the samples were taken immediately after the dilution.
(O) Electrochemistry. Cyclic voltammetry was carried out on a
BAS 100A electrochemical analyzer at the desired pH with a three-
electrode system consisting of a gold working electrode (electrode area
) 0.02 cm2), a coiled platinum counter electrode, and a saturated
calomel electrode as the reference. The voltammetric measurements
were performed with a potentiostat (Princeton Applied Research 173)
in conjunction with a triangular wave generator (Princeton Applied
Research 175) and recorded on a X-Y recorder (Kipp and Zonen). Prior
to each experimental run, the gold working electrode was cleaned with
a mixture of concentrated sulfuric acid and chromic acid, washed with
purified, deionized distilled water, dried by N2 flow, and transferred
rapidly to the electrochemical cell containing the test solution. The
experiments were carried out in Ar-purged solutions under an Ar
atmosphere at room temperature. For the cyclic voltammetry of 3ox, 8,
9, and 13red, the buffer solution used in the assay was 0.1 M potassium
phosphate (pH 6.74 ( 0.01) with I ) 0.3 (KCl). For the pH titration
of 1ox, the buffers used were as described for the UV-vis assay. Sample
solutions of quinones were prepared in 10 mL of the same potassium
phosphate buffer at a final concentration of 10-3-10-5 M. All sample
solutions were freshly prepared and subjected to the measurement
immediately.
C3), 7.573 (2H, d, J ) 9.2 Hz), 8.245 (2H, d, J ) 9.2 Hz), 16.132 (1H,
+
br s, -OH); MS (EI) m/z 342 (M+, 100%), 192 (M+ - C6H4N3O2
,
42%); HRMS (EI) C18H22N4O3 (M+) calcd 342.16973, obsd 342.16920.
(L) Synthesis of a Phenylhydrazine Adduct of 4 (11). A stirred
solution of phenylhydrazine hydrochloride (10.5 g, 72.5 mmol) in
glacial acetic (20 mL) was heated to 100 °C, and then 4 (2.3 g, 14.5
mmol) was added as an acetic acid solution. The reaction mixture was
stirred at 100 °C for 30 min and then cooled to room temperature.
Hexanes (200 mL) were added, and the precipitated unreacted phen-
ylhydrazine hydrochloride salt was removed by vacuum filtration. The
precipitate was washed five times with 50 mL portions of hexanes.
The combined hexane layer was washed three times with water and
then dried over MgSO4. Removal of the solvent in vacuo yielded a
dark orange-brown solid, which was dissolved in a minimum volume
of chloroform and applied to silica gel column chromatography. Elution
with chloroform and ethyl acetate (4/1 v/v) and subsequent removal of
the solvent gave 11 as dark red needles (350 mg, 10% yield): 1H NMR
(CDCl3) δ 1.278 (3H, t, J ) 7.6 Hz, -CH2CH3), 2.628 (2H, q, J ) 7.6
Hz, -CH2CH3), 6.372 (1H, s, ring proton at C3), 7.397 (1H, m), 7.484
(2H, m), 7.599 (1H, s, ring proton at C6), 7.776 (2H, d), 13.868 (1H,
1
s, -OH); H NMR [(CD3)2CO] δ 1.204 (3H, t, J ) 7.6 Hz), 2.664
(2H, q, J ) 7.6 Hz), 6.783 (1H, s, ring proton at C3), 7.368 (1H, s,
ring proton at C6), 7.442-7.543 (3H, m), 8.317 (1H, s, -OH), 11.848
(1H, s, -OH); 13C NMR [(CD3)2CO] δ 13.2, 23.4, 114.6, 117.7, 122.0,
129.4, 130.8, 135.8, 139.0, 147.0, 148.3, 151.1.
(M) Deuterium Exchange of the Ring Proton of 1ox, 4, and 8.
NMR samples of quinones (4 mg) were prepared in D2O (0.5 mL) and
CD3OD (0.5 mL). Spectra were taken at different time intervals.
Changes in integration of the proton signal were used to monitor the
deuterium exchange of the protons.
(N) Acid-Base Titration. The acid-base dissociation constants
were determined in HCl for pH 0-1.0, H3PO4 + KH2PO4 for pH 1.5-
3.0, CH3COOH + CH3COOK for pH 3.0-6.0, KH2PO4 + K2HPO4
for pH 5.7-7.8, Tris for pH 8.0-8.8, KHCO3 + K2CO3 for pH 9.0-
11.0, K2HPO4 + KOH for pH 9.1-12.0, and KOH for pH 11.9-13.2.
All buffer solutions had a concentration of 0.1 M, and their ionic
strength was adjusted to 0.3 with KCl.
pKas of 1ox: The stock solution of 1ox was prepared by oxidation of
the catechol (1red) in situ. 1red‚HCl (2.71 mg, 11.0 mmol) was dissolved
in 1 mL of a 0.1 M potassium phosphate buffer (pH 6.7, I ) 0.3 with
KCl) and subjected to autoxidation in air, resulting in a quinone stock
solution of 1ox (1.10 × 10-2 M). A 10 µL aliquot of the stock solution
was then diluted with 990 µL of the appropriate buffer with the desired
pH value. UV-vis spectra of the samples were taken immediately (7
s) after the dilution.
pKas of 1red: A stock solution of 1red prepared from 1red‚HCl (3.46
mg, 14.1 mmol) dissolved in 1 mL of 0.1 N HCl. A buffer solution (3
mL) in a quartz cell fused to an 8 cm graded seal tube, which was
tightly closed with a septum rubber cap, was degassed by purging with
O2-free Ar gas for 15 min. O2 was removed from the Ar gas by passage
through an alkaline pyrogallol solution. An excess amount of NaBH4
(23.5 molar equiv) was added to all buffers at pH >8. Twenty
microliters of the stock solution of 1red was then added by a gastight
syringe into the degassed buffer solution. The spectra were taken
immediately after the dilution.
