Model insect cuticle sclerotization: Reactions of catecholamine quinones with the nitrogen-centered nucleophiles imidazole and N-acetylhistidine

Article abstract of DOI:10.1006/bioo.1997.1065

The catecholamines N-acetyldopamine (NADA) and N-β-alanyldopamine (NBAD) are two precursors for quinonoios used as sclerotizing agents in insect cuticle. This study focused on the reaction pathways of the quinones of NADA and NBAD by using two nitrogencen

Full text of DOI:10.1006/bioo.1997.1065

BIOORGANIC CHEMISTRY 25, 179–202 (1997)
ARTICLE NO. BH971065
Model Insect Cuticle Sclerotization: Reactions of Catecholamine
Quinones with the Nitrogen-Centered Nucleophiles
Imidazole and N-Acetylhistidine
Xin Huang,* Rongda Xu,† M. Dale Hawley,* and Karl J. Kramer‡
*Department of Chemistry and †Department of Entomology, Kansas State University,
Manhattan, Kansas 66506; and ‡Grain Marketing and Production Research Center,
Agricultural Research Service, United States Department of Agriculture,1
Manhattan, Kansas 66502
Received March 24, 1997
The catecholamines N-acetyldopamine (NADA) and N-ͱ-alanyldopamine (NBAD) are
two precursors for quinonoios used as sclerotizing agents in insect cuticle. This study focused
on the reaction pathways of the quinones of NADA and NBAD by using two nitrogen-
centered nucleophiles, imidazole and N-acetylhistidine, to model cuticular proteins containing
histidyl residues. The quinones were prepared by electrochemical oxidation, using either a
coulometric microcell or a flow-through cell. The reactions of the quinones with the
nucleophiles were investigated at physiological pH using electrochemical, chromatographic,
and spectroscopic methods. The major products were purified by semipreparative liquid
chromatography and identified by mass spectrometry and nuclear magnetic resonance
spectroscopy to be nucleophilic addition products of the quinones with the nucleophiles
bonded to two carbons in the aromatic ring. The predominant products for both nucleophiles
were C6 adducts of NADA and NBAD. C2 adducts of N-acetylhistidine were minor
products.
1997 Academic Press
INTRODUCTION
Insect cuticle sclerotization or hardening is a vital process required for the survival
of insects (1–4). During sclerotization, the chemical and mechanical properties and
the appearance of the cuticle are changed: the water content decreases, the matrix
proteins become much more resistant to extraction and degradation, and the cuticle
becomes stiff and hard. This process has been under extensive study for several
decades and especially so in recent years. In a recent model for the insect sclerotiza-
tion process, catechols are first oxidized to their corresponding electrophilic
o-quinones by cuticular phenoloxidases (1, 5). The o-quinones are subsequently
isomerized, first to p-quinone methides by quinone isomerases and then to Ͱ,ͱ-
dehydrocatechols by quinone methide isomerases. The Ͱ,ͱ-dehydrocatechols are
then oxidized to form the corresponding dehydrocatechol o-quinones, which are
subsequently isomerized into p-quinone methides. All these quinonoid intermedi-
1 The Agricultural Research Service, USDA, is an equal opportunity/affirmative action employer and
all agency services are available without discrimination.
179
0045-2068/97 $25.00
Copyright 1997 by Academic Press
All rights of reproduction in any form reserved.

180
HUANG ET AL.
ates can act as sclerotizing agents and react directly with nucleophilic groups of
cuticular proteins to form covalent bonds. The three different pathways involving
reactions of nucleophiles with quinones, quinone methides, and Ͱ,ͱ-dehydrocate-
chols are denoted as quinone tanning, quinone methide sclerotization, and Ͱ,ͱ-
sclerotization, respectively (1, 6, 7). The protein-bonded catechols, then, might be
oxidized again and form another round of sclerotizing agents, which can react with
nucleophilic groups in other protein molecules. This second addition results in
covalent cross-linking between polypeptide chains via aromatic or side-chain car-
bons of catechols. In addition to these covalent cross-links, replacement of water
in cuticle with catechols or their polymeric oxidation products and formation of
numerous noncovalent interactions might also further stabilize the protein–chitin
matrix of the cuticle (1, 8).
Catechols that act as immediate precursors for quinonoid-sclerotizing agents in
insect cuticle are primarily N-acyl derivatives of dopamine, such as N-acetyldopa-
mine (NADA)2 and N-ͱ-alanyldopamine (NBAD) (1). Although other catechols,
for example, 3,4-dihydroxyphenylalanine, 3,4-dihydroxyphenylacetic acid, N-acetyl-
norepinephrine, and N-ͱ-alanylnorepinephrine, have also been detected in extracts
of insect cuticles, their roles in sclerotization are less certain (2). Until now, convinc-
ing evidence for precursor roles was available only for NADA and NBAD. NADA
was found to be present in several insect species in 1962 and has been demonstrated
to be incorporated into cuticle during the sclerotization process (9, 10). In 1982,
NBAD was discovered to be a major precursor for tanning agents in stiff brown
cuticles (11, 12). Although the involvement of NADA and NBAD in sclerotization
has been established, the mechanisms (Scheme 1) by which these compounds serve
as precursors for quinonoios used as sclerotizing agents have not been fully eluci-
dated and some aspects remain controversial.
The nucleophilic groups of cuticular proteins that react with the quinonoid agents
have been suggested to be Ͱ-amino groups, ␧-amino groups of lysyl residues, and
imidazole nitrogens of histidyl residues (1, 13). Other nucleophilic groups of cuticu-
lar proteins, such as thiol, hydroxyl, and carboxyl groups, might participate in cross-
linking as well, but their involvement is probably minor (1). Direct evidence for
the existence of covalent bonds between catecholamines and nucleophilic groups
in cuticular proteins had not been obtained until solid-state nuclear magnetic reso-
nance (NMR) was used to examine the C–N covalent linkages in the pupal cuticle
of the tobacco hornworm (14–17). It was determined by double cross-polarization
magic-angle-spinning 13C NMR that the histidyl nitrogens of cuticular protein were
attached to aromatic ring carbons of a catecholamine in pupal exuviae prepared from
Manduca sexta larvae injected with both [1,3-15N2]histidine and [ring-13C6]dopamine
2 Abbreviations used: NADA, N-acetyldopamine; NBAD, N-ͱ-alanyldopamine; NMR, nuclear mag-
netic resonance spectroscopy; NAcH, N-acetylhistidine; DA, dopamine; DMF, N,N-dimethylformamide;
MS, mass spectrometry; LC, liquid chromatography; MALDI-MS, matrix-assisted laser desorption mass
spectrometry; EC, electrochemical detection; 6-Imid-NADA, 6-(N-imidazole)-N-acetyldopamine; 6-
Imid-NBAD, 6-(N-imidazole)-N-ͱ-alanyldopamine; HMQC, heteronuclear multiple quantum coher-
ence; TOCSY, total correlation spectroscopy; 6-NAcH-NADA, 6-(N-acetylhistidyl)-N-acetyldopamine;
2-NAcH-NADA, 2-(N-acetylhistidyl)-N-acetyldopamine; 6-NAcH-NBAD, 6-(N-acetylhistidyl)-N-ͱ-
alanyldopamine; 2-NAcH-NBAD; 2-(N-acetylhistidyl)-N-ͱ-alanyldopamine.

