On Pyridopyrazinol Chemistry: Synthesis of Chemiluminescent Substances

Article abstract of DOI:10.1055/a-1396-8607

Our work on new chemiluminescent substances related to the marine luciferin coelenterazine (λ _max= 465 nm) led us to attempt the synthesis of four nitrogen-rich pyridopyrazine-bearing analogues. Accordingly, the preparation of the corresponding benzyl-bearing pyridopyrazinols is studied. By varying the conditions for the condensation of phenylpyruvic acid with 1,2-diaminopyridine or 3,4-diaminopyridine, all the possible pyridopyrazin-2-ol regioisomers are isolated and properly characterized, including by means of crystallographic studies. The ensuing syntheses of the halogenated pyridopyrazines are fraught with difficulties ranging from extensive decomposition to an unexpected ring contraction. In one instance, the inherently reductive mixture of phosphorus oxychloride and phosphorus trichloride provides 2-benzyl-3-chloropyrido[2,3- b ]pyrazine. This precursor is then transformed into the target O -acetylated luciferin (6,8-dibenzylimidazo[1,2- a ]pyrido[3,2- e ]pyrazin-9-yl acetate). The 'benzo' derivative of this analogue (i.e., 2,12-dibenzylimidazo[1′,2′:1,6]pyrazino[2,3- c ]isoquinolin-3-yl acetate) is also prepared and the chemiluminescence emission spectra of these compounds are determined in a phosphate buffer (λ _max= 546 and 462 nm).

Full text of DOI:10.1055/a-1396-8607

SYNTHESIS
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart
2021, 53, A–H
paper
A
en
G. Gagnot et al.
Paper
Synthesis
On Pyridopyrazinol Chemistry: Synthesis of Chemiluminescent
Substances
Glwadys Gagnota,b,c
0
0
0
0
-
0
0
0
2
-
5
7
7
8
-9
9
5
X
NH2
O
OH
OH
O
Pierre Legrandd
Amira Tadrosa,b
0
0
0
0-0
0
0
3
-2
4
3
1
-
2
2
5
5
N
+
NH2
O
N
N
Fatima Ezzahra Hibtie
Alessia Quatelae
coelenterazine
N
λmax = 465 nm
H
(by chemiluminescence)
HO
O
Yves L. Janin*a,b
0
0
0
0
-
0
0
0
3
-
3
0
1
9
-
9
8
4
2
O
N
N
a Unité de Chimie et Biocatalyse, Département de Biologie Structurale et
Chimie, Institut Pasteur, 28 rue du Dr Roux, 75724 Paris Cedex 15, France
b Unité Mixte de Recherche 3523, Centre National de la Recherche Scien-
tifique, 28 rue du Dr Roux, 75724 Paris Cedex 15, France
yves.janin@cnrs.fr
N
N
F
F
N
N
H
λ
max = 513 nm
N
λ
max = ?
H
c Université de Paris, 12 rue de l’école de Médecine, 75006 Paris, France
d Synchrotron Soleil, L’Orme des Merisiers, Saint-Aubin, Gif-sur-Yvette,
France
e HORIBA FRANCE SAS, 14 Boulevard Thomas Gobert, Passage Jobin Yvon
CS45002, 91120 Palaiseau, France
Corresponding Author
Received: 21.01.2021
Accepted after revision: 22.02.2021
Published online: 22.02.2021
v-coelenterazine (1b) or the heteroatom-bearing analogues
4–6. In these cases, the structural alterations provided a de-
gree of electron enrichment of the imidazo[1,2-a]pyrazine
DOI: 10.1055/a-1396-8607; Art ID: ss-2021-z0028-op
nucleus and, contrary to the blue-hued derivative 3 (max
=
Abstract Our work on new chemiluminescent substances related to
the marine luciferin coelenterazine (max = 465 nm) led us to attempt
the synthesis of four nitrogen-rich pyridopyrazine-bearing analogues.
Accordingly, the preparation of the corresponding benzyl-bearing pyri-
dopyrazinols is studied. By varying the conditions for the condensation
of phenylpyruvic acid with 1,2-diaminopyridine or 3,4-diaminopyridine,
all the possible pyridopyrazin-2-ol regioisomers are isolated and proper-
ly characterized, including by means of crystallographic studies. The en-
suing syntheses of the halogenated pyridopyrazines are fraught with
difficulties ranging from extensive decomposition to an unexpected
ring contraction. In one instance, the inherently reductive mixture of
phosphorus oxychloride and phosphorus trichloride provides 2-benzyl-
3-chloropyrido[2,3-b]pyrazine. This precursor is then transformed into
the target O-acetylated luciferin (6,8-dibenzylimidazo[1,2-a]pyrido[3,2-
e]pyrazin-9-yl acetate). The ‘benzo’ derivative of this analogue (i.e.,
2,12-dibenzylimidazo[1′,2′:1,6]pyrazino[2,3-c]isoquinolin-3-yl acetate)
is also prepared and the chemiluminescence emission spectra of these
compounds are determined in a phosphate buffer (max = 546 and 462
nm).
462 nm), their chemiluminescence emission spectra were
shifted toward redder wavelengths.2 On the other hand,
with other types of analogues, we recently achieved a rath-
er modest chemiluminescence shift with a max of 513 nm
in the case of compound 2b, which features two electron-
attracting fluorine atoms.1 This result drove us to envisage
the synthesis of the four possible aza analogues 7–10 in an
attempt to study the effect of an additional nitrogen on
their chemiluminescence in comparison with compounds
2a,b.
In the event of a tangible red shift of their chemilumi-
nescence spectra, this would pave the way for a research
program aimed at an extensive alteration of the catalytic
sites of coelenterazine-using luciferases so that these com-
pounds would become actual substrates, and thus provide a
red-shifted bioluminescent reporting system. Such an ap-
proach has actually been successful in the case of a firefly-
based luciferin/luciferase bioluminescence system.3 Accord-
ingly, we focused on the preparation of the four halogenat-
ed pyridopyrazines 11–14 which, as we have demonstrated
in other instances, are good intermediates for construction
of the imidazole component of imidazo[1,2-a]pyrazin-
3(7H)-ones as well as many related heterocyclic systems.1,4
Synthetic access to the chloropyridopyrazines 11 and 12
was sought via chlorination of the corresponding isomeric
2-benzylpyrido[3,4-b]pyrazin-3-ol (17) and 3-benzylpyri-
do[3,4-b]pyrazin-2-ol (20). As depicted in Scheme 1, to pre-
pare these substances, we started from two reports describing
Key words chemiluminescence, coelenterazine, imidazo[1,2-a]pyraz-
in-3(7H)-one, pyridopyrazine, luciferin, heterocycles, rearrangement
We recently reported on the preparation and chemilu-
minescence properties of an array of heterocyclic analogues
of the luciferin coelenterazine (1a), such as compounds 2a,b
depicted in Figure 1.1 Because of their imidazolone compo-
nent, such heterocyclic systems will, upon an oxidative de-
carboxylation process, lead to the production of a photon.1
Similarly, reports have described the synthesis and chemi-
or bio-luminescence properties of altered luciferins such as
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H
Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany

B
G. Gagnot et al.
Paper
Synthesis
ppm as well as two broad singlets at 11.05 and 9.50 ppm)
and appeared to reach a stable 1:1.6 ratio after 48 hours. To
conclude this series of unexpected results, all our attempts
to obtain the corresponding chloropyridopyrazines 11 or
12, from 17 or 20, using a variety of methods unfortunately
failed as extensive decompositions were the sole results.