(P) Reaction of 4-Ethylphenol or 4-Ethylcatechol with N-Methyl-
n-butylamine in the Presence of Tyrosinase. A 10 µL aliquot of a
stock solution of 4-ethylphenol or 4-ethylcatechol (20 mM in 2 mM
HCl) and 0.1 mL of a stock solution of N-methyl-n-butylamine (0.2 M
in 0.1 M potassium phosphate buffer, pH 7.2) were added to a quartz
cuvette containing 0.876 mL of a 0.1 M potassium phosphate buffer
(pH 7.2) which had been preequilibrated at 25 ( 0.2 °C. A 14 µL
aliquot of a mushroom tyrosinase stock solution (14 µg/mL) was then
added to the cuvette. The spectra of the reaction mixture were
periodically recorded with a HP photodiode array spectrophotometer.
(Q) Reaction of 4-Ethylcatechol and N-Methyl-n-butylamine in
the Presence of NaIO4. A 10 µL aliquot of a stock solution of
4-ethylcatechol (20 mM in 2 mM HCl) and a 10 µL aliquot of the
stock solution of N-methyl-n-butylamine (0.2 M in 0.1 M potassium
phosphate buffer, pH 7.2) were added to a quartz cuvette containing
0.97 mL of 0.1 M potassium buffer (pH 7.2) that had been preequili-
brated at 25 ( 0.2 °C. A 10 µL aliquot of a NaIO4 stock solution (0.2
M in 0.1 M potassium phosphate buffer, pH 7.2) was then added to
the cuvette. The spectra of the reaction mixture were periodically
recorded with a HP photodiode array spectrophotometer.
Results and Discussion
Synthesis and Characterization of LTQ Model Com-
pounds. (A) Synthesis of 1ox. 4-n-Butylamino-5-ethyl-1,2-
benzoquinone (1ox) is a simplified analogue of the LTQ cofactor
having the same functional groups but lacking a peptidyl
backbone. Initially, we attempted to isolate 1ox from the reaction
of n-butylamine with 4-ethyl-1,2-benzoquinone. This strategy
proved unsuccessful as the chemistry is not clean, and a number
of side products formed in the reaction mixture. Ultimately, 1ox
was prepared from 6-nitropiperonal via a five-step synthesis,
as shown in Scheme 1. The stability of 1ox varies depending on
the conditions under which it is prepared. Decomposition of
1ox was observed when concentrated in organic solvents
(precipitates as dark brown mass). We found that 1ox can be
pKas of 3ox: 3red‚HCl (5.1 mg, 19.6 mmol) was dissolved in H2O
containing 1 mL of CH3CN. The solution was neutralized by adding 1
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6117

A R T I C L E S
Mure et al.
Scheme 1. Synthesis of 1ox
isolated by lyophilization of an aqueous solution at neutral pH
following air oxidation of the catechol (1red).
The 1H NMR of 1ox showed two proton resonances at 5.140
and 6.154 ppm, corresponding to the quinonoid ring protons.
The resonance at 6.154 ppm showed a long-range coupling to
the methylene proton of the ethyl side chain at C5 and thus was
assigned to H6 (J ) 1.5 Hz). The same coupling was seen for
H6 of 4, a TPQ model compound with an ethyl side chain at
C5. The resonance at 5.140 ppm was consequently assigned as
H3. The two ring protons of an o-quinone with a methoxy side
chain (7) had been previously assigned as H3 at 5.760 ppm and
H6 at 6.200 ppm, respectively.37 The H3 of 1ox is 0.62 ppm
upfield shifted compared to that of 7, reflecting the electron-
donating nature of the nitrogen lone pair of the n-butylamino
side chain. It should be noted that the ring protons of an anionic
TPQ model compound (4) have chemical shifts very close to
those of 1ox. However, in contrast to 4, no deuterium exchange
of H3 was observed for 1ox (see below).
(B) UV-Vis Spectra of Model Compounds. Perhaps the
most accessible feature of quinoproteins is their prominent
visible absorption spectra. However, these spectra can often be
misleading when attempting to assign the nature of the quinon-
oid cofactor. The UV-vis spectra of LTQ model compounds
1ox and 3ox are shown in Figure 1 together with those of TPQ
model compounds (4, 5) and p-quinones bearing an n-butyl-
amino side chain (8, 9). The spectrum of 1ox displays an
absorption band with a λmax at 298 nm and a broad absorption
band with a λmax around 504 nm, which is an excellent match
to that of LTQ in LOX.27,48 The ratio of the absorbances at 298
and 504 nm (A298/A504) of 1ox is 3.9. Both λmax of 1ox are ca. 20
nm red shifted compared to those of 4. At pH 7.0, the visible
spectrum of 1ox is close to that of 4, so this alone cannot be
relied upon for cofactor identification (Figure 1a).
Figure 1. (a) UV-vis spectra of 1ox (s) and 4 (- - -). (b) UV-vis spectra
of 1ox (s) and 3ox (- - -). (c) UV-vis spectra of 8 (s) and 9 (- - -). All
spectra are for 0.1 mM of the appropriate model compound in 0.1 M
potassium phosphate (pH ) 7.0, I ) 0.3 with KCl).