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
181
S
CHEME 1
(14). In addition, a linkage between a histidyl nitrogen and the ͱ-carbon in the
side-chain of catechols in the pupal exuviae from larvae that had been injected
with both [1,3-15N2]histidine and [ͱ-13C6]dopamine was found by rotational echo
double resonance 13C NMR (15). The results of these solid-state NMR studies are
consistent with a sclerotization mechanism in which quinones and quinone methides
derived from N-acylcatecholamines form covalent bonds with functional groups of
cuticular proteins. However, the exact structures of the histidyl-catechol ring or
side-chain carbon addition adducts could not be determined by the solid-state
NMR studies.
In vitro evidence for cuticle-catalyzed formation of adducts between NADA and
amino acid derivatives was obtained from the analysis of products derived from
their incubation with cuticle or cuticular enzymes (18–21). The structures of a few
of these model adducts have also been determined. Recently, four histidyldopamine
adducts were isolated and characterized from an acid hydrolysate of M. sexta pupal
cuticle (22). These adducts apparently are formed from Michael 1,4- and 1,6-addition
reactions with quinonoios used as sclerotizing agents. Thus, the imidazole side chain
of histidine residues in cuticular proteins are nucleophiles critical for sclerotization.
Although knowledge about the chemistry of cuticular sclerotization has increased
substantially in recent years, many aspects are still unclear. For example, the identi-
ties of most quinonoios used as sclerotizing agents and products have not been
directly established and the kinetics of the reactions between the sclerotizing agents
and functional groups of proteins have not been determined. The objective of this

182
HUANG ET AL.
study was to elucidate the reaction pathways of the quinones of the catecholamines
NADA and NBAD, which are immediate precursors for sclerotizing agents in insect
cuticle, by using amino acid derivatives to model the functional groups in cuticular
protein. Two nitrogen-centered nucleophiles, imidazole and N-acetylhistidine
(NAcH), were employed.
EXPERIMENTAL
Syntheses of Catecholamines
Although some NADA was obtained commercially (Sigma Chemical Co., St.
Louis, MO),3 most was synthesized from dopamine (DA) using a procedure modified
from that of Andersen (23). First, 0.80 g DA hydrochloride (Sigma Chemical Co.)
was dissolved in 40 ml of 10% potassium tetraborate that had been purged with
nitrogen. Then, 0.44 ml of 99% acetic anhydride (Fisher Scientific Co., Pittsburgh,
PA) was slowly added over a period of 30 min to the DA hydrochloride solution
under a nitrogen atmosphere. To verify the completeness of the acetylation reaction,
1-Ȑl aliquots of the product mixture were subjected to thin-layer chromatography
on silica gel sheets (Eastman Kodak Co., Rochester, NY) using methanol as the
developing solvent. The retention factors of DA and NADA were 0.40 and 0.83,
respectively. After all DA had reacted, 2 ml of concentrated formic acid (Fisher
Scientific Co., Pittsburgh, PA) were added to the product mixture to precipitate
the boric acid. After filtration through Whatman No. 1 paper, the supernatant was
subjected to gel filtration chromatography on a column packed with Bio-Gel P-2
(Bio-Rad Laboratories, Richmond, CA) using 0.02% formic acid as the mobile
phase. The effluent was monitored spectroscopically at a wavelength of 280 nm.
Fractions were collected and analyzed by analytical liquid chromatography (LC)
for the presence of NADA. Those fractions showing only NADA were pooled and
lyophilized. NADA was obtained as a white powder in 80% overall yield. Mass
spectrometry (MS), NMR, and LC analyses confirmed that the desired product had
been prepared.
NBAD was synthesized from DA using a method modified from Kramer et al.
(24) and Yamasaki et al. (25). First, 0.96 g DA hydrochloride was dissolved in
17.5 ml N,N-dimethylformamide (DMF) (Fisher Scientific Co., Pittsburgh, PA) that
˚
had been deaerated with nitrogen and predried with molecular sieves (size: 5 A).
The molecular sieves had been activated by heating at 600ЊC for a period of 24 h
in the stream of nitrogen. N-Methylmorpholine (0.55 ml) (Sigma Chemical Co.)
and N-t-butyloxycarbo-ͱ-alanine-N-hydroxysuccinimide ester (17.2 g) (Chemalog,
Chemical Dynamics Co., NJ) were dissolved in a second 17.5-ml volume of DMF.
The two DMF solutions were combined and stirred for 48 h under a nitrogen
atmosphere in an ice water bath. 1-(2-Aminoethyl)piperazine (0.26 ml) (Aldrich
Chemical Co., St. Louis, MO) was then added to the product mixture in an ice
water bath, after which the solution was allowed to warm to room temperature.
3 Mention of a proprietary product does not constitute a recommendation by the USDA.

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
183
Another 0.52 ml of 1-(2-aminoethyl)piperazine was added to the mixture, which
was then stirred for another hour. The product mixture was acidified with 100 ml
of 0.25
M
H2SO4 and extracted with 100 ml of ethyl acetate. Extraction of the
aqueous layer with 30 ml of ethyl acetate was repeated three times. The extracts
were subsequently pooled, washed with saturated NaCl solution three times, and
dried over sodium sulfate under nitrogen overnight. The ethyl acetate extract was
then filtered through Whatman paper and reduced to a volume of about 10 ml
using a rotary evaporator, at which time a white solid formed. The suspension was
filtered and the solid was collected and allowed to dry under vacuum for about
4 h. The dried solid was dissolved in 4 ml of a HCl : 1,4-dioxane (1 : 2) solution and
left at room temperature for 3 h. The solution was reduced to a small volume
using a rotary evaporator. Diethyl ether was added to the concentrated solution in
sufficient volume to ensure precipitation of NBAD · HCl, which was collected by
filtration, washed with diethyl ether, and dried under vacuum overnight. The yield
was 68% (0.90 g). MS and NMR analyses confirmed that the desired product had
been prepared.
Preparation of Catecholamine Quinones
Electrochemical oxidation was used to prepare the quinones. If chemical oxida-
tion were to be used to prepare quinones, excess oxidizing agents would be generally
employed to ensure complete oxidation of the catecholamines. However, if quinones
thus obtained were to react with nucleophiles directly and form adducts, then the
presence of the excess oxidizing agent might cause complications by oxidizing the
adducts. In contrast, the electrochemical oxidation method results in the preparation
of quinones that are devoid of excess oxidizing agent.
Since the quinones are relatively stable at low pH, the electrosynthesis of the
quinones was conducted at pH 2.0 in the absence of nucleophiles. The supporting
electrolyte was 0.01
N
HCl and 0.09
N
KCl. The potential for oxidation of NADA
or NBAD was controlled at 0.7 V vs Ag/AgCl (saturated KCl). A custom-made,
three-electrode coulometer (26) was used to apply a constant potential for electroly-
sis and to give instantaneous digital coulomb and analog current readouts. A
Hewlett–Packard (Palo Alto, CA) 3390A integrator was used to monitor the elec-
trolysis current. Either a coulometric microcell or a coulometric flow-through cell
was used for the oxidation.
Coulometric microcell. A coulometric microcell was designed to prepare small
quantities of quinones (Fig. 1) (27). Volumes of catecholamine solutions used
for each batch preparation ranged from 0.4 to 1.5 ml. The large surface area
to volume ratio of the working electrode and efficient mixing of the solutions
permit rapid and quantitative coulometric conversion. Separation of the auxiliary
and working electrodes by the membrane filter prevents mixing of the anode and
cathode products, thereby permitting complete electrolysis. The close proximity of
the reference electrode to the working electrode minimizes uncompensated iR
drop and allows accurate control of the potential under high-current conditions.
In addition, the small volume of the sample compartment and short electrolysis
time make the cell suitable for use when starting materials are available in