Similarly, methylation trials under basic conditions, or us-
ing methyl orthocarboxylates,6 led to complex mixtures of
compounds.
OH
O
O
N
N
R
N
N
R
N
N
H
H
HO
coelenterazine (1a)
2a: R = H
2b: R = F
OH
O
O
N
N
N
N
X
Y
N
H
As depicted in Scheme 2, our attempts to prepare the
chloropyridopyrazines 13 and 14 were somewhat more
successful. Again, a regioselective synthesis of the corre-
sponding 2-benzylpyrido[2,3-b]pyrazin-3-ol (22) and 3-
benzylpyrido[2,3-b]pyrazin-2-ol (23) relied on the reported
mechanistic investigations explaining the orientation of the
condensations between 2,3-diaminopyrazine (21) and ethyl
pyruvate.5 Of note is that if these reports described rather
optimistic yields and selectivity, a subsequent publication
using these reagents reported far more sober results.7 In
our case, from 2,3-diaminopyrazine (21) and phenylpyruvic
acid (16), we obtained mixtures of the target isomers 22
and 23 depending on the reaction conditions. Heating the
diamine 21 and phenylpyruvic acid (16) in ethanol at 85 °C
in a closed reactor for 26 hours (conditions i) gave the pyri-
dopyrazine isomers 22 and 23 in 40% and 15% isolated
yields, respectively. As depicted in the ORTEP representa-
tion, an X-ray diffraction analysis again firmly established
the structure of compound 22. Moreover, the more polar
imidazopyridine derivative 24 was also detected in the
crude reaction products. Under acidic conditions (condi-
tions ii), the proportion of isomers was inversed and com-
pounds 22 and 23 were isolated in 4% and 33% yields, re-
spectively. Moreover, we could also isolate in this trial 7% of
the rather insoluble imidazopyridine derivative 24. Inter-
estingly, upon treatment of compound 24 with sodium
N
H
v-coelenterazine (1b)
HO
7: X = N, Y = CH
8: X = CH, Y = N
O
N
O
N
N
X
Y
N
N
X
H
3: X = CH2
4: X = O
5: X = S
N
H
9: X = N, Y = CH
10: X = CH, Y = N
6: X = Se
N
Cl
X
N
Cl
X
Y
N
Y
N
11: X = N, Y = CH
12: X = CH, Y = N
13: X = N, Y = CH
14: X = CH, Y = N
Figure 1 Structures of the compounds 1–10 and intermediates 11–14
regioselective condensations between 3,4-diaminopyrazine
(15) and pyruvic acid or ethyl pyruvate in which the acidity
of the reaction medium is actually governing the hydration
state of the pyruvate and thus the orientation of the reac-
tion.5 Accordingly, we tried these conditions using phen-
ylpyruvic acid (16) instead of pyruvic acid. When heating
compounds 15 and 16 in ethanol, the reaction led to the
isolation of 34% of the pyrido[3,4-b]pyrazin-2-ol 17, the
structure of which was firmly established by means of X-
ray diffraction, as depicted here with the corresponding
ORTEP representation.
Moreover, a sizable amount of the highly insoluble acid
18 was also obtained, however, this compound could only
be partially characterized (by 1H NMR and HRMS) as it
turned out to decompose in the course of recrystallization
attempts. A drastically different result was obtained when
heating compounds 15 and 16 in ethanol along with 3
equivalents of hydrochloric acid (2 N). Indeed, the obtained
thick precipitate turned out to be the hydrochloride salt of
the imidazopyridine 19a (featuring a CH signal at 6.74
ppm), which was isolated in an 83% yield and fully charac-
terized. Interestingly, upon basic treatment of this com-
pound, the target isomeric 2-benzylpyrido[3,4-b]pyrazin-3-
ol (20) was then isolated in a 66% yield after a simple ex-
traction. A microanalysis was actually required to confirm
this structure since, in solution, the pyridopyrazin-3-ol 20
(characterized by a CH2 signal at 4.18 ppm and a broad sin-
glet at 12.59 ppm) slowly isomerized into the free base
form of imidazopyridine 19b (displaying a CH signal at 6.45
H
H
N
NH2
NH2
N
O
i, 16
N
CO2H
N
N
N
H
+
N
15
HO
O
17
18
ii
O
16
H
N
OH
N
O
N
iii
N
N
H
N
·HCl
19a
20
DMSO-d6
OH
N
N
N
H
ORTEP of 17
19b
Scheme 1 Reagents and conditions: (i) EtOH, reflux; (ii) EtOH, HCl (2 N,
3 equiv); (iii) (1) EtOH, NaOH, reflux, 3 minutes, (2) NH4Cl.
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H

C
G. Gagnot et al.
Paper
Synthesis
As depicted in Scheme 3, we also undertook the prepa-
ration of the benzo derivative 35, which is structurally re-
lated to compound 27. To achieve this, we envisaged a cop-
per chloride catalyzed condensation between the readily
available8 4-bromoisoquinolin-3-amine (28) and phenylala-
nine (29) to give the 1,4-dihydropyrazinoisoquinoline de-
rivative 31. The first report describing similar transforma-
tions mentions the use of 1% of copper chloride, and in
most of the examples, the yields described for the corre-
sponding ‘dihydro’ derivative are quite remarkable.9 How-
ever, in our specific case, we had to resort to the use of 10%
of copper chloride to achieve a degree of transformation
along with rather low yields. Moreover, we always ob-
served, by LC/MS and 1H NMR spectroscopy, the occurrence
of a sizable amount of the aromatized derivative 32, which
is in accordance with related results described in a later
publication.10 Along with these two compounds, we also
noticed the occurrence of the aminoisoquinoline 30, plausi-
bly resulting from a high temperature DMSO-based reduc-
tion process.11 Eventually, we employed a different set of re-
ported conditions12 involving copper oxide instead of cop-
per chloride, a somewhat larger excess of the amino acid, a
much lower reaction temperature of 90 °C, and an extended
reaction time of 40 hours. In our case, from phenylalanine
(29) and 4-bromoisoquinolin-3-amine (28), this led to the
expected mixture of the dihydro and the aromatized deriv-
atives 31 and 32 along with some unreacted 28. As fully de-
tailed below, a procedure was then designed to isolate 31
and 32 from the reaction mixture; treatment of the dihydro
derivative 31 with N-bromosuccinimide in ethanol, fol-
lowed by a filtration, gave an additional amount of the aro-
matized product 32. With this two-stage procedure an
overall yield of 30% of compound 32 was thus achieved
O
O
OH
N
NH2
i or ii
+
NH2
16
21
H
N
H
N
O
N
O
N
OH
N
+
+
N
N
N
N
H
22
23
24
i: n.i.
ii: 7%
i: 40%
ii: 4%
i: 15%
ii: 33%
iv
iii
N
N
N
Cl
13 (55%)
ORTEP of 22
v
O
O
O
O
N
N
NH
vi
N
N
N
N
N
26 (12%)
27 (33%)
Scheme 2 Reagents and conditions: (i) EtOH, 85 °C; (ii) EtOH, HCl (2 N),
reflux; (iii) PCl3, POCl3, 80 °C; (iv) POCl3 or PhPOCl2 or SOCl2, see text;
(v) PhCH2CHNH2CO2Et (25), Pd(OAc)2, BINAP, Cs2CO3, MeCN, 90 °C; (vi)
(1) NaOH, THF, 20 °C, (2) Ac2O, 20 °C.
hydroxide, and contrary to its isomers 19, this imidazopyri-
dine mostly underwent extensive decomposition. Chlorina-
tion of the hydroxypyridopyrazine isomer 22 was initially
problematic, although upon LC/MS monitoring we observed
the occurrence of side compounds plausibly arising from
oxidation processes. Accordingly, we undertook this reac-
tion in a closed vessel in the presence of phosphorus tri-
chloride as a reducing agent, being very much compatible
with the phosphorus oxychloride used for this reaction.
