Boron Dipyrromethene (BODIPY) as Electron-Withdrawing Group in Asymmetric Copper-Catalyzed [3+2] Cycloadditions for the Synthesis of Pyrrolidine-Based Biological Sensors

Article abstract of DOI:10.1002/adsc.201901465

In this work, we describe the use of Boron Dipyrromethene (BODIPY) as electron-withdrawing group for activation of double bonds in asymmetric copper-catalyzed [3+2] cycloaddition reactions with azomethine ylides. The reactions take place under smooth conditions and with high enantiomeric excess for a large number of different substituents, pointing out the high activation of the alkene by using a boron dipyrromethene as electron-withdrawing group. Experimental, theoretical studies and comparison with other common electron-withdrawing groups in asymmetric copper-catalyzed [3+2] cycloadditions show the reasons of the different reactivity of the boron dipyrromethene derivatives, which can be exploited as a useful activating group in asymmetric catalysis. Additional experiments show that the so obtained pyrrolidines can be employed as biocompatible biosensors, which can be located in the endosomal compartments and do not present toxicity in three cell lines. (Figure presented.).

Full text of DOI:10.1002/adsc.201901465

Title: Boron Dipyrromethene (BODIPY) as Electron-Withdrawing Group
in Asymmetric Copper-Catalyzed [3+2] Cycloadditions for the
Synthesis of Pyrrolidine-Based Biological Sensors
Authors: Thomas Rigotti, Juan Asenjo-Pascual, Ana Martín-Somer,
Paula Milán Rois, Marco Cordani, Sergio Díez Tendero,
Alvaro Somoza, Alberto Fraile, and Jose Julian Aleman Lara
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: Adv. Synth. Catal. 10.1002/adsc.201901465
Link to VoR: http://dx.doi.org/10.1002/adsc.201901465

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Boron Dipyrromethene (BODIPY) as Electron-Withdrawing
Group in Asymmetric Copper-Catalyzed [3+2] Cycloadditions
for the Synthesis of Pyrrolidine-Based Biological Sensors
Thomas Rigotti,a,f Juan Asenjo-Pascual,a,f Ana Martín-Somer,b Paula Milán Rois,c
Marco Cordani,c Sergio Díaz-Tendero,b,d,e Álvaro Somoza,c Alberto Frailea,d,* and José
Alemána,d,*
a
Department of Organic Chemistry (module 01), Universidad Autonoma de Madrid, Cantoblanco, 28049 Madrid, Spain.
Webpage: www.uam.es/jose.aleman. E-mail: alberto.fraile@uam.es; jose.aleman@uam.es
Department of Chemistry (module 13), Universidad Autonoma de Madrid, Cantoblanco, 28049 Madrid, Spain.
IMDEA Nanociencia, Cantoblanco, 28049 Madrid, Spain.
Institute for Advanced Research in Chemical Sciences (IAdChem), Universidad Autónoma de Madrid, 28049 Madrid,
b
c
d
Spain.
e
Condensed Matter Physics Center (IFIMAC), Universidad Autónoma de Madrid, Cantoblanco, 28049 Madrid, Spain.
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract. In this work, we describe the use of Boron cycloadditions show the reasons of the different reactivity
Dipyrromethene (BODIPY) as electron-withdrawing group of the boron dipyrromethene derivatives, which can be
for activation of double bonds in asymmetric copper- exploited as a useful activating group in asymmetric
catalyzed [3+2] cycloaddition reactions with azomethine catalysis. Additional experiments show that the so obtained
ylides. The reactions take place under smooth conditions and pyrrolidines can be employed as biocompatible biosensors,
with high enantiomeric excess for a large number of which can be located in the endosomal compartments and
different substituents, pointing out the high activation of the do not present toxicity in three cell lines.
alkene by using a boron dipyrromethene as electron-
withdrawing group. Experimental, theoretical studies and
comparison with other common electron-withdrawing
groups in asymmetric copper-catalyzed [3+2]
Keywords: asymmetric catalysis; pyrrolidines; Electron-
Withdrawing Group; bioimaging cell; Frontier Molecular
Orbitals; BODIPY; cycloaddition
modulates the nucleation of their secondary structures.