In contrast to TPQ, no spectral changes were observed for
ox over the physiological pH range (4-8). Methylation of the
The delocalization was confirmed for 4 and TPQ in CAO by
resonance Raman spectroscopy where the C3 ring proton was
shown to exchange with solvent water.52 In the model chemistry,
NMR was also used to confirm that H3 of 4 could exchange
with solvent.36,37 As the UV-vis spectra of 1ox and 3ox are close
to that of 4, one might presume that LTQ has similar electronic
properties to TPQ. We looked for deuterium exchange of the
quinonoid ring protons on 1ox, 4, and 8 in D2O and CD3OD by
1H NMR spectroscopy. For 1ox, only the NH proton of the
1
nitrogen of 1ox (3ox) resulted in a 20 nm red shift of both
absorption maxima, and the ratio of A316/A528 was reduced to
2.6 (Figure 1b). A similar red shift was observed upon
N-methylation (9) of a p-quinone with an n-butylamino side
chain (8) (Figure 1c). These results suggest that 1ox exists as a
neutral o-quinone in which the NH proton is retained on the
n-butylamino side chain. At pH 7, 4 exists as a monoanion
where the C2 hydroxyl group is deprotonated.36,38 The red color
of 4 results from the delocalization of this negative charge into
the quinone ring (eq 1).
(52) Moenneloccoz, P.; Nakamura, N.; Steinebach, V.; Duine, J. A.; Mure, M.;
Klinman, J. P.; Sandersloehr, J. Biochemistry 1995, 34, 7020-7026.
9
6118 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

LTQ Cofactor of Lysyl Oxidase
A R T I C L E S
n-butylamino side chain showed exchange with deuterium, and
no incorporation of deuterium on the quinonoid ring was
detected even after 7 days of incubation. The same exchange
pattern was observed for 8. In contrast, the proton at C-3 (H3)
of 4 disappeared rapidly (within the minimal acquisition time
<2-3 min) in D2O. In CD3OD, the exchange was much slower,
taking hours to complete. This is because 4 exists predominantly
as the neutral p-quinone form in methanol.
These observations are consistent with resonance Raman
spectroscopic studies of LTQ in LOX, which showed that the
quinonoid ring proton was not exchangeable with D2O.48 The
absence of exchange at pH 7 implies that LTQ and 1ox exist as
only the o-quinone under these conditions, i.e., that there are
no resonance forms which provide a pathway for the exchange
of the ring protons.
Figure 2. (a) Absorption spectra of 1red at pH 1.65 (- - -), 1red at pH 7.25
(s), 1red at pH 10.36 (O), and 1red at pH 12.97 (b), where [1red] ) 9.3 ×
105 M. (b) Spectroscopic pKa determination for 1red. The line was drawn
1
redH+
by a least-squares fit using the equation A310 ) {(ꢀ310
[H+]3
+
-
2-
ꢀ310 Kred1[H+]2
+
ꢀ310 Kred Kred2[H+]
+
ꢀ310
K
1Kred Kred3)/
1red
1red
1
1red
2
red
([H+]3 + Kred1[H+]2 + Kred Kred2[H+] + Kred Kred Kred3)}[1red]T.
1
1
2
electron-donating n-butylamino side chain at C4. Once depro-
tonation has occurred, it is possible that the proton on the
monoanion form of 1red is equally shared between the oxygens
at C1 and C2.
(C) Acid-Base Properties of 1red. The pKa values for 1red
3
(pKred1, pKred2, pKred in eq 2) were determined spectrophoto-
metrically under anaerobic conditions. In acidic solution, O2 is
removed from the buffer by purging with Ar gas prior to the
addition of the hydrochloride salt of 1red. Above pH 8, it was
necessary to add an excess of NaBH4, in addition to removing
O2, to reverse the oxidation of 1red as this reaction becomes
very rapid at alkaline pH. At pH <4, 1red shows a λmax at 282
nm which red shifts with increasing pH (300 nm at pH 7, 310
nm at pH 13) as shown in Figure 2a. Two pKas, calculated to
be 5.97 ( 0.05 and 9.39 ( 0.07, respectively, were obtained
by least-squares fitting of the absorbance change at 310 nm
(Figure 2b). The former pKa is assigned to the deprotonation of
the amino group of the n-butylamino side chain (pKred1), and
the latter is assigned to one of the hydroxyl groups (pKred2),
presumably at C2. The pKa of the remaining hydroxyl group
(pKred3) is estimated to be greater than 13 but could not be
accurately determined by the UV-vis spectroscopic pH titration
(see Electrochemistry). The value of pKred1 is very close to that
of the amino group of the aminoresorcinol form of TPQ (5.88).36
The protonation of this amino group yielded spectroscopic
changes similar to those observed for 1red, i.e., a blue shift in
λmax resulting in no absorbance above 300 nm. This analogy
was used, in part, for the assignment of pKred1. At neutral pH,
1red has a λmax at 300 nm which is 10 nm blue shifted when
compared to the aminoresorcinol form of TPQ. Further, 1red
lacks the secondary features below 280 nm, which are observed
(D) Acid-Base Properties of 1ox and 3ox. A decrease in
pH from 4 to 1 resulted in a bleaching of the characteristic broad
absorption band of 1ox at 504 nm (Figure 3a). A plot of A504 vs
pH was fit to the titration curves for the dissociation of a single
proton, allowing the calculation of a pKa of 1.65 ( 0.04 (Figure
3b). This corresponds to that of the amino side chain of 1ox
1
(pKox in eq 3). 3ox showed a similar spectral change upon
acidification and gave an acidic pKa value very close to that of
1ox (pKa ) 1.96 ( 0.03) (data not shown).
2
for the aminoresorcinol form of TPQ at pH 7-9. pKred was
assigned to the C2 hydroxyl group because the C1 hydroxyl
It was difficult to determine whether 1ox has any pKa at
group has a decreased pKa as a result of it being para to the
alkaline pHs (above 9) due to the instability of this species under
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6119

A R T I C L E S
Mure et al.
