Latent fluorophores based on a Mannich cyclisation trigger

Article abstract of DOI:10.1016/j.tetlet.2006.06.138

We report the synthesis of a strongly blue-fluorescent pyrazino-benz[e]indole derivative, and its utility for enzyme sensing in biological assays through an original Mannich cyclisation triggered fluorescence unveiling.

Full text of DOI:10.1016/j.tetlet.2006.06.138

Tetrahedron Letters 47 (2006) 6229–6233
Latent fluorophores based on a Mannich cyclisation trigger
a
a
b
´
Guillaume Clave, Aude Bernardin, Marc Massonneau,
Pierre-Yves Renarda,* and Anthony Romieua,*
a
`
IRCOF/LHO, Equipe de Chimie Bio-Organique, UMR 6014 CNRS, INSA de Rouen et Universite´ de Rouen, 1, rue Tesnieres,
76131 Mont-Saint-Aignan Cedex, France
b
ˆ
´
QUIDD, Technopole du Madrillet – ESIGELEC, Avenue Galilee, BP 10024, 76801 Saint Etienne du Rouvray, France
Received 19 May 2006; revised 26 June 2006; accepted 27 June 2006
Available online 14 July 2006
Abstract—We report the synthesis of a strongly blue-fluorescent pyrazino-benz[e]indole derivative, and its utility for enzyme sensing
in biological assays through an original Mannich cyclisation triggered fluorescence unveiling.
Ó 2006 Elsevier Ltd. All rights reserved.
Latent fluorophores (or pro-fluorophores) are stable
probes that unmask their intense fluorescence only by
a user-designated chemical reaction. They are especially
useful tools for basic research in the biological sciences.1
Contrary to other fluorescence based probes, those
latent fluorophores display a unique selectivity and limited
interferences associated with the probe concentration,
excitation intensity and emission sensitivity. Thus
increasing interest and efforts have been devoted lately
to the development of such latent fluorophores. Sophis-
ticated molecules that release a fluorescent coumarine,
rhodamine or benzo[a]phenoxazine derivative upon acti-
vation by an enzyme or reactive oxygen species have
been developed for detecting hydrolytic enzyme activi-
ties (esterase, protease, epoxide hydrolase, phosphatase,
etc.) in biological environments,2–8 or imaging cellular
metabolites.9–11 Furthermore, new fluorogenic dyes hav-
ing longer excitation and emission wavelength maxima
(in the far-red and near-infrared ranges) are expected
to be powerful tools for in vivo imaging applications.8,12
In this letter, we report our first results towards such a
protease detection probe. The aim is to use a non fluo-
rescent amide which, upon protease unveiling of the free
amine will yield a fluorescent product through an origi-
nal approach based on an intramolecular Mannich cyc-
lisation reaction.13 In this first approach, as a proof of
concept, we used a new aminonaphthalene derived short
wavelength fluorophore since, among the multitude of
available UV light-excitable dyes, aminonaphthalene
derivatives exhibit fluorescence properties, which are
often sensitive to the environment. Some of them such
as Dansyl 1 and ANTS 2 are currently used as cell
tracers or tools for protein structural studies.14
N
SO3-
HO3S
NH3+
SO3H
O
S O
1, Dansyl
2, ANTS
Furthermore, it had previously been demonstrated that
the quaternisation or protonation of the amino group
on substituted naphthalene derivatives resulted in a dra-
matic reduction in fluorescence intensity.15,16 We thus
selected alkylated 1,1,2-trimethylbenz[e]indole deriva-
tives such as iminium salt 3 as the pro-fluorophores,
since they should be poorly fluorescent. We then foresaw
that the hydrolysis of the carboxamide should generate
the free amine 4, which will spontaneously cyclise by
an intramolecular reaction onto the iminium moiety,
to give the highly fluorescent pyrazino-benz[e]indole
derivative 5 (Fig. 1). To our knowledge, this heterocyclic
compound has never been described to date. Only the
synthesis of compounds 6 and 7 which exhibit some
structural similarities with 5 was reported during the last
Keywords: Pro-fluorophore; Fluorescent probe; Penicillin G acylase.