Those spatial arrangements tune their biological
behavior and protect them against degradation. Such
changes are especially relevant in therapeutic
peptides, where their selectivity and activity can be
enhanced.[7] Moreover, proline core is present in the
structure of kainoid natural neurotoxins such as -
Kainic acid (I),[8] (-)-Domoic acid (II),[9] and
Acromelic acid (III),[10] which promote potent
stimulation of the central nervous system, cause brain
damage, and neurological disorders (Figure 1).
Another related family of synthetic molecules such as
IV[11] has been identified as potent hepatitis C virus
(HCV) inhibitors (Figure 1).
Plenty of novel strategies and methods have been
developed in recent years to synthesize these
heterocycles. Among them, catalytic asymmetric 1,3-
dipolar cycloadditions of azomethine ylides with
activated alkenes have turned out to be one of the
most straightforward and efficient methods for the
preparation of enantioenriched prolines,[12] from
which, and by simple transformations, it is possible to
obtain highly functionalized pyrrolidines.[13]
Introduction
Boron dipyrromethene derivatives (BODIPYs) are a
very remarkable family of fluorescent dyes that have
been employed as powerful tools for labeling
strategies in biochemistry and molecular biology.[1]
The reasons for their success are related to their high
cell
permeability,
impressive
spectroscopic
properties[2] as well as excellent robustness and
chemical- and photo-stability.[2a,3] Owing to their
unique optical and chemical properties, BODIPYs
have been used as biological sensors and for cell
imaging.
However,
although
some
recent
methodologies allow the direct labelling of
biomolecules with the BODIPY unit,[4] generally, it is
necessary to use approaches that are based on
conventional coupling reactions from two
functionalized substrates,[5] thus requiring tedious
modifications of the target biomolecule, including the
corresponding linker.
Proline and its derivatives constitute a remarkable
family of natural and synthetic compounds with very
interesting chemical and bioactive applications.[6]
Furthermore, proline plays a crucial role in peptides
by limiting their conformational freedom, which
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
allowing their use as dipolarophiles in [3+2]
cycloadditions.[14] Nevertheless, the range of suitable
dipolarophile partners is still limited, and most of
them are related to the common EWGs in asymmetric
catalysis.
Since BODIPY core is known for its electron-
withdrawing character,[15] and can be used as
fluorescent moiety for bioimaging techniques,[1] we
wondered if it would be possible to use this
interesting group as EWG for alkene activation in
asymmetric [3+2] cycloaddition reactions. Taking
into account that BODIPYs and pyrrolidines moieties
are very important structures, this [3+2] cycloaddition
will give access to enantiomerically enriched
pyrrolidines incorporating a fluorophore dye (Scheme
1b). In this work, we show that BODIPY units can be
used as EWG in asymmetric catalysis for the
synthesis of pyrrolidines thus adding a new EWG to
the range of suitable dipolarophiles in [3+2]
Figure 1. Different biologically active proline derivatives
described in the literature.
cycloaddition
reactions.
Quantum
chemistry
calculations, comparing the reactivity with other
common EWGs, demonstrate that BODIPY is able to
activate the double bond better than any other.
Besides, we show that the so obtained pyrrolidines
incorporating a fluorophore moiety (BODIPY) could
be applied to live-cell imaging in three different
cellular cell lines.
Usually, strong electron-withdrawing groups
(EWG) (e.g. nitro, ester, sulfone, amide, phosphonate,
etc) at the alkene are needed to achieve the desired
reactivity (Scheme 1a). In a very recent report, the
activation of a double bond was achieved by a very
electron-poor aryl group, such as nitrophenyl styrenes,
Scheme 1. Comparison of the previously used approaches and present work (M/L*: Metal/Chiral Ligands).