Figure 3. (a) Absorption spectra of (b) 1ox at pH 0.34, (s) 1ox at pH
7.32, and (O) 1ox at pH 11.75. [1ox] ) 1.1 × 10-4 M. (b) Spectroscopic
Figure 4. (a) Cyclic voltammogram of 1ox at pH 6.7 at a scan speed of
100 mV s-1. [1ox] ) 9.8 × 10-5 M. The direction of the scan is from positive
to negative. (b) Em-pH diagram for 1ox. The line was drawn by a least-
squares fit using the equation Em ) E0 + (RT/2F) log{([H+]4 + Kred1[H+]3
pKa determination for 1ox. The line was drawn by a least-squares fit using
1
oxH+
1ox
1ox
1
-
the equation A504 ) {(ꢀ504
[H+]2 + ꢀ504 Kox1[H+] + ꢀ504 Kox Kox2)/
1
([H+]2 + Kox1[H+] + Kox Kox2)}[1ox]T.
1
1
2
1
+ Kred Kred2[H+]2 + Kred Kred Kred3[H+])/([H+]2 + Kox1[H+] + Kox Kox2)}.
basic conditions. The UV-vis spectra of 1ox were acquired
within 7 s from mixing of the stock solution with the appropriate
buffer. Upon increasing pH, a blue shift of the λmax from 504
to 460 nm was observed (Figure 3a). A pKa of 10.33 ( 0.07
was calculated from the absorbance changes at 504 nm, possibly
corresponding to the proton of the n-butylamino side chain
(Figure 3b). However, there was an interfering but distinct
absorbance change from bleaching of the characteristic visible
absorbances observed concomitant with an increase in absor-
bance at 282 nm and isosbestic points at 290 and 250 nm (data
not shown). This latter change was irreversible and also
accelerated with increasing pH (t1/2 ) 500 s at pH 10.8). As
this occurs outside of the physiological pH range, we did not
pursue the characterization of the reaction product in this study.
Table 1. Midpoint Potentials of Model Compounds at pH 6.7
model compd
Em (V) vs SCE
model compd
Em (V) vs SCE
1ox
3ox
4
-0.182a
-0.037
-0.181
+0.109b
7
8
9
-0.032b
-0.331
-0.290
-0.108
6
13ox
a Reference 27. b Reference 37.
V). These results suggest that methylation of the n-butyl side
chain of 1ox has a significant effect on the electronic nature of
1
the quinonoid ring, as discussed in the context of its H NMR
properties.
Electrochemical redox potentials of 1ox were measured as a
function of pH. Between pH 1.10 and pH 13.95, cyclic
voltammograms showing a single process for the reduction and
oxidation of 1ox were obtained. Between pH 1 and pH 3, ∆Ep
was 30 mV, which is in agreement with the theoretical value
for a reversible concerted two-electron reaction. With increasing
pH, the ∆Ep value increased (maximum of 45 mV) to give a
quasi-reversible cyclic voltammogram. No decomposition of 1ox
was observed within the time scale of the CV scans.
3ox has no pKa above 10 but was also unstable, undergoing time-
dependent spectral changes similar to those observed for 1ox.
(E) Electrochemistry. The cyclic voltammogram of 1ox at
pH 6.7 (Figure 4a) reveals a single process for the oxidation
and reduction to give a midpoint potential of 1ox of -0.182 V
(vs SCE). The peak separation of the anodic and cathodic peaks
(∆Ep) is 35 mV, which is very close to the theoretical value for
a reversible concerted two-electron reaction (30 mV).
In Table 1, the midpoint potentials of model compounds at
pH 6.7 are summarized. 1ox and 4 have very similar redox
potentials as might be expected since the cofactors that they
model catalyze similar reactions (i.e., primary amine oxidation).
Although 8 shows a UV-vis absorbance spectrum similar to
those of 1ox and 4, its Em is ca. -0.1 V negative when compared
to 1ox and 4. N-Methylation of the n-butylamino side chain of
Figure 4b is a plot showing the pH dependence of the
midpoint potentials of the overall 2e- 1ox/1red couple. The solid
1
line was generated using the values of pKas of 1ox (pKox and
pKox2) and 1red (pKred1 and pKred2) calculated from the UV-vis
spectroscopic titrations (see Figures 3b and 2b). The third pKa
of 1red (pKred3) was then calculated to be 12.96 ( 0.19 from the
least-squares curve fitting. When the fitting was performed with
2
3
1
ox resulted in a large positive shift of Em (+0.145 V), yielding
pKox and pKred as variables (fit not shown), they were
calculated as 10.52 ( 0.09 and 13.42 ( 0.35, respectively. The
value of pKox2 matches very well with that obtained from UV-
a value for 3ox very close to that of 7. A positive shift of Em
was also observed for 9 vs 8, but the ∆Em was small (0.041
9
6120 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

LTQ Cofactor of Lysyl Oxidase
A R T I C L E S
Scheme 2. Intermediates Deduced from Em-pH Studiesa
vis spectroscopic pH titration (Figure 3b). Further, a good fit
of the data could not be obtained without including pKox
2
,
suggesting that this pKa is not an artifact of the irreversible time-
dependent bleaching seen in the UV-vis spectra at high pH.
The fit of the Em vs pH plot (Figure 4b) is consistent with
five linear regions with slopes of -60 mV/pH (pH 1-1.6), -90
mV/pH (pH 2-5.8), -60 mV/pH (pH 6-9.3), -30 (pH 9.6-
10.8), and -60 mV/pH (pH 11.4-13.3), respectively. Scheme
2 shows a summary of the species involved in the Em-pH
diagram of 1ox. Below pH 2, the protonated 1red is oxidized to
the protonated quinone in a 2H+, 2e- process to give a slope
of -60 mV/pH. Between pH 2 and pH 5.8, the slope of -90
mV/pH corresponds to a 3H+, 2e- redox process due to the
acid dissociation of the n-butylamino side chain of 1ox (pKox1).