This combination proved to be rewarding and we were able
to isolate compound 13 in a 55% yield. On the other hand,
the use of these conditions, or many others, to obtain the
corresponding chlorinated derivative 14 from compound 23
all failed. In every case, the imidazopyridine 24 resulting
from a remarkable rearrangement was the sole product de-
tected by 1H NMR analysis of the reaction (more is suggest-
ed regarding this observation in the conclusion). Neverthe-
less, from the halogenated pyridopyrazine 13 we undertook
the preparation of the O-acetylated derivative 27, the stabi-
lized precursor of the target compound 9. This proceeded
via a Buchwald–Hartwig N-arylation of phenylalanine ethyl
ester (25) to give compound 26 in a rather modest 12%
yield. Next, a saponification followed in situ by the addition
of an excess of acetic anhydride gave the O-acetylated de-
rivative 27 in a 33% yield after chromatography.
H
N
N
O
N
NH2
Br
N
NH2
+
i
N
H
28
30
31
ii
H
N
N
O
N
N
N
Cl
iii
iv
+
N
32
33
O
O
O
N
O
N
NH
N
N
N
N
v
N
34
35
Scheme 3 Reagents and conditions: (i) PhCH2CHNH2CO2H (29), Cu2O,
K3PO4, DMSO, 90 °C; (ii) NBS, EtOH, 20 °C; (iii) PCl3, POCl3, 80 °C; (iv)
PhCH2CHNH2CO2Et (25, as a free base), NMP, 120 °C, 4 d; (v) (1) NaOH,
THF, 20 °C, (2) Ac2O, 20 °C.
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H

D
G. Gagnot et al.
Paper
Synthesis
from compound 28. Chlorination of compound 32, using
the same combination of PCl3 and POCl3 mentioned above,
then gave the halogenated derivative 33 in 61% yield. The
next step turned out to be problematic as the Buchwald–
Hartwig N-arylation of phenylalanine ethyl ester (25) with
compound 33, using our previously reported conditions,4
invariably led to extensive decomposition. It was only by
heating a solution of the phenylalanine ethyl ester (25) (as
the free base) and 33 in N-methylpyrrolidone (NMP) with-
out any palladium catalyst for four days at 120 °C that a de-
gree of N-arylation was achieved and compound 34 was
isolated in a 10% yield along with 25% of the starting mate-
rial 33. Interestingly, no cyclized product (the target lucifer-
in) could be observed in this step although we previously
reported such a reaction in another difficult instance under
similar reaction conditions.1 From this ester, a final step
gave the target O-acetylated luciferin 35, albeit in a rather
modest 29% yield as chromatography turned out to be nec-
essary.
tions into benzimidazoles.13 In our case, we suggest the
mechanism depicted in Scheme 4 to account for this exper-
imental observation.
Cl
Cl
N
N
O
N
O
P
P
Cl
Cl
O
O
i
ii
N
N
N
23
POCl2
36
37
N
N
O
O
O
PCl3
PCl3
iii
iv
O
N
NH
N
N
POCl2
POCl2
38
39
H
N
Cl
Cl
Cl
Cl
Cl
O
P
P
O
N
O
v
vi
O
N
N
POCl2
N
N
POCl2
40
41
OH
N
vii
Unexpectedly and probably because of the inherent ba-
sicity of their neighboring nitrogen, the acetyl functional-
ities of compounds 27 and 35 could not be hydrolyzed un-
der the acidic conditions we had previously used [EtOH, HCl
(10 equiv), DMSO at 50 °C for two hours],4 and even at-
tempts under harsher (acidic) conditions or longer reaction
times failed. This problem could be circumvented as the
acetyls of compounds 27 and 35 turned out to be readily
hydrolyzed in a phosphate buffer (pH 7.4), and as expected,
the resulting luciferin immediately reacted with oxygen to
decompose and produce a photon. Accordingly, the deter-
mination of the chemiluminescence emission spectra of
these two analogues was conducted using their O-acetylat-
ed derivatives immediately after their addition to this buf-
fer at a low (respectively 2.5 M and 5 M) final concentra-
tion. The use of an Aqualog® (HORIBA Instruments, Inc.), a
spectrometer endowed with a sensitive CCD detector (Wa-
ter-Raman SNR > 20,000:1, RMS method), was again1 cru-
cial, as was the addition of cetyltrimethylammonium bro-
mide (2.5 mmol) to the buffer which led to a significant en-
hancement of their signals. As outlined in the experimental
section, the max values measured for the chemilumines-
cence of compounds 27 and 35 were 546 nm and 462 nm,
respectively.
In conclusion, our attempts to prepare the four possible
aza-analogues of luciferin 2a met with quite a lot of unex-
pected problems. If the preparation and isolation of the
benzylpyridopyrazinol precursors 17, 20, 22 and 23 could
be achieved in rather modest yields, their transformation
into the four corresponding halogenated derivatives turned
out to be possible in only one instance. The rather unex-
pected rearrangement of compound 23 into the imidazopy-
ridine 24 upon treatment with reagents such as POCl3 or
SOCl2 actually provides some insights into the reactivity of
such heterocyclic systems and is actually reminiscent of
various previously reported quinoxalinone ring contrac-
N
N
H
24
Scheme 4 Suggested mechanism for the rearrangement of 23 into 24.
Reagents and conditions: (i) first addition of POCl3, (– HCl); (ii) second
addition of POCl3, (– HCl); (iii) cyclization (+ HCl); (iv) and (v) rearrange-
ment of the carbon–nitrogen bond; (vi) HCl elimination; (vii) hydrolysis
(upon treatment) of the two phosphoryl dichlorides.
The depicted second addition of phosphorus oxychlo-
ride [step (ii)] on the nitrogen of compound 23 to give inter-
mediate 37 could be the key accounting for the fragility of
the pyrazine ring and its contraction into the imidazole
component of compound 24 via penta-coordinated phos-
phorus-containing intermediates 38–40. A similar process
can also be written with thionyl chloride leading to a tetra-
coordinated sulfur intermediate.