184
HUANG ET AL.
F
IG. 1. Diagram of the microelectrochemical cell. The sample compartment is a polypropylene tube
containing a prefilter and a 0.2-Ȑm-pore size nylon 66 membrane at the tip. The working electrode is
a piece of platinum gauze (20 ϫ 50 mm) that was folded and shaped into a small sphere (diameter, ȁ5
mm). The reference electrode is a AgCl-coated silver wire placed inside a glass tube containing a
saturated KCl solution; a 0.2-Ȑm-pore size filtration membrane was used as the divider and sealed inside
the tube tip. The auxiliary electrode is a platinum foil (5 ϫ 10 mm). The stir bars in the sample
compartment and the auxiliary electrode compartment are 5 and 20 mm, respectively.
limited quantity and/or when relatively reactive compounds are to be synthesized.
This microcell was used for the electrosyntheses of reactive quinones that
were immediately used for kinetics studies and subsequent reactions with
nucleophiles.
Flow-through coulometric cell. To synthesize sufficient quantities of catechola-
mine–nucleophile adducts for structural analysis and characterization, relatively
large-scale electrosyntheses of catecholamine quinones were required. Since
catecholamine quinones are reactive, batch preparation is not desirable when a
long electrolysis time is required. Hence, a flow-through coulometric cell was
constructed that permits electrolysis of a relatively large amount of catecholamine
and allows the quinone to react with a nucleophile shortly after formation of
the quinone (Fig. 2) (27). Catecholamine solutions were introduced from a sample
reservoir by gravity flow into the sample compartment and were electrolyzed to
form quinones at a controlled potential (0.8 V). The quinone effluent was added
to a stirred solution of the nucleophile to form adducts at pH 7.0. In this flow-
through cell, rapid and efficient coulometric conversion was achieved with a
large ratio of working electrode surface area to solution volume, a concentric
arrangement of the working and the auxiliary electrodes, and a nylon membrane
that separated the working and the auxiliary electrode compartments. Both a
low uncompensated iR drop and accurate potential control were the result of
the close proximity of the tip of the reference electrode to the working electrode.
The cell design allowed the catecholamine quinones to react with nucleophiles

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
185
F
IG. 2. Diagram of the flow-through electrochemical cell. The working electrode is a platinum gauze
that was tightly folded and placed in the cellulose dialysis tubing. Rubber septa were inserted into both
ends of the dialysis tubing. A tight seal between the dialysis tubing and the rubber septa was attained
by addition of rubber o-rings that encircled the ends of the dialysis tubing. The Teflon tubing tip of the
reference electrode was inserted into the dialysis tubing and positioned close to the upper portion
of working electrode. The auxiliary electrode encircled the dialysis tubing. The auxiliary electrode
compartment contained about 4.5 ml of 0.1
M HCl solution. Solutions were introduced by gravity flow
into the dialysis tubing. The direction of fluid flow was from the bottom to the top of the cell. The dead
volume of the sample compartment in the dialysis tubing was about 0.3 ml.
outside of the cell almost immediately after their formation. This minimizes
quinone decay, which might occur during long-term electrolysis of a large quantity
of catecholamines in a batch cell. It also precludes electrochemical oxidation of
adducts that would have occurred if the catecholamine were electrolyzed in
the presence of nucleophiles. This cell is efficient for relatively large-scale
electrochemical syntheses and is especially useful when the electrogenerated
intermediates are highly reactive and must be mixed with other substrates shortly
after their formation to produce the desired products.

186
HUANG ET AL.
Syntheses of Catecholamine Adducts
Catecholamine adducts were synthesized at pH 7 by adding the effluent from
the flow-through coulometric cell, which contained NADA or NBAD quinone at
pH 2, into a solution of a nucleophile (vide supra). Purification of adducts formed
in the reaction mixtures was carried out using either gel filtration or semiprepara-
tive LC.
Gel filtration of the product mixtures from the reactions of quinones and imidazole
was performed using a column packed with Bio-Gel P-2 (Bio-Rad Laboratories).
The mobile phase used was 0.05% formic acid. The effluent containing an adduct
in dilute formic acid was lyophilized to yield the adduct as a white solid.
Purification using semipreparative LC was performed as described previously
(28). A binary mobile phase system was used that consisted of solvents A and B.
Solvent A was 150 m
solvent B was 50% methanol (uv cutoff: 204 nm), 180 m
M
formic acid and 30 m
M
ammonium formate (pH 3.0), whereas
formic acid, and 8 mM
M
ammonium formate (pH 3.0). For adducts of NBAD quinone with NAcH, the
gradient employed was 0–15 min, 100% solvent A; and 15–30 min, linear gradient
from 0 to 50% solvent B.
Analytical Liquid Chromatography
Analytical LC was conducted to analyze the compositions of product mixtures.
The liquid chromatography system and the uv/vis and electrochemical detectors
were the same as those used previously (28). Separation was achieved on a Micro-
sorb-MV C18 stainless steel column (5 Ȑm, 4.6 ϫ 250 mm) (Rainin Instrument Co.,
Inc., Woburn, MA) using a flow rate of 1 ml/min. As was reported previously,
during LC analysis of quinones, stainless steel in the columns, including the inlet
and outlet frits, causes on-column reduction of quinones, which might lead to
misidentification of chromatographic peaks (29). In the current LC studies of prod-
uct mixtures from reactions of quinones with nucleophiles, samples were injected
into columns after the quinones had been completely consumed by nucleophiles.
Thus, on-column reduction of quinones was not a concern.
The binary mobile phase system, which consisted of solvents A and B, was
used (see the preceding section). For the product mixtures produced from NADA
quinone reacting with imidazole, the gradient was 0–15 min, 80% solvent A and
20% solvent B; and 15–30 min, linear gradient from 20 to 80% solvent B. For the
product mixtures from the reaction of NBAD quinone with imidazole, the gradient
was 0 min, 90% solvent A and 10% solvent B; 0–15 min, linear gradient from 10
to 14% solvent B; 15–20 min, linear gradient from 14 to 25% solvent B; and 20–25
min, linear gradient from 25 to 40% solvent B. For the product mixtures from
NBAD quinone reacting with NAcH, the gradient was 0 min, 100% solvent A; 0–20
min, linear gradient from 0 to 20% solvent B; and 20–30 min, linear gradient from
20 to 30% solvent B.
UV/Vis Spectroscopy
In addition to on-line uv/vis detection conducted during analytical LC, uv/vis
spectra of adducts in 0.01
M
HCl solution and uv/vis spectral changes associated
with reactions of NADA quinone or NBAD quinone with various nucleophiles