Between pH 6 and pH 9.3, the n-butylamino side chain of
1red is deprotonated (pKred ) 5.97) and undergoes a 2H+,
1
2e- oxidation to 1ox, resulting in a slope of -60 mV/pH.
Between pH 9.6 and pH 10.8, the second acid dissociation of
1
red (pKred2 ) 9.39) gives a slope of 30 mV/pH, corresponding
to a 1H+, 2e- process. Between pH 11.4 and pH 13.3, the second
2
acid dissociation of 1ox (pKox ) 10.33) gives a slope of 60
mV/pH, corresponding to a 2H+, 2e- process. Above pH 13.3,
the quinol is fully deprotonated (pKred3 ) 13.4) to give a slope
of 30 mV/pH, corresponding to a 1H+, 2e- process.
Between pH 8.0 and pH 8.5, LOX shows maximum activity
in the oxidation of n-hexylamine.53 Under these conditions, from
this study it can be seen that LTQ will most likely undergo the
2H+, 2e- redox process of 1ox/1red (Scheme 2). The Em value
for the substrate-reduced form of TPQ, aminoresorcinol/
iminoquinone couple, was determined to be -0.133 V vs SCE
at pH 6.7, which is only slightly positive compared to that for
TPQox/TPQred (see Table 1).36 The similarity in Em values
between 1ox and 4 suggests that their substrate-reduced forms
may be similar in their reactivities toward O2.
a The species in brackets are attributed to those on LOX between pH
8.0 and pH 8.5, conditions of maximal LOX activity.
Synthesis and Characterization of the Phenylhydrazine
(2) and 4-Nitrophenylhydrazine Adducts (10) of 1ox. The
identification of LTQ in LOX was achieved by following the
same experimental procedures used for identifying TPQ in
CAOs.26,27 LOX was inhibited with phenylhydrazine to co-
valently modify LTQ to form a stable phenylhydrazine adduct.
The modified LTQ-containing peptide was isolated and then
characterized by Edman degradation, mass spectrometry, and
resonance Raman spectroscopy. To identify the structure of
derivatized LTQ, a model compound (2) mimicking the
proposed cofactor structure was synthesized. Resonance Raman
spectra of 2 were identical to that of the modified cofactor-
containing peptide, confirming the presence of LTQ in LOX.27
The position of the nucleophilic addition to LTQ has been
proposed to be at C1. We confirmed that 2 results from the
addition of the hydrazine to the C1 oxygen using a combination
of HMQC, HMBC, and NOESY experiments. The details of
these analyses are provided in the Supporting Information. 2
can undergo hydrazone-azo tautomerization (eq 4), and the
NMR results suggest that the azo form is predominant in the
equilibrium for 2. This is similar behavior to that observed for
11 and 12, the phenyl and 4-nitrophenylhydrazine adducts of
the TPQ model compound, 4.36
The UV-vis spectra of 2 and 11 at different pHs are shown
in Figures 5a and 6a, respectively. At neutral pH, 2 shows λmax
at 446 nm that is red shifted (16 nm) vs 11. At high alkaline
pH (>13), both exhibit very similar absorbance properties with
λ
max at 486 nm. The pKa values for these spectral changes were
calculated to be 11.90 ( 0.04 for 2 (from the absorbance
changes at 446 and 486 nm shown in Figure 5c) and 12.83 (
0.17 for 11 (from the absorbance changes at 430 and 486 nm
shown in Figure 6d). These are assigned as the acid dissociation
of the hydroxyl group at C-2 of 2 and 11 respectively, where
both participate in a hydrogen-bonding interaction with an
adjacent azo group (eqs 5 and 6).
As discussed below, at neutral pH, 11 exists as a monoanion
in contrast to 2, which is a neutral species; deprotonation from
the monoanion of 11 makes the dissociation of second hydroxyl
group harder. In this way, the 2-hydroxyl group of 2 has a pKa
value about 1 pH unit lower than that of the 4-hydroxyl group
of 11.
2 does not show a significant spectral change between pH
3.11 and pH 7, and the λmax remains at 446 nm. In contrast, a
(53) Gacheru, S. N.; Trackman, P. C.; Kagan, H. M. J. Biol. Chem. 1988, 263,
16704-16708.
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6121

A R T I C L E S
Mure et al.
large spectral change was observed for 11 in this pH range. At
pH 4.38, 11 shows a broad absorption band with λmax at 388
nm. This shifts to 430 nm, with an isosbestic point, upon
increasing the pH, giving a pKa value of 5.93 ( 0.03 (Figure
6b). This indicates that 2 exists as a neutral species but 11 exists
as a monoanion at physiological pH. 2 shows a large spectral
change associated with the pKa (2.12; Figure 5b) of the
n-butylamino side chain, namely, a large red shift of the λmax
at 446 nm to 478 nm. Although 11 does exhibit a red shift of
the λmax at 388 nm to 430 nm by protonation of the hydrazone-
azo group, it takes place below pH 2 (pKa ) -0.28; Figure
6a,d).