*
Corresponding authors. Tel.: +33 2 35 52 24 14; fax: +33 2 35 52 29
59 (P.-Y.R.); tel.: +33 3 35 52 24 15 (A.R.); e-mail addresses: pierre-
yves.renard@univ-rouen.fr; anthony.romieu@univ-rouen.fr
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.06.138

6230
G. Clave´ et al. / Tetrahedron Letters 47 (2006) 6229–6233
30 years.17,18 However, no data about their spectro-
scopic properties are available.
With the goal in mind to develop this new class of effi-
cient and structurally simple fluorophores suitable for
various biological applications, we have first explored
the synthesis of 5 and studied its fluorescence properties
in detail. The pyrazino-benz[e]indole derivative was pre-
pared via two different convenient two-step procedures.
As shown in Scheme 1, the starting material 1,1,2-tri-
methyl-1H-benz[e]indole19,20 undergoes quaternisation
with either (3-bromopropyl)phthalimide or 3-bromo-
propylamine hydrobromide to give the iminium quater-
nary salts 8 and 9, respectively.21,22 As expected, those
two iminium salts did not display significant fluores-
cence properties. Removal of the phthalimide protective
group of 8 was achieved by treatment with hydrazine
monohydrate in methanol. Under these neutral condi-
tions, the iminium ion was readily trapped by the re-
leased primary amino group to give the fluorophore 5
which was isolated in good yield by silica gel chromato-
graphy. Interestingly, this intramolecular Mannich reac-
tion also occurred when quaternary salt 9 was treated
with a slight excess of K2CO3 in methanol. The structure
of 5 was confirmed by detailed measurements, including
MALDI-TOF mass spectrometry and NMR analyses.23
As expected, this intramolecular addition was found to
be subject to Baldwin’s rules concerning the cyclisation
aptitude of trigonal systems.24 Indeed, the same reaction
conditions applied to the aminoethyl derivative 10 led
O
N
N
Br
NH2.HBr
sealed tube, 140˚C
Br
O
CH3CN, reflux
N+
N+
Br
Br
NH2.HBr
O
9, 74%a
N
O
8, 38%
K2CO3, MeOH
50%
N2H4, MeOH
80%
H
N
N
5
ayield of crude product
Scheme 1. Synthesis of the new fluorophore.
N+
O
Br
to a 7:3 mixture of iminium salt 11 and diazabicyclic
compound 12 (Scheme 2). This result corroborates the
empirical rule namely the 5-endo-trig ring closure is less
favoured than the 6-endo-trig process leading to the
quantitative formation of 5.
N
H
R
R-C(O)-N
cleavage
3, non-fluorescent
Mannich cyclization
- HBr
H
N+
N
N
Br
NH2
N+
4, non-fluorescent
5, fluorescent
Br
NH2
11
K2CO3, MeOH
H
N
+
N+
60%, 11/12 7:3
Br
N
NH
NH2.HBr
N
10
NH
N
7
6
Figure 1. Principle of a pro-fluorophore unmasked by Mannich
cyclisation and structures of analogues of 5 already reported in the
literature.
12
Scheme 2. Mannich cyclisation of aminoethyl derivative 10.

G. Clave´ et al. / Tetrahedron Letters 47 (2006) 6229–6233
Table 1. Spectral properties of 5 in three solvents
6231
Solvent
kmax,abs (nm)
kmax,em (nm)
Stokes shift (nm)
e (dm3 molꢁ1 cmꢁ1
)
Relative QYa
Phosphate buffer
EtOH
Aqueous TFA 0.1%
245
253
266, 276
455
435
b
210
182
b
28,157
27,864
10,309, 10,123
0.95
0.25
b
a Determined at 25 °C by using DL-tryptophan in phosphate buffer (QY = 0.12, according to PhotochemCAD database available free of charge via
the internet at http://omlc.ogi.edu/spectra/) or anthracene in EtOH (QY = 0.27, according to Ref. 29) as standard.
b No detectable fluorescence under these acidic conditions.
