Core-Modified Coelenterazine Luciferin Analogues: Synthesis and Chemiluminescence Properties

Article abstract of DOI:10.1002/chem.202004311

In this work on the design and studies of luciferins related to the blue-hued coelenterazine, the synthesis of heterocyclic analogues susceptible to produce a photon, possibly at a different wavelength, is undertaken. Here, the synthesis of O-acetylated derivatives of imidazo[1,2-b]pyridazin-3(5 H)-one, imidazo[2,1-f][1,2,4]triazin-7(1 H)-one, imidazo[1,2-a]pyridin-3-ol, imidazo[1,2-a]quinoxalin-1(5 H)-one, benzo[f]imidazo[1,2-a]quinoxalin-3(11 H)-one, imidazo[1′,2′:1,6]pyrazino[2,3-c]quinolin-3(11 H)-one, and 5,11-dihydro-3 H-chromeno[4,3-e]imidazo[1,2-a]pyrazin-3-one is described thanks to extensive use of the Buchwald–Hartwig N-arylation reaction. The acidic hydrolysis of these derivatives then gave solutions of the corresponding luciferin analogues, which were studied. Not too unexpectedly, even if these were “dressed” with substituents found in actual substrates of the nanoKAZ/NanoLuc luciferase, no bioluminescence was observed with these compounds. However, in a phosphate buffer, all produced a light signal, by chemiluminescence, with extensive variations in their respective intensity and this could be increased by adding a quaternary ammonium salt in the buffer. This aspect was actually instrumental to determine the emission spectra of many of these luciferin analogues.

Full text of DOI:10.1002/chem.202004311

Accepted Article
Accepted Article
Title: Core-modified coelenterazine luciferin analogues, synthesis and
chemiluminescence properties
Authors: Glwadys Gagnot, Vincent Hervin, Eloi P Coutant, Sophie
Goyard, Yves Jacob, Thierry Rose, Fatima Ezzahra Hibti,
Alessia Quatela, and Yves Janin
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202004311
Link to VoR: https://doi.org/10.1002/chem.202004311
01/2020

10.1002/chem.202004311
Chemistry - A European Journal
Core-modified coelenterazine luciferin analogues, synthesis and chemiluminescence
properties
Glwadys Gagnot,ab Vincent Hervin,a Eloi P. Coutant,a Sophie Goyard,c Yves Jacob,d Thierry
Rose,c Fatima Ezzahra Hibti,e Alessia Quatela,e Yves L. Janina*
a: Unitꢀ de Chimie et Biocatalyse, Institut Pasteur, UMR 3523, CNRS, 28 rue du Dr. Roux,
75724 Paris Cedex 15, France.
b: Université de Paris, 12 rue de l'école de Médecine, 75006 Paris, France.
c: Center for Innovation and Technological Research, Institut Pasteur, 25 rue du Dr. Roux, 75724
Paris Cedex 15, France.
d: Unité de Génétique Moléculaire des Virus à ARN, Institut Pasteur, UMR 3569, CNRS, 28 rue
du Dr. Roux, 75724 Paris Cedex 15, France.
e: HORIBA FRANCE SAS, 14 Boulevard Thomas Gobert, Passage Jobin Yvon CS45002, 91120
Palaiseau, France
Abstract
In our work on the design and studies of luciferins related to the blue-hued coelenterazine, we
undertook the synthesis of heterocyclic analogues susceptible to produce a photon, possibly at a
different wavelength. We describe here the synthesis of O-acetylated derivatives of imidazo[1,2-
b]pyridazin-3(5H)-one,
imidazo[2,1-f][1,2,4]triazin-7(1H)-one,
imidazo[1,2-a]pyridin-3-ol,
imidazo[1,2-a]quinoxalin-1(5H)-one,
benzo[f]imidazo[1,2-a]quinoxalin-3(11H)-one,
imidazo[1',2':1,6]pyrazino[2,3-c]quinolin-3(11H)-one
and
5,11-dihydro-3H-chromeno[4,3-
e]imidazo[1,2-a]pyrazin-3-one thanks to an extensive use of the Buchwald-Hartwig N-arylation.