Of note in this work is the use of the inherently reduc-
tive mixture of phosphorus oxychloride and phosphorus
trichloride to achieve the preparation of compounds 13 and
33. We are not aware of any reports describing the advan-
tages of using such reaction conditions. Concerning the
chemiluminescence study, even if the two prepared O-
acetylated proluciferin analogues 27 and 35 could not be
hydrolyzed under acidic conditions, we were able to deter-
mine their emission spectra by adding them to a neutral
phosphate buffer, which led in situ to the hydrolysis of their
acetyl moiety. Interestingly, from compound 27, the lucifer-
in 9 turned out be endowed with yellow-colored chemilu-
minescence spectra (max of 546 nm), being even better
than the green-emitting compound 2b (max = 513 nm). On
the other hand, in the same phosphate buffer, the benzo de-
rivative 35 gave a blue signal at 462 nm, thus negating any
gain from the insertion of the nitrogen in this heterocyclic
system. In conclusion, the shift in emission observed from
the O-acetylated analogue 27 is encouraging, although the
chemistry to generate further derivatives will need to
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H

E
G. Gagnot et al.
Paper
Synthesis
address the difficulties described in this report. In any case,
the determined max experimental values should again4 be
useful for the design and validation of computer-based
models14 of such light-producing decarboxylation reactions.
Condensation of 3,4-Diaminopyrazine (15) and Phenylpyruvic
Acid (16) under Neutral Conditions
In a 60 mL tube fitted with a Teflon-covered screw cap, 3,4-diamino-
pyrazine (15) (1.82 g, 16.67 mmol) and phenylpyruvic acid (16) (2.73
g, 16.67 mmol) were dispersed in ethanol (40 mL, dried over 4 Å mo-
lecular sieves). The tube was closed tightly and heated at 85 °C for 24
hours. The resulting suspension was dispersed in boiling ethanol (100
mL), left to cool and filtered, and this material was subjected again to
this treatment to give, after drying, compound 18 (0.98 g) as a white
powder which could not be further purified, probably because of the
onset of an oxidative decarboxylation characterized by the appear-
ance in LC/MS of a compound with a mass of m/z = 210. Both filtrates
were collected, concentrated to dryness and purified by chromatogra-
phy over silica gel (dichloromethane/ethanol, 97:3 to 94:6) to yield
compound 17 (1.37 g, 34%) as a white powder after dispersion of the
corresponding fraction in boiling ethanol (100 mL) and filtration fol-
lowing cooling.
When specified, the anhydrous solvents used were purchased. Exper-
iments under inert atmospheres were carried out by purging the
glassware with a stream of dry argon. Then, an argon balloon, fitted
with a needle, was used to insure a positive pressure of inert gas
during the reaction. Column chromatography was performed either
on Merck silica gel 60 (0.035–0.070 mm) or on neutral alumina con-
taining 1.5% of added water using a solvent pump and an automated
collecting system driven by a UV detector set to 254 nm, unless re-
quired otherwise. Sample deposition was carried out by absorption of
the mixture to be purified on a small amount of the solid phase fol-
lowed by its deposition on the top of the column. Unless stated other-
wise, a purity of at least 95% was obtained for all the compounds by
means of chromatography or recrystallization, and this level of purity
was established by TLC, LC/MS and NMR spectroscopy.
3-Benzylpyrido[3,4-b]pyrazin-2-ol (17)
Note: the structure of this compound (recrystallized in ethanol) was
further established by X-ray crystallography.
The X-ray-based structural elucidation of compounds 17 (CCDC
2056217) and 22 (CCDC 2056386) was achieved using the beamline
PROXIMA-2 at Synchrotron SOLEIL. The data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/structures. 1H NMR and 13C NMR spectra were
recorded on a Bruker Avance 400 spectrometer at 400 MHz and 100
MHz, respectively. Chemical shifts () are given in ppm with respect
to the TMS signal and cross-coupling constants (J) are given in hertz.
Low-resolution mass spectra were obtained on an Agilent 1100 series
LC/MSD system using an atmospheric electrospray ionization system
or an Agilent 1200 series LC/MSD system employing an Agilent Jet-
Stream atmospheric electrospray ionization system. High-resolution
mass spectrometry (HRMS) was performed using a Waters Micro-
mass Q-Tof with an electrospray ion source. Elemental analysis were
performed by the Service de Microanalyse de l’ICSN, CNRS, Gif sur
Yvette, France.
1H NMR (400 MHz, DMSO-d6):  = 12.61 (br s, 1 H), 8.86 (br s, 1 H),
8.45 (d, J = 5.5 Hz, 1 H), 7.35–7.27 (m, 3 H), 7.23–7.18 (m, 2 H), 4.14 (s,
2 H).
13C NMR (100 MHz, DMSO-d6):  = 162.8, 155.3, 150.2, 149.0, 137.9,
137.4, 129.7, 128.8, 128.6, 126.9, 39.5.
HRMS (ESI): m/z [M + H]+ calcd for C14H12N3O: 238.0980; found:
238.0975.
2-Benzyl-2,3-dihydro-1H-imidazo[4,5-c]pyridine-2-carboxylic
Acid (18)
1H NMR (400 MHz, DMSO-d6, still containing some EtOH):  = 9.20 (br
s, 1 H), 7.48 (d, J = 6.1 Hz, 1 H), 7.23–7.21 (m, 2 H), 7.18–7.09 (m, 3 H),
6.96 (s, 1 H), 6.89 (br s, 1 H), 6.17 (d, J = 6.2 Hz 1 H), 3.03 (m, 2 H).
13C NMR (100 MHz, DMSO-d6): not soluble enough.
HRMS (ESI): m/z [M + H]+ calcd for C14H14N3O2: 256.1086; found:
256.1081.
The time-dependent chemiluminescence signals were recorded using
an Aqualog (HORIBA Instruments, Inc.) equipped with a thermoelec-
trically cooled charge-coupled device (CCD) array detector. The sys-
tem uses a xenon arc light source to excite samples and two detectors
for simultaneous absorbance, transmission and emission analysis was
adapted to the chemiluminescence analysis by blocking the excitation
source. The 2D emission spectra were recorded from 250 to 800 nm
using 5 nm steps, high gain and 30 s integration times for emission
detection. A detector accumulation of 2 was chosen to improve the
signal-to-noise ratio and increase the photon counts. Chemilumines-
cence data were corrected for the instrumental emission spectral re-
sponse, detector dark currents and blank sample emission. All sam-
ples were analyzed using fluorescence quartz cuvettes (1 cm path
length), and the signal was recorded immediately after addition of ei-
ther 1 or 2 L of a 5 mM stock solution in DMSO of the O-acetylated
luciferin in 2 mL of the phosphate buffer DPBS GIBCO 1X, which also
contained 2.5 mM of cetyltrimethylammonium bromide. The chemi-
luminescence data were analyzed using Aqualog software based on
OriginLab® and the signal was smoothed by 4 points fast Fourier
transform (FFT).
Note: the growing occurrence of another compound (m/z = 210 and,
amongst others, a singlet 1H signal at 4.23 ppm) was noticed when
trying to further purify this batch using a third dispersion.
1-(1H-Imidazo[4,5-c]pyridin-2-yl)-2-phenylethen-1-ol Hydrochlo-
ride Salt (19a)
A mixture of 3,4-diaminopyrazine (15) (2.26 g, 20.70 mmol), phen-
ylpyruvic acid (16) (3.39 g, 20.70 mmol), ethanol (40 mL) and 2 N hy-
drochloric acid (31 mL, 62.12 mmol) was heated to reflux for 2 h. This
was left to cool and the resulting precipitate was filtered off, washed
with ethanol and dried under vacuum at 60 °C to yield pure com-
pound 19a (4.75 g, 83%) as the hydrochloride salt.