[Cu(CH3CN)4][PF6] (15 mol%), triethylamine (30
mol%) as base, toluene as solvent and several chiral
ligands (L1-L6) (15 mol%) at room temperature
(Table 1, entries 1-6). It should be noted that in all
trials a complete exo selectivity was obtained
(dr>98:2). The best result was achieved using
Walphos (L6) as the ligand (entry 6, Table 1),
obtaining the exo-3a adduct in 90% ee and complete
Results and Discussion
Synthesis of BODIPY pyrrolidines, screening and
scope. We began our investigations carrying out the
cycloaddition reaction between iminoester 1a and
alkenyl-BODIPY 2a in the presence of
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
diastereoselectivity (dr>98:2) but with low yield. In
order to improve this result, we tested different
solvents (see S.I. for further details), obtaining the
desired product in 85% yield and with a 96% of
enantiomeric excess (ee) when the reaction was
performed in DCM (entry 7). Then, the influence of
the catalyst loading on the reaction outcome was
evaluated. The use of 10 mol% of catalyst maintained
the ee values previously obtained (entry 8), but a
lower catalytic loading (5 mol%) provoked a slight
decrease in the final enantioselectivity (90% ee, entry
9, Table 1). Considering these results, we decided to
decrease the temperature to 4 ºC, achieving the final
adduct exo-3a in high yield (89%) and excellent
enantioselectivity (98%) (entry 10, table 1).[16]
However, the use of even lower catalyst loading (2.5
mol%, entry 11, table 1) gave low ee.
Table 1. Catalysts screening and optimization of reaction conditions.[a]
Entry
L (%mol)
Solvent
T (ºC) Yield (%)[c] Exo/endo[d] ee (%)[e]
1
2
3
4
5
6
7
8
L1 (15)
L2 (15)
L3 (15)
L4 (15)
L5 (15)
L6 (15)
L6 (15)
L6 (10)
L6 (5)
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
DCM
DCM
DCM
DCM
DCM
25
25
25
25
25
25
25
25
25
4
64
58
60
49
34
33
85
nd
nd
89
nd
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
-84
6
-75
-70
-76
90
96
97
90
98
90
9
10
11
L6 (5)
L6 (2.5)
4
[a] All the reactions were performed on a 0.07 mmol scale of 2a using a dry solvent (0.5 mL)
[b]
[c]
under air atmosphere.
CuPF6: [Cu(CH3CN)4][PF6].
Isolated yield after flash-
chromatography. [d] Ratio determined by H NMR spectroscopy of the crude mixture.
1
[e]
Determined by chiral HPLC.
Once the best conditions were determined (entry
10, Table 1), we carried out the scope of the reaction
using different iminoesters 1 and alkenyl-BODIPYs 2
(Table 2). Firstly, we tested the influence of the
electronic properties and the steric effect of
substituents at the dipole precursor 1. The presence of
an electron-donating group (EDG, p-MeO, 1b) or an
EWG (p-CF3, 1c) in the aryl moiety yielded
pyrrolidines 3b-c in high yield (79-84%) and high ee
(93-98%), which indicates that the reaction tolerates
imines with different electronic properties. A more
sterically hindered imine such as 1d allowed the
synthesis of 3d with excellent enantioselectivity and
without erosion of the final yield (88%). Similarly,
the cycloaddition reactions of imines 1e-f, containing
a heteroaromatic ring, gave the corresponding
adducts 3e and 3f with high ee. The reaction also
proceeded with very high enantioselectivity (98% ee)
but with a lower yield (40%) with the cycloalkyl
substituent (1g). The reaction worked properly even
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
when the ester group of the imine was substituted for
an amide group, obtaining the pyrrolidine 3h in very
high yield and excellent ee (83% yield, 98% ee).
Then, a variety of different alkenyl-BODIPYs 2d-f
were also studied (third and fourth row). Indeed,
electron-donating and electron-withdrawing groups
on the aryl moiety of the dipolarophile were tolerated,
leading to pyrrolidines 3i (p-MeO), 3j (p-CF3) and 3k
(p-Cl) with very high enantioselectivities (93-97%
ee). The presence of two ethyl groups at C-3 and C-5
positions of the BODIPY core did not affect the
stereoselectivity (91% ee) but provoked a decrease of
the yield (53%). Finally, we were delighted to find
that alkenyl-BODIPY 2f, which contains a primary
alkyl chain substituent, also led to the pyrrolidine 3m
in high yield and very high enantioselectivity. In
addition, pyrrolidine 3n with a quaternary center was
obtained, but in low yield and enantiomeric excess.