The acidic hydrolysis of these derivatives then gave solutions of the corresponding luciferin
analogues which we studied. Not too unexpectedly, even if these were “dressed” with
substituents found in actual substrates of the nanoKAZ/NanoLuc luciferase, no bioluminescence
was observed with these compounds. However, in a phosphate buffer, all produced a light signal,
by chemiluminescence, with extensive variations in their respective intensity and this could be
increased by adding a quaternary ammonium salt in the buffer. This aspect was actually
instrumental to determine the emission spectra of many of these luciferin analogs.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Introduction
Chemiluminescence as well as bioluminescence,[1] its luciferase-catalyzed version, are based on
remarkable oxidation reactions of luciferins leading to products in an excited electronic state,
which can then relax either by non-radiative processes or by a photon emission.[2] This
phenomenon has found numerous applications, especially for detection purposes. In life sciences,
a great variety of bioluminescent reporting systems have thus been constructed out of a luciferin
and a luciferase. Amongst them, systems based on the marine luciferin coelenterazine (1) and the
corresponding luciferases or photoproteins from a remarkably diverse number of sea dwelling
species[1a, 3] are increasingly used.[4] The mechanism at the source of the photon emission by
these systems is depicted in scheme 1. The first step is a reaction with dioxygen leading to the
peroxide 2 which, unless stabilized,[5] undergoes a rearrangement into the endoperoxide 3 and
then decomposes into carbon dioxide and coelenteramide (4). The latter product occurs in an
excited electronic state which can relax back to its ground state via the emission of a photon at a
“short” wavelength corresponding to a blue color (between 450 and 480 nm), the most visible
undersea. Contrary to firefly luciferins, which require an ATP-based activation, this oxidation can
also be achieved without a luciferase in the presence of oxygen under a variety of conditions and
will lead, via the same mechanism,[6] to a usually far less efficient photon emission phenomenon
which is called chemiluminescence. Some current avenue of research are focusing on the design
of luciferin/luciferase reporting systems, still based on coelenterazine using enzymes, but
emitting a photon at a higher wavelength which is more suitable for in vivo studies.[7] This is due
to the fact that a red photon will be detectable through the living tissues of a small animal,
contrary to a blue one. In the present report, we planned on the use of mutants of luciferases such
as nanoKAZ/NanoLuc, which is already accepting a rather large array of imidazo[1,2-
8]
b]pyridazin-3(5H)-ones as substrates,[7i,
and designed analogues featuring an altered
heterocyclic core which we also “dressed” with the phenyl/bisbenzyls residues found in actual
substrates of the nanoKAZ/NanoLuc luciferase. We thus sought accesses to unprecedented
derivatives of imidazo[1,2-b]pyridazin-3(5H)-one 5, imidazo[2,1-f][1,2,4]triazin-7(1H)-one 6 or
imidazo[1,2-a]pyridin-3-ol 7 which could theoretically undergo an electronic cascade leading to
the occurrence of hydroperoxide intermediates as depicted for coelenterazine (1). The synthesis
of a “deaza” coelenterazine analog featuring this last heterocyclic core has actually been reported
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
previously.[9] We also prepared some imidazo[1,2-a]quinoxalin-1(5H)-ones 8 which have been
the subject of an untranslated patent for their chemiluminescence properties although no mention
of the wavelength of the emitted photon appears to be mentioned.[10] A single derivative featuring
this ring system has also been claimed in a more recent patent.[11] The synthesis of analogues
featuring
a
benzo[f]imidazo[1,2-a]quinoxalin-3(11H)-one
skeleton
9
or
imidazo[1',2':1,6]pyrazino[2,3-c]quinolin-3(11H)-one 10 were also explored as a similar
heterocyclic system is found in the structure of v-coelenterazine (11), a luciferin analogue which
is a substrate of the soft coral Renilla reniformis luciferase[12] and some of its mutants,[7b] as well
as the photoprotein found in the Aequorea jellyfish.[13] Finally, we replaced the additional
pyridine ring of compound 11 by a pyran ring and designed an access to an unprecedented 5,11-
dihydro-3H-chromeno[4,3-e]imidazo[1,2-a]pyrazin-3-one 12 derivative.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Scheme 1. Luciferase-based oxidative decarboxylation of coelenterazine (1) and potential light-
producing ring systems investigated in this report
Syntheses
We have previously reported improvements of the nanoKAZ/NanoLuc-based reporting system
using an original synthetic pathway leading to enhanced luciferin analogues.[8c, 14] As depicted in
scheme 2, these compounds were obtained from chloropyrazines 13 and amino esters 14 via a
key N-arylation step to give the stable esters 15a. Upon treatment with sodium hydroxide under
an inert atmosphere, these gave the corresponding sodium salts 15b and the addition of an excess
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
of acetic anhydride led, via the imidazo[1,2-a]pyrazin-3(7H)-ones 16, to the far more stable O-
acetylated luciferins 17. Then, when required, the proluciferins 17 can be hydrolyzed back into
the luciferins 16 under mild conditions to give, after further dilution in appropriate buffer, ready-
to-use solution for bioluminescence-based studies.
Scheme 2. Previously reported[8c, 14b] synthesis of O-acetyl imidazo[1,2-a]pyrazin-3(7H)-one
luciferins.
Preparation of an imidazo[1,2-b]pyridazin-3(5H)-one derivative. As depicted in scheme 3, the
O-acetylated compound 26 featuring an imidazo[1,2-b]pyridazine ring system was synthesized in
5 steps from phenylpyridazine 18, readily available from 3-chloro-6-methoxypyridazine and
phenylboronic acid using a Suzuki-Miyaura coupling reaction.[15] From this compound, a
regioselective deprotonation using lithium diisopropylamide (LDA) followed by the addition of
benzaldehyde led to alcohol 19. The reduction of the hydroxyl function to produce compound 21
turned out to be fraught with difficulties as an over-hydrogenation was often observed. An
acceptable yield was eventually found via a palladium-catalyzed hydrogenation of the bromine-
bearing 3-hydroxypyridazine 20 readily prepared by treatment of alcohol 19 with hydrogen
bromide in acetic acid. From the reduced compound 21, the 3-chloropyridazine 22 was then
obtained upon its treatment with hot phenylphosphonic dichloride. Then, we tried to prepare the
N-arylated derivative of phenylalanine ethyl ester (23) using our previously published method.[8c,
14b]
Unexpectedly, under these conditions the corresponding ester 24 directly underwent a
cyclization into the corresponding imidazo[1,2-b]pyridazin-3(5H)-one 25. However, due its
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
instability under basic conditions, this intermediate immediately decomposed upon exposure to
dioxygen. Accordingly, an excess of acetic anhydride was added to the reaction mixture at the
end of this chemical step leading to the far more stable O-acetylated imidazo[1,2-b]pyridazine
derivative 26 which was isolated in a 58% yield from 3-chloropyridazine 22.
Scheme 3. i: 1) LDA, THF, - 78 °C, 2) PhCHO. ii: HBr/AcOH, 90 °C. iii: H2, Pd/C, iPrOH, 20
°C. iv: PhPOCl2, 100 °C. v: Pd(OAc)2, BINAP, Cs2CO3, toluene, 90 °C.