1H NMR (400 MHz, DMSO-d6):  = 14.40 (br s, 1 H), 11.62 (s, 1 H),
10.72 (br s, 1 H), 8.15 (dd, J = 6.6, 1.0 Hz, 1 H), 8.03 (d, J = 1.0 Hz, 1 H),
7.56 (m, 2 H), 7.47 (m, 2 H), 7.41 (d, J = 6.6 Hz, 1 H), 7.36 (m, 1 H), 6.74
(s, 1 H).
13C NMR (100 MHz, DMSO-d6):  = 157.6, 141.6, 136.1, 133.8, 129.3
(two signals), 128.4, 127.0, 125.0, 124.6, 111.5, 109.1.
HRMS (ESI): m/z [M + H – HCl]+ calcd for C14H12N3O: 238.0980; found:
238.0985.
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H

F
G. Gagnot et al.
Paper
Synthesis
Anal. Calcd for C14H11N3O + HCl: C, 61.43; H, 4.42; N, 15.35; O, 5.85.
Found: C, 61.23; H, 4.42; N, 15.41; O, 6.14.
gave pure compound 22 (0.11 g, 4%). In another trial, extensive dis-
persion of the initial precipitate in ethyl acetate (100 mL) and water
(100 mL) followed by filtration of the remaining insoluble material
gave a solid which was purified by chromatography over silica gel (di-
chloromethane/ethanol, 97:3) to give a fraction (0.98 g) containing
the imidazo[4,5-b]pyridine 24. This was further purified by disper-
sion in boiling ethanol (40 mL) to give a pure sample of 24 (0.32 g, 7%)
as described below.
2-Benzylpyrido[3,4-b]pyrazin-3-ol (20)
Compound 19a (0.19 g, 0.69 mmol) and sodium hydroxide (0.2 g, 5
mmol) were brought to reflux in ethanol/water (1:1, 20 mL). This was
left to cool, neutralized with solid ammonium chloride diluted in wa-
ter and extracted with ethyl acetate (2 × 50 mL). The organic layer was
washed with brine (25 mL), dried over magnesium sulfate and con-
centrated to dryness to yield pure compound 20 (0.11 g, 66%) as a
beige powder. If 1H and 13C NMR analysis were run using a freshly
made sample of compound 20, the following spectra were obtained:
1H NMR (400 MHz, DMSO-d6):  = 12.59 (br s, 1 H), 8.63 (d, J = 0.6 Hz,
1 H), 8.40 (d, J = 5.3 Hz, 1 H), 7.65 (dd, J = 5.3, 0.6 Hz, 1 H), 7.34–7.27
(m, 4 H), 7.22 (m, 1 H), 4.18 (s, 2 H).
2-Benzylpyrido[2,3-b]pyrazin-3-ol (22)
Note: as described above, the structure of this compound (recrystal-
lized in ethanol) was further established by X-ray crystallography. In
1
addition, the H spectrum was similar to the substance obtained us-
ing a different (but plausibly acid-containing) synthetic approach,
which gave very optimistic yields that we could not reproduce.15
1H NMR (400 MHz, DMSO-d6):  = 12.84 (br s, 1 H), 8.48 (dd, J = 4.7,
1.7 Hz, 1 H), 8.14 (dd, J = 7.8, 1.7 Hz, 1 H), 7.37–7.27 (m, 5 H), 7.22 (m,
1 H), 4.15 (s, 2 H).
13C NMR (100 MHz, DMSO-d6):  = 164.1, 156.3, 150.1, 144.3, 137.5,
136.7, 129.7, 128.8, 127.3, 126.9, 102.1, 39.3.
13C NMR (100 MHz, DMSO-d6):  = 167.0, 154.6, 143.9, 139.1, 137.1,
135.7, 129.8, 128.8, 128.5, 127.0, 121.5, 39.7.
Following about 24 hours in the DMSO-d6 solution, the occurrence of
a second product could be seen and subtraction of the signals de-
scribed above gave the following spectra compatible with the free
base form of compound 19:
HRMS (ESI): m/z [M + Na]+ calcd for C14H11N3ONa: 260.0800; found:
260.0805.
1H NMR (400 MHz, DMSO-d6):  = 11.05 (br s, 1 H), 9.50 (br s, 1 H),
7.99 (s, 1 H), 7.92 (d, J = 5.5 Hz, 1 H), 7.52 (d, J = 7.9 Hz, 2 H), 7.43–7.39
(m, 2 H), 7.28–7.25 (m, 1 H), 7.06 (d, J = 5.5 Hz, 1 H), 6.44 (s, 1 H).
3-Benzylpyrido[2,3-b]pyrazin-2-ol (23)
1H NMR (400 MHz, DMSO-d6):  = 12.51 (br s, 1 H), 8.48 (dd, J = 4.5,
1.7 Hz, 1 H), 7.69 (dd, J = 8.2, 1.7 Hz, 1 H), 7.51 (dd, J = 8.1, 4.5 Hz, 1 H),
7.37–7.20 (m, 5 H), 4.18 (s, 2 H).
13C NMR (100 MHz, DMSO-d6):  = 164.7, 154.6, 145.3, 143.0, 137.4,
129.8, 128.8, 128.3, 126.9, 125.2, 124.6, 39.6.
13C NMR (100 MHz, DMSO-d6):  = 158.8, 144.5, 135.6, 135.2, 134.6,
129.2, 129.0, 128.7, 127.1, 122.8, 108.6, 105.2.
HRMS (ESI): m/z [M + H]+ calcd for C14H12N3O: 238.0980; found:
238.0983.
Note: when using formic acid as the vector, a peak at m/z = 256.1074
was also present.
HRMS (ESI): m/z [M + H]+ calcd for C14H12N3O: 238.0980; found:
238.0971.
Anal. Calcd for C14H11N3O: C, 70.87; H, 4.67; N, 17.71; O, 6.74. Found:
C, 70.39; H, 4.81; N, 17.63; O, 6.77.
1-(1H-Imidazo[4,5-b]pyridin-2-yl)-2-phenylethen-1-ol (24)
1H NMR (400 MHz, DMSO-d6):  = 11.06 (br s, 1 H), 9.20 (br s, 1 H),
7.80 (dd, J = 4.9, 1.4 Hz, 1 H), 7.55 (m, 2 H), 7.39 (m, 2 H), 7.24 (m, 1
H), 7.16 (dd, J = 7.7, 1.4 Hz, 1 H), 6.80 (dd, J = 7.7, 4.9 Hz, 1 H), 6.41 (m,
1 H).
13C NMR (100 MHz, DMSO-d6):  = 158.6, 141.6, 135.4, 129.8, 129.3,
129.2, 128.5, 127.0, 121.5, 121.1, 116.3, 105.1.
Condensation of 2,3-Diaminopyrazine (21) and Phenylpyruvic
Acid (16) under Neutral Conditions
In a 60 mL tube fitted with a Teflon-covered screw cap, 1,2-diamino-
benzene (21) (1.39 g, 12.73 mmol) and phenylpyruvic acid (16) (2.09
g, 12.73 mmol) were heated at 85 °C in ethanol (25 mL) for 26 hours.