The pH-dependent spectral shifts of 11 at alkaline pH have
been used to assign the presence of TPQ as a cofactor in
CAOs.54 In these experiments, the spectra of the hydrazone-
inhibited (either phenyl or 4-nitrophenyl) enzyme at pH 7 were
compared to those in 2 M KOH, and a shift in λmax from ∼442
to ∼482 nm was taken as a positive identifier of TPQ. However,
as a similar pattern was seen for both 2 and 11 under these
conditions, this test alone cannot distinguish between the TPQ
and LTQ cofactors in enzymes. We suggest that, in addition to
using the basic pH shift, the spectra of the modified cofactors
should also be monitored under acidic conditions. By analogy
to the model systems, derivatized LTQ should show a red shift
in the λmax from 446 to 478 nm between pH 3 and pH 1, giving
a spectrum distinct from that of derivatized TPQ (λmax ) 388
Figure 5. (a) Absorption spectra of (- - -) 2 at pH 0.23, (b) 2 at pH 3.11,
(s) 2 at pH 7.01, and (O) 2 at pH 13.43. [2] ) 2.3 × 10-5 M. (b)
Spectroscopic determination of acidic pKa for 2. The line was drawn by a
2H+
1
least-squares fit using the equation A478 ) {(ꢀ478 [H+] + ꢀ4782 Ka )/
1
([H+] + Ka )}[2]T. (c) Spectroscopic determination of basic pKa for 2: (b)
absorbance at 486 nm and (O) absorbance at 446 nm. The line was drawn
2-
2
by a least-squares fit using the equations A486 ) {(ꢀ4862[H+] + ꢀ486 Ka )/
2
2-
2
2
([H+] + Ka )}[2]T and A446 ) {(ꢀ4462[H+] + ꢀ446 Ka )/([H+] + Ka )}[2]T.
nm). Further reducing the pH below 1 should shift the λmax of
derivatized TPQ to 470 nm while leaving that of derivatized
LTQ unchanged.
4-Nitrophenylhydrazine is generally used in preference to
phenylhydrazine in detection of TPQ in CAOs, as the adduct
resulting from the former has stronger UV-vis spectral features
than that from the latter.54 In contrast to 12, 10 was found to be
unstable in aqueous solution, and it was not possible to
accurately determine the pKa values of 10, where an approximate
value of 11.7 was measured (vs 12.2 for 1254). In aqueous
solution, 10 has a λmax at 480 nm at 10 s after dilution from the
CH3CN stock solution, but time-dependent changes of the
spectrum were observed at pH <9 (λmax at 480 nm decreased
and a new absorption built in at 600 nm). 10 did not show this
spectral change in organic solvents and gave a λmax of
approximately 460 nm. Above pH 12, 10 was stable in aqueous
solution and showed a λmax at 572 nm, showing the similar base-
(54) Janes, S. M.; Palcic, M. M.; Scaman, C. H.; Smith, A. J.; Brown, D. E.;
Dooley, D. M.; Mure, M.; Klinman, J. P. Biochemistry 1992, 31, 12147-
12154.
9
6122 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

LTQ Cofactor of Lysyl Oxidase
A R T I C L E S
Figure 6. (a) Absorption spectra of (- - -) 11 at pH -0.33, (b) 11 at pH 4.38, (s) 11 at pH 7.10, (×) 11 at pH 10.16, and (O) 11 at pH 13.18. [11] )
1.9 × 10-5 M. (b) Spectroscopic determination of acidic pKa (pKa ) for 11. The line was drawn by a least-squares fit using the equation A470
)
1
11H+
{(ꢀ470
[H+] + ꢀ47011Ka1)/([H+] + Ka1)}[11]T. (c) Spectroscopic determination of the second pKa (pKa2) for 11: (b) absorbance at 350 nm and (O) absorbance
11- 2 2
at 430 nm. The line was drawn by a least-squares fit using the equations A350 ) {(ꢀ35011[H+] + ꢀ350 Ka )/([H+] + Ka )}[11]T and A430 ) {(ꢀ43011[H+] +
11-
2
2
3
ꢀ430 Ka )/([H+] + Ka )}[11]T. (d) Spectroscopic determination of basic pKa (pKa ) for 11: (b) absorbance at 486 nm and (O) absorbance at 430 nm. The
11-
112-
3
3
11-
112-
3
line was drawn by a least-squares fit using the equations A486 ) {(ꢀ486 [H+] + ꢀ486 Ka )/([H+] + Ka )}[11]T and A430 ) {(ꢀ430 [H+] + ꢀ430 Ka )/
3
([H+] + Ka )}[11]T.
Scheme 3. Proposed Pathway for Biogenesis of LTQa
a Steps: (i) autocatalytic oxidation with O2 and Cu2+ or exogenous enzyme catalysis; (ii) O2 oxidation or exogenous enzyme catalysis; (iii) 1,4-addition;
(iv) O2 oxidation.
induced purple shift associated with the 4-nitrophenylhydrazine
adduct of 4 (12).54 Because of the unassigned spectral changes
of 10 at neutral pH, derivatization with phenylhydrazine is
recommended for identification and quantification of the LTQ
cofactor in enzyme systems.
Model Study on Biogenesis of LTQ. The biogenesis of TPQ
has been extensively studied. It is well accepted that TPQ is
formed from a tyrosine residue at the active site via an
autocatalytic mechanism, requiring Cu(II) at the active site and
O2, but no other enzymes.32,35,55 At this juncture there is little
known about the mechanism of the biogenesis of LTQ, mainly
due to the absence of an expression system yielding an active
enzyme in suitable quantities for detailed spectroscopic studies.
We have proposed a mechanism based upon that of TPQ
biogenesis,27 in which the hydroxylation of a tyrosine residue
at the active site, either autocatalytic or enzymatic, is followed
by formation of dopaquinone, subsequent nucleophilic addition
of ꢀ-amino group of a lysine side chain, and air oxidation to
yield LTQ (Scheme 3).