Preparation of an imidazo[2,1-f][1,2,4]triazin-7(1H)-one derivative. As depicted in scheme 4, to
reach an imidazo[2,1-f][1,2,4]triazine analogue the corresponding 6-hydroxytriazine 30 was
made in 3 steps from hippuric acid (27) using a previously described method.[16] Unexpectedly,
transformation of the 6-hydroxytriazine 30 into the corresponding 6-chlorotriazine 31 turned out
to be difficult as low yield and/or decomposition were observed when trying various chlorination
agents: thionyl chloride, phosphorus oxychloride, or phenylphosphonic dichloride. The use of
mild conditions (phosphorus oxychloride diluted in acetonitrile in the presence of solid potassium
carbonate at reflux) led to a rather modest 19-22% yield of the target compound 31. The ensuing
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
N-arylation of phenyl alanine ethyl ester (23) also turned out to be problematic as an extensive
decomposition was observed. Out of the many side products occurring, a minute amount (4%) of
the pyrrolotriazine 32, resulting from a condensation of acetonitrile, was isolated. Switching to
toluene at 90 °C actually led to the isolation of the expected amino ester 33 but in a very poor
11% yield. Moreover, the target compound 33 turned out to be unstable in open air under various
conditions, such as when left in solution in CDCl3, as seen by LC/MS analysis (m/z = 453 and
455). An alternative was eventually designed as we found that the slow uncatalyzed N-arylation
of phenylalanine ethyl ester (23) by compound 31 was also possible. Indeed, upon heating at 120
°C these compounds in N-methylpyrrolidine (NMP) using triethylamine as a base for 14 days led
directly to a solution containing the unstable luciferin 34. Its in situ treatment with an excess of
acetic anhydride and additional triethylamine allowed the isolation of the target proluciferin 35 in
an 8% overall yield from compound 31.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Scheme 4. i: PhCHO, AcONa, Ac2O, 65 - 90 °C. ii: NH2NH2, EtOH, 20 °C. iii: NaOH 1.5N,
reflux. iv: POCl3, K2CO3, MeCN, reflux. v: Pd(OAc)2, BINAP, Cs2CO3, MeCN, 60 °C. vi:
Pd(OAc)2, BINAP, Cs2CO3, toluene, 90 °C. vii: NEt3, NMP, 120 °C, 14 days. viii: Ac2O, NEt3,
20 °C.
Preparation of an imidazo[1,2-a]pyridin-3-ol derivative. The original preparation of compound
42 depicted in scheme 5 started with the readily available 2-methoxypyridine derivative 36.[17] It
turned out that a regiospecific Negishi reaction was possible even when using 1.5 equivalents of
benzylzincbromide, and the C-3 benzylated derivative 37 was isolated in a 47% yield. The
structure of this compound was established by long distance NMR correlation. From this
intermediate, a Suzuki-Miyaura coupling provided the phenyl in position 5 and a cleavage of its
methoxy group led to the 2-hydroxypiridine derivative 39. Chlorination of this compound turned
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
out to require harsher conditions than expected and even heating the reaction at 150 °C for 22
hours gave an overall 56% of the 2-chloropyridine 40 from the C-3 benzylated derivative 37. The
ensuing N-arylation step with phenylalanine ethyl ester (23) gave 60% of compound 41. From
this ester, a hydrolysis using sodium hydroxide followed by a cyclisation in the presence of an
excess of acetic anhydride led to 32% of the target O-acetylated imidazo[1,2-a]pyridine 42.
Scheme 5. i: PhCH2ZnBr, Pd(OAc)2, XPhos, THF, 60 °C. ii: PhB(OH)2, Pd(OAc)2, SPhos,
dioxane/H2O 5:1, 90 °C. iii: 33% HBr/AcOH, 100 °C. iv: PhPOCl2, 150 °C. vi: PheOEt (23),
Pd(OAc)2, BINAP, Cs2CO3, toluene, 90 °C. vii: 1) NaOH, THF, 20 °C, 2) Ac2O, 20 °C.
Preparation of an imidazo[1,2-a]quinoxalin-1(5H)-one derivative. As depicted in scheme 6, an
original preparation of luciferin analogues featuring an imidazoquinoxaline ring system was
developed. For this, we used the quinoxalinol 45a made from the condensation between o-
phenylenediamine (43a) and phenylpyruvic acid (44) in the presence of hydrochloric acid and 2-
mercaptoethanol[18] which led to 79% of 45a. Of interest is the fact that a control experiment
without 2-mercaptoethanol still provided 62% of compound 45a. From this quinoxalinol, the
corresponding 2-chloroquinoxaline 46a was then obtained in a 90% yield and this was followed
by its N-arylation with phenylalanine ethyl ester (23) to give 71% of compound 47a. The ensuing
ester hydrolysis using sodium hydroxide followed by a cyclisation using an excess of acetic
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
anhydride finally gave 69% of the target O-acetylated imidazo[1,2-a]quinoxaline 48a. Although a
more extensive purification procedure was required to obtain 46% of compound 46b from the
difluorinated o-phenylenediamine 43b, the same synthetic pathway led to the analogue 48b in a
12% overall yield. Of note was the far slower N-arylation of phenylalanine ethyl ester (23) with
the fluorine-bearing 2-chloroquinoxaline 46b as only an 11% yield of the N-arylester 47b was
obtained in 12 hours of heating whereas a 38% yield was achieved when heating for 27 hours.
Scheme 6. i: EtOH, HSCH2CH2OH, 2N H3O+Cl-, reflux. ii: PhPOCl2, 100 °C. iii: Pd(OAc)2,
BINAP, Cs2CO3, toluene, 90 °C. iv: 1) NaOH, THF, 20 °C, 2) Ac2O, 20 °C.