The resulting solution was concentrated to dryness and the residue
purified by chromatography over silica gel (dichloromethane/ethanol,
98:2 to 96:4) to give, in that order of elution, compound 22 (1.21 g,
40%) and its isomer 23 (0.48 g, 15%) as described below.
HRMS (ESI): m/z [M + H]+ calcd for C14H12N3O: 238.0980; found:
238.0969.
2-Benzyl-3-chloropyrido[2,3-b]pyrazine (13)
Condensation of 2,3-Diaminopyrazine (21) and Phenylpyruvic
Acid (16) under Acidic Conditions
In a 60 mL tube fitted with a Teflon-covered screw cap, compound 23
(0.4 g, 1.68 mmol) was dispersed in a mixture of phosphorus oxychlo-
ride (4 mL) and phosphorus trichloride (2 mL). The tube was closed
and heated at 80 °C in an oil bath for 40 minutes. The resulting solu-
tion was diluted in ethyl acetate (100 mL), poured onto an excess of
ice and stirred for 15 minutes. The organic layer was made basic with
1 N ammonia, washed with water (30 mL) and brine (30 mL), dried
over magnesium sulfate and concentrated to dryness. The residue was
purified by chromatography over silica gel (cyclohexane/ethyl acetate,
3:1) to give compound 13 (0.35 g, 81%) as a white solid. Note: as men-
tioned in scheme 2, a 55% yield was obtained when running the reac-
tion on 0.15 g of compound 23 and heating 12h.
A mixture of 1,2-diaminobenzene (21) (1.32 g, 12.10 mmol), phen-
ylpyruvic acid (16) (2.00 g, 12.10 mmol), and 2 N hydrochloric acid
(18.1 mL, 36.29 mmol) was heated to reflux in ethanol (20 mL) for 26
hours. The resulting solution was cooled, diluted with water (100 mL),
and made basic with solid sodium hydrogen carbonate. The resulting
precipitate was filtered off and the filtrate was extracted with ethyl
acetate (2 × 50 mL). The organic layer was washed with water (30 mL)
and brine (30 mL), dried over magnesium sulfate and concentrated to
dryness. This residue and the precipitate described above were both
adsorbed on silica gel and purified by chromatography over silica gel
(dichloromethane/ethanol, 98:2) to give an unclean fraction contain-
ing compound 22 (0.25 g) and then a clean fraction of compound 23
(0.95 g, 33%). Further purification of the first fraction by a second
chromatography over silica gel (cyclohexane/ethyl acetate, 2:1 to 1:0)
1H NMR (400 MHz, CDCl3):  = 9.13 (dd, J = 4.2, 1.9 Hz, 1 H), 7.73 (dd,
J = 8.3, 4.2 Hz, 1 H), 7.73 (dd, J = 8.3, 1.9 Hz, 1 H), 7.34 (m, 5 H), 4.57 (s,
2 H).
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H

G
G. Gagnot et al.
Paper
Synthesis
13C NMR (100 MHz, CDCl3):  = 155.9, 154.1, 150.6, 149.4, 137.9,
2-Benzylpyrazino[2,3-c]isoquinolin-3-ol (32)
136.2, 136.1, 129.2, 128.6, 127.0, 125.4, 41.8.
HRMS (ESI): m/z [M + H]+ calcd for C14H11N3Cl: 256.0641; found:
256.0634.
To a 60 mL tube fitted with a Teflon-covered screw cap were added
compound 28 (1.61 g, 7.21 mmol), racemic phenylalanine (29) (3.6 g,
21.6 mmol), potassium phosphate (4.6 g, 7.21 mmol) and copper(I)
oxide (0.05 g, 0.36 mmol). Dry DMSO (15 mL) was added, the tube
closed and the contents heated using an oil bath at 90 °C for 40 hours.
The resulting orange suspension was dispersed in a mixture of water
(100 mL), ethyl acetate (50 mL), ammonium chloride (1.1 g) and 30%
ammonia (10 mL) and stirred in open air for 30 minutes. The resulting
precipitate was filtered, washed with water (2 × 50 mL) and ethyl ace-
tate (2 × 50 mL) and then dried under vacuum at 50 °C to give pure
compound 32 (0.38 g, 18%) as a white powder. The filtrate was ex-
tracted with ethyl acetate (2 × 50 mL). The organic layer was washed
with water (30 mL) and brine (30 mL), dried over magnesium sulfate
and concentrated to dryness. Chromatography over silica gel (cyclo-
hexane/ethyl acetate, 2:1) of the resulting residue gave an impure
fraction containing the dihydro derivative 31 (0.47 g). This was dis-
persed in ethanol, N-bromosuccinimide was added (0.43 g) and the
suspension was stirred for 90 minutes. The resulting precipitate was
filtered, washed with ethanol and dried under vacuum at 50 °C to give
additional compound 32 (0.26 g, 12%).
Ethyl (2-Benzylpyrido[2,3-b]pyrazin-3-yl)phenylalaninate (26)
To a 20 mL sealable vial fitted with a septum were added compound
13 (0.31 g, 1.21 mmol), phenylalanine ethyl ester hydrochloride salt
(25) (0.28 g, 1.21 mmol), cesium carbonate (1.3 g, 3.87 mmol), palla-
dium diacetate (0.013 g, 0.06 mmol) and BINAP (0.053 g, 0.084
mmol). The air was replaced by argon and, under an inert atmo-
sphere, dry acetonitrile (4 mL) was added via syringe. The mixture
was stirred and heated at 60 °C for 12 hours. The resulting suspension
was diluted in ethyl acetate (75 mL), filtered, the filtrate was concen-
trated to dryness and the residue purified by chromatography over
silica gel (cyclohexane/ethyl acetate, 2:1) to give compound 26 (0.06
g, 12%) as a colorless oil.
1H NMR (400 MHz, CDCl3):  = 8.86 (dd, J = 4.5, 2.0 Hz, 1 H), 8.25 (dd,
J = 8.1, 2.0 Hz, 1 H), 7.38 (dd, J = 8.1, 4.5 Hz, 1 H), 7.29 (m, 3 H), 7.20–
7.14 (m, 5 H), 6.89 (m, 2 H), 5.64 (d, J = 7.4 Hz, 1 H), 5.38 (m, 1 H), 4.24
(s, 2 H), 4.14 (m, 2 H), 3.34 (dd, J = 14.2, 5.4 Hz, 1 H), 3.17 (dd, J = 14.2,
5.2 Hz, 1 H), 1.22 (t, J = 7.2 Hz, 3 H).
1H NMR (400 MHz, DMSO-d6):  = 13.1 (s, 1 H), 9.29 (s, 1 H), 8.60 (m,
1 H), 8.20 (m, 1 H), 7.93 (m, 1 H), 7.68 (m, 1 H), 7.41 (m, 2 H), 7.32 (m,
2 H), 7.23 (m, 1 H), 4.25 (s, 2 H).
13C NMR (100 MHz, DMSO-d6):  = 160.2, 156.5, 154.5, 139.7, 137.8,
133.3, 132.7, 129.8, 128.8, 128.7, 126.9, 126.8, 126.4, 121.4, 119.9,
39.5.