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
187
and/or in various buffers were recorded in a 1.0-cm quartz cuvette using a Hewlett–
Packard (Palo Alto, CA) HP 8452A diode array spectrophotometer. For studies
of reactions of quinones with nucleophiles, once electrolysis of varying concentra-
tions of NADA or NBAD in 0.01
M
HCl and 0.09
M
KCl was completed, 0.195 ml
of electrogenerated NADA quinone or NBAD quinone was added to 0.455 ml of
either a nucleophile solution or a buffer in the quartz cuvette to give 0.650 ml of
a 3 : 10 diluted quinone solution at the desired nucleophile or buffer concentration
and pH. The uv/vis measurements within the wavelength range of 200 to 800 nm
were initiated 15 s after mixing and conducted for 10 min at 15-s intervals. The cell
temperature was controlled at 25ЊC by using a temperature-regulated circulating
water bath.
The spectral data were used to estimate the initial rate constants for the reaction
of NADA quinone or NBAD quinone with imidazole or N-acetylhistidine. To do
this, the absorbance changes at 396 nm with time were extracted from each set of
spectra. To compensate for instrument drift from one spectrum to the next, the
absorbance at 396 nm was corrected by subtracting the absorbance at 700 nm,
where there is no absorption from reactants or products. Regression analyses of
log(corrected absorbance at 396 nm) vs time were performed to obtain the pseudo-
first-order rate constants (Table 1). For the reaction between a quinone and nucleo-
phile, the reaction rate law is expected to be
reaction rate ϭ k[quinone][nucleophile] ϭ kЈ[quinone].
[1]
Cyclic Voltammetry, Mass Spectrometry, and Nuclear Magnetic
Resonance Spectroscopy
The cyclic voltammetry and NMR systems and conditions used were the same
as described previously (28). Electrospray MS of adducts was carried out either at
the National Institute of Environmental Health Sciences (Research Triangle Park,
NC) or at the University of Kansas (Lawrence, KS). A few adducts were character-
ized by matrix-assisted laser desorption mass spectrometry (MALDI-MS) at Kansas
State University.
RESULTS
Spectroscopic Studies of Reactions of N-Acetyldopamine and
N-ͱ-Alanyldopamine Quinones with Imidazole
The uv/vis spectroscopic studies of the kinetics of the reaction of NADA and
NBAD quinones with imidazole were performed at various pH values and molar
ratios. In all cases studied, the analytical concentration of imidazole was at least
10 times greater than that of the quinone, such that the kinetics behavior of the
quinones was investigated under pseudo-first-order conditions.
The uv/vis spectra that were recorded for 0.3 m
M
NADA quinone in 200 m
M

188
HUANG ET AL.
F
IG. 3. Spectral changes associated with the reaction of 0.3 mM NADA quinone with 200 mM imidazole
at pH 7.0. Although spectral measurements were conducted at 15-s intervals from 15 to 600 s after
mixing of NADA quinone and imidazole solutions, for clarity only every other spectrum is shown here.
(Inset) Absorbances at 236, 284, and 396 nm vs time.
imidazole at pH 7.0 are shown in Fig. 3, with three absorbance–time curves ap-
pearing as an inset. The absorbance at the ␭
of the quinone, 396 nm, decreases
max
with time, indicating consumption of the quinone, whereas the absorbances at 236
and 284 nm increase, indicating the formation of a new product(s). The major
product was determined to be an adduct of the corresponding catecholamine qui-
none with imidazole (vide infra). Similar changes in absorbances were also observed
for quinones of various concentrations in the presence of various concentrations
of imidazole and at different pH values. Comparable results were obtained for the
kinetics behavior of NBAD quinone in the presence of imidazole (data not shown).
A broad absorption from ca. 300 to 600 nm is observed in the uv/vis spectra of
some NBAD quinone reaction systems, such as in the case of 0.3 m
M
NBAD
quinone in the presence of 50 m imidazole at pH 7 (Fig. 4). This result indicates
M
that another product(s) is forming in addition to the adduct of NBAD and imidazole.
When the ratio of imidazole to NBAD quinone decreases or when the pH value

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
189
F
IG. 4. Spectral changes associated with appearance of an unidentified species during the decay of
0.3 m NBAD quinone in the presence of 50 m imidazole at pH 7.0. Although spectral measurements
M
M
were conducted at 15-s intervals from 15 to 600 s after mixing of NBAD quinone and imidazole solutions,
for clarity only every other spectrum is shown here. The unidentified species has a broad absorption
from ca. 300 to 600 nm. (Inset) Absorbances at 236, 284, 396, and 500 nm vs time.
increases, formation of the minor product becomes more evident (data not shown).
A similar phenomenon also occurs for both NADA and NBAD quinones in buffer
solutions in the absence of imidazole (unpublished data). Apparently, this phenome-
non is due to a side reaction(s) of the quinones, which competes with the addition
reaction of imidazole.
The effects of quinone concentrations, imidazole concentration, and pH on rate
constants for the quinone reactions were examined and the expected rate law (Eq.
[1]) was confirmed. First, the pseudo-first-order rate constants for reactions of 0.15,
0.3, 0.6, and 0.9 m
imidazole at pH 7.0 were found to be independent of the quinone concentration.
Second, the pseudo-first-order rate constants for reactions of 0.3 m NADA qui-
none (Table 1) or NBAD quinone with 20, 50, 100, and 200 m imidazole at pH
7.0 showed the expected linear dependence on the concentration of imidazole.
M
NADA quinone (Table 1) or NBAD quinone with 50 m
M
M
M