Preparation of benzo[f]imidazo[1,2-a]quinoxalin-3(11H)-one derivatives. As depicted in scheme
7, an original[13, 19] synthesis of such analogues was designed via the use of three distinct
Buchwald-Hartwig N-arylation reactions. The first one allowed the N-arylation of phenylalanine
ethyl ester (23) by the 2-bromonaphthalene derivatives 49a and 49b. Rather long reaction time
were required: 22 hours at 105 °C and 64 hours at 100 °C, to secure respectively 67 and 54%
yield of compounds 50a and 50b. Bromination of the resulting products led selectively to the bis-
brominated derivative 51a and the 2-bromo-4-fluoro derivative 51b. This was followed by a
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
quantitative amidification of these esters, using methanolic ammonia in a pressure vessel at 100
°C, which gave the amides 52a and 52b. From these, an intramolecular Buchwald-Hartwig
reaction was envisaged using Xantphos as the palladium ligand.[20] However, from the bis-
brominated derivative 52a, trials pointed out an extensive decomposition and very little
occurrence of compounds 53a or 54a, probably due to an un-selective palladium insertion in the
carbon-bromine bounds. On the other hand, from the 2-bromo-4-fluoro amide 52b, the same
reaction led to a chromatographic fraction containing mixture of compound 53b, resulting from
the planned cyclization, and compound 54b resulting from the aromatization of some 53b. This
aromatization was then completed using peracetic acid as the oxidant and compound 54b was
isolated in 54% yield from 52b. The ensuing chlorination with hot phenylphosphonic dichloride
gave 67% of compound 55 which was subjected to the last Buchwald-Hartwig reaction to yield
71% of the N-arylated phenylalanine ethyl esters 56. The last two steps, hydrolysis of the ester
function followed in situ by a cyclization with an excess of acetic anhydride, gave 65% of the
target O-acetylated luciferins analogue 57.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Scheme 7. i: PheOEt (23), Pd(OAc)2, BINAP, Cs2CO3, toluene, 100 - 105 °C. ii: NBS, CH2Cl2,
0-20 °C. iii: 7N NH3/MeOH, 100 °C. iv: Pd(OAc)2, Xantphos, Cs2CO3, THF, 120 °C. v: 35%
AcOOH/AcOH, AcOEt, 20 °C. vi: PhPOCl2, 100 °C. vii: PheOEt (23), Pd(OAc)2, BINAP,
Cs2CO3, MeCN, 60 °C. viii: 1) NaOH, THF, 20 °C, 2) Ac2O, 20 °C.
Preparation of an imidazo[1',2':1,6]pyrazino[2,3-c]quinolin-3(11H)-one derivative. The
synthetic pathway described above was used again to prepare the nitrogen-bearing analogue,
starting with 4-bromoquinoline (58) as depicted in scheme 8. If the first four steps, including the
first two Buchwald-Hartwig reactions, proceeded without any difficulties, the aromatization step
of compound 62 required sulfur in boiling decaline and gave 67% of the target compound 63. In
the next step, chlorination in hot phenylphosphonic dichloride to obtain compound 65 turned out
to be very slow, even at 100 °C. Raising the temperature to 150 °C led to 2.6% of the unexpected
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
compound 64 resulting from an oxidative dimerization process along with 17% of the target
compound 65. It is only by running the reaction as thoroughly as possible under an argon
atmosphere and adding potassium carbonate (presumably to quench any evolving hydrochloric
acid) that compound 65 could be obtained in a 38% yield although 19% of the dimer 64 and some
unreacted material (less than 8%) were also isolated. This unexpected methylene reactivity,
somehow reminiscent of what we previously observed with the chlorotriazine 31 depicted in
scheme 4, probably had also some effect on the last Buchwald-Hartwig N-arylation as we could
isolate the resulting α-amino ester 66 in only 43% yield. Moreover, a rather high incidence of
decomposition was observed in the course of the last synthetic steps and the target O-acetylated
luciferin analogue 67 was obtained in only a 35% yield.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Scheme 8. i: PheOEt (23), Pd(OAc)2, BINAP, Cs2CO3, toluene, 95 °C. ii: NBS, CH2Cl2, 0-20 °C.
iii: 7N NH3/MeOH, 100 °C. iv: Pd(OAc)2, Xantphos, Cs2CO3, THF, 120 °C. v: S8, decaline,
reflux. vi: PhPOCl2, K2CO3, 150 °C, argon. vii: PheOEt (23), Pd(OAc)2, BINAP, Cs2CO3,
MeCN, 60 °C. viii: 1) NaOH, THF, 20 °C, 2) Ac2O, 20 °C.
Preparation of a 5,11-dihydro-3H-chromeno[4,3-e]imidazo[1,2-a]pyrazin-3-one derivative. In an
attempt to design an alternative pathway to benzo[f]imidazo[1,2-a]quinoxalin-3(11H)-one 14, we
initially contemplated to use our previously reported preparation of chloropyrazines starting with
a 1,4-addition of phenyl alanine ethyl ester (23) as a free base on nitrostyrene.[8c, 14b] Accordingly,
as depicted in scheme 9, we started with the bridged derivative 3-nitro-1,2-dihydronaphthalene
(68a), readily prepared by nitration of dihydronaphtalene using tetranitromethane.[21]
Unexpectedly and despite many trials, no 1,4-addition was observed with this substrate. On the
other hand, this 1,4-addition readily took place on 3-nitro-2H-chromene (68b)[22] to give the
pyran-bearing 1,4-adduct 69b. We do not have a satisfying explanation for this difference of
reactivity although the only example reported of a nitrogen-bearing nucleophilic 1,4-addition on
3-nitro-1,2-dihydronaphthalene (68a) involved a far more nucleophilic specie.[23] In any case,
from compound 69b, a reduction of its nitro function using zinc and hydrochloric acid followed
by a thermal cyclisation of the resulting amine gave the tetrahydrochromeno[3,4-b]pyrazin-
3(4H)-one 70 as a mixture of diastereoisomers. Many attempts were made to achieve its
aromatization, using sulfur in boiling 1,3-dichlorobenzene, iodine as well as other oxidants with a
very limited success.[14a] We thus resorted to the preparation of the quite unstable nitrone 71.