13C NMR (100 MHz, CDCl3):  = 171.6, 152.0, 151.5, 151.0, 147.7,
137.2, 135.9, 135.0, 131.4, 129.3, 129.1, 128.6, 127.4, 126.8, 120.4,
61.5, 54.3, 41.3, 37.1, 14.2.
HRMS (ESI): m/z [M + H]+ calcd for C25H25N4O2: 413.1978; found:
413.1980.
HRMS (ESI): m/z [M + H]+ calcd for C18H14N3O: 288.1137; found:
288.1136.
6,8-Dibenzylimidazo[1,2-a]pyrido[3,2-e]pyrazin-9-yl Acetate (27)
Compound 26 (0.06 g, 0.14 mmol) and powdered sodium hydroxide
(0.023 g, 0.58 mmol) were added to a 100 mL flask. The air was re-
placed by argon and dry THF (0.2 mL) was added via syringe. The mix-
ture was stirred for 19 hours and acetic anhydride (0.2 mL, 2.18
mmol) was then added via syringe. The resulting suspension was
stirred for 2 hours, diluted in water, stirred for 15 minutes and ex-
tracted with ethyl acetate (2 × 50 mL). The organic layer was washed
with water (30 mL) and brine (30 mL), dried over magnesium sulfate
and concentrated to dryness. The residue was purified by chromatog-
raphy over silica gel (cyclohexane/ethyl acetate, 4:1) to give com-
pound 27 (0.02 g, 33%) as a pale pink glass.
1H NMR (400 MHz, CDCl3):  = 8.52 (dd, J = 4.7, 1.7 Hz, 1 H), 8.31 (dd,
J = 8.2, 1.7 Hz, 1 H), 7.60 (m, 2 H), 7.47 (dd, J = 8.2, 4.7 Hz, 1 H), 7.38–
7.18 (m, 8 H), 4.63 (s, 2 H), 4.22 (s, 2 H), 2.39 (s, 3 H).
13C NMR (100 MHz, CDCl3):  = 168.7, 155.4, 146.9, 139.5, 138.4,
137.4, 137.3, 134.5, 134.4, 131.9, 131.1, 129.8, 129.0, 128.4, 128.3,
126.6, 126.4, 122.1, 39.6, 33.1, 20.4.
2-Benzyl-3-chloropyrazino[2,3-c]isoquinoline (33)
In a 60 mL tube fitted with a Teflon-covered screw cap, compound 32
(1.4 g, 4.59 mmol) was dispersed in a mixture of phosphorus oxychlo-
ride (10 mL) and phosphorus trichloride (5 mL). The tube was closed
and heated at 120 °C in an oil bath for 23 hours. The resulting solution
was diluted in ethyl acetate (100 mL), poured onto an excess of ice and
stirred for 15 minutes. The organic layer was made basic with 1 N am-
monia, washed with water (2 × 30 mL) and brine (30 mL), dried over
magnesium sulfate and concentrated to dryness. The residue was pu-
rified by chromatography over silica gel (cyclohexane/ethyl acetate,
4:1) to give compound 33 (0.90 g, 61%) as a white solid.
1H NMR (400 MHz, CDCl3):  = 9.55 (s, 1 H), 9.11 (m, 1 H), 8.17 (m, 1
H), 8.02 (m, 1 H), 7.91 (m, 1 H), 7.44 (m, 2 H), 7.34 (m, 2 H), 7.26 (m, 1
H), 4.64 (s, 2 H).
13C NMR (100 MHz, CDCl3):  = 158.0, 154.0, 148.4, 146.5, 136.7,
133.7, 133.2, 132.2, 129.9, 129.3, 128.6, 128.5, 128.2, 126.9, 123.8,
41.6.
HRMS (ESI): m/z [M + H]+ calcd for C25H21N4O2: 409.1664; found:
409.1667.
HRMS (ESI): m/z [M + H]+ calcd for C18H13ClN3: 306.0798; found:
306.0793.
4-Bromoisoquinolin-3-amine (28)
Ethyl (2-Benzylpyrazino[2,3-c]isoquinolin-3-yl)phenylalaninate
(34)
The procedure described on page 67 of a patent8 was followed and
compound 28 (7.09 g, 87%) was obtained as a yellow powder after a
purification by chromatography over silica gel (cyclohexane/ethyl ac-
etate, 4:1).
1H NMR (400 MHz, CDCl3):  = 8.81 (s, 1 H), 7.93 (m, 1 H), 7.81 (m, 1
H), 7.67 (m, 1 H), 7.34 (m, 1 H), 5.03 (br s, 2 H).
In a 60 mL tube fitted with a Teflon-covered screw cap, compound 33
(0.43 g, 1.41 mmol) and freshly extracted phenylalanine ethyl ester
(25) (0.55 g, 2.81 mmol) were dissolved in dry N-methylpyrrolidone
(4 mL). The tube was closed and heated using an oil bath at 120 °C for
four days. The resulting black solution was diluted in saturated sodi-
um hydrogen carbonate (150 mL) and was extracted with ethyl ace-
tate (2 × 50 mL). The organic layer was washed with water (30 mL)
and brine (30 mL), dried over magnesium sulfate and concentrated to
13C NMR (100 MHz, CDCl3):  = 151.7, 150.5, 137.0, 131.9, 128.1,
124.8, 123.8, 123.6, 97.6.
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H

H
G. Gagnot et al.
Paper
Synthesis
dryness. The residue was purified by a chromatography over silica gel
(cyclohexane/ethyl acetate, 3:1) to give compound 34 (0.14 g, 10%) as
an oil. In another chromatographic fraction, pure starting material 33
(0.14 g, 25%) was recovered.
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/a-1396-8607. Su
p
p
ortingInformationS
u
p
p
ortingI
n
formati
o
n
1H NMR (400 MHz, CDCl3):  = 9.40 (s, 1 H), 9.00 (m, 1 H), 8.05 (m, 1
H), 7.88 (m, 1 H), 7.69 (m, 1 H), 7.28 (m, 5 H), 7.19 (m, 3 H), 6.94 (m, 2
H), 5.46 (m, 2 H), 4.33 (s, 2 H), 4.17 (m, 2 H), 3.38 (dd, J = 14.0, 4.9 Hz
1 H), 3.22 (dd, J = 14.0, 5.3 Hz, 1 H), 1.24 (t, J = 7.3 Hz, 3 H).
13C NMR (100 MHz, CDCl3):  = 172.0, 155.8, 151.2, 147.2, 144.6,
136.2, 135.8, 134.3, 131.3, 129.4, 129.0, 128.7, 128.6, 128.4, 127.9,
127.2, 127.0, 126.8, 126.7, 122.3, 61.4, 54.4, 41.1, 37.4, 14.1.
References
(1) Gagnot, G.; Hervin, V.; Coutant, E. P.; Goyard, S.; Jacob, Y.; Rose,
T.; Hibti, F. E.; Quatela, A.; Janin, Y. L. Chem. Eur. J. 2021, 27,
2112.
(2) (a) Inouye, S.; Shimomura, O. Biochem. Biophys. Res. Commun.