Table 2. Scope for the synthesis of BODIPY pyrrolidines 3 from different imines 1 and double bonds 2.[a]
[a] All reactions were performed in 0.1 mmol scale of 2 using dry DCM (0.75 mL) under air atmosphere.
Isolated yields were calculated after flash-chromatography. ee were determined by chiral HPLC. [b] Yield
determined by 1H NMR using an internal standard.
The spatial arrangement of all the substituents of
the pyrrolidine moiety was determined from the
absolute configuration of the asymmetric centers of
pyrrolidine 3k, which was unequivocally assigned as
2R, 3R, 4R, 5S by X-Ray crystallographic analysis
(see bottom-right, Table 2).[17] This configuration can
be explained by an exo approach, (the more favorable
when copper salts are used)[12b] of the dipolarophile 2
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
to the bottom prochiral face (2Re,4Si) of dipole 1 (see
explanation in next section).
favorable way for the dipole to coordinate to the
catalyst complex. As it can be seen in Figure 2a, once
the metalloazomethine ylide is formed in presence of
the base (dipole), it can coordinate with Cu/L6 in two
different ways: complex I or II; both of them having
the top prochiral face blocked by the 3,5-
ditrifluoromethylaryl substituents of the Walphos
ligand. The Gibbs free energy difference (G)
between I and II is 4.15 kcal/mol, being complex I
the most stable.
DFT calculations: BODIPY as EWG,
thermodynamics and kinetics.[18] In order to shed
light into the reaction mechanism, additional
experiments and DFT calculations were carried
out.[19] We modeled the reaction between dipole 1a
and dipolarophile 2a in the presence of the catalyst
complex: [Cu(CH3CN)4][PF6]/Walphos (Cu/L6). We
started the computational study analyzing the most
Figure 2. a) Optimized structures (at B3LYP/6-31G(d,p) level of theory) for the two possible orientations of the complex
between the catalyst and dipole: Cu/L6/dipole (see Figure S4 in the Supporting Information for more details). The Gibbs
free energy difference (G) between them is 4.15 kcal/mol, being complex I the most stable. (b) Stabilizing interactions
present in complex I. (c) Mass spectrum of a mixture of Cu, ligand L6 and imine 1a (see S.I. for full mass-spectrum).
This could be attributed to a stabilizing -
interaction between the phenyl group of the imine
and a 3,5-(CF3)2Ph ring of the ligand, which is absent
in complex II (Figure 2b). In addition, an
intramolecular hydrogen bond between the carbonyl
oxygen of the imine and a hydrogen of the other 3,5-
(CF3)2Ph ring of the ligand could further stabilize
complex I. In order to check that a complex
Cu/L6/1a was formed and stable, we carried out a
m/z 993.04 amu, which corresponds to the Cu/L6
complex (Figure 2c). This evidences the existence
and stability of the Cu/L6/1a complex.
In a second step, we studied the attack of several
dipolarophiles to the previously formed dipole-
catalyst complex (complex-I). Frontier Molecular
Orbital Theory is often employed to explain the
reactivity on this kind of reactions.[20] Thus, the
HOMO-LUMO gap energy difference between the
HOMO of the dipole and the LUMO of the
dipolarophile can be related to the reaction rate
constant, being faster the reaction as the HOMO-
mass
spectroscopy
study
(see
Supporting
Information). We detected a peak at m/z 1170.11 amu
corresponding to the Cu/L6/1a complex and a peak at
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
LUMO gap decreases. Experimental conversions,[21]
after 30 minutes at r.t., and values of HOMO-LUMO
gap for different dipolarophiles are shown in Figure 3
together with the LUMO of the dipolarophiles
considered.
Figure 3. Conversions and HOMO-LUMO gap for the reaction of Cu/L6/dipole with different dipolarophiles. We have
used Kohn-Sham orbitals computed at B3LYP/6-31G(d,p) level of theory. The LUMO of the different dipolarophiles is
also shown in the bottom.
In general, we found a good correspondence
between conversion and HOMO-LUMO gap.