From this crude compound, a dehydration reaction using sodium hydroxide[14b] gave the desired
5H-chromeno[3,4-b]pyrazin-3-ol 72 in 52% yield from compound 70. The preparation of the
corresponding chloro derivative using hot phenylphosphonic dichloride only led to minute
amount of this compound. Accordingly, we used a milder alternative[24] via the preparation of the
triflate ester 73, and a displacement reaction with sodium bromide which gave the bromo
derivative 74 in 40% yield from the chromenopyrazinol 72. From compound 74, the ensuing N-
arylation of phenylalanine ethyl ester (23) led to 70% of the expected ester 75 and the final steps
provided 80% of the target O-acetylated luciferin analogue 76.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Scheme 9. i: PheOEt (23; free base), neat, 0 °C. ii: 1) Zn dust, 37% H3O+Cl-, dioxane, 0-20 °C,
2) neat, 140 °C. iii: 35% AcOOH/AcOH, AcOEt, 20 °C. iv: NaOH, EtOH, 20 °C. v: Tf2O, NEt3,
CH2Cl2, 20 °C. vii: NaBr, TfOH, DMF, 120 °C. vii: PheOEt (23), Pd(OAc)2, BINAP, Cs2CO3,
toluene, 85 °C. viii: 1) NaOH, THF, 20 °C, 2) Ac2O, 20 °C.
Hydrolysis of the O-acetylated luciferin analogues and chemiluminescence properties.
The acid hydrolysis of the O-acetylated analogues 26, 35, 42, 48a-b, 57, 67 and 76, using a
mixture of DMSO, ethanol and hydrochloric acid at 50 °C for two hours gave the luciferins 77-85
depicted in figure 1. Of note is that, in the specific case of compound 35, which gave compound
79, this reaction required 4 hours. The LC/MS monitoring of these reactions pointed out the
occurrence of clean solutions of the expected compounds 78-85 along with, in the case of
compounds 83 and 85, a slow decomposition taking place over 6 weeks even when the solution
was stored at -20 °C. Moreover, hydrolysis of compound 26, occurred along with the formation
of a small proportion of two dimers (m/z = 781). A study of this oxidative process pointed out
that it took place in the course of the hydrolysis of 26 but further heating of the resulting solution,
even in the presence of air or when adding far more hydrochloric acid did not lead to any
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
additional dimerization of compound 78. Solution of the compounds 1, 86 and 87 were also used
as references.
Figure 1. Structures of the luciferin analogues 77-87.
From the resulting solution of compound 78-85, no tangible bioluminescence was observed with
compound 78-85 when using n nanoKAZ/NanoLuc luciferase under the usual conditions,[14b]
contrary to coelenterazine (1), bisdeoxycoelenterazine (77),[8b] furimazine (86)[8a] or the enhanced
analogue 87.[8c] On the other hand, the chemiluminescence properties of these compounds was
somewhat more rewarding. In the past, the chemiluminescence properties of coelenterazine (1) or
some of its analogues have been the focus of many studies.[25] The final concentration of the
compounds used in these experiments is always low, varying between 1 µM,[25e] 3 µM,[25i] 10
µM[25a, c, d, f] or in two instances 50 µM.[25g, h] We quickly found out the reason as, at much higher
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
concentration (i.e.: 1 mM final), we observed remarkable wavelength shift which could be only
explained by resonance energy transfers to fluorescent substances arising from the light
production process. Interestingly, this requirement for low concentration was not found necessary
for the study of the chemiluminescence of firefly luciferins.[26] For our part, as depicted in table 1,
for every compound made, their addition to a phosphate buffer (DPBS; pH = 7.4), leading to 5
µM final concentration, led to a chemiluminescence signal as measured by a HIDEX Triathler
Gamma Counter. Of note is that the signal intensity varied widely according to the compound
studied. Moreover, we confirmed the ammonium-based surfactant enhancement effect reported
by Shimomura[1a] since quite remarkable intensity improvement was systematically achieved
when adding of 2.5 mMol of cetyltrimethylammonium bromide (CTAB). Concerning the
products resulting from this chemiluminescence, LC/MS analysis, using ammonium carbonate as
a buffer, pointed out the occurrence of the expected ions typical of the light-producing reaction
depicted in scheme 1 (see supplementary info). In the case of coelenterazine (1, m/z = 424), the
ion corresponding to coelenteramide (4, m/z = 412) and, as reported,[27] the amine (m/z = 278)
were observed. Concerning the emission spectra of the strong and medium emitters, the use of an
Aqualog® (HORIBA Instruments, Inc.), a spectrometer endowed with a sensitive CCD detector
(Water-Raman SNR > 20,000:1 - RMS method), gave good results but only upon adding 2.5
mMol of CTAB in the phosphate buffer. Concerning the weak emitters, such as compounds 78-
81, we have yet to find a device sensitive enough to determine their emission spectra.
Interestingly, even at a 5 µM final concentration a degree of signal quench was observed in one
instance (compound 87) but this was alleviated when measuring the spectra at a 2.5 µM final
concentration.
Table 1. Chemiluminescence spectra of the luciferins
Imax
DPBS
4.000
65.000
200
160
250
215
660
Imax
+2.5M CTAB
1.100.000
2.400.000
28.000a
λmax
Cpd
465
462
n.s.
n.s.
n.s.
n.s.