190
HUANG ET AL.
TABLE 1
Rate Constants for NADA Quinone/Imidazole System
[NADA
quinone]
Pseudo-first-order
Second-order
rate const.d
(
CImid
[Imid]eq
rate const.c
a
b
1
No.
(m
M)
(m
M
)
pH
(m
M
)
(sϪ1) ϫ 103
MϪ sϪ1) ϫ 102
1
2
3
4
5
6
7
8
9
0.15
0.30
0.60
0.90
0.30
0.30
0.30
0.30
0.30
50
50
50
50
20
100
200
50
7.0
7.0
7.0
7.0
7.0
7.0
7.0
6.1
8.0
25
25
25
25
10
50
100
5.6
45.5
2.1
2.3
2.2
2.8
1.1
4.1
7.0
0.5
3.7
8.4
9.2
8.8
11
11
8.2
7.0
9.3
8.1
50
a Cimid is the analytical concentration of imidazole, i.e., the sum of the equilibrium concentrations of
imidazole and its conjugate acid, imidazolium ion.
b [Imid]eq , the equilibrium concentration of imidazole, is calculated using the corresponding analytical
concentration and the pKa of imidazolium ion, which is 7.0.
c The pseudo-first-order rate constants were obtained from slopes of plots for regression analyses of
log(absorbance at 396 nm) vs time.
d The second-order rate constants were obtained by dividing the corresponding pseudo-first-order
rate constants by the equilibrium concentrations of imidazole at the specified pH.
Third, the results obtained for reactions of 0.3 m
M
NADA quinone (Table 1) or
NBAD quinone with 50 m imidazole at pH 6.1, 7.0, and 8.0 show that pseudo-
M
first-order rate constants increase with pH in this pH range. This is expected, since
the fraction of imidazole that is not protonated increases with increasing pH. The
pKa of imidazolium ion is 7.0. At each pH, the equilibrium concentration of imidaz-
ole, i.e., the effective concentration of imidazole as a nucleophile was calculated
from the corresponding analytical concentration of imidazole and the pKa of imida-
zolium ion (Table 1). From the equilibrium concentration of imidazole and the
pseudo-first-order rate constant (kЈ), the second-order rate constant for reaction
of the quinone with the imidazole (k) was obtained by using k ϭ kЈ/[Imidazole]
2
1
1
(Table 1). The average second-order rate constants are 9.0 ϫ 10Ϫ MϪ sϪ for the
2
1
1
reaction of NADA quinone with imidazole and 7.4 ϫ 10Ϫ MϪ sϪ for the reaction
of NBAD quinone with imidazole.
Liquid Chromatographic Studies of Product Mixtures from Reactions of
N-Acetyldopamine and N-ͱ-Alanyldopamine Quinones with Imidazole
The compositions of the product mixtures obtained from mixing 90 Ȑl of 1 m
M
NADA quinone or NBAD quinone with 225 Ȑl of 40 or 400 m imidazole (1 : 100 or
M
1 : 1000 molar ratio) at pH 7.0 were analyzed by LC with uv/vis and electrochemical
detection (EC) under different gradient conditions. LC-EC (reduction) chromato-
grams for product mixtures showed no discernible peaks (data not shown), which
indicates that the quinone had been consumed completely. In the corresponding

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
191
F
IG. 5. LC-EC (oxidation) chromatogram of the product mixture from the reaction of NADA quinone
with imidazole (1 : 1000 molar ratio) at pH 7.0.
LC-EC (oxidation) chromatograms for the products from reactions of NADA
quinone with imidazole (Fig. 5), there is one major peak at 6.8 min and a minor
peak at 17.4 min. The reaction product that gives rise to the peak at 6.8 min is
denoted here as Imid-NADA. The peak at 17.4 min is due to NADA. For the
product mixtures from reactions of NBAD quinone with imidazole, the LC-EC
(oxidation) chromatograms obtained using a different gradient exhibit one major
peak at 5.4 min and a minor peak at 12.7 min (data not shown). The peak at 5.4
min is due to reaction product, which is denoted as Imid–NBAD, whereas the peak
at 12.7 min is due to NBAD.
The amounts of the adduct and the catecholamine produced in either the NADA
quinone system or the NBAD quinone system were found to be dependent upon
the molar ratio of the initial concentrations of the quinone and imidazole. The
[Imid–NADA] : [NADA] ratios were approximately 4 : 1 and 26 : 1 when the
[NADA quinone] : [imidazole] ratios were 1 : 100 and 1 : 1000, respectively. The

192
HUANG ET AL.
[Imid–NBAD] : [NBAD] ratios were 2 : 1 and 25 : 1 when [NBAD quinone] : [imidaz-
ole] ratios were 1 : 100 and 1 : 1000, respectively.
Identification and Characterization of Imidazole Adducts of N-Acetyldopamine
and N-ͱ-Alanyldopamine
In order to identify Imid–NADA and Imid–NBAD, the adducts were synthesized
and purified for structural analysis and characterization. Their molecular masses
were determined by electrospray MS. The former has MHϩ at m/z ϭ 262 and the
latter has MHϩ at m/z ϭ 291. These m/z values are the same as the theoretical
values for monoaddition adducts of imidazole with the corresponding quinones.
The structure of the Imid–NADA adduct was determined by 1H NMR spectro-
scopic studies (data not shown). The 1H spectrum shows two well-resolved singlets
in the aromatic region at 6.81 and 6.64 ppm. This result indicates that neither
o-coupling nor m-coupling occurs between these two protons. Hence, Imid–NADA
must be the C6 mono-addition adduct, 6-N-imidazole-N-acetyldopamine (6-Imid–
NADA). The chemical shift assignments for this adduct are as follows: ͳ (ppm) ϭ
7.59 (s, 1H, H2Ј), 7.10 [s (br.), 1H, H5Ј], 7.02 (d, 1H, J ϭ 7.5 Hz, H4Ј), 6.81
(s, 1H, H2), 6.64 (s, 1H, H5), 3.12 (t, 2H, J ϭ 7.5 Hz, H8), 2.36 (t, 2H, J ϭ 7.5 Hz,
H7), 1.77 (s, 3H, H11) (see Scheme 2 for structure). The structure of Imid–NBAD
was not determined, but it is assumed to be the C6 monoaddition adduct,
6-N-imidazole-N-ͱ-alanyldopamine (6-Imid–NBAD).
The uv/vis spectra of 6-Imid–NADA and 6-Imid–NBAD, which were taken
during LC of the product mixtures from the reactions of NADA quinone and
NBAD quinone with imidazole, show that both adducts have a ␭
at 284 nm
max
(data not shown).
Cyclic Voltammetric Studies of N-ͱ-Alanyldopamine in the Presence
of N-Acetylhistidine
The electrochemical behavior of NBAD in the presence of NAcH in 0.1
phosphate buffer at pH 7.0 was studied at various molar ratios and scan rates. In
the cyclic voltammogram obtained for 0.3 m NBAD in the presence of 0.4
M
M
M
NAcH at a scan rate of 100 mV/s, the first cycle of the voltammogram of NBAD
consists of an anodic peak at 0.184 V and a cathodic peak at 0.134 V (Fig. 6). The
anodic peak is due to the two-electron oxidation of NBAD to NBAD quinone and
the cathodic peak is due to the reduction of the NBAD quinone to NBAD. On
subsequent cycles another redox couple emerges at a potential that is 75 mV more
positive than the anodic and cathodic peaks for the NBAD/NBAD quinone couple.
The relative magnitudes of the new peaks increase with either increasing scan
number or decreasing scan rate, whereas the relative peak heights for the NBAD/
NBAD quinone couple decrease. These results indicate that NBAD quinone under-
goes a relatively slow chemical reaction with NAcH under these conditions to give
one or more products that are oxidized at more positive potentials. The rate of the
reaction is too slow to be measured accurately by cyclic voltammetry.