Absorption and fluorescence properties for the pyr-
azino-benz[e]indole derivative 5 in phosphate buffer,
aqueous acidic trifluoroacetic acid and ethanol are
summarised in Table 1. The absorption and emission
spectra of 5 in phosphate buffer, as typical examples,
are displayed in Figure 2. Under the simulated physio-
logical conditions, this short-wavelength fluorophore
exhibits a strong blue fluorescence with an emission
maximum around 450 nm and, noteworthingly, a large
Stokes’ shift (210 nm). Moreover, a particularly high
quantum yield was obtained, which could be ascribed
mainly to two factors: firstly the strong rigidity of the
tetracyclic system, and secondly the efficient rehybrid-
ization through an intramolecular charge transfer
(RICT) process in the singlet excited state (S1) directed
from the N-1 of the pyrazinyl moiety to the naphthalene
ring. The observed sensitivity of fluorescence of 5
towards solvent polarity is also consistent with emission
from a polar S1 intermediate resulting from this charge
transfer process.25,26 Furthermore, as a first encouraging
result for the latent fluorescent unveiling strategy, under
acidic conditions, 5 did not produce significant fluores-
cence.23 Indeed, protonation at N-3 favours the pyrazi-
nyl ring opening leading to the formation of a
nonfluorescent iminium salt.à The disappearance of the
low-energy emission band for this latter compound
was explained by the lack of an enlarged electronic delo-
calisation within the benzindoleninium moiety.
co-solvent such as acetone or methanol,28 essential to
solubilise pro-fluorophore 15 in enzymatic assays per-
formed with high concentrations (ꢀ1 mM) in a fluoro-
genic substrate.§ However, 15 was found to be
perfectly soluble under simulated physiological condi-
tions in the concentration range (1–50 lM) suitable for
fluorometric or colorimetric enzymatic assays. Since
the selective enzymatic deprotection of the phenylaceta-
mide group of 15 generates the free primary amine 4, we
expected that it will spontaneously cyclise to give the
fluorescent pyrazino-benz[e]indole derivative 5. Scheme
3 outlines the preparation of 15. Acylation of 3-bromo-
propylamine with phenylacetyl chloride affords a 9:1
mixture of phenylacetamide derivative 13 and oxazi-
nium salt 14 which was used in the subsequent reaction
without further purification. Quaternisation reaction of
1,1,2-trimethyl-1H-benz[e]indole with 13/14 proceeds
smoothly to generate the targeted fluorogenic substrate
15 in 50% yield after work up and purification by col-
umn chromatography.