One of the intriguing aspects of LTQ biogenesis concerns
the directionality of nucleophilic attack by a lysine side chain
on the dopaquinone intermediate. In theory, lysine could add
to dopaquinone via either a 1,2-, 1,4-, or 1,6-pathway. Naturally
occurring o-quinones undergo nucleophilic addition by amino
acids such as lysine, cysteine, and histidine, and this addition
initiates polymerization. This process is known as quinone
(55) Schwartz, B.; Olgin, A. K.; Klinman, J. P. Biochemistry 2001, 40, 2954-
2963.
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6123

A R T I C L E S
Mure et al.
tanning and is a source of pigmentation in plants, insects, and
animals and also glue and varnish formation in mollusks.56,57
Insect cuticles contain quinones that are cross-linked with
histidine side chains and are the cause of sclerotization of the
exoskeleton.45,58 Cross-links between glutathione and o-quinone
have been detected in flesh fly (Sarcophaga peregrina) infected
with Escherichia coli, where this product is proposed to be
involved in a self-defense mechanism against bacterial infec-
tion.43,59 Studies of the reactions seen in insects have shown
that the quinones are derivatized via both 1,4- and 1,6-addition
to the ring (in which the latter is preferred) and also by addition
to their side-chain carbons.60
(A) Synthesis and Characterization of 13red. To definitively
assign LTQ as the product of a 1,4-addition of lysine to tyrosine,
we synthesized the alternative 1,6-addition product (13red
)
starting from 5-nitrovaniline in seven steps. Details of the
synthesis are given in the Supporting Information. 13red has an
absorption maximum at 266 nm at pH 7.2 which is 34 nm blue
shifted compared to 1red (Figure 7a). 13red undergoes air
oxidation, but the product is unstable. To obtain a clean UV-
vis spectrum of the oxidized product of 13red, the reaction was
carried out in the presence of NaIO4, where the oxidation of
13red was completed within 2 s. The oxidized product has a
λmax at 476 nm and approximately 740 nm (Figure 7b), but it
was very unstable and underwent bleaching to yield a featureless
spectrum within 10 min (Figure 7a).
Nonetheless, it was possible to obtain a quasi-reversible cyclic
voltammogram (∆Ep ) 28 mV) of the transient oxidized form
(by air oxidation) of 13red, yielding a midpoint potential of
-0.106 V vs SCE (Figure 7c). The same value of Em is
measured using 13red as a starting product, suggesting that the
oxidized product is the quinone (data not shown). This value is
about 80 mV positive when compared to 1ox. The instability of
13ox may only be applicable to solution chemistry, and such an
adduct could be stabilized in the enzyme active site. However,
the λmax of 13ox is quite different from 1ox, confirming the
assignment of LTQ as the 1,4-addition product. Due to the
instability of 13ox, it was not possible to further characterize
this analogue of LTQ.
(B) Formation of 3ox from 4-Ethylphenol and N-Methyl-
n-butylamine. To understand further how LTQ is formed from
a tyrosine and a lysine residue in the active site of LOX,
reactions were studied between amines and phenols or catechols
in the presence of tyrosinase or NaIO4 as an oxidant. This
methodology is based on an earlier study where a number of
amino acid cross-linked quinones were generated in situ by a
reaction with catechols in the presence of tyrosinase.61,62
When 4-ethylphenol, a model compound for a tyrosine
residue, is oxidized in the presence of tyrosinase and N-methyl-
n-butylamine, 3ox is formed as the sole product (identified and
quantitated by its UV-vis spectrum; Figure 8a). Tyrosinase
catalyzes the hydroxylation of phenol to catechol and the
oxidation of catechol to the o-quinone. The o-quinone reacts
Figure 7. (a) UV-vis spectrum of 13red at pH 6.7 (- - -). Offset: Slow
decomposition of 13ox following NaIO4 oxidation of 13red as followed by
the bleaching at 476 and 730 nm (2 s, 13 s, 30 s, 1 min, 2 min, 3 min, 5
min, 10 min). (b) Comparison of UV-vis spectra of 13ox (s) and 1ox
(- - -). All spectra are for 0.1 mM of the appropriate model compound in
0.1 M potassium phosphate (pH ) 6.7, I ) 0.3 with KCl). (c) Cyclic
voltammogram of 13ox at pH 6.7 at a scan speed of 100 mV s-1. 13ox was
in situ generated from air oxidation of 13red. The direction of the scan is
from positive to negative. Background (buffer) was subtracted.
with the amine and, in the absence of amine, accumulates and
gradually decomposes. Clearly, the reaction between the o-
quinone and N-methyl-n-butylamine is rapid as only 3ox is
detected when the amine is present, indicating a strong prefer-
ence for 1,4-addition. When 4-ethylcatechol was used in place
of 4-ethylphenol, the final product was similar to 3ox; however,
the spectrum of the final reaction mixture showed small
differences (Figure 8b), suggesting that this reaction was not
quite as clean. A similar result was obtained using NaIO4 in
place of tyrosinase with 4-ethylcatechol and N-methyl-n-
butylamine. We suggest that the best way to prepare 3ox in situ
is to start from 4-ethylphenol and to use tyrosinase as a catalyst
for oxidation.
(56) Waite, J. H. Comp. Biochem. Physiol. B 1990, 97, 19-29.
(57) Burzio, L. A.; Waite, J. H. Biochemistry 2000, 39, 11147-11153.
(58) Sugumaran, M. Bioorg. Chem. 1987, 15, 194.