513b
438
1
77
78
79
80
81
82
83
1.400
3.200
602.600a
2.400.000
19 millionc
26.000
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
2.000
4.600.000
13.000
596.000
38 millionc
1.400.000
6.100.000
460
458
472
460d
84
85
86
87
4.100
a : slow increase over 2 minutes
b: presence of a small peak at 400 nm and a shoulder at 570 nm
c : saturation of the HIDEX Triathler
d : at a 2.5 µM substrate concentration
n.s.: no signal exploitable
Conclusion
In this study, we made an array of compounds which all produced a photon upon their reaction
with oxygen via, as confirmed by LC/MS analysis, a mechanism plausibly similar to the one
described for imidazo[1,2-a]pyrazin-3(7H)-one luciferins. Many of these analogues were
prepared using our previously reported[8c, 14b] synthetic pathway although unexpected limits were
encountered in their preparations. These limitations drove us to use and develop alternative
conditions to produce the target compounds, notably using the fact that the N-arylations could
sometime take place without a palladium-based catalyst along with a (not too) strong base or the
observation that some the resulting esters could undergo a cyclisation into the target luciferin just
upon heating. Again, trapping these relatively unstable substances under their O-acetylated form
was instrumental in their proper isolation and the success of this work. Concerning their
chemiluminescence properties when adding these compounds to a phosphate buffer, a very
variable degree of signal intensities was observed. Of interest is the observation that a very strong
chemiluminescence is observed for the most rigid analogues such as compound 83 and 85 but
unexpectedly not for the “aza” analogue 84. Aside from a degree of instability (in comparison
with other derivatives) which was noted for the O-acetylated precursors of compounds 78-80, it is
more difficult to identify a reason for their very low chemiluminescence. Concerning the
emission profiles of the compounds endowed with a chemiluminescence of sufficient intensity,
the λmax were mainly situated in the blue region (438-472 nm) with the notable exception of the
difluorinated derivative 82, endowed with a green signal at 513 nm. Previous and far more
extensive, studies on the chemiluminescence properties of coelenterazine (1) have reported on the
large impact on the signal wavelength when using a protic or an aprotic organic solvents as well
when using a base.[25a] In our case we chose to use only a protic one (water) and a mild base (a
phosphate buffer) since we observed little difference between the chemiluminescence (472 nm)
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
and bioluminescence (455 nm;[8c] with NanoLuc) of the reference compound furimazine (1). We
thus retained only this set of condition to qualify the emission wavelength of these analogues as a
plausibly predictive parameter of their eventual bioluminescence properties (if a luciferase,
derived from nanoKAZ/NanoLuc, could be designed to consume them). However, the relatively
modest chemiluminescence λmax shifts achieved here do not meet the criteria to initiate a research
program aiming at designing a luciferase capable of catalyzing the light-producing combustion of
one of these analogues. In any case, the results obtained are providing some additional clues on
how to red-shift this type of luciferin and, probably more important, the experimental data
presented here should be useful for the design and validation of computer-based models[2] of
these light-producing reactions.
Acknowledgement
This work was supported by the Agence Nationale de la Recherche (ANR), grant ANR-11-
CRNT-0004, in the context of the investment program ‘GLOBAL CARE’, an association of the
Instituts Carnot ‘Pasteur-Maladies Infectieuses’, ‘Curie-Cancer’, ‘Voir et Entendre’, ‘Institut du
Cerveau et de la moelle Épinière’ and the ‘Consortium pour l’Accꢀlꢀration de l’Innovation et de
son Transfert dans le domaine du Lymphome’ (CALYM). G.G. acknowledges a PhD fellowship
from the Université Paris Descartes, Sorbonne Paris Cité and E.P.C. was supported by a
fellowship from Global Care. This project also benefited from the Valoexpress funding call of the
Institut Pasteur. Prof. Christian Bréchot and Drs. Muriel Delepierre and Daniel Larzul are
acknowledged for their interest and support. Finally, we thank Florence Miller from Berthold
Technologies for her support and Aline Rifflet from the Unité de Biologie et Génétique at the
Institut Pasteur for her great help in LC/MS experiments.
Bibliography
[1] a) O. Shimomura, Bioluminescence: Chemical Principles And Methods Revised Edition,
World Scientific, Singapore, 2012, p. 1-468; b) A. C. Love and J. A. Prescher, Cell Chem.
Biol. 2020, 27, 904-920.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
[2] M. Vacher, I. Fdez Galván, B. W. Ding, S. Schramm, R. Berraud-Pache, P. Naumov, N.
Ferré, Y. J. Liu, I. Navizet, D. Roca-Sanjuán, W. J. Baader and R. Lindh, Chem. Rev. 2018,
118, 6927-6974.
[3] a) P. J. Herring, J. Biolumin. Chemilumin. 1987, 1, 147-163; b) S. Inouye, Y. Sahara-Miura,
M. Nakamura and T. Hosoya, Protein Expr. Purif. 2020, 171, 105615; c) M. Tessler, J. P.
Gaffney, A. G. Oliveira, A. Guarnaccia, K. C. Dobi, N. A. Gujarati, M. Galbraith, J. D.
Mirza, J. S. Sparks, V. A. Pieribone, R. J. Wood and D. F. Gruber, Sci. Rep. 2020, 10,
17724; d) M. Bessho-Uehara, W. R. Francis and S. H. D. Haddock, Mar. Biol. 2020, 167,
doi.org/10.1007/s00227-00020-03706-w.
[4] a) H. Hoshino, Expert Opin. Drug Dis. 2009, 9, 374-389; b) A. Roda, M. Guardigli, E.
Michelini and M. Mirasoli, Trends Analyt. Chem. 2009, 28, 307-322; c) T. Ozawa, H.