However, for the derivative 2a with the BODIPY unit
as EWG, the low HOMO-LUMO gap (1.06 eV) does
not correspond to the low conversion found
experimentally (25%). The BODIPY is a better EWG
than the nitro group since it triggers a larger decrease
of the LUMO energy than the nitro group (and
consequently the HOMO-LUMO gap is smaller for
the BODIPY derivative). However, conversion to
product is one quarter of that obtained with -
nitrostyrene. Therefore, these differences between
LUMO and HOMO cannot explain the observed
trend and other factors must influence the reactivity,
affecting the conversion.
This leads us to evaluate whether kinetic factors
control the reactivity. Figure 4a shows the potential
energy surface (PES) for the [3+2] cycloaddition
reaction of ylide 1a with 2a in the presence of
catalyst (Cu/L6). Figure 4b shows the corresponding
PES using -nitrostyrene as dipolarophile in the
presence of catalyst (Cu/L6) and ylide 1a. Therefore,
a direct comparison of the reaction of BODIPY 2a
and -nitrostyrene with 1a is shown (Figure 4).
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
Figure 4. Potential energy surface for the [3+2] cycloaddition reaction of ylide 1a in the presence of catalyst (CuPF6/L6)
with BODIPY alkene 2a (a), or with -nitrostyrene (b). PES computed at SMD(Dichloromethane)/wB97X-D/6-
311+G(d,p)//B3LYP/6-31G(d,p) level of theory. Relative Gibbs free energies are referred to the separate reactants
(ylide 1a, catalyst and dipolarophile ((a) BODIPY alkene 2a, (b) -nitrostyrene)). From left to right: separate
reactants, PAC, 1st transition state (TS), intermediate, 2nd TS and final pyrrolidine product. A table with the distances
of the 1st and 2nd C-C forming bonds for key structures is also shown.
The zero corresponds to the separated catalyst,
dipole and dipolarophile. The first step is the
exothermic formation of complex I between the
catalyst and the dipole (Cu/L6/ylide-1a) as
previously stated, which is 30 kcal/mol more stable
than the separate reactants. Then, complex I forms,
together with the dipolarophile (2a or -nitrostyrene),
a pre-association complex (PAC),[22] which is about 5
kcal/mol above in energy.
cycloaddition reaction is stepwise in both cases. Thus,
the first energy barrier corresponds to the formation
of the first new C-C bond leading to an intermediate.
The second energy barrier corresponds to cyclization
of the intermediate through the formation of a 2nd C-C
bond, yielding the final products. Also in both cases
this second step is reversible since the product is less
stable than the intermediate. However, there are
several quantitatively essential differences between
the reaction mechanisms of the two considered
dipolarophiles. For BODIPY derivative, the rate-
The mechanism for BODIPY substituted alkene 2a
and -nitrostyrene is qualitatively similar. The [3+2]
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
limiting step is the formation of the 2nd C-C bond. On
the contrary, for -nitrostyrene, the highest barrier
corresponds to 1st C-C bond formation being this the
rate-limiting step.
pyrrolidines-BODIPYs 3a-n showed excellent
absorption and emission properties (see Figure 7b),
similar to those described in the literature for other
BODIPY derivatives,[3] we thought that it would be
very interesting to evaluate the applicability of these
pyrrolidine-BODIPYs 3 for live-cell imaging. We
interrogated representative derivatives (3a and 3h)
for their ability to penetrate in three cellular lines,
MEL 202 (uveal melanoma cell line), PANC-1
(pancreatic cancer cell line) and MCF7 (breast cancer
cell line), and their propensity for nonspecific
background staining. The results obtained with
BODIPYs 3a and 3h were compared with the
derivative 4,[23] to determinate the influence of the
pyrrolidine moiety and to show the advantage of
having this structure joined to the BODIPY core.
First, we studied their toxicity by incubation during
24 h in the three cell lines mentioned at different
concentrations. The BODIPY derivatives 3a, 3h and
4 were non-toxic up to 10 µM (Figure 6 and S6 in the
Supporting Information). We observed a decrease in
the viability of the cells when treated with the
compounds at high concentrations (100 µM).
However, this result is due to the presence of DMSO,
required to solubilize the compounds at such
concentration, since the viability reduction is the
same when exposed only to DMSO in the three cell
lines tested. Therefore, we selected 10 µM
concentration for the next experiments.