In contrast to N-methyl-n-butylamine, when n-butylamine was
used as the nucleophile, the reaction was not clean, and multiple
products were suggested by the UV-vis and NMR spectra (data
not shown). 1ox, if formed, is not the only product, concomitant
with the expectation that the o-quinone would react in multiple
(59) Akiyama, N.; Hijikata, M.; Kobayashi, A.; Yamori, T.; Tsuruo, T.; Natori,
S. Anticancer Res. 2000, 20, 357-362.
(60) Christensen, A. M.; Schaefer, J.; Kramer, K. J.; Morgan, T. D.; Hopkins,
T. L. J. Am. Chem. Soc. 1991, 113, 6799-6802.
(61) Mason, H. S.; Peterson, E. W. Biochim. Biophys. Acta 1965, 111, 134-
146.
(62) Rzepecki, L. M.; Waite, J. H. Anal. Biochem. 1989, 179, 375-381.
9
6124 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

LTQ Cofactor of Lysyl Oxidase
A R T I C L E S
amounts of TPQ (data not shown). Similar results have been
observed for other 4-alkyl-1,2-benzoquinones which are resistant
to hydration.67 It has been shown that the precursor, the Cu2+
-
free/Tyr-containing form of CAOs, folds in an identical manner
to the mature protein (Cu2+/TPQ form)35,63 and that the
biogenesis of TPQ is initiated upon addition of metal ion.35,64,65
Cu2+ is proposed to play a key role in this step, providing an
66
electrophilic center for the activation of tyrosine toward O2
and for the reduction of the pKa of the nucleophilic water that
generates TPQ from dopaquinone.33,36,67
Conclusions
The synthesis of a model compound (1ox) for the LTQ
cofactor of LOX together with a number of homologues has
been presented. The characterization of these compounds,
including their spectroscopic, acid-base, and redox properties,
provides the groundwork for future studies of both LOX and
its isozymes, as well as models of these physiologically
important enzymes. The UV-vis spectroscopic properties of
the phenylhydrazine adduct of 1ox and 4, 2, and 11 respectively,
have been compared. The spectrum of 2 is similar to that of 11
at high pH, whereas low pH titrations indicate a different
ionization for 2; in this manner, the low-pH UV-visible spectral
shift of 2 that is absent in 11 provides a means of distinguishing
the LTQ cofactor from TPQ. Synthesis and characterization of
a 1,6-adduct between dopaquinone and butylamine (13) confirm
the assignment of LTQ as a 1,4-addition product. The non-
specificity of primary amine addition to dopaquinone in solution
suggests that the protein environment plays a key role in
directing the nucleophilic addition of the ꢀ-amino group of a
lysyl side chain toward the correct position of the putative
o-quinone intermediate in LTQ biogenesis. Further studies will
focus on developing a model system to study the amine
oxidation and reaction with LOX inhibitors.
Figure 8. (a) Formation of 3ox from the reaction of N-methyl-n-butylamine
and 4-ethylphenol catalyzed by mushroom tyrosinase. (b) Comparison of
UV-vis spectra of authentic 3ox (s) with species formed from the oxidation
of 4-ethylphenol (b) or 4-ethylcatechol (- - -) in the presence of N-methyl-
n-butylamine. [4-Ethylphenol] ) 2 × 10-4 M, [4-ethylcatechol] ) 2 ×
10-4 M, [N-methyl-n-butylamine] ) 2 × 10-2 M, and [mushroom
tyrosinase] ) 0.2 µg/mL, in potassium phosphate buffer, pH 7.2.
ways with primary amines (1,2-, 1,4-, and 1,6-addition). tert-
Butylamine did not yield any addition product, and isopropyl-
amine reacted slowly and gave results similar to those of
n-butylamine. In this particular model system, it is necessary
to use N-methyl-n-butylamine to prevent the formation of a
Schiff base (1,2-addition product).
Acknowledgment. We thank Dr. Graham Ball for acquiring
2D NMR spectra and Prof. Marcin Majda for allowing use of
the BAS 100A electrochemical analyzer. M.M. also thanks Dr.
Julian Limburg for critical reading of the manuscript. Supported
by a grant from the NIH (GM39296 to J.P.K.).
These results suggest that the specific 1,4-cross-linking of
tyrosine with the ꢀ-amino group of lysine at the active site of
LOX will be carefully controlled to prevent undesired products.
The tyrosine-derived dopaquinone intermediate is generally
accepted to be the precursor of the TPQ cofactor. To control
the selectivity of nucleophilic attack, it is likely that the precursor
protein is fully folded with the correct positioning of the lysyl
side chain, prior to the initiation of oxidative chemistry on the
precursor tyrosine. This will prevent unwanted chemistry
involving the o-quinone intermediate. An analogous situation
exists for the final step in the biogenesis of the TPQ cofactor,
i.e., a Michael addition of hydroxide (water) to a dopaquinone
intermediate. In aqueous solution, at neutral pH, the o-quinone
does not undergo this reaction at an appreciable rate but, instead,
slowly decomposes over the course of 3 h, with only trace
Supporting Information Available: The experimental details
of the synthesis of 13red and the NMR characterization of 2 and
10. This material is available free of charge via the Internet at
http://pubs.acs.org.
JA0214274
(63) Chen, Z.; Schwartz, B.; Williams, N. K.; Li, R.; Klinman, J. P.; Mathews,
F. S. Biochemistry 2000, 39, 9709-9717.
(64) Matsuzaki, R.; Fukui, T.; Sato, H.; Ozaki, Y.; Tanizawa, K. FEBS Lett.
1994, 351, 360-364.
(65) Cai, D. Y.; Klinman, J. P. J. Biol. Chem. 1994, 269, 32039-32042.
(66) Schwartz, B.; Dove, J. E.; Klinman, J. P. Biochemistry 2000, 39, 3699-
3707.
(67) Mandal, S.; Lee, Y.; Purdy, M. M.; Sayre, L. M. J. Am. Chem. Soc. 2000,
122, 3574-3584.
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6125