Yoshimura and S. B. Kim, Anal. Chem. 2013, 85, 590-609; d) E. Michelini, L. Cevenini,
M. M. Calabretta and A. Roda, Anal. Bioanal. Chem. 2014, 406, 5531-5539; e) G. Thouand
and R. Marks, Bioluminescence: Fundamentals and Applications in Biotechnology -
Volume 1-3 (Advances in Biochemical Engineering/Biotechnology), 2014, p; f) C. G.
England, E. B. Ehlerding and W. Cai, Bioconjug. Chem. 2016, 27, 1175-1187; g) Z. M.
Kaskova, A. S. Tsarkova and I. V. Yampolsky, Chem. Soc. Rev. 2016, 45, 6048-6077; h) T.
Jiang, L. Du and M. Li, Photochem. Photobiol. Sci. 2016, 15, 466-480; i) L. Mezzanotte,
M. van 't Root, H. Karatas, E. A. Goun and C. W. G. M. Löwik, Trends Biotechnol. 2017,
35, 640-652; j) C. M. Rathbun and J. A. Prescher, Biochemistry 2017, 56, 5178-5184; k) H.
W. Yeh, T. Wu, M. Chen and H. W. Ai, Biochemistry 2019, 58, 1689-1697; l) H. W. Yeh
and H. W. Ai, Annu. Rev. Anal. Chem. 2019, 12, 129-150; m) V. V. Krasitskaya, E. E.
Bashmakova and L. A. Frank, Int. J. Mol. Sci. 2020, 10.3390/ijms21207465 ; n) A.
Lopreside, M. M. Calabretta, L. Montali, M. Zangheri, M. Guardigli, M. Mirasoli and E.
Michelini, Luminescence 2020, doi.org/10.1002/bio.3948.
[5] J. F. Head, S. Inouye, K. Teranishi and O. Shimomura, Nature 2000, 405, 372-376.
[6] a) T. Goto, S. Inoue and S. Sugiura, Tetrahedron Lett. 1968, 9, 3873-3876; b) K. Teranishi
and O. Shimomura, Anal. Biochem. 1997, 249, 37-43.
[7] a) T. Curie, K. L. Rogers, C. Colasante and P. Brûlet, Mol. Imaging 2007, 6, 30-42; b) A. M.
Loening, A. M. Wu and S. S. Gambhir, Nat. Methods 2007, 8, 641-643; c) G. Giuliani, P.
Molinari, G. Ferretti, A. Cappelli, M. Anzini, S. Vomero and T. Costa, Tetrahedron Lett.
2012, 53, 5114-5118; d) K. Suzuki, T. Kimura, H. Shinoda, G. Bai, M. J. Daniels, Y. Arai,
M. Nakano and T. Nagai, Nat. Commun. 2016, 7, 13718; e) K. A. Rumyantsev, K. K.
Turoverov and V. V. Verkhusha, Sci. Rep. 2016, 6, 36588; f) K. M. Grinstead, L. Rowe, C.
M. Ensor, S. Joel, P. Daftarian, E. Dikici, J.-M. Zingg and S. Daunert, PLoS One 2016, 11,
e0158579; g) H. W. Yeh, O. Karmach, A. Ji, D. Carter, M. M. Martins-Green and H. W.
Ai, Nature Methods 2017, 14, 971-974; h) J. Hiblot, Q. Yu, M. D. B. Sabbadini, L.
Reymond, L. Xue, A. Schena, O. Sallin, N. Hill, R. Griss and K. Johnsson, Angew. Chem.,
Int. Ed. 2017, 56, 14556-14560; i) A. Shakhmin, M. P. Hall, T. Machleidt, J. R. Walker, K.
V. Wood and T. A. Kirkland, Org. Biomol. Chem. 2017, 15, 8559-8567; j) A. Taylor, J.
Sharkey, A. Plagge, B. Wilm and P. Murray, Contrast Media Mol. Imaging 2018, 2514796;
k) D. C. Alcobia, A. I. Ziegler, A. Kondrashov, E. Comeo, S. Mistry, B. Kellam, A. Chang,
J. Woolard, S. J. Hill and E. K. Sloan, iScience 2018, 6, 280-288; l) R. Nishihara, R.
Paulmurugan, T. Nakajima, E. Yamamoto, A. Natarajan, R. Afjei, Y. Hiruta, N. Iwasawa,
S. Nishiyama, D. Citterio, M. Sato, S. B. Kim and K. Suzuki, Theranostics 2019, 13, 2646-
2661; m) M. Abe, R. Nishihara, Y. Ikeda, T. Nakajima, M. Sato, N. Iwasawa, S.
Nishiyama, R. Paulmurugan, D. Citterio, S. Kim and K. Suzuki, Chembiochem 2019, 20,
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
1919-1923; n) H. W. Yeh, Y. Xiong, T. Wu, M. Chen, A. Ji, L. X. and H. W. Ai, ACS
Chem. Biol. 2019, 17, 959-965; o) F. Weihs and H. Dacres, Trends Analyt. Chem. 2019,
116, 61-73; p) H. Kobayashi, L. P. Picard, A. M. Schönegge and M. Bouvier, Nat. Protoc.
2019, 1084-1107.
[8] a) M. P. Hall, J. Unch, B. F. Binkowski, M. P. Valley, B. L. Butler, M. G. Wood, P. Otto, K.
Zimmerman, G. Vidugiris, T. Machleidt, M. B. Robers, H. A. Benink, C. T. Eggers, M. R.
Slater, P. L. Meisenheimer, D. H. Klaubert, F. Fan, L. P. Encell and K. V. Wood, ACS
Chem. Biol. 2012, 7, 1848-1857; b) S. Inouye, J. Sato, Y. Sahara-Miura, S. Yoshida, H.
Kurakata and T. Hosoya, Biochem. Biophys. Res. Commun. 2013, 437, 23-28; c) E. P.