As expected, this compound was found to be nonfluo-
rescent especially in phosphate buffer.23 It was then as-
sayed with its target amidase, immobilised PA. Figure
3 showed the fluorescence emission time course for the
enzyme catalysed hydrolysis.– After addition of PA to
the substrate solution, a strong fluorescent signal gener-
ated at 455 nm indicated the catalytic cleavage of the
amide bond and release of free amine which spontane-
ously cyclised yielding highly fluorescent tetracyclic
derivative 5. Furthermore, no nonspecific cleavage of
the probe was detected in a control reaction where 15
was incubated only with the PA buffer. Further evidence
of the selectivity of the PA-initiated cleavage was pro-
vided by RP-HPLC analysis of the crude reaction mix-
ture of an independent enzymatic assay performed at a
larger scale. Indeed, after 4 h of incubation, a major
peak corresponding to the pyrazino-benz[e]indole deriv-
ative 5 was observed (yield of 5 estimated at 95%).23
Moreover, since 5 displayed its highest fluorescence effi-
ciency in simulated physiological medium, it makes this
new fluorophore a perfect tool for enzyme activities
sensing. We thus next investigated the potential biolog-
ical utility of fluorophore 5. With this goal in mind, and
as a first example, we prepared a new fluorogenic sub-
strate of penicillin amidase (penicillin G acylase, PA),
15. PA is the key enzyme in the industrial manufacturing
of semisynthetic penicillins. The enzyme catalyses the
hydrolysis of a number of structurally diverse amides
of the general form R–CO–NH–R0, where R0 can be
varied substantially. At the acyl position (R–CO), how-
ever, phenylacetyl is the superior residue.27 PA was cho-
sen since this enzyme is readily available under an
immobilised form which tolerates the use of an organic
In conclusion, we have presented the first synthesis and
fluorescence characteristics of an original pyrazino-benz-
[e]indole derivative. Preliminary experiments with a
model enzyme have clearly shown that Mannich cyclisa-
tion-triggered latent fluorophores releasing this deriva-
RICT process is promoted by the enlarged electronic delocalisation
caused by the conjugation of the lone pair of N-1 atom with p
electrons of the naphthalene ring.
à Conversion into the iminium salt was confirmed by RP-HPLC
analyses and after co-injection with a standard independently
prepared from 1,1,2-trimethyl-1H-benz[e]indole and 3-bromopropyl-
amine hydrobromide. Its structure is similar to that of compound 9,
except for the nature of counter-ion (TFA instead of Brꢁ).
§ This was the case for our enzymatic assays followed by RP-HPLC
and performed at a concentration of 2.7 mM.
– The use of a nonsoluble biocatalyst and the absence of a continuous
stirring (to avoid fluorescence signal disturbances) explain that this
enzymatic reaction has taken a long time (24 h) to go to almost
completion.

6232
G. Clave´ et al. / Tetrahedron Letters 47 (2006) 6229–6233
efforts are in progress to improve the water solubility
of this UV light-excitable dye and to shift its absorption
maximum to complex proteic media compatible wave-
lengths (ꢀ350 nm) especially by introducing sulfonate
groups onto the naphthalene ring. Thus, it will be possi-
ble to get efficient substitutes of short-wavelength fluo-
rophores commercially available from Invitrogen such
as Alexa FluorÒ 350 and Marina BlueÒ dyes.14
Acknowledgements
´ ˆ
We thank Dr. Jerome Leprince (U413 INSERM/
Figure 2. Normalised absorption and emission spectra of
phosphate buffer at 25 °C.
5 in
IFRMP 23) and Annick Leboisselier (IRCOF) for their
contribution to the MALDI-TOF mass spectrometry
measurements and the determination of elemental anal-
yses, respectively.
O
+
NH2.HBr
Br
Cl
Supplementary data
DIEA, CH3CN
Detailed synthetic procedures, spectroscopic and photo-
physical characterisations of 5 and 15. Supplementary
data associated with this article can be found, in the
online version, at doi:10.1016/j.tetlet.2006.06.138.
O
O
+
N
H
N+
Br
Br
13
References and notes
14
95%, 13/14 9:1
1. Chen, X.; Sun, M.; Ma, H. Curr. Org. Chem. 2006, 10,
477–489.
2. Grant, S. K.; Sklar, J. G.; Cummings, R. T. J. Biomol.
Screening 2002, 7, 531–540.
3. Goddard, J.-P.; Reymond, J.-L. Curr. Opin. Biotechnol.
2004, 15, 314–322.
N
4. Goddard, J.-P.; Reymond, J.-L. Trends Biotechnol. 2004,
22, 363–370.
sealed tube, 140˚C
5. Chandran, S. S.; Dickson, K. A.; Raines, R. T. J. Am.
Chem. Soc. 2005, 127, 1652–1653.
6. Wang, Z.-Q.; Liao, J.; Diwu, Z. Bioorg. Med. Chem. Lett.
2005, 15, 2335–2338, and references cited therein.