1997, 233, 349. (b) Giuliani, G.; Cappelli, A.; Vomero, S.; Anzini,
M. WO2011007314, 2011. (c) Giuliani, G.; Molinari, P.; Ferretti,
G.; Cappelli, A.; Anzini, M.; Vomero, S.; Costa, T. Tetrahedron
Lett. 2012, 53, 5114. (d) Yuan, M.-L.; Jiang, T.-Y.; Du, L.-P.; Li, M.-
Y. Chin. Chem. Lett. 2016, 27, 550. (e) Yeh, H. W.; Karmach, O.; Ji,
A.; Carter, D.; Martins-Green, M. M.; Ai, H. W. Nat. Methods
2017, 14, 971. (f) Giuliani, G.; Merolla, A.; Paolino, M.; Reale, A.;
Saletti, M.; Blancafort, L.; Cappelli, A.; Benfenati, F.; Cesca, F.
Photochem. Photobiol. 2021, 97, in press; DOI: 10.1111/php.13347.
(3) Yao, Z.; Zhang, B. S.; Steinhardt, R. C.; Mills, J. H.; Prescher, J. A.
J. Am. Chem. Soc. 2020, 142, 14080.
(4) (a) Coutant, E. P.; Goyard, S.; Hervin, V. O.; Gagnot, G.; Baatallah,
R.; Rose, T.; Jacob, Y.; Janin, Y. L. Org. Biomol. Chem. 2019, 17,
3709. (b) Coutant, E. P.; Gagnot, G.; Hervin, V. O.; Baatallah, R.;
Goyard, S.; Jacob, Y.; Rose, T.; Janin, Y. L. Chem. Eur. J. 2020, 26,
948.
HRMS (ESI): m/z [M + H]+ calcd for C29H27N4O2: 463.2129; found:
463.2124.
2,12-Dibenzylimidazo[1′,2′:1,6]pyrazino[2,3-c]isoquinolin-3-yl
Acetate (35)
Compound 34 (0.14 g, 0.30 mmol) and powdered sodium hydroxide
(0.05 g, 1.2 mmol) were added to a 100 mL flask. The air was replaced
by argon and dry THF (3 mL) was added via syringe. The mixture was
stirred for 19 hours and acetic anhydride (0.43 mL, 4.51 mmol) was
then added via syringe. The resulting suspension was stirred for 2
hours, diluted in water (50 mL), stirred for 15 minutes and extracted
with ethyl acetate (2 × 50 mL). The organic layer was washed with
water (30 mL) and brine (30 mL), dried over magnesium sulfate and
concentrated to dryness. The residue was purified by chromatogra-
phy over silica gel (cyclohexane/ethyl acetate, 6:1) to give compound
35 (0.04 g, 29%) as a beige solid.
(5) (a) Abasolo, M. I.; Bianchi, D.; Atlasovitch, F.; Gaozza, C.;
Fernandez, B. M. J. Heterocycl. Chem. 1990, 27, 157.
(b) Bekerman, D. G.; Abasolo, M. I.; Fernandez, B. M. J. Hetero-
cycl. Chem. 1992, 29, 129.
1H NMR (400 MHz, CDCl3):  = 9.07 (s, 1 H) 9.04–9.02 (m, 1 H), 8.01–
7.99 (m, 1 H), 7.87 (m, 1 H), 7.67 (m, 3 H), 7.41 (m, 2 H), 7.35 (m, 4 H),
7.26 (m, 2 H), 4.75 (s, 2 H), 4.28 (s, 2 H), 2.42 (s, 3 H).
(6) Janin, Y. L.; Huel, C.; Flad, G.; Thirot, S. Eur. J. Org. Chem. 2002,
1763.
(7) Rombouts, F. J. R.; Tresadern, G.; Buijnsters, P.; Langlois, X.;
Tovar, F.; Steinbrecher, T. B.; Vanhoof, G.; Somers, M.; Andrés, J.-I.;
Trabanco, A. A. ACS Med. Chem. Lett. 2015, 6, 282.
13C NMR (100 MHz, CDCl3):  = 168.9, 153.3, 150.3, 138.5, 137.7,
134.9, 134.8, 134.1, 134.0, 131.5, 131.4, 129.9, 129.0, 128.4, 128.3,
127.5, 127.4, 126.5, 126.3, 125.4, 123.5, 39.5, 33.4, 20.5 (one signal
missing).
HRMS (ESI): m/z [M + H]+ calcd for C29H23N4O2: 459.1821; found:
459.1822.
(8) Kato, T.; Sakamoto, T.; Kubo, A.; Sawamoto, D. EP2896613,
2016.
(9) Tanimori, S.; Kashiwagi, H.; Nishimura, T.; Kirihata, M. Adv.
Synth. Catal. 2010, 352, 2531.
(10) Scarry, S. M.; Lovell, K. M.; Frankowski, K. J.; Bohn, L. M.; Aubé, J.
J. Org. Chem. 2016, 81, 10538.
(11) (a) An, Z.; She, Y.; Yang, X.; Pang, X.; Yan, R. Org. Chem. Front.
2016, 3, 1746. (b) Rajasekar, S.; Adarsh, T. P.; Tharmalingam, N.;
Andivelu, I.; Mylonakis, E. ChemistrySelect 2019, 4, 2281.
(c) Chu, L.; Yue, X.; Qing, F. L. Org. Lett. 2010, 12, 1644.
(d) Deguchi, Y.; Kono, M.; Koizumi, Y.; Izato, Y.; Miyake, A. Org.
Process Res. Dev. 2020, 24, 1614.
Conflict of Interest
The authors declare no conflict of interest.
Funding Information
G.G. acknowledges a Ph.D. fellowship from the Université Paris Des-
cartes, Sorbonne Paris Cité. This project also benefited from the Valo-
(12) Han, C.; Kelly, S. M.; Cravillion, T.; Savage, S. J.; Nguyen, T.;
Gosselin, F. Tetrahedron 2019, 75, 4351.
express funding calls of the Institut Pasteur.
U
n
i
versiét
P
ari
s
D
escartes()Instiut
t
P
asteur()
(13) (a) Mamedov, V. A. RSC Adv. 2016, 6, 42132. (b) Mamedov, V. A.;
Zhukova, N. A.; Kadyrova, M. S.; Syakaev, V. V.; Beschastnova, T.
N.; Buzyurova, D. N.; Rizvanov, I. K.; Latypov, S. K.; Sinyashin, O.
G. J. Org. Chem. 2019, 84, 13572.
(14) Vacher, M.; Fdez Galván, I.; Ding, B. W.; Schramm, S.; Berraud-
Pache, R.; Naumov, P.; Ferré, N.; Liu, Y. J.; Navizet, I.; Roca-San-
juán, D.; Baader, W. J.; Lindh, R. Chem. Rev. 2018, 118, 6927.
(15) Wang, S. L.; Ding, J.; Jiang, B.; Gao, Y.; Tu, S. J. ACS Comb. Sci.
2011, 13, 572.
Acknowledgment
We thank Dr. Ahmed Haouz of the PF6 core facility of the Institut Pas-
teur for crystal freezing and data collection. We acknowledge SOLEIL
(Saint-Aubin, France) for provision of synchrotron radiation facilities
and we would like to thank the staff of PROXIMA-2 for assistance in
using beamline. We also thank Dr. Florence Miller from Berthold
Technologies for her support.
© 2021. Thieme. All rights reserved. Synthesis 2021, 53, A–H