The intermediate is markedly more stable for 2a
(BODIPY) than for -nitrostyrene. Moreover, -
nitrostyrene intermediate has about the same energy
as the PAC while BODIPY intermediate is
significantly more stable than its PAC (about 9
kcal/mol). This could be explained by the more
significant charge separation found for the BODIPY
derivative (see Figure S5 in the Supporting
Information). Since BODIPY is a better EWG than
nitro, the negative charge located in the alkene
moiety is larger for the first one (0.74 vs. 0.60,
respectively). Concomitantly, the positive charge
located in the catalyst moiety is equivalently more
substantial. The greater charge separation leads to a
more significant dipole moment in the BODIPY
alkene (21.6 D vs. 16.2 D) and a larger electrostatic
stabilization of the BODIPY intermediate.
We can also explain these quantitative differences
using the orbital picture. Thus, the 1st step is favored
for BODIPY 2a because the HOMO-LUMO gap is
smaller (1.06 vs. 1.27 eV for 2a and nitrostyrene,
respectively). However, this interaction leads to the
formation of a very stable intermediate compared to
-nitrostyrene. In addition, orbital overlap for
formation of the product from the intermediate (2nd
C-C bond) is much better for -nitrostyrene (see
Figure 5), making this process -and the overall
reaction- faster for the latter one. Furthermore, -
nitrostyrene product is more stable than BODIPY
product 3a. Probably due to higher steric congestion
at the BODIPY product.
Figure 6. BODIPYs toxicity in PANC-1 cell line. The
highest DMSO concentration employed was 2.5 %, the
same DMSO amount present in the samples at 100 µM
concentration. (a) BODIPY 3a, (b) BODIPY 3h, (c)
BODIPY 4.
Figure 5. Frontier molecular orbitals in the intermediate of
the reaction pathway computed at the B3LYP/6-31G(d,p).
(a) BODIPY alkene 2a; (b) -nitrostyrene.
Overall, BODIPY is able to activate the double
bond in the first C-C bond formation better than other
typical EWGs used for [3+2] cycloaddition reactions
in spite of its bulkiness. However, the second C-C
bond requires more energy due to the high
stabilization of the intermediate provided by the
BODIPY group.
We studied BODIPYs accumulation in the cell
using lysotracker, as an endosomal probe (Figure 7a).
The images show that BODIPYs were colocalized
with the lysotracker in PANC-1, revealing that
BODIPYs were accumulated in endosomal
compartments. Moreover, the internalization of the
BODIPY derivatives was also studied by flow
cytometry (Figure 7c). In this assay, the peaks at the
right imply better uptake by the cells. Therefore, the
BODIPY pyrrolidine derivatives as probes for
imaging. Taking into account that the synthesized
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
internalization in PANC-1 was higher with BODIPY
3a than BODIPY 3h, and the latter was better
uptaken than BODIPY 4 (Figure 7c). The same
behavior was observed in MEL 202 and MCF-7 cell
lines (see Supporting Information).
Figure 7. (a) BODIPYs localization in PANC-1 cells. In the first column, BODIPY fluorescence is showed in green, in the
second one a lysotracker Red labeling to show the endosomal compartment and in the third column the merge. Row a
corresponds to untreated PANC-1 cells. Row b to PANC-1 cells treated with 10 µM of BODIPY 3a. Row c to PANC-1
cells treated with 10 µM of BODIPY 3h and Row d to PANC-1 cells treated with 10 µM of BODIPY 4. (b) Absorption
and emission (dash-line) spectra of BODIPYs 3a-n (see S.I. for details). (c) BODIPY uptake in PANC-1. The further to
the right, the better the uptake.
This study highlights the potential use of the newly
synthesized pyrrolidines-BODIPYs for imaging
lysosomes in cells. All of them were localized in the
endosomal compartment but the cellular uptake was
very different between the different BODIPYs,
obtaining the best result with the proline BODIPY 3a
in the three different tumoral cells.
separation between the catalyst/L6/dipole complex
and the dipolarophile, leading to a great electrostatic
stabilization. The derivatives were tested in three
different cell lines revealing a selective accumulation
in the lysosomes. Therefore, due to their inherent
excellent fluorescent properties, the synthesized
BODIPY derivatives could be used for the imaging of
these organelles in live cells.
