ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
REFERENCES
(Received for publication, August 21, 1995)
From the
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
REFERENCES
Bacterial luciferase catalyzes the emission of visible light
from the reaction of reduced flavin, molecular oxygen, and an n-alkyl aldehyde. The mechanism of the reaction was probed by
measuring the electronic effects of various substituents at the
8-position of the flavin ring system. Substituent effects were obtained
for CH
-, Cl-, CHO-, CHS-, F-, and
H- on the rate of formation and decay of the hydroperoxyflavin
intermediate and the time courses for the emission of visible light.
The rate constant for the decay of light emission increases for the
series Cl < F < H < CHS < CH <
CHO. These results are not compatible with a standard
Baeyer-Villiger type mechanism for the chemical transformation, but
they are consistent with a decrease in the electron density at the
reaction center of the flavin moiety during the rate-limiting step of
the reaction.
Bacterial luciferase catalyzes the reaction of FMNH
, (
)O, and a long chain aliphatic aldehyde to
yield riboflavin-5`-monophosphate (FMN), the carboxylic acid, and
blue-green light (Baldwin and Ziegler, 1992). The mechanism proceeds
through the formation of a stable C4a-hydroperoxyflavin intermediate by
the reaction of FMNH with molecular oxygen, presumably by
way of a one-electron transfer to yield a radical pair consisting of a
flavin radical and superoxide and subsequent recombination to form the
C4a-hydroperoxyflavin intermediate (Bruice, 1984). This intermediate
has been isolated using low temperature chromatography (Hastings et
al., 1973) and characterized by C-NMR spectroscopy
(Vervoort et al., 1986). In the absence of aldehyde, this
intermediate slowly decomposes to FMN and H
O.
In the presence of aldehyde, the C4a-hydroperoxyflavin intermediate is
presumed to react with the aldehyde to form a C4a-hydroperoxyhemiacetal
intermediate. The formation of the hydroperoxyhemiacetal intermediate
from FMNH, O and aldehyde (RCHO) is shown in Fig. SI. This intermediate subsequently rearranges to form a
4a-hydroxyflavin intermediate and the carboxylic acid. Finally, the
C4a-hydroxyflavin intermediate dehydrates to form FMN and
HO.
Figure SI: Scheme I.
This enzyme system has been studied for over 30
years, and although most of the intermediates involved in the reaction
have been isolated and identified, the mechanism by which the excited
state is formed is still a subject of much debate. At least seven
mechanisms have been proposed for the light-emitting reaction, but some
of these can now be eliminated because they are contrary to
experimental observations. From a mechanistic point of view, one very
important result was obtained by Suzuki et al.(1983).
Utilizing 
O, they demonstrated the
incorporation of one atom of labeled oxygen into the carboxylic acid
product. The three most likely mechanisms that are consistent with this
result are the Baeyer-Villiger mechanism, the dioxirane mechanism and
the chemically initiated electron exchange luminescence (CIEEL)
mechanism. These three mechanisms all assume the formation of a
C4a-hydroperoxyhemiacetal intermediate, but they describe different
ways for the rearrangement of this complex and formation of the excited
state. The first chemical mechanism for the luciferase reaction was
proposed by Eberhard and Hastings(1972) and consists of a
Baeyer-Villiger rearrangement of the C4a-hydroperoxy-hemiacetal
intermediate with migration of hydrogen to yield the carboxylic acid
and the luciferase-bound flavin in the singlet excited state. This
mechanism was later modified to include the formation of the
C4a-hydroxyflavin intermediate as the excited state, since it is
currently believed that this latter species is the primary emitter in
the luciferase reaction (Kurfurst et al., 1984). The
Baeyer-Villiger mechanism is shown in Fig. SII. Cyclohexanone
monooxygenase, an enzyme that catalyzes the oxidation of a variety of
substrates including aldehydes, has been shown to proceed through a
Baeyer-Villiger mechanism (Schwab et al., 1983; Branchaud and
Walsh, 1985). In this enzyme, a flavin C4a-hydroperoxide intermediate
is also the active oxygen transfer agent.
Figure SII: Scheme II.
The dioxirane mechanism provides an alternative to the Baeyer-Villiger mechanism. This mechanism involves the decomposition of the C4a-hydroperoxyhemiacetal intermediate to form a dioxirane and the C4a-hydroxyflavin (Raushel and Baldwin, 1989). The decay of the dioxirane by one of several mechanisms could lead to the formation of the light-emitting species. The dioxirane mechanism is depicted in Fig. SIII. There are at least two potentially chemiluminescent pathways for the breakdown of the dioxirane, both sufficiently energetic to populate the first singlet excited state of the flavin. In one of these pathways, the dioxirane could undergo an homolytic cleavage of the oxygen-oxygen bond to yield a biradical, which could subsequently rearrange to form the carboxylic acid in either the triplet or singlet state. The singlet or triplet state of the carboxylic acid could interact with and excite the C4a-hydroxyflavin intermediate prior to light emission.
The CIEEL mechanism describes the decomposition of the C4a-hydroperoxyhemiacetal intermediate via a series of one-electron transfer reactions. Schuster(1979) has demonstrated that a CIEEL mechanism can account for certain chemiluminescent and bioluminescent reactions. In this process, a caged radical anion/cation pair is generated, and charge annihilation then populates the excited singlet state of the activator. For the reaction catalyzed by luciferase, it is possible to describe the production of either an excited flavin hydroxide or an excited acid or acylium cation. The CIEEL mechanism for the luciferase reaction proposed by Eckstein et al.(1993) is shown in Fig. SIV. A CIEEL mechanism can also be proposed for the decomposition of the dioxirane species. In such a mechanism, the dioxirane would receive an electron from the flavin donor to form a radical ion pair. The radical anion of the dioxirane could then yield the radical anion of the carboxylic acid. Back transfer of the electron to the donor would then yield the excited state of the fluorophore.
Figure SIV: Scheme IV.
It has been shown that 8-substituted flavin derivatives are substrates for the luciferase reaction (Watanabe et al., 1978). The structure of the FMN derivatives is shown in Fig. Z1. The position at C-8 is believed to be exposed to the solvent (Schopfer et al., 1981; Chen and Baldwin, 1984) so that modifications at this position should not alter the protein-flavin interactions to any significant extent. It has also been shown that substitutions at this position alter significantly the electronic properties (i.e. redox potential) of the flavin. Previous studies directed at determining the effect of these flavin analogs have been limited to simple measurements of bioluminescence intensity and rate of decay of light emission. It is clear that these substitutions should also affect the rates by which the various flavin intermediates react with oxygen and aldehyde substrates. For example, the nucleophilicity of the C4a-hydroperoxyflavin anion should vary with the different substitutions, and as a consequence, the rate of reaction of this intermediate with the aldehyde substrate and the rate of decomposition of the formed C4a-flavin peroxyhemiacetal intermediate should differ from one flavin analog to the other. In this investigation the kinetics of the luciferase-catalyzed reaction with various 8-substituted flavin derivatives have been studied in detail. The rates of formation and decay of the reaction intermediates have been measured. In addition the rate of formation of visible light and the effect of these substitutions on the individual steps of the reaction mechanism have been determined.
Figure Z1: Structure I.
)The
riboflavin derivatives were phosphorylated according to the method of
Scola-Nagelschneider and Hemmerich(1976). The 8-H-FMNH derivative was a generous gift from Professor Franz
Müller (Sandoz Agro, Ltd.). All the FMN analogs
were further purified by column chromatography. All purified flavins
moved as single spots on thin-layer chromatography and were
characterized by H,
O-FMN (Fritz et al., 1987), 28,500 MS-FMN (Moore et al., 1979), and 12,500 MThe anaerobic enzyme
solutions were prepared using an all glass vacuum system (Williams et al., 1979) by several cycles of evacuation and
equilibration with nitrogen gas. Purification of the nitrogen gas was
performed by passing the gas over a heated column of BASF catalyst
R3-11 (Chemical Dynamics Corp./Kontes Glass Co.). The anaerobic
luciferase-FMNH solutions were made by mixing the anaerobic
enzyme solutions with reduced flavin under a nitrogen atmosphere, and
the mixture was then transferred to the stopped-flow instrument using
an air-tight Hamilton syringe. The kinetic experiments were carried out
using a stopped-flow apparatus from Hi-Tech Ltd. (model SF-51)
connected to an HP-300 series computer. A glass cut-off filter (type GU
380) was used for absorbance measurements at 380 nm to avoid
interference by the bioluminescence during the course of the reaction.
The stopped-flow instrument was equipped with a rapid scanning device
(MG-3000) designed to collect a complete spectrum (200-800 nm)
every 90 ms. The stopped-flow experiments were carried out under the
same conditions as described previously (Abu-Soud et al.,
1992, 1993). A stock solution of 0.1 M decanal was freshly
prepared before use. The time courses for the various kinetic
experiments were fit to one or more of the following rate equations
using a nonlinear least-square procedure contained in the software
supplied by Hi-Tech Ltd.
represents the sum of two independent exponentials
for a parallel process. and describe the time
courses for a sequential process (X Y
Z) monitoring the formation of Y and Z,
respectively. In these equations, k
and k are the first-order rate constants, t is time, A and B are amplitude factors, and e is 2.718. Ten individual traces were collected and averaged
to improve the signal-to-noise ratio.
All of the kinetic data collected from the stopped-flow studies were transferred to a Silicon Graphics workstation and subsequently analyzed with extensively modified forms of the KINSIM (Barshop et al., 1983) and FITSIM (Zimmerle and Frieden, 1989) programs, using the comprehensive kinetic model that appears in Fig. SV. The microscopic rate constants that appear in these kinetic models were estimated by comparison of the experimental time courses for product formation with the calculated time courses derived by numerical integration of the appropriate differential equations with the KINSIM program. The rate constants were first estimated graphically until the simulated time courses matched the experimental data as closely as possible. The final values were then adjusted, and the error limits were obtained using the automated FITSIM routine that calculates the best values by minimization of the difference between the experimental and simulated data using an iterative nonlinear least squares procedure. The standard error for each individual rate constant has been estimated to be less than 15% using the FITSIM program.
Figure SV: Scheme V.
with O. Fig. 1A shows the absorption spectra for the reaction of
8-Cl-FMNH with O. Spectrum I corresponds to that of the reduced form, and spectrum F
corresponds to the fully oxidized form. Similar spectra were collected
for all of the 8-substituted flavin derivatives studied.
Single-wavelength experiments were done to measure the kinetics of the
nonenzymatic decomposition of the 8-substituted FMNH derivatives. The wavelengths at which the absorbance changes were
measured were the following: 380 and 445 nm for FMNH, 365
and 445 nm for 8-Cl-FMNH, 370 and 472 nm for
8-CHS-FMNH, 330 and 445 nm for
8-CHO-FMNH, 350 and 448 nm for
8-H-FMNH, and 356 and 435 nm for 8-F-FMNH. In
general, these wavelengths correspond to the absorbance maxima of the
corresponding 8-X-FMN. The time courses for the appearance of
8-X-FMN measured at both wavelengths were very similar, and
each had a distinctive lag followed by a single exponential. Fig. 2shows the time course for the reaction of 15 µM 8-Cl-FMNH with 120 µM O measured at 365 and 445 nm. The time courses for this reaction
were fit to with values for k and k of 3.6 s derivatives were also fit to with values for k and k of 4.6 s, 3.8
sS-FMNH, 3.4 sO-FMNH, 5.0
s, and 2.2 s.
Figure 1:
Rapid scanning absorption spectra for
the autooxidation of 8-Cl-FMNH and for the formation and
decomposition of the E
8-Cl-FMNOOH intermediate. A, absorption spectra following the reaction of
8-Cl-FMNH (30 µM) with air-equilibrated buffer
(120 µM O) at 25 °C. Spectrum I was taken 7 ms after mixing. Additional spectra were taken at
90-ms intervals for 900 ms. Not all spectra are shown. B,
absorption spectra following the reaction of luciferase (75
µM) and 8-Cl-FMNH (15 µM) with
air-equilibrated buffer (120 µM O) at 25
°C. Spectrum I was taken 23 ms after mixing. Additional
spectra were taken at 0.9-s intervals for 8.5 s. Not all spectra are
shown.
Figure 2:
Time
courses for the reoxidation of 8-Cl-FMNH. The reaction was
initiated by mixing 8-Cl-FMNH (15 µM) with
O (120 µM), and the reaction was monitored at
365 () and 445 nm (
). The solid lines represent the
experimental data, and the symbols represent the fit of the
data to with values of k and k of 3.6 and 7.5 s,
respectively.
was rapidly mixed
with oxygen is shown in Fig. 1B. The formation of the
8-Cl-FMNOOH intermediate was essentially complete prior to the
collection of the first spectrum (spectrum I). This
intermediate has an absorption maximum at approximately 365 nm. This
also corresponds to an isosbestic point for the hydroperoxyflavin
intermediate and oxidized flavin. The subsequent change in absorbance
at 445 nm was due to the decomposition of this intermediate to 8-Cl-FMN
and HO. The time course for the formation of
the E8-CHO-FMNOOH complex is presented in Fig. 3. This reaction was initiated by the mixing of the E8-CHO-FMNH complex with
O. The final enzyme and flavin concentrations were 75 and
15 µM, respectively. At a final O concentration of 120 µM, the pseudo-first-order rate
constant, obtained from a fit of the data to , is 190
s8-CHO-FMNOOH is directly
proportional to the concentration of O. A fit of these data
to gives a second-order rate constant of 1.5 
10
M 10M, 8.8 10
MS-FMNH, 5.6 10M, and 1.0 10M. It was not possible to measure the rate for the
8-F-FMNH derivative.
Figure 3:
Time course for the formation of the E8-CHO-FMNOOH intermediate. The reaction was
monitored at 365 nm after mixing a solution of luciferase (75
µM) and 8-CHO-FMNH (15
µM) with O (120 µM). The
experimental data (solid line) were fit to with a
first-order rate constant of 190 s as indicated by
the filled circles. The inset shows the effect of
O
concentration on the pseudo-first-order rate constant for
the formation of E8-CHO-FMNOOH. A
second-order rate constant of 1.5 10M s
was obtained from a
fit of these data to .
Figure 5:
Time courses for the bioluminescence with
8-CHO-FMNH. The light emission was measured
after mixing a solution of luciferase (75 µM) and
8-CHO-FMNH (15 µM) with various
amounts of n-decanal (15 (
), 30 (), 60 (
),
80 (
), 100 (
), and 400 () µM) in
air-equilibrated buffer. The symbols represent portions of the
experimental data, while the solid lines represent the
simulated time courses using the rate constants that appear in Table 1and the model in Fig. SV.
In the absence of aldehyde, the
decomposition of the 8-substituted FMNOOH intermediates to
8-X-FMN and HO was studied. The time
course for the absorbance change at 445 nm after mixing E8-CHO-FMNH with O is shown in Fig. 4. The data were fit to with
a first-order rate constant of 12.4 s. No
significant difference in the decomposition rates was observed upon
variation of O
concentration. The first-order rate constant
for the decomposition of the hydroperoxyflavin intermediates was 0.10
s for FMNH
, 0.07 s for 8-Cl-FMNH
, 0.14 s for
8-CH
S-FMNH, 0.2 s for
8-H-FMNH
, and 0.85 s for
8-F-FMNH
.
Figure 4:
Time course for the decomposition of E8-CHO-FMNOOH to 8-CHO-FMN and
HO when monitored at 445 nm. The reaction was
initiated by mixing a solution of luciferase (75 µM) and
8-CHO-FMNH (15 µM) with
air-equilibrated buffer (120 µM O). The
experimental data (solid line) were fit to with a
first-order rate constant of 12 s as indicated by
the filled circles.
8-CHO-FMNH complex was
mixed with increasing amounts of n-decanal in air-equilibrated
buffer are shown in Fig. 5. The rate of light emission reaches a
maximum in approximately 0.1 s and then decays exponentially over a
period of about 0.5 s. Similar experiments were performed with the
other 8-substituted flavin analogs. The relative maximum light
intensity for the various 8-substituted flavin derivatives at a given
concentration of enzyme, aldehyde, and flavin was not determined. The
main difference between the various analogs is the rate with which the
emission of light decays. For all the flavin analogs, the light decay
can be accurately fit to with a rate constant of 4.4
sO-FMN, 0.50 sS-FMN,
0.025 s
for the
8-substituted flavin derivatives were taken from Eckstein et
al.(1993). These dissociation constants were obtained by the
method of Eadie-Hoftstee using the maximal light intensity in standard
assays at different reduced flavin concentrations. The second-order
rate constants (k) for the formation of E8-X-FMNOOH from E8-X-FMNH and O were obtained from the time courses that
followed the mixing of solutions of luciferase and 8-substituted
FMNH derivatives with various concentrations of
O. In all cases, the variation of the pseudo-first-order
rate constant with the oxygen concentration was linear, and the plots
intersected the axes near the origin. The rate constants for the
processes associated with the binding of aldehyde to the E8-X-FMNOOH complexes through the chemical
formation of E8-X-FMNOH (k-k
) were established from
the time courses of the light emission that followed the mixing of
air-equilibrated decanal with mixtures of enzyme and
8-X-FMNH. The rate constants for the decomposition
of the E8-X-FMNOOH complexes to FMN and
HO were obtained by monitoring the formation of
8-X-FMN at the corresponding absorption maxima in the absence
of decanal. The time course for this decay could be fit to a
single-exponential (). The rate constants for the
dehydration of E8-X-FMNOH and the rate
constants associated with aldehyde binding were only determined for
FMNH as described by Abu-Soud et al.(1992). The
rate constants associated with the binding of aldehyde to free enzyme (k to
8-X-FMN and HO in the presence of
O (k (15 µM) with O (120 µM) to .Substitution at the 8-position of the isoalloxazine ring of the flavin mononucleotide markedly alters the kinetics of the bacterial luciferase reaction. A linear free-energy relationship analysis can therefore be conducted to determine the effect of these flavin substitutions on the individual steps of the reaction mechanism. However, not all of the calculated rate constants listed in Table 1show linear plots when analyzed by the Hammett equation(6) .
The Hammett plot for the calculated rate constants for the
formation of the 4a-hydroperoxyflavin intermediate and for
decomposition of this intermediate is shown in Fig. 6. The rate
constant for the formation of the 4a-hydroperoxyflavin intermediate
exhibits a good fit to the Hammett equation (r = 0.90) with a value of of -1.5 (panel
A). On the other hand, the rate of decomposition of this
intermediate shows a very poor fit (r
=
0.15; panel B). This latter result might indicate the more
complex nature of this process. The substituent constants used for
these plots are the classic 
; no special substituent
constants were used. It is reasonably well established that substituent
effects in biological systems can also reflect a combination of
electronic, steric, and hydrophobic contributions to the observed rate
constants. The failure to obtain a good correlation with these rate
constants, based only on electronic contributions, might therefore be a
reflection of further contributions from steric and hydrophobic
factors. In some cases it is possible to assess the importance of these
other contributions by performing a multiple regression analysis
(Shorter, 1978). However, it was not possible to perform this type of
analysis with the data obtained in this study, since a three-parameter
correlation (electronic, steric, and hydrophobic) would not be
statistically valid with the five sets of rate constants determined
here for the limited number of available flavin analogs.
Figure 6:
Hammett plots for the rates of formation
and decomposition of the 4a-hydroperoxyflavin intermediate. Correlation
of the second-order rate constant for the formation (panel A)
and first-order rate constant for the decomposition (panel B)
of the 4a-hydroperoxyflavin intermediate with the electronic
substituent constant .
The Hammett
equation could also be applied with success to the rate of decay of
light emission. The semilog plot of k/k
versus is shown in Fig. 7. The
slope of this plot, , has a value of -4. A negative value
for this slope is a direct indication that the reaction in question is
facilitated by increasing the electron density at the reaction center.
The value obtained here is a very large value for
, typical of
processes involving ionic transition states. Baeyer-Villiger reactions
of benzaldehydes with substituted perbenzoic acids under conditions
where a hydrogen is the migrating group show small positive values of
(0.2-0.6) (Ogata and Sawaki, 1972). The results obtained
here are therefore inconsistent with a Baeyer-Villiger type mechanism
for the bioluminesence reaction catalyzed by the bacterial luciferase.
Figure 7:
Correlation of the rate of decay of
bioluminescence with the electronic substituent constant
.
The results obtained from the linear free-energy relationship
analysis described above are, however, in accordance with the formation
of a C4a-hydroperoxyflavin radical cation intermediate, as suggested by
the CIEEL mechanism (Fig. SIV). The CIEEL mechanism also
predicts that the oxidation potential of the flavin should affect the
rate of decay for the bioluminescence reaction. Semilog plots of the
light emission decay rate against the two-electron redox potential of
the FMN/1,5-FMNH couples and the one-electron redox
potential for the oxidation of 4a,5-dihydroflavin model compounds to
the radical cation as determined by Eckstein et al.(1993) are
shown in Fig. 8, A and B. There is a linear
relationship between the two redox potential values and the light
emission decay rate. The decay of light emission is fastest with FMN
analogs of the lowest oxidation potential. This result is an indication
that in the rate-limiting step there is a decrease in the electron
density at the reaction center of the flavin moiety. Substituents that
can donate electron density favor this process. This is in agreement
with the results obtained with the Hammett analysis.
Figure 8:
Correlation of the rates of decay of light
emission with the flavin redox potential. The redox potentials used are
of the 8-substituted FMN/FMNH couples (panel A)
and the one-electron oxidation potentials of the
8-substituted-4a,5-dihydroflavin analogs determined by Eckstein et
al.(1993) (panel B).
It is not possible with the information gained through these studies to differentiate between the CIEEL mechanism depicted in Fig. SIVand that involving the dioxirane species. It has been recently shown that dioxiranes can participate in one-electron transfer reactions. Nelsen et al.(1993) have demonstrated that unsaturated hydrazines react with dimethyldioxirane to form N-methylhydrazinium acetate. The mechanism of the reaction involves a single-electron transfer from the hydrazine to the dioxirane to initially form an ion pair; methyl transfer then generates the final product.
A similar mechanism can be drawn for the luciferase reaction and is shown in Fig. SVI. In this mechanism, the 4a-hydroperoxyhemiacetal intermediate decomposes to form a dioxirane and the 4a-hydroxyflavin intermediate. This process is identical to that of the oxidation of ketones by monopersulfuric acid. Dioxiranes have been successfully isolated and characterized in nonbiological model systems. In these oxidations, it is important to notice that the reaction proceeding through a Baeyer-Villiger type mechanism competes with the reaction pathway forming the dioxirane species, which prevails at neutral pH. After formation of the dioxirane and the 4a-hydroxyflavin intermediate, electron transfer from the flavin to the dioxirane produces an ion pair, consisting of the 4a-hydroxyflavin radical cation and the dioxirane radical anion. The production of this flavin hydroxide radical cation should be more favorable with substituents that can donate electron density to the newly formed positively charged center. This is in agreement with the results of the Hammett analysis. The formation of the radical pair would be the slow step in the reaction mechanism. In the next step, rearrangement of the dioxirane radical anion with transfer of the hydrogen atom to the N5 of the flavin would then result in the production of the excited state of the flavin hydroxide and the carboxylic acid. The excited hydroxyflavin then decays to ground state with the production of a photon of light. The flavin hydroxide then dehydrates and deprotonates to yield the final FMN product.
Figure SVI: Scheme VI.
This mechanism is also in agreement with the
results obtained for the deuterium kinetic isotope effect experiments.
The deuterium isotope effect for light formation and decay of light
emission is approximately 1.6. ()This value is somewhat
larger than expected for a secondary isotope effect but might indicate
that the breakage of the carbon-hydrogen bond does not happen in
the rate-determining step. In the mechanism depicted here, the
rate-determining step is the formation of the radical pair, but actual
breakage of the carbon-hydrogen bond is involved in the formation
of the excited state intermediates.
Although the mechanism presented in Fig. SVIprovides a reasonable pathway for the decomposition of the C4a-hydroperoxyhemiacetal intermediate and the generation of an excited state, there are potentially two problems with the dioxirane mechanism. One is that dioxiranes of aldehydes have not yet been identified or isolated, and second, attempts to generate light in the presence of oxidizable electron donor activators have not been successful. On the other hand, it has been established that the rearrangement of a dioxirane into a carboxylic acid is a highly exothermic process. The energy difference between a dioxirane and the corresponding carboxylic acid is approximately 80 kcal/mol (Adam et al., 1989). A CIEEL type mechanism for the generation of light from the decomposition of dioxiranes has been proposed (Adam et al., 1989), but it awaits experimental support.
) and
FMN, reduced and oxidized riboflavin 5`-phosphate; FMNOOH, flavin
C4a-hydroperoxide; FMNOOR, flavin C4a-hydroperoxyhemiacetal; FMNOH,
flavin C4a-hydroxide; RCHO, n-alkylaldehyde; bis-Tris,
2-[bis-(hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol;
CIEEL, chemically initiated electron exchange luminescence.))
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | Biochemistry and Molecular Biology Education |