7. Huang, S.-T.; Lin, Y.-L. Org. Lett. 2006, 8, 265–268.
8. Ho, N.-H.; Weissleder, R.; Tung, C.-H. Tetrahedron 2006,
62, 578–585.
9. Chang, M. C. Y.; Pralle, A.; Isacoff, E. Y.; Chang, C. J. J.
Am. Chem. Soc. 2004, 126, 15392–15393.
10. Miller, E. W.; Albers, A. E.; Pralle, A.; Isacoff, E. Y.;
Chang, C. J. J. Am. Chem. Soc. 2005, 127, 16652–16659.
11. Xu, K.; Tang, B.; Huang, H.; Yang, G.; Chen, Z.; Li, P.;
An, L. Chem. Commun. 2005, 5974–5976.
12. Tung, C.-H.; Zeng, Q.; Shah, K.; Kim, D.-E.; Schellin-
gerhout, D.; Weissleder, R. Cancer Res. 2004, 64, 1579–
1583.
N+
Br
O
PhAc moiety
cleaved by PA
N
H
15, 50%
Scheme 3. Synthesis of fluorogenic substrate of PA.
Enzyme
Control
150
100
50
13. Overman, L. E.; Ricca, D. J. In Comprehensive Organic
Synthesis; Trost, B. M., Flemming, I., Heathcock, C. H.,
Eds.; Pergamon: Oxford, 1991; Vol. 2, pp 1007–1046.
14. Haugland, R. P. In Handbook of Fluorescence Probes and
Research Products; Molecular Probes: Eugene, OR, USA,
2002; pp 66–73, 554–572.
500
1000
Time (min)
Figure 3. Fluorescence assay for the hydrolysis of 15 (1 lM) with
immobilised PA (6.1 U) in phosphate buffer (50 mM, pH 7.5). The
fluorescence excitation and emission were 240 and 455 nm,
respectively.
15. Seliskar, C. J.; Brand, L. J. Am. Chem. Soc. 1971, 93,
5405–5414.
16. Seliskar, C. J.; Brand, L. J. Am. Chem. Soc. 1971, 93,
5414–5420.
17. Degutis, J.; Sackus, A. Khim. Geterotsikl. Soedin. 1987,
227–230.
tive could be used as proteolytic enzyme sensors (detec-
tion and quantification) in biological assays. Further

G. Clave´ et al. / Tetrahedron Letters 47 (2006) 6229–6233
6233
18. Sackus, A.; Degutis, J.; Medeksiene, H. Khim. Geterotsikl.
Soedin. 1990, 815–817.
24. Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734–
736.
19. Portoghese, P. S.; Sultana, M.; Takemori, A. E. J. Med.
Chem. 1990, 33, 1714–1720.
20. Kappe, T.; Fruhwirth, F.; Roschger, P.; Jocham, B.;
Kremsner, J.; Stadlbauer, W. J. Heterocycl. Chem. 2002,
39, 391–397.
25. Lahmani, F.; Zehnacker-Rentien, A.; Coudert, L. H.;
Zachariasse, K. A. J. Phys. Chem. A 2003, 107, 7364–
7372.
26. Cuquerella, M. C.; Miranda, M. A.; Bosca, F. J. Phys.
Chem. A 2006, 110, 2607–2612.
21. Narayanan, N.; Patonay, G. J. Org. Chem. 1995, 60,
2391–2395.
22. Narayanan, N.; Strekowski, L.; Lipowska, M.; Patonay,
G. J. Org. Chem. 1997, 62, 9387.
27. Ninkovic, M.; Riester, D.; Wirsching, F.; Dietrich, R.;
Schwienhorst, A. Anal. Biochem. 2001, 292, 228–233.
28. Grether, U.; Waldmann, H. Chem. Eur. J. 2001, 7, 959–
971.
23. See Supplementary data.
29. Melhuish, W. H. J. Phys. Chem. 1961, 65, 229–235.