Conclusion
Experimental Section
In summary, we have reported an easy and versatile
approach for preparation of pyrrolidine-BODIPY
derivatives in high yields and excellent
stereoselectivities by an asymmetric metal-catalyzed
[3+2] cycloaddition reaction. Also, it has been shown
that the BODIPY group is capable of activating
double bonds in this type of reactions acting as an
electron-withdrawing group. DFT studies show that,
in this specific [3+2] cycloaddition, the BODIPY
moiety, although it should work as a better EWG, is
less reactive than a nitro group. This is because the
intermediate generated in the first step of the
cycloaddition is very stable in the case of the
BODIPY-alkene because of the high charge
General procedure for the [3+2] cycloaddition
An oven-dried 10 mL vial equipped with a magnetic stir
bar was charged with [Cu(CH3CN)4]PF6 (0.005 mmol, 5
mol%), (R)-Walphos ligand L6 (0.006 mmol, 6 mol%) and
the corresponding BODIPY substrate 2 (0.1 mmol). Then,
anhydrous DCM (0.75 mL) were added, followed by the
corresponding imine 2 (0.15 mmol, 1.5 equiv). The vial
was closed with a PTFE/rubber septum and the reaction
mixture was stirred at 4 °C for 15 minutes prior adding
triethylamine (0.03 mmol, 30 mol%). Stirring was
maintained at the same temperature for 12 hours. The
solvent was removed under reduced pressure and the crude
mixture was purified by flash chromatography (eluent
indicated in each case) affording the corresponding
pyrrolidines 3.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
Methyl
(2R,3R,4R,5S)-4-(5,5-difluoro-5H-44,54-
Flow cytometry assay. After the treatment the cells were
trypsinized for 5 min, then the process stopped with
complete medium. The cells were washed with PBS and
centrifugated with 177× g for 5 min in an Eppendorf
centrifuge 5804 R (Eppendorf, Hamburg, Germany). The
process was done twice. After that, the cells were analyzed
in a Beckman Coulter Cytomics 500 Flow Cytometer
(Beckman Coulter, Indianapolis, IN, USA) using 20000
cells.
dipyrrolo[1,2-c:2',1'-f][1,3,2] diazaborinin-10-yl)-3,5-
diphenylpyrrolidine-2-carboxylate (3a)
Pyrrolidine 3a was obtained following the general
procedure, from BODIPY 2a and imine 1a, after
purification by flash chromatography (CyHex:EtOAc =
2:1) as a red solid (42 mg, 89% yield). [α]20 = - 139 (c =
D
0.035, CHCl3). 1H NMR (500 MHz, CDCl3, 278 K): δ 7.80
(br s, 1H), 7.73 (br s, 1H), 7.62 (br s, 1H), 7.45 – 7.41 (m,
2H), 7.33 – 7.18 (m, 8H), 6.65 (br s, 2H), 6.20 (br s, 1H),
4.87 (d, J = 9.8 Hz, 1H), 4.68 (d, J = 9.4 Hz, 1H), 4.33 (t, J
= 9.5 Hz, 1H), 3.87 (t, J = 9.7 Hz, 1H), 3.25 (s, 3H), 3.02
(br s, 1H). 13C NMR (126 MHz, CDCl3, 278 K): δ 172.6,
148.2, 145.8, 141.8, 139.5, 139.0, 137.2, 132.3, 129.7,
129.0, 128.7, 128.5, 127.8, 127.7, 127.6, 126.4, 118.8,
118.0, 72.5, 65.9, 59.7, 58.8, 51.9. 19F NMR (282 MHz,
CDCl3, 298 K): δ -145.6 – -145.9 (m, 2F). HRMS (ESI+):
Acknowledgements
This work was supported by the Spanish Government (RTI2018-
095038-B-I00,
CTQ2016-76061-P,
SAF2017-87305-R),
Comunidad de Madrid (IND2017/IND-7809), and co-financed by
European Structural and Investment Fund. We acknowledge the
generous allocation of computing time at the CCC (UAM).