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
193
S
CHEME 2
UV/Vis Spectroscopy of Reactions of N-ͱ-Alanyldopamine Quinone
with N-Acetylhistidine
Ultraviolet/visible spectroscopic studies of the kinetics of the reaction of NBAD
quinone with NAcH solutions (pH 7.0) were also performed at different molar
ratios under pseudo-first-order conditions. In the uv/vis spectra that were recorded

194
HUANG ET AL.
F
IG. 6. Cyclic voltammogram of NBAD in the presence of NAcH in 0.1
The scan was initiated in the positive-going direction from Ϫ0.2 V at a rate of 100 mV/s. Concentrations of
NBAD and NAcH were 0.3 and 400 m , respectively.
M phosphate buffer at pH 7.0.
M
for 0.3 m
M
NBAD quinone in 200 m
M
NAcH at pH 7.0 (Fig. 7), the absorbance
at the ␭max of the quinone, 396 nm, decreases with time, indicating consumption of
the quinone, whereas the absorbances at 244 and 284 nm increase with time, indicat-
ing the formation of a new product(s). The uv/vis spectra for 0.3 m
in the presence of other concentrations of NAcH at pH 7.0 and for NBAD quinone
of different concentrations in 200 m NAcH at pH 7.0 were also recorded. All
M
NBAD quinone
M
spectra show decreases in absorbance at 396 nm and increases in absorbances at
242 and 284 nm. The major product formed from the reaction of NBAD quinone
with NAcH was determined to be an adduct of NBAD quinone with NAcH (vide
infra). In addition, as seen in the NBAD quinone/imidazole system, a broad absorp-
tion from ca. 300 to 600 nm that is due to a minor unidentified product(s) is evident
in the NBAD quinone/NAcH system (Table 2). Again, the relative yield of the
unknown product(s) is related to the ratio of NAcH to NBAD quinone as well as
the pH. Either a decrease in the ratio of NAcH to NBAD quinone or an increase
in pH caused the relative yield of the unknown product(s) to increase.
Rate constants for the reaction of NBAD quinone with NAcH were estimated
using the same procedures as those for the reactions of NADA quinone or NBAD
quinone with imidazole (Table 2). Similar results were obtained. First, the results
obtained for 0.15, 0.3, 0.6, and 0.9 m
M
NBAD quinone with 200 m
M
NAcH at
pH 7.0 showed independence of the pseudo-first-order rate constant on quinone

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
195
F
IG. 7. Spectral changes associated with the reaction of 0.3 m
M
NBAD quinone with 200 m
M
NAcH
at pH 7.0. Although spectral measurements were conducted at 15 s intervals from 15 to 600 s after
mixing of NBAD quinone and NAcH solutions, for clarity only every other spectrum is shown here.
(Inset) Absorbances at 244, 284, and 396 nm vs time.
concentration. Second, the results obtained for 0.3 m
M
NBAD quinone with 20,
50, 100, and 200 m NAcH at pH 7.0 showed that the pseudo-first-order rate
M
constants have a linear dependence on the concentration of NAcH (Fig. 8). The
positive intercept is due to intrinsically slow reactions of NADA quinone or NBAD
quinone with unidentified components of the solvent system. Third, since the pKa
of the conjugate acid of NAcH is 7.08, the equilibrium concentration of NAcH,
which is the effective concentration of NAcH as a nucleophile, is only 45% of the
analytical concentration of NAcH at pH 7.0. When the pseudo-first-order rate
constants were divided by the equilibrium concentration of NAcH, an average value
2
1
1
of 5.7 ϫ 10Ϫ MϪ sϪ was obtained for the second-order rate constant (Table 2).
Liquid Chromatography of Product Mixture from Reaction of
N-ͱ-Alanyldopamine Quinone with N-Acetylhistidine
The composition of the product mixture obtained from mixing 90 Ȑl of 1 m
M
NBAD quinone with 225 Ȑl of 40 m NAcH (1 : 100 molar ratio) at pH 7.0 was
M

196
HUANG ET AL.
TABLE 2
Rate Constants for NBAD Quinone/NAcH System
[NBAD
quinone]
Pseudo-first-order
Second-order
rate const.d
(
CNAcH
[NAcH]eq
rate const.c
a
b
1
No.
(m
M)
(mM
)
pH
(m
M)
(sϪ1) ϫ 103
MϪ sϪ1) ϫ 102
1
2
3
0.15
0.30
0.60
0.90
0.30
0.30
0.30
200
200
200
200
20
7.0
7.0
7.0
7.0
7.0
7.0
7.0
90.8
90.8
90.8
90.8
9.1
4.7
4.9
4.6
4.9
0.7
1.4
2.5
5.2
5.4
5.1
5.4
7.3
6.2
5.5
4e
5e
6e
7
50
100
22.7
45.4
a CNAcH is the analytical concentration of imidazole, i.e., the sum of the equilibrium concentrations
of NAcH and its conjugate acid.
b [NAcH]eq , the equilibrium concentration of NAcH, is calculated from the corresponding analytical
concentration and pKa of NAcH’s conjugate acid, which is 7.08.
c The pseudo-first-order rate constants were obtained from slopes of plots for regression analyses of
log(absorbance at 396 nm) vs time.
d The second-order rate constants were obtained by dividing the corresponding pseudo-first-order
rate constants by the equilibrium concentrations of NAcH at the specified pH.
e In these cases, a broad absorption from ca. 300 to 600 nm is evident in the spectra of NBAD quinone
in the presence of imidazole.
analyzed by LC with uv/vis and EC detection. The LC-EC (reduction) chromato-
gram showed no discernible peaks (data not shown), which indicates that the qui-
none had been consumed completely. In the corresponding LC-EC (oxidation)
chromatogram there are three major peaks resolved (Fig. 9). The peak at 18.7 min
is due to NBAD, whereas the peaks at 14.9 and 7.3 min are due to two reaction
products, which are denoted here as NAcH–NBAD-I and NAcH–NBAD-II, re-
spectively. Again, the amounts of the adducts and the catecholamine produced
were related to the molar ratio of the initial concentrations of the quinone and
imidazole. The [NAcH–NBAD-I and NAcH–NBAD-II] : [NBAD] ratios were ap-
proximately 3 : 1 and 7 : 1 when the [NBAD quinone] : [NAcH] ratios were 1 : 100
and 1 : 1000, respectively.
Identification and Characterization of N-Acetylhistidine Adducts of
N-ͱ-Alanyldopamine
In order to identify NAcH–NBAD-I and NAcH–NBAD-II, the adducts were
synthesized and purified for structural analysis and characterization. Their molecular
masses were determined by MALDI-MS. The results indicate that both are mono-
addition adducts of NBAD (theoretical value: m/z ϭ 420).
The structure of NAcH–NBAD-I was determined by NMR spectroscopic studies.
1
For an H NMR spectrum obtained in H2O : D2O (90 : 10) at 30ЊC, the chemical
shift assignments are as follows: ͳ (ppm) ϭ 8.45 (s, 3H, NH13), 8.34 (s, 1H, H2Ј),
7.96 [s (br.), 1H, NH9], 7.94 (d, 1H, NH8Ј), 7.27 (s, 1H, H5Ј), 6.91 (s, 2H, H2, H5),