Coutant, G. Gagnot, V. O. Hervin, R. Baatallah, S. Goyard, Y. Jacob, T. Rose and Y. L.
Janin, Chem. Eur. J. 2020, 26, 948-958.
[9] Y. Lu, R. Hendra, A. Oakley and P. A. Keller, Tetrahedron Lett. 2014, 55, 6212-6215.
[10] K. Teranishi, (Mie University), JP 2007254352, 2007
[11] M. Hall, T. Kirkland, T. Machleidt, A. Shakhmin and J. R. Walker, (Promega),
US2018030059, 2018
[12] S. Inouye and O. Shimomura, Biochem. Biophys. Res. Commun. 1997, 233, 349-353.
[13] O. Shimomura, B. Musicki and Y. Kishi, Biochem. J. 1988, 251, 405-410.
[14] a) Y. L. Janin, E. P. Coutant, V. Hervin, G. Gagnot, Y. Jacob, S. Goyard and T. Rose,
(Institut Pasteur, CNRS), WO 2018197727, 2018; b) E. P. Coutant, S. Goyard, V. O.
Hervin, G. Gagnot, R. Baatallah, T. Rose, Y. Jacob and Y. L. Janin, Org. Biomol. Chem.
2019, 17, 3709-3713.
[15] I. Parrot, Y. Rival and C. G. Wermuth, Synthesis 1999, 1163-1168.
[16] C. G. Wermuth, J. J. Bourguignon, I. Morin, P. Renard, M. Devissaguet, J. F. de La Faverie
and G. Adam, (Adir et compagnie), US 5254548, 1993
[17] E. Spinner and J. C. B. White, J. Chem. Soc. B: Phys. Org. 1966, 991-992.
[18] a) R. J. Biediger, B. Dupre, L. K. Hamaker, W. W. Holland, J. M. Kassir, W. Li, R. V.
Market, N. Nguyen, I. L. Scott, C. Wu and E. R. Decker, (Rockey, Milnamow and Katz,
Ltd.), US 20030199692, 2003; b) H. Arakawa, J. Ishida, M. Yamagushi and M. Nakamura,
Chem. Pharm. Bull. 1990, 38, 3491-3493.
[19] T. Hosoya, R. Iimori, S. Yoshida, Y. Sumida, Y. Sahara-Miura, J. Sato and S. Inouye, Org.
Lett. 2015, 17, 3888-3891.
[20] P. J. Manley and M. T. Bilodeau, Org. Lett. 2004, 6, 2433-2435.
[21] M. R. Michaelides, Y. Hong, S. DiDomenico Jr, E. K. Bayburt, K. E. Asin, D. R. Britton, C.
W. Lin and K. Shiosaki, J. Med. Chem. 1997, 40, 1585-1599
[22] a) D. Dauzonne and R. Royer, Synthesis 1984, 348-349; b) A. H. Clark, J. D. McCorvy, V. J.
Watts and D. E. Nichols, Bioorg. Med. Chem. 2011, 19, 5420-5431.
[23] K. Imagawa, E. Hata, T. Yamada and T. Mukaiyama, Chem. Lett. 1996, 25, 291-292.
[24] a) A. R. Katritzky, J. Li, C. Stevens and D. J. Ager, Org. Prep. Proced. Int. 1994, 26, 439-
444; b) S. Murata, K. Kiguchi and T. Sugimoto, Heterocycles 1998, 48, 1255-1274.
[25] a) K. Teranishi and T. Goto, Bull. Chem. Soc. Jpn. 1990, 63, 3132-3140; b) T. Hirano, Y.
Gomi, T. Takahashi, K. Kitahara, C. F. Qi, I. Mizoguchi, S. Kyushin and M. Ohashi,
Tetrahedron Lett. 1992, 39, 5771-5774; c) K. Teranishi, M. Hisamatsu and T. Yamada,
Luminescence 1999, 14, 297-302; d) H. Kondo, T. Igarashi, S. Maki, H. Niwa, H. Ikeda and
T. Hirano, Tetrahedron Lett. 2005, 46, 7701-7704; e) Y. Takahashi, H. Kondo, S. Maki, H.
Niwa, H. Ikeda and T. Hirano, Tetrahedron Lett. 2006, 47, 6057-6061; f) T. Hirano, Y.
Takahashi, H. Kondo, S. Maki, H. Kojima, H. Ikeda and H. Niwa, Photochem. Photobiol.
Sci. 2008, 7, 197-207; g) R. Saito, T. Hirano, S. Maki and H. Niwa, J. Photochem.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Photobiol A: Chemistry 2014, 293, 12-25; h) M.-L. Yuan, T.-Y. Jiang, L.-P. Du and M.-Y.
Li, Chin. Chem. Lett. 2016, 27, 550-554; i) J. M. Lourenço, J. C. G. Esteves da Silva and L.
Pinto da Silva, J. Luminescence 2018, 194, 139-145.
[26] R. Saito, T. Kuchimaru, S. Higashi, W. L. Shijia, M. Kiyama, S. Iwano, R. Obata, T. Hirano,
S. Kizaka-Kondoh and S. A. Maki, Bull. Chem. Soc. Jpn. 2019, 92, 608-618.
[27] O. Shimomura and F. H. Johnson, Biochemistry 1972, 11, 1602-1608.
This article is protected by copyright. All rights reserved.

10.1002/chem.202004311
Chemistry - A European Journal
Can extensive alteration of the heterocyclic core of the coelenterazine ring system, a blued-hued
luciferin of marine origin, lead to chemiluminescent substances and can this also “red-shift” their
emission spectra?
Key words:
Homogeneous catalysis; Synthesis design; Chemiluminescence; Luciferin; Bioluminescence
This article is protected by copyright. All rights reserved.