Financial support from the Spanish Ministry of Economy and
Competitiveness, through the ‘‘Maria de Maeztu’’ Program of
Excellence in R&D (MDM-2014-0377), is also acknowledged.
Asociación Española Contra el Cáncer, and IMDEA Nanociencia
acknowledge support from the 'Severo Ochoa' Programme for
Centres of Excellence in R&D (MINECO, Grant SEV-2016-0686).
F. E. and A. G. thank the Spanish Government for FPI-PhD.
fellowships and E.M.A for FPU-PhD fellowship. P.M.R thanks
the Ministry of Economy, Industry and competitiveness of Spain
for the FPI grant (BES-2017.082521). A.M.S. thanks CAM for a
postdoctoral contract (2016-T2/IND-1660).
calculated for C27H25BF2N3O2 , [M+H]+ = 472.2002; found
+
= 472.2008. The enantiomeric excess was determined by
HPLC on a Chiralpak IB column: n-Hex/i-PrOH 40:60,
flow rate 1.0 mL/min, λ = 495 nm, major = 12.06 min, minor
= 7.06 min (98% ee).
Computational details
Geometry optimizations, orbital energies, harmonic
frequency calculations, thermodynamic corrections and
intrinsic reaction coordinate (IRC) calculations were
computed with B3LYP functional combined with Pople’s
double- basis set 6-31G(d,p), which includes polarization
functions. Harmonic vibrational frequencies were
computed to characterize minima and transition states (TS)
and IRCs to verify connectivity between TSs and adjacent
minima at the same level of theory - B3LYP/6-31G(d,p).
More accurate values for the final energies were computed
by means of single point calculations with -B97X-D
functional and a larger, triple-, basis set including diffuse
functions, 6-311+G(d,p), over the geometries previously
optimized. The range-separated hybrid functional -B97X-
D includes Grimme’s D2 dispersion correction, absent in
B3LYP functional. The effect of the solvent (CH2Cl2) was
also taken into account using the SMD continuum
solvation model.
References
[1] a) S. Kolemen, E. U. Akkaya, Coord. Chem. Rev. 2018,
354, 121-134; b) S. Krajcovicova, J. Stankova, P.
Dzubak, M. Hajduch, M. Soural, M. A. Urban, Chem.
Eur. J. 2018, 24, 4957-4966; c) R. Lincoln, L. E.
Greene, W. Zhang, S. Louisia, G. Cosa, J. Am. Chem.
Soc. 2017, 139, 16273-16281; d) T. Kowada, H. Maeda,
K. Kikuchi, Chem. Soc. Rev. 2015, 44, 4953-4972; e) P.
Rivera-Fuentes, S. J. Lippard, Acc. Chem. Res. 2015,
48, 2927-2934; f) Y. Ni, J. Wu, Org. Biomol. Chem.
2014, 12, 3774-3791.
Biological studies
Viability assays: alamar blue. After treating the cells
with the solutions of the compounds, a stock solution of
resazurin sodium salt (Sigma-Aldrich, St. Louis, MO,
USA) (1 mg/mL) in PBS was diluted 1% (v/v) in complete
RPMI medium and added to the cells. After 3 h in the
incubator (37 ˚C), the fluorescence was measured at 25 ˚C
in a plate reader Synergy H4 Hybrid reader (BioTEK) λex
= 550 nm, λem = 590 nm. The fluorescent intensity
measurements were processed using the following
equation: % Cell viability = ((Sample data - Negative
control) / (Positive control - Negative control)) ×100. The
positive control corresponds with untreated cells. A
resazurin solution without cells was used as negative
control.
[2] a) G. Ulrich, R. Ziessel, A. Harriman, Angew. Chem.
Int. Ed. 2008, 47, 1184-1201; b) N. Boens, V. Leen, W.
Dehaen, Chem. Soc. Rev. 2012, 41, 1130-1172; c) A.
Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891-4932.
[3] N. Boens, B. Verbelen, W. Dehaen, Eur. J. Org. Chem.
2015, 6577-6595.
[4] a) W. Wang, M. M. Lorion, O. Martinazzoli, L.
Ackermann, Angew. Chem. Int. Ed. 2018, 57, 10554-
10558; b) L. Mendive-