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
197
F
IG. 8. The dependence of the rate constants for the reaction of NBAD quinone with NAcH on
NAcH concentration at pH 7.0. The NBAD quinone concentration was 0.3 m , whereas the analytical
concentrations of NAcH were 20, 50, 100, and 200 m , respectively. The pseudo-first-order rate constants
M
M
were obtained from slopes of plots for regression analyses of log(absorbance at 396 nm) vs time.
4.45 (m, 1H, H7Ј), 3.23 (m, 1H, H6Јa), 3.17 (m, 2H, H8), 3.16 (m, 2H, H12), 3.05
(m, 1H, H6Јb), 2.56 (m, 2H, H11), 2.52 (m, 2H, H7), and 2.01 (s, 3H, H10Ј). These
assignments were confirmed by two-dimensional heteronuclear multiple quantum
coherence (HMQC) and total correlation spectroscopy (TOCSY) NMR experi-
ments. The resonance signals of the two phenyl protons are overlapped at 6.91 ppm
1
in the H spectrum, but are more resolved in the HMQC spectrum. In order to
1
1
resolve this overlapped signal in the H NMR spectrum, a H experiment was
conducted in an alternative solvent containing acetone-d6 and D2O (10 : 1). The
two phenyl protons exhibited two well-resolved singlets at 6.81 and 6.71 ppm (data
not shown). Since neither o-coupling nor m-coupling is observed, the resonances
of 6.91 ppm in H2O/D2O are assigned to H2 and H5. Therefore, NAcH–NBAD-I
must be the C6 mono-addition adduct, 6-N-(N-acetylhistidyl)-N-ͱ-alanyldopamine
(6-NAcH–NBAD) (see Scheme 2 for structure).
The structure of NAcH–NBAD-II was also elucidated by 1H NMR spectrometry.
1
Several proton resonance peaks in the H spectrum of this adduct are present in
pairs, which may be due to a slow conformational change in the structure over the

198
HUANG ET AL.
F
IG. 9. LC-EC (oxidation) chromatogram of the product mixture from the reaction of NBAD quinone
with NAcH (1 : 100 molar ratio) at pH 7.0.
NMR time scale. This feature and the pattern of the resonance signals are parallel
to those of 2-NAcH–NADA (28). Therefore, NAcH–NBAD-II is assigned as the
C2 monoaddition adduct, 2-N-(N-acetylhistidyl)-N-ͱ-alanyldopamine (2-NAcH–
NBAD). The chemical shift assignments are as follows: ͳ (ppm) ϭ 8.41 (s, 1H,
H2Ј), 7.22 and 7.18 (2s, 1H, H5Ј), 6.93 and 6.92 (2s, 1H, H5), 6.91 (s, 1H, H6), 4.22
and 4.16 [dd (J ϭ 7.2, 4.9 Hz), dd (J ϭ 8.4, 5.2 Hz), 2H, H7Ј], 3.30 and 3.22 (2m,
1H, H6Јa), 3.15 (m, 2H, H8), 3.12 and 2.98 (2m, 2H, H12), 2.83 (m, 1H, H6Јb), 2.54
(m, 1H, H7), 2.45 and 2.31 (2m, 2H, H11), and 1.90 and 1.87 (2s, 3H, H10Ј) (see
Scheme 2 for structure).
The cyclic voltammetric behavior of adducts 6-NAcH–NBAD and 2-NAcH–
NBAD in 0.1
M
phosphate buffer at pH 7.0 was examined at a scan rate of 200
mV/s (Fig. 10). The results demonstrate that both adducts are oxidized at a more
positive potential than NBAD and that the redox behavior of neither one is as
electrochemically reversible as that of NBAD.
Ultraviolet/visible spectra of 6-NAcH–NBAD and 2-NAcH–NBAD were ac-
quired during LC analysis of the product mixture from reactions of NBAD quinone

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
199
F
IG. 10. Cyclic voltammograms of NBAD and its NAcH adducts in 0.1 M phosphate buffer at pH
7.0. The scan was initiated in the positive-going direction from Ϫ0.2 V at a rate of 200 mV/s. (A)
NBAD, (B) 6-NAcH–NBAD, and (C) 2-NacH–NBAD.

200
HUANG ET AL.
with NAcH (data not shown). Both 6-NAcH–NBAD and 2-NAcH–NBAD have
a uv ␭max at 284 nm. To estimate the relative molar ratio of the adducts 6-NAcH–
NBAD and 2-NAcH–NBAD that formed during the reaction of NBAD with NAcH
at a 1 : 100 molar ratio, a LC-UV chromatogram at 284 nm was obtained by extracting
absorbances at 284 nm from the uv/vis spectra recorded during LC with on-line
UV detection (data not shown). If the two monoaddition products are assumed to
have equal molar absorption coefficients at 284 nm, then the relative molar ratio
of 6-NAcH–NBAD to 2-NAcH–NBAD is estimated to be 6 : 1 on the basis of the
areas of corresponding peaks in the LC-UV chromatogram.
DISCUSSION
Pathways for the nucleophilic additions of imidazole to NADA and NBAD
quinones and for the nucleophilic additions of NAcH to NBAD are proposed in
Scheme 2. Although imidazole and NAcH are both nitrogen-centered nucleophiles
and yield the C6 adducts of NADA and NBAD as major products, imidazole does
not yield the C2 adducts of NADA and NBAD in detectable amounts, whereas
NAcH gives the C2 adducts as minor products. The ratios of the C6 to C2 NAcH
adducts are approximately 7 : 1 for NADA quinone (28) and 6 : 1 for NBAD quinone
when the [NAcH] : [NADA or NBAD quinone] ratio is 100.
The formation of a C2 NAcH adduct but not a C5 NAcH adduct with the quinone
is intriguing. Both C2 and C5 of the quinone ring can be electrophilic reaction sites
in Michael 1,6-addition reactions. However, C2 is expected to be less favored than
C5, because the alkyl group at C1 is an electron-donating substituent that causes
C2 to be less electropositive than C5. Also, the presence of the catecholamine side
chain causes steric hinderance for nucleophilic attack at C2. We have no explanation
for the regioselectivity of this addition reaction.
The reaction rates of imidazole with NADA and NBAD quinones are only
slightly faster than those of NAcH. The uv/vis kinetics showed that the second-
order rate constants for NADA quinone/imidazole, NBAD quinone/imidazole, and
2
2
2
1
1
NBAD quinone/NAcH systems are 9.0 ϫ 10Ϫ , 7.4 ϫ 10Ϫ , and 5.7 ϫ 10Ϫ MϪ sϪ ,
respectively (vide supra). If the equilibrium concentration of NAcH at pH 7 is
taken into account, then the second-order rate constant for NADA quinone/NAcH
2
1
1
system is 4.3 ϫ 10Ϫ MϪ sϪ (28). The absence of the side chain of imidazole may
account for its more rapid rate of reaction than NAcH. Based on these results, the
rate of nucleophilic addition of histidyl residues of protein to catecholamine qui-
nones would be expected to be slower than that of free NAcH, unless the protein
has a special conformation to facilitate the reaction of one relative to the other.
The formation of 6-NAcH–NADA is consistent with a previous study in which
NADA and NAcH were incubated with insect cuticle from late fifth instar larvae
of Hyalophora cecropia (19). However, 2-NAcH–NADA, which was detected in
our study as a minor product, was not identified in the cuticle-catalyzed coupling
study (19). This observation might be because the C2 adduct either was present as
one of the several unidentified products in the cuticle-catalyzed system or it was
not formed under enzymatic conditions. The C2 position of the aromatic ring of

REACTION OF QUINONES WITH IMIDAZOLE AND N-ACETYLHISTIDINE
201
the quinones is more sterically hindered than C6, which is consistent with peak
splitting in the NMR spectra of the 2-NAcH–NADA adduct. Moreover, in contrast
to the results that are reported here, a side-chain product was detected when NADA
and NAcH were incubated with insect cuticle (18). The absence of a side-chain
adduct in our study indicates that nonenzymatic isomerization of the quinone to
the quinone methide does not occur as rapidly as nucleophilic addition at ring
carbons of the quinone. To our knowledge no adducts from reactions of NBAD
quinone with amino acid derivatives have been previously characterized, although
it was suggested in an earlier study in which NBAD and NACySH were incubated
with insect cuticle that one of several unknown compounds resolved by chromatog-
raphy was due to such an adduct (5).
Two types of side reactions of NADA and NBAD quinones have been observed.
The first type involves formation of a minor product(s) that has a broad uv/vis
absorption in the range of 300 to 600 nm (Fig. 4). The second side reaction is the
reduction of quinones by an unknown reductant to regenerate NADA and NBAD,
as demonstrated by results of LC (Figs. 5 and 9). Both side reactions also are
observed for NADA or NBAD quinones in buffer solutions in the absence of
nucleophiles (unpublished data), but they are suppressed when other more reactive
nucleophiles are present. These results indicate that there is competition for the
quinones between nucleophilic addition reactions and side reactions. One possible
side reaction is addition of a component of the solvent/buffer system such as water
to the quinone to generate a catecholamine adduct that is more easily oxidized
than the original catecholamine. If this reaction were to occur, then unreacted
quinone would become involved in a redox reaction and reduced to the original
catecholamine.
In conclusion, this study has delineated the reactions of NADA and NBAD
quinones with the nitrogen-centered nucleophiles, imidazole and NAcH. Together
with results from the recent structural elucidation of four natural cross-links in the
insect exoskeleton (22), these data provide definitive evidence for the participation
of histidyl residues of cuticular protein in the sclerotization of the insect exoskeleton
and are consistent with the previously obtained solid-state NMR results that indi-
cated the presence of C–N linkages in cuticle but did not reveal the exact structures
(14). The reactions of NADA with NAcH in the presence of cuticle were previously
found to yield C6 and C7 adducts (18, 20). However, we have shown that nucleophilic
additions at C2 of the catecholamine quinones by NAcH also occur. No formation
of side-chain adducts occurs in our model systems, indicating that spontaneous
quinone isomerization is not favored in the absence of an appropriate enzymatic
system. The kinetics for the reactions of NADA and NBAD quinone with the two
nucleophiles were obtained for the first time, which provide information about the
rates of reactions between sclerotizing agents and histidyl groups in cuticular pro-
teins.
ACKNOWLEDGMENTS
The authors extend thanks to Dr. Kenneth Tomer, Dr. Todd Williams, and Mr. Gary Radke for their
assistance in mass spectrometry and Dr. Om Prakash for his assistance in NMR spectroscopy. We also
thank Dr. T. L. Hopkins, Dr. Maryanne Collinson, and Mr. T. Morgan for assistance with this study

202
HUANG ET AL.
and for reviewing the manuscript. This research was supported in part by National Science Foundation
Grants DCB-9019400, MCB-9418129, and CHE-9216101.
REFERENCES
1. Hopkins, T. L., and Kramer, K. J. (1992) Annu. Rev. Entomol. 37, 273–302.
2. Andersen, S. O., Peter, M. G., and Roepstorff, P. (1996) Comp. Biochem. Physiol. 113B, 689–705.
3. Pryor, M. G. M. (1940a) Proc. R. Soc. London Ser. B 128, 378–393.
4. Pryor, M. G. M. (1940b) Proc. R. Soc. London Ser. B 128, 393–407.
5. Sugumaran, M., Saul, S. J., and Dali, H. (1990) Arch. Insect Biochem. Physiol. 15, 255–269.
6. Sugumaran, M. (1988) Adv. Insect Physiol. 21, 179–231.
7. Ricketts, D., and Sugumaran, M. (1994). J. Biol. Chem. 269, 22217–22221.
8. Hackman, R. H., and Goldberg, M. (1977) Insect Biochem. 7, 175–184.
9. Karlson, P, and Sekeris, C. E. (1962) Nature 195, 183–184.
10. Andersen, S. O. (1971) Insect Biochem. 1, 157–170.
11. Hopkins, T. L., Morgan, T. D., Aso, Y., and Kramer, K. J. (1982) Science 217, 364–366.
12. Hopkins, T. L., Morgan, T. D., and Kramer, K. J. (1984) Insect Biochem. 14, 533–540.
13. Sugumaran, M., and Lipke, H. (1982) Proc. Natl. Acad. Sci. USA 79, 2480–2484.
14. Schaefer, J., Kramer, K. J., Garbow, J. R., Jacob, G. S., Stejskal, E. O., Hopkins, T. L., and Speirs,
R. D. (1987) Science 235, 1200–1204.
15. Christensen, A. M., Schaefer, J., Kramer, K. J., Morgan, T. D., and Hopkins, T. L. (1991a) J. Am.
Chem. Soc. 113, 6799–6802.
16. Christensen, A. M., Schaefer, J., and Kramer, K. J. (1991b). Magn. Reson. Chem. 29, 418–421.
17. Kramer, K. J., and Hopkins, T. L. (1995) Insect Biochem. Mol. Biol. 25, 1067–1080.
18. Andersen, S. O., Jacobsen, J. P., Roepstorff, P., and Peter, M. G. (1991) Tetrahedron Lett. 32, 4287–
4290.
19. Andersen, S. O., Peter, M. G., and Roepstorff, P. (1992a) Insect Biochem. Mol. Biol. 22, 459–469.
20. Andersen, S. O., Jacobsen, J. P., and Roepstorff, P. (1992b) Insect Biochem. Mol. Biol. 22, 517–527.
21. Wybrandt, G. B., and Andersen, S. O. (1994) Biochim. Biophys. Acta 1201, 15–18.
22. Xu, R., Huang, X., Hopkins, T. L., Kramer, K. J. (1997) Insect Biochem. Mol. Biol. 27, 101–108.
23. Andersen, S. O. (1980) in Cuticle Techniques in Arthropods (Miller, T. A., Ed.), pp. 200–201,
Springer-Verlag, New York.
24. Kramer, K. J., Nuntnarumit, C., Aso, Y., Hawley, M. D., and Hopkins, T. L. (1983) Insect Biochem.
13, 475–479.
25. Yamasaki, N., Aso, Y., and Tsukamoto, T. (1990) Agric. Biol. Chem. 54, 833.
26. Van Swaay, M. (1978) J. Chem. Ed. 55, A7–A12.
27. Huang, X. (1996) Ph.D. Dissertation, Kansas State University, Manhattan.
28. Xu, R., Huang, X., Morgan, T. D., Prakash, O., Kramer, K. J., and Hawley, M. D. (1996) Arch.
Biochem. Biophys. 329, 56–64.
29. Xu, R., Huang, X., Kramer, K. J., and Hawley, M. D. (1995) Anal. Biochem. 231, 72–81.