Kinetics of inhibition of firefly luciferase by dehydroluciferyl-coenzyme A, dehydroluciferin and l-luciferin

Article abstract of DOI:10.1039/c0pp00379d

The inhibition mechanisms of the firefly luciferase (Luc) by three of the most important inhibitors of the reactions catalysed by Luc, dehydroluciferyl-coenzyme A (L-CoA), dehydroluciferin (L) and l-luciferin (l-LH₂) were investigated. Light production in the presence and absence of these inhibitors (0.5 to 2 μM) has been measured in 50 mM Hepes buffer (pH = 7.5), 10 nM Luc, 250 μM ATP and d-luciferin (d-LH₂, from 3.75 up to 120 μM). Nonlinear regression analysis with the appropriate kinetic models (Henri-Michaelis-Menten and William-Morrison equations) reveals that L-CoA is a non-competitive inhibitor of Luc (K_i = 0.88 ± 0.03 μM), L is a tight-binding uncompetitive inhibitor (K_i = 0.00490 ± 0.00009 μM) and l-LH₂ acts as a mixed-type non-competitive-uncompetitive inhibitor (K_i = 0.68 ± 0.14 μM and αK_i = 0.34 ± 0.16 μM). The K_m values obtained for L-CoA, L and l-LH₂ were 16.1 ± 1.0, 16.6 ± 2.3 and 14.4 ± 0.96 μM, respectively. L and l-LH₂ are strong inhibitors of Luc, which may indicate an important role for these compounds in Luc characteristic flash profile. L-CoA K_i supports the conclusion that CoA can stimulate the light emission reaction by provoking the formation of a weaker inhibitor.

Full text of DOI:10.1039/c0pp00379d

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Photochemical &
Photobiological Sciences
Cite this: Photochem. Photobiol. Sci., 2011, 10, 1039
www.rsc.org/pps
PAPER
Kinetics of inhibition of firefly luciferase by dehydroluciferyl-coenzyme A,
dehydroluciferin and L-luciferin†
Lu´ıs Pinto da Silva and Joaquim C. G. Esteves da Silva*
Received 16th December 2010, Accepted 19th February 2011
DOI: 10.1039/c0pp00379d
The inhibition mechanisms of the firefly luciferase (Luc) by three of the most important inhibitors of
the reactions catalysed by Luc, dehydroluciferyl-coenzyme A (L-CoA), dehydroluciferin (L) and
L-luciferin (L-LH2) were investigated. Light production in the presence and absence of these inhibitors
(0.5 to 2 mM) has been measured in 50 mM Hepes buffer (pH = 7.5), 10 nM Luc, 250 mM ATP and
D-luciferin (D-LH2, from 3.75 up to 120 mM). Nonlinear regression analysis with the appropriate kinetic
models (Henri–Michaelis–Menten and William–Morrison equations) reveals that L-CoA is a
non-competitive inhibitor of Luc (Ki = 0.88 0.03 mM), L is a tight-binding uncompetitive inhibitor
(Ki = 0.00490 0.00009 mM) and L-LH2 acts as a mixed-type non-competitive-uncompetitive inhibitor
(Ki = 0.68 0.14 mM and aKi = 0.34 0.16 mM). The Km values obtained for L-CoA, L and L-LH2 were
16.1 1.0, 16.6 2.3 and 14.4 0.96 mM, respectively. L and L-LH2 are strong inhibitors of Luc, which
may indicate an important role for these compounds in Luc characteristic flash profile. L-CoA Ki
supports the conclusion that CoA can stimulate the light emission reaction by provoking the formation
of a weaker inhibitor.
imunoassays, bioimaging, biosensing and is used as a gene
Introduction
reporter.1,2,6–12
Firefly luciferase (EC 1.13.12.7) is an enzyme that catalyzes the
oxidation of firefly luciferin (LH2), giving rise to light in a two-
step reaction:1,2 the first step involves the formation, from D-LH2
(1)
Luc + LH2 + ATP ꢀ Luc·LH2–AMP + PPi
and adenosine-5¢-triphosphate (ATP), of an adenylyl intermediate
Luc·LH2–AMP + O2 → Luc + AMP + CO2 + oxyluciferin +
(D-LH2-AMP) (eqn (1)). The second step consists on the oxidation
of D-LH2-AMP, and the release of adenosine-5¢-monophosphate
(AMP), CO2 and oxyluciferin (OxyLH2) (eqn (2)). The light
emitter is formed in an excited singlet state S1, decaying to the
ground state with the emission of visible light (550–570 nm).
This system is known for its efficiency when compared with
chemiluminescence. For many years the efficiency of this reaction
was thought to be of 88%,3,4 but a recent work performed by Niwa
et al.5 determined the quantum yield of several beetle luciferase
to be 45–61%. Despite the sharp decrease, these new values still
strongly support the study and the development of practical
applications for this bioluminescence system. Currently, it has
gained numerous bioanalytical, biomedical and pharmaceutical
applications, among others. More specifically, it is involved in
the analytical determination of ATP, in microbial detection,
(2)
photon
Due to this broad range of applications, it is crucial to have
a deep knowledge of the inhibitors of Luc and their inhibition
parameters. These compounds may have a non-negligible effect on
the bioluminescence reaction and can lead to erroneous results in
Luc-catalyzed applications, as for example in reporter-gene assays.
The current knowledge of Luc inhibition was recently reviewed by
one of the present authors.2
The in vitro emission of light follows, at relatively high substrate
concentration, a flash pattern, which starts with an initial flash
that quickly decays to a low basal level. This characteristic
of light emission is caused by an accumulation of inhibitory
products.1,2,6 Two of the most well characterized inhibitors of the
bioluminescence reaction are OxyLH2 (Ki = 0.50 0.03 mM), the
reaction product, and dehydroluciferyl-adenylate (L-AMP).2,13 L-
AMP results from the oxidation of luciferyl-adenylate (LH2-AMP)
in a dark reaction also catalyzed by Luc, and acts as fast tight-
binding competitive inhibitor (Ki = 3.8 0.7 nM) being responsible
for the typical flash profile.13 Besides its role as light-production
inhibitor, this LH2-AMP derivative can be used by Luc as a
substrate, in another dark reactions:
Centro de Investigac¸a˜o em Qu´ımica (UP), Departamento de Qu´ımica e
Bioqu´ımica, Faculdade de Cieˆncias da Universidade do Porto, R. Campo
Alegre 687, 4169-007, Porto, Portugal. E-mail: jcsilva@fc.up.pt; Fax: +351-
220402659; Tel: +351-220402569
† Electronic supplementary information (ESI) available: Experimental
results obtained for the chemical characterization of L-CoA, L-LH2 and
L. See DOI: 10.1039/c0pp00379d
(3)
Luc·L–AMP + CoA ꢀ Luc + L–CoA + AMP
This journal is
The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 1039–1045 | 1039
©

View Online
evaporated under a stream of N2. The addition of a mixture of CoA
(81 mg) and dimethyl sulfoxide (2 ¥ 3 mL) to the reaction mixture,
and posterior contact with atmospheric oxygen resulted on the
formation of L-CoA. The initial objective for this synthesis was
the obtaining of D-LH2-CoA, but contact with oxygen lead to the
formation of this oxidized derivative. Its purity was assessed by
LC-MS and RP-HPLC, and the synthesised compounds have the
following percentage of purity: L-CoA 88.4%, L 95.0% and L-LH2
90.5% (ESI†). Taking into consideration the percentage of purity
of the synthesised L-CoA, L and L-LH2, calculated as indicated
in the ESI, they were weighed accurately and dissolved in water
and Hepes 0.5 M (pH 7.5), respectively, to a final concentration of
250 mM, 167 mM and 208 mM respectively.
(4)
Luc·L–AMP + PPi → Luc + L + ATP
The reaction of L-AMP with coenzyme A (CoA) and inor-
ganic pyrophosphate (PPi) has an antagonizing effect on light
inhibition, as it results in dehydroluciferyl-coenzyme A (L-CoA)
and dehydroluciferin (L), which experimental evidence indicates
that these compounds are weaker Luc inhibitors than L-AMP.14–16
As L results from the oxidation of LH2 and ATP, through the
formation of L-AMP and its posterior pyrophosphorolysis,15,16 it
was initially considered as the light emitter, instead of OxyLH2.17
Its chemical synthesis18 permitted the clarification of L role in
the bioluminescence reaction, and its incubation with Luc and
ATP-Mg2+ characterized it as a Luc inhibitor.14 Some groups19–22
hypothesized a competitive mechanism of inhibition for L with
respect to D-LH2 with a Ki of about 0.10–1.20 mM.23 However
there is still no evidence to support any inhibition mechanism for
L.
CoA does not participate in the “classic” bioluminescence
reaction, but due to its light-production stimulating effect14 is
now added to Luc commercial assays. It is thought that this
effect is achieved by the conversion of L-AMP into L-CoA, a
supposed much weaker inhibitor. However, no study to date has
ever elucidated its inhibition constant and mechanism.
The in vitro synthesis of LH2 leads to two enantiomers, D and
L, according to the cysteine isomer used.1,6,8 D-LH2 is regarded as
the only isomer to naturally produce light,1,6 while the L-isomer is
an inhibitor.24,25 Lembert25 studied its inhibitory mechanism and
suggested that L-LH2 acts as a competitive inhibitor with a Ki of
3–4 mM. However, and despite Lembert’s important contribution
to the existing literature, more accurate values are needed in order
to clarify the importance of this isomer in this bioluminescent light
profile.
Luc-catalysed light production assays
All the enzyme reactions took place at ambient temperature (24–
27 ◦C) and were performed at least in triplicate. The biolumines-
cence tests were performed in a homemade luminometer using a
Hamamatsu HC135-01 photomultiplier tube. All light reactions
were carried out in 50 mM Hepes buffer and pH 7.5. The reaction
was initiated by the injection of D-LH2 (3.75–120 mM – volumes
of 50 mL) into a transparent assay tube, by simple reagent mixing,
containing a mixture of ATP (250 mM), MgCl2 (2 mM) and Luc
(10 nM). In some experiments, this latter mixture was supple-
mented with L, L-LH2 or L-CoA (in the concentration range be-
tween 0.5 and 2 mM). All the indicated concentrations refer to the
final volume of 200 mL. The light was integrated and recorded in
0.1 s intervals. L, L-LH2 and L-CoA solutions were protected from
light at all time by covering the tubes with aluminium foil.
The steady-state initial velocities were determined by incubating
Luc with serial dilutions of the inhibitors for 3 min at room
temperature, followed by the addition of D-LH2. The reaction was
allowed to continue for 3 min. Reaction rates were determined
from the increase in light production over the first milliseconds
after the start of the reaction in the linear part of the flash profile.
In the current study we provide a detailed kinetic model of
the inhibition exerted by these three compounds by measuring
the light production in their presence and absence. The results
obtained in this work suggest an important role for L and L-
LH2 in Luc flash profile due to their strong inhibitory character.
The discovery that L-CoA is indeed a weaker inhibitor than L-
AMP further supports the thiolytic mechanism regarding CoA
stimulating effect on light production.
Inhibition models and data analysis
Analysis of the steady-state kinetics of the inhibition of Luc by
L-CoA and L-LH2. The apparent Michaelis constant (Kmapp) and
the apparent maximum velocity (Vmaxapp) values were determined
from the nonlinear least-squares fit of the Henri–Michaelis–
Menten equation:32
Experimental
Vmapapx [S]
Kmapp +[S]
Materials
(5)
n =
A stock solution of commercial Luc (Sigma; L9506) was prepared
by dissolving the lyophilized powder in Hepes buffer 0.5 M, pH
7.5 (15 mg lyophilisate per mL; 60 mM Luc) and stored in small
Dissociation constant values (Ki) for L-CoA were estimated
from the nonlinear least-squares best fit of the non-competitive
inhibition model:32
◦
aliquots at -20 C to prevent self-degradation. Its concentration
was confirmed by UV spectroscopy at 278 nm using the extinction
coefficient of 45 560 M-1 cm-1.26 D-LH2, ATP and Hepes were also
purchased from Sigma.
Vmax
[S]
n =
[I]
(6)
[S]+ Km
1+
Ki
L and L-LH2 were chemically synthesized and purified as
described previously.18,27–30 The chemical synthesis of L-CoA was
based on the chemical synthesis of LH2-CoA performed by Fraga
et al.31 D-LH2 (50 mg) in THF (2 mL) was mixed with triethylamine
(0.75 mL) and ethyl chloroformate (0.51 mL). The reaction was
left at 25 ◦C for 5 h, and the reaction mixture volume was
where Km is the Michaelis constant and Vmax the maximum
velocity.
Dissociation constant values (Ki) for L-LH2 were estimated from
the nonlinear least-squares best fit of the mixed-type inhibition
model:33
1040 | Photochem. Photobiol. Sci., 2011, 10, 1039–1045 This journal is
The Royal Society of Chemistry and Owner Societies 2011
©

View Online
Table 1 Kinetic parameters for the inhibition of Luc by L-CoA
Vmax
[S]
[I]
1+ aK
[L-CoA]/mM
Kmapp/mM
Vmaxapp/¥10-5 RLU s-1
i
n =
(7)
Km 1+ [I]
[I]
1+ aK
i
(
K
i
)
0
15
18
17
18
21
1
1
2
2
2
2.29 0.3
1.56 0.12
1.17 0.15
0.84 0.1
0.70 0.06
[S]+
0.50
1.0
1.5
2.0
where Km is the Michaelis constant and Vmax the maximum
velocity.
Ki/mM
0.88 0.03
2.34 0.04
16.1 1.0
Analysis of the steady-state kinetics of the inhibition of Luc by L.
The Kiapp values for L were determined according to the Williams
and Morrison equation:32
V
max/¥10-5 RLU s-1
Km/mM
decrease in the maximum of light intensity and in the initial
velocity under steady-state conditions, confirming its inhibitory
properties.
[E]−[I]−Kiapp
+
([E]−[I]−Kiapp )2 + 4[E]Kiapp
n
n0
(8)
=
2[E]
L-CoA was found to inhibit Luc in a non-competitive fashion
with a Ki of 0.88 0.03 mM (Table 1). This value was determined
from a nonlinear fit of eqn (6) to steady-state initial velocities
obtained with several concentrations of D-LH2, L-CoA and satu-
rating concentrations of ATP (Fig. 2a). This inhibition followed a
characteristic hyperbolic Henri–Michaelis–Menten pattern.
where [E] is the active enzyme concentration, [I] is the total
inhibitor concentrations, Kiapp is the overall dissociation constant,
v is the initial reaction velocity at inhibitor concentration [I], and v0
is the initial velocity in the absence of the inhibitor. If the inhibitor
presents different affinity for the enzyme alone and for the complex
enzyme-substrate, Kiapp is related to the dissociation constant of
the inhibitor (Ki) by the equation:32
[S]+ Km
Kiapp
=
Km
[S]
(9)
+
Ki aKi
where [S] is the competing substrate concentration, Ki the inhibi-
tion constant with respect to the unbound enzyme and aKi the
inhibition enzyme referring to the enzyme–substrate complex.
Data analysis
All data were analyzed by nonlinear regression analysis using eqn
(5) to (9) with Graphpad software package (GraphPad Prism 5.01
for Windows).
Results and discussion
L-CoA inhibition kinetics
Fig. 1 shows the first three seconds of typical Luc flash profiles.
The flash profile in the presence of L-CoA shows a considerable
Fig. 2 (a) The dependence of the initial velocity on the concentration of
D-LH2 at different concentrations of L-CoA:(ꢀ) -0.5 mM, (᭡) -1.0 mM,
(᭢) -1.5 mM, (᭜) -2.0 mM, and without inhibitor ( ). The lines represent a
᭹
least squares best fit of the Henri–Michaelis–Menten equation to the data.
app
app
The fit parameters, Km and Vmax are represented on Table 1. (b) The
app
dependence of 1/Vmax on L-CoA concentration. All experiments were
performed using 50 mM Hepes buffer (pH 7.5) at ambient temperature as
described in experimental. The Luc concentration was 10 nM.
The secondary plot (Fig. 2b; 1/Vmaxapp as a function of inhibitor
concentration) shows that an increase in L-CoA concentration
resulted in a linear decrease of Vmaxapp. These results indicate that
L-CoA is a non-competitive inhibitor towards D-LH2.
Lineweaver–Burk plot (Fig. 3a) exhibit a linear trend typical of
a non-competitive inhibitor, with the same intercept on the 1/[D-
LH2] axis, further underlining the non-competitive behaviour of
this compound. These results demonstrate that L-CoA binds with
identical affinity to the free enzyme and the enzyme–substrate
complex, and that this binding impairs the light-production
Fig. 1 Light emission data from standard bioluminescence activity assays
of Luc performed in the presence of (ꢀ) L-CoA 0.5 mM, (᭡) L-LH2
0.5 mM, ( ) L 0.5 mM and in the absence (᭜) of inhibitors. D-LH 15
᭹
2
mM was injected (at t = 0 s) into mixtures containing ATP 250 mM, Luc
10 nM and Mg2+ 2 mM in Hepes buffer (pH = 7.5) pre-incubated with
inhibitor.
This journal is
The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 1039–1045 | 1041
©

View Online
bioluminescence reaction does suffer a large conformation change
during its two-step reaction. A mutagenesis work performed by
Branchini et al.36 indicated that two different and opposite C-
domain lysine residues are each one essential to only one of the
two-steps of this reaction. Thus, this indicates the occurrence of a
large C-domain rotation when the product of the adeynlation step
becomes the substrate for the second step of light-production.
Thus, taking these data in consideration L-CoA could inhibit
the bioluminescence reaction, not by hindering D-LH2 binding
site, but by having the ability to bind to the two conformations
of Luc, and so, compete with the substrates of the two-steps.
However, it should be said that the major Luc mutants used in
this study showed bioluminescence spectra different from the wild-
type, indicating that the mutations produced structures that could
be too different to be compared with the wild-type. Moreover, the
role of the two lysines was assessed by calculating the steady-state
constants of LH2-AMP and L. Although there is evidence that
support the use of LH2-AMP,35 our work demonstrates that L
do not bind at the same site of D-LH2, and so the adenylation
of these two compounds may not occur at the same conditions.
Thus, more data is needed about Luc conformation during the
bioluminescence reaction prior to more conclusive explanations.
Also, the relatively weak inhibitory character of L-CoA when
in comparison with L-AMP (Ki = 3.8 0.7 nM)13 is in good
agreement with the experimental work that indicates that the
stimulating effect exerted by CoA on the bioluminescence reaction
is caused by its interaction with L-AMP and subsequent formation
of L-CoA.14
Fig. 3 (a) Lineweaver–Burk plot of inhibition of Luc by L-CoA with
respect to D-LH2. L-CoA concentrations: (ꢀ) 2.0, (ꢀ) 1.5, (᭢) 1.0, (
)
᭺
0.5 and ( ) 0 mM. (b) Dixon plot of the same data used in plot (a). D-LH
᭹
2
concentration: (ꢀ) 120, ( ) 60, (᭜) 30, (᭢) 15, (᭡) 7.5 and (ꢀ) 3.75 mM.
᭹
reaction. It is known that Luc has acyl-CoA synthetase activity,
being part of the acyl-adenylate/thioester-forming superfamily
of enzymes.24,34 The members of this family catalyze a two-step
reaction: first exist an initial adenylation of a carboxylate, forming
a adenylate intermediate; in the second step, this intermediate
is commonly involved in the formation of a thioester. Recent
experimental evidence indicate that these enzymes use a 140◦
domain rotation to present opposing faces of a dynamic C-
terminal domain to active site for the different partial reactions.34
On the contrary, structural analysis of the bioluminescence reac-
tion indicates that there are only minor conformational changes
during this catalysis, and that the Luc active site remains in
the conformation corresponding to the adenylation step during
light production.35 Analysis with Luc complexed with ATP, an
adenylyl analogue and AMP and OxyLH2, generating a series of
“snapshots” of the bioluminescence reaction, demonstrated that
the different complexes are in similar conformations. Thus, if we
have an enzyme complexed with one of substrates, the intermediate
and the two products in the same conformation, it is reasonable
to assume that this reaction does not need large conformational
changes to occur. So, there are indications that L-CoA can bind to
different conformations of Luc, corresponding to the absence or
presence of D-LH2, causing a dynamic domain rotation that leads
to the more stable Luc conformation for CoA-thioesters. As there
are no major changes in Luc structure during the bioluminescence
reaction, the stabilization of the active site should be essential
to the phenomenom and so the change of conformation induced
by L-CoA could hinder the binding of D-LH2 to the enzyme,
accounting for the observed inhibition. Alternatively, there are
evidences that contradict the structural analysis and state that the
L inhibition kinetics
The flashprofile inthe presenceofL (Fig. 1) shows asharp decrease
in the maximum of light intensity and in the initial velocity under
steady-state conditions. This decrease is more pronounced than
for L-CoA and L-LH2, illustrating the strong inhibitory potency
of this LH2 derivative. Moreover, this inhibitory power greatly
resembles the inhibitory profile of L-AMP, which only differs
from the bioluminescence reaction adenylyl intermediate by its
dehydroluciferin moiety. This evidence indicates a tight-binding
inhibition mechanism for L.
The Kiapp values were estimated by nonlinear regression using
the Morrison equation which accounts for the tight-binding inhi-
bition (eqn (8) and Fig. 4) and are shown in Table 2. The inhibition
mechanism of L was determined by plotting the IC50 value,
concerning each substrate concentration, as a function of D-LH2
concentration (Fig. 5a). The obtained graphical representation
is typical of a non-competitive inhibition mechanism when a <
1, also termed as mixed-type inhibition.33 The true Ki and aKi
values were determined by non-linear fitting of the Kiapp values
to eqn (9), as a function of D-LH2 concentration (Fig. 5b). The
values obtained were Ki = • and aKi = 0.00490 0.00009 mM, thus
indicating L basically acts as an uncompetitive inhibitor towards
D-LH2.
An uncompetitive inhibition mechanism suggest that L binds
only to the enzyme–substrate complex and do not compete with
the substrate for the active site, suggesting the presence of a
different binding site. This may sound strange, considering the
obvious structural similarities of L with D-LH2, which indicate
that even if there was a presence of a new binding site during the
1042 | Photochem. Photobiol. Sci., 2011, 10, 1039–1045 This journal is
The Royal Society of Chemistry and Owner Societies 2011
©

View Online
13
Table 2 Kinetic parameters for the inhibition of Luc by L
inhibitor towards D-LH2 and a similar compound to L) does
not bind in this supposed binding site. Another confusing fact
regarding this subject is that L-AMP13 and L-CoA, two L-
derivatives, can bind to the free enzyme. So, the only explanation
that appears to be logical is that L does not bind to a different
binding site, but at the active site. The different affinities can
be explained as it is known that the Luc active site suffers
changes in its conformation during the bioluminescence reaction,
changing from a more open to a more closed structure.36 This
more closed conformation possibly can better accommodate the
different orientation of the carboxylic group of L. On the contrary,
L-CoA and L-AMP can accommodate this carboxylic group in the
free enzyme, possibly because their CoA and AMP moieties are so
much bigger and have so many more points of interaction with the
active site in their structure that they may “force” the positioning
of the L moiety in the enzyme.
[L]/mM
Kiapp/mM
120
60.0
30.0
15.0
7.5
0.00070 0.00001
0.0045 0.0004
0.0073 0.0006
0.0074 0.0006
0.0100 0.0023
0.0327 0.0020
3.75
aKi/mM
Km
0.00490 0.00009
16.6 2.3
Alternatively, if we consider the work of Branchini et al.,36 the
mechanism of inhibition of L could be explained by the rotation
of the C-domain of Luc during the bioluminescence reaction. This
could indicate that L only binds to the conformation of the second-
step of the bioluminescence reaction. However, there are some
aspects of this mutagenesis work that need some clarification,
as stated above in the present paper, that prevent its use in a
more conclusive explanation. The study of the inhibition of L
with respect to LH2-AMP could also be of pivotal importance for
the clarification of this topic.
Fig. 4 Plot of fractional velocity as a function of L concentration at
different D-LH2 concentrations. The inhibition assays were conducted with
D-LH : ( ) 120 mM, (ꢀ) 60 mM, (᭡) 30 mM, (᭢) 15 mM, (᭜) 7.5 mM, (
)
᭹
2
᭺
3.75 mM and ATP 250 mM in 50 mM Hepes buffer (pH 7.5) at room
temperature. Enzyme concentration was 10 nM. L concentration was
varied from 0 to 1.9 mM.
L-LH2 inhibition kinetics
The flash profile in the presence of L-LH2 shows also a decrease
in the maximum of light intensity and in the initial velocity under
steady-state conditions, confirming its inhibitory properties. This
decrease is intermediate between the observed for L-CoA and L,
indicating that this inhibitor is stronger than L-CoA and weaker
than L.
This inhibition followed a characteristic hyperbolic Henri–
Michaelis–Menten pattern, revealing that the addition of L-LH2
app
to the reaction mixtures causes a decrease in both Vmax and
Kmapp, indicating an uncompetitive mechanism for L-LH2 (Fig. 6a
and Table 3). However, the Lineweaver–Burk plot (Fig. 6b) shows
that the slope is also affected, determining a mixed-type inhibition
for L-LH2 with a Ki of 0.68 0.14 mM and a aKi of 0.34 0.16
mM. The values was determined from a nonlinear fit of eqn (7) to
steady-state initial velocities obtained with several concentrations
of D-LH2, L-LH2 and saturating concentrations of ATP (Fig. 5a).
Table 3 Kinetic parameters for the inhibition of Luc by L-LH2
[L-LH2]/mM
Kmapp/mM
Vmaxapp/¥10-5 RLU s-1
0
14.8 1.9
10.1 1.4
9.7 1.0
8.8 0.6
6.7 1.1
1.027 0.036
0.484 0.017
0.282 0.007
0.174 0.003
0.084 0.003
0.5
0.9
1.4
1.8
Fig. 5 (a) The effect of D-LH2 concentration (3.75 to 120 mM) on the
IC50 values. (b) Secondary plot of Kiapp as a function of D-LH2. Kiapp values
were obtained from the nonlinear fit of Fig. 4.
bioluminescence reaction, L should still bind to the active site.
Furthermore, even if we consider the different orientations of the
carboxylic group between these two molecules as a impediment
for the binding of L, it is not clear as why OxyLH2 (a competitive
Ki/mM
a
aKi/mM
Km/mM
0.68 0.14
0.50 0.13
0.34 0.16
14.4 0.96
This journal is
The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 1039–1045 | 1043
©

View Online
its D-isomer. The existence of a lower aKi value indicates that this
molecule can still bind to the active site even after the formation
of the binary complex Luc·LH2-AMP, and that the active site is
present in a more favorable conformation for their interaction. We
know that L-LH2 can be used by Luc to form L-LH2-CoA to the
contrary of D-LH2,24 that the Luc-catalyzed formation of CoA-
thioesters is a more favorable reaction than the bioluminescence
reaction14 and that there are some changes in Luc conformations
during the formation of the adenylyl intermediate.35 So, it is
reasonable to deduce that these changes result in a more favorable
active site structure to the compounds that are involved in the
formation of CoA-thioesters, as L-LH2, thus resulting in higher
affinities for the enzyme.
Tables 1, 2 and 3 show that the Km values obtained for L-CoA
(16.1 1.0 mM), L (16.6 2.3 mM) and L-LH2 (14.4 0.95 mM)
agree with each other and the Km reported by Branchini et al.37
(15 mM).
Conclusions
The present work gives, for the first time, accurate Ki values and
kinetic mechanisms for L-CoA, L and L-LH2, indicating a possible
importance of L and L-LH2 for the characteristic in vitro flash
profile of Luc.
Fig. 6 The dependence of the initial velocity on the concentration of
d-LH2 at different concentrations of L-LH2:(ꢀ) -0.5 mM, (᭡) -0.9 mM,
(᭢) -1.4 mM, (᭜) -1.8 mM, and without inhibitor ( ). The lines represent
᭹
a least squares best fit of the Henri–Michaelis–Menten equation to the
app
app
The mechanism of inhibition by L was described for the first
time in the present work. It acts as a tight-binding uncompetitive
inhibitor of Luc with respect to substrate D-LH2 (Ki = 4.90
0.09 nM). In this type of inhibition the population of free soluble
inhibitor is significantly depleted by the formation of the enzyme–
inhibitor complex. This suggests that L binds almost irreversibly to
the Luc active-site arresting the normal light-producing reactions.
As the main difference of the structure of L, in comparison with
D-LH2, is the orientation of the carboxylic group, this fast tight-
interaction with Luc could be due to the formation of a more stable
hydrogen bond network, a sufficiently fast process to account to
this inhibition mechanism. Since this inhibitor could be formed as
a side product of Luc catalyzed light reactions, as indicated in eqn
(4), and D-LH2 could be oxidized into this compound our data
indicate that L could have an important role in the typical flash
profile.
Previous experimental observations have shown that the strong
inhibition of Luc caused by L-AMP could be easily antagonised by
addition of CoA, resulting in the formation of L-CoA. It is thought
that the thiolytic reaction is much faster than the bioluminescent
reaction and that the L-CoA is a much less powerful Luc inhibitor
than L-AMP.14 The mechanism of L-CoA inhibition was studied
in this work. It was found to be a non-competitive inhibitor with
respect to D-LH2, less potent (Ki = 0.88 0.03 mM) than L-AMP,13
and both the magnitude of L-CoA inhibition and its kinetics of
inhibition are in agreement with the stimulating effect exerted by
CoA on the bioluminescence reaction.
data. The fit parameters, Km and Vm are represented in Table 3. (b)
Lineweaver–Burk plot of inhibition of Luc by L-LH2 with respect to D-LH2.
L-LH2 concentrations: (ꢀ) 1.8, (ꢀ) 1.4, (᭢) 0.9, ( ) 0.5 and ( ) 0 mM.
᭹
᭺
All experiments were performed using 50 mM Hepes buffer (pH 7.5) at
ambient temperature as described in experimental. The Luc concentration
was 10 nM.
Lineweaver–Burk plot (Fig. 6b) exhibits a linear trend typical
of a mixed-type inibitor, with different intercepts on the 1/[D-
LH2] and the 1/Vi axis, and different slopes. Furthermore, the
plots cross to the left of the 1/Vi axis and below the 1/[D-LH2]
axis. This is typical of a form of mixed inhibition known as non-
competitive-uncompetitive inhibition. So, these results show that
L-LH2 can bind both to the free enzyme and to the enzyme-
substrate complex, but with different affinities. Thus, this type
of inhibition can have some different explanations. First, we can
consider a different binding site for L and D-LH2, with the binding
site conformation of the L-isomer becoming more favorable to
their association during the bioluminescence reaction. We can
also consider two binding sites for L-LH2, one on the free enzyme
and another that is formed during the bioluminescence reaction.
However, it is unlikely that these two isomers bind only at different
sites due to their obvious structural resemblance. Moreover, even
if we consider that the opposite conformation of the carboxylic
group that these molecules present can impair the binding in these
different sites, we cannot explain why OxyLH2 binds only to the
D-LH2 active site. As was referred in the introduction section,
this LH2-derivative is a competitive inhibitor with respect to D-
LH2,13 and it is very similar to the (D/L)-isomers (presenting only a
thiazolone moiety instead of a thiazoline-carboxylic acid one1). So,
it should have the ability for binding wherever the isomers can. Its
competitive inhibition mechanism towards D-LH2 indicates that
we are dealing with only one binding site for this type of molecule.
Therefore, the more logical explanation for this situation is that
the Ki value refers to a competitive inhibition of L-LH2 towards
L-LH2 was found to be an mixed-type non-competitive-
uncompetitive inhibitor with a Ki = 0.68 0.14 mM and aKi =
0.34 0.16 mM. This value, along with its kinetics of inhibition,
suggests that this inhibitor is not a key component of the fast
decay of bioluminescence. However, the lack of knowledge of
the conversion of D-LH2 into L-LH2 at bioluminescence reaction
conditions, and its relevant Ki suggest that L-LH2 could have an
important role on the in vitro flash profile.
1044 | Photochem. Photobiol. Sci., 2011, 10, 1039–1045 This journal is
The Royal Society of Chemistry and Owner Societies 2011
©

View Online
luciferase luminescence because they act as substrates and not as
allosteric effectors, FEBS J., 2008, 275, 1500–1509.
Abbreviations
16 R. Fontes, A. Dukhovich, A. Sillero and M. A. Gu¨nther Sillero,
Synthesis of dehydroluciferin by firefly luciferase, Effect of dehydrolu-
ciferin, coenzyme A and nucleoside triphosphates on the lumines-
cente reaction, Biochem. Biophys. Res. Commun., 1997, 237, 445–
450.
17 H. Fraga, Firefly luminescence: A historical perspective and re-
cent developments, Photochem. Photobiol. Sci., 2008, 7, 146–
158.
AMP
ATP
L
Adenosine-5¢-monophosphate
Adenosine-5¢-triphosphate
Dehydroluciferin;
L-AMP,
dehydroluciferyl-
adenylate
Dehydroluciferyl-CoA
Firefly luciferase; (D/L)-LH2, firefly luciferin
L-CoA
Luc
LH2-AMP Luciferyl-adenylate; PPi, inorganic pyrophosphate
RLU Relative light units; OxyLH2, oxyluciferin
18 L. J. Bowie, Synthesis of firefly luciferin and structural analogs, Methods
Enzymol., 1978, 57, 15–28.
19 B. J. Gates and M. DeLuca, The production of oxyluciferin during
the firefly luciferase ligh reaction, Arch. Biochem. Biophys., 1975, 169,
616–621.
20 J. J. Lemasters and C. R. Hackenbrock, Kinetics of product inhibition
during firefly luciferase luminescence, Biochemistry, 1977, 16, 445–447.
21 R. T. Lee and W. D. McElroy, Effects of 5¢-adenylic acid on firefly
luciferase, Arch. Biochem. Biophys., 1971, 145, 78–81.
22 R. T. Lee, J. L. Denburg and W. D. McElroy, Substrate-binding
properties of firefly luciferase. II ATP-binding site, Arch. Biochem.
Biophys., 1970, 141, 38–52.
Acknowledgements
Financial support from Fundac¸a˜o para a Cieˆncia e a Tecnologia
(FCT, Lisbon) (FSE-FEDER) (Project PTDC/QUI/71366/2006)
is acknowledged.
23 J. L. Denburg, R. T. Lee and W. D. McElroy, Substrate-binding
properties of firefly luciferase. I luciferin-binding site, Arch. Biochem.
Biophys., 1969, 134, 381–394.
24 M. Nakamura, S. Maki, Y. Amano, Y. Ohkita, K. Niwa, T. Hirano,
Y. Ohmiya and H. Niwa, Firefly luciferase exhibits bimodal action
depending on the luciferin chirality, Biochem. Biophys. Res. Commun.,
2005, 331, 471–475.
25 N. Lembert, Firefly luciferase can use L-luciferin to produce light,
Biochem. J., 1996, 317, 273–277.
26 B. R. Branchini, R. A. Magyar, K. M. Marcantonio, K. J. Newberry,
J. G. Stroh, L. K. Linz and M. H. Murtiashaw, Identification of
a Firefly Luciferase Active Site Peptide Using a Benzophenone-
based Photooxidation Reagent, J. Biol. Chem., 1997, 272, 19359–
19364.
27 H. Fraga, J. C. G. Esteves da Silva and R. Fontes, pH opposite
effects on synthesis of dinucleoside polyphosphates and on oxidation
reactions catalyzed by firefly luciferase, FEBS Lett., 2003, 543, 37–
41.
28 E. H. White, M. G. Steinmetz, J. D. Miano, P. D. Wildes and R.
Morland, Chemi-luminescence and bioluminescence of firefly luciferin,
J. Am. Chem. Soc., 1980, 102, 3199–3208.
29 B. R. Branchini, M. H. Murtiashaw, R. A. Magyar, N. C. Portier, M. C.
Ruggiero and J. G. Stroh, Yellow-green and red firefly bioluminescence
from 5, 5-dimethyloxyluciferin, J. Am. Chem. Soc., 2002, 124, 2112–
2113.
30 E. H. White, H. Whorter, G. F. Field and W. D. McElroy, Analogs of
firefly luciferin, J. Org. Chem., 1965, 30, 2344–2348.
31 H. Fraga, R. Fontes and J. C. G. Esteves da Silva, Synthesis of luciferyl
coenzyme A: a new bioluminescent substract for firefly luciferase in the
presence of AMP, Angew. Chem., 2005, 117, 3493–3495.
32 R. A. Copeland, Enzymes, 2nd Edition, 2000, Wiley.
33 T. Palmer, Understanding Enzymes, 2nd Edition, 1985, Ellis Horwood
Limited.
34 A. M. Gulick, Conformational dynamics in the Acyl-CoA synthetases,
adenylation domains of non-ribosomal peptides synthetases, and firefly
luciferase, ACS Chem. Biol., 2009, 4, 811–827.
35 T. Nakatsu, S. Ichiyama, J. Hiratake, A. Saldanha, N. Kobashi, K.
Sakata and H. Kato, Structural basis for the spectral difference in
luciferase bioluminescence, Nature, 2006, 440, 372–376.
36 B. R Branchini, T. L. Southworth, M. H. Murtiashaw, S. R. Wilkinson,
N. F. Khattak, J. C. Rosenberg and M. Zimmer, Mutagenesis Evidence
that the Partial Reactions of Firefly Bioluminescence Are Catalyzed
by Different Conformations of the Luciferase C-Terminal Domain,
Biochemistry, 2005, 44, 1385–1393.
37 B. R. Branchini, M. M. Hayward, S. Bamford, P. M. Brennan and
E. J. Lajiness, Naphthyl- and Quinolylluciferin: Green and Red Light
Emitting Firefly Luciferin Analogues, Photochem. Photobiol., 1989, 49,
689–695.
Notes and references
1 S. M. Marques and J. C. G. Esteves da Silva, Firefly bioluminescence.
A mechanistic approach of luciferase catalyzed reactions, IUBMB Life,
2009, 61, 6–17.
2 J. M. M. Leita˜o and J. C. G. Esteves da Silva, Firefly luciferase
inhibition, J. Photochem. Photobiol., B, 2010, 101, 1–8.
3 H. H. Seliger and W. D. McElroy, Spectral emission and quantum
yield of firefly bioluminescence, Arch. Biochem. Biophys., 1960, 88, 136–
141.
4 H. H. Seliger and W. D. McElroy, Colors of firefly bioluminescence-
enzyme configuration + species specificity, Proc. Natl. Acad. Sci. U. S.
A., 1964, 52, 75–81.
5 K. Niwa, Y. Ichino, S. Kumata, Y. Nakajima, Y. Hiraishi, D. Kato,
V. R. Viviani and Y. Ohmiya, Quatum Yields and Kinetics of the
Firefly Bioluminescence Reaction of Beetle Luciferases, Photochem.
Photobiol., 2010, 86, 1046–1049.
6 S. Inouye, Firefly luciferase: an adenylate-forming enzyme for
multicatalytic functions, Cell. Mol. Life Sci., 2009, 67, 387–
404.
7 T. C. Doyle, S. M. Burns and C. H. Contag, In vivo bioluminescence
imaging for integrated studies of infection, Cell. Microbiol., 2004, 6,
303–317.
8 A. Roda, P. Pasini, M. Mirasoli, E. Michelini and M. Guardigli,
Biotechnological application of bioluminescence and chemilumines-
cence, Trends Biotechnol., 2004, 22, 295–303.
9 K. E. Luker and G. D. Luker, Applications of bioluminescence imaging
to antiviral research and therapy: multiple luciferase enzymes and
quantitation, Antiviral Res., 2008, 78, 179–187.
10 F. Fan and K. V. Wood, Bioluminescence assays for high-throughput
screening, Assay Drug Dev. Technol., 2007, 5, 127–136.
11 B. R. Branchini, T. R. Southworth, N. F. Khattak, E. Michelini
and A. Roda, Red and green emitting firefly luciferase mutants for
bioluminescent reporter application, Anal. Biochem., 2005, 345, 140–
148.
12 P. Billard and M. S. DuBow, Bioluminescence-based assay for detection
and characterization of bacteria and chemicals in clinical laboratories,
Clin. Biochem., 1998, 31, 1–14.
13 C. Ribeiro and J. C. G. Esteves da Silva, Kinetics of inhibition of firefly
luciferase by oxyluciferin and dehydroluciferyl-adenylate, Photochem.
Photobiol. Sci., 2008, 7, 1085–1090.
14 H. Fraga, D. Fernandes, R. Fontes and J. C. G. Esteves da Silva,
Coenzyme A affects firefly luciferase luminescence because it acts as a
substrate and not as an allosteric effector, FEBS J., 2005, 272, 5206–
5216.
15 R. Fontes, D. Fernandes, F. Peralta, H. Fraga, I. Maio and J. C. G.
Esteves da Silva, Pyrophosphate and tripolyphosphate affect firefly
This journal is
The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 1039–1045 | 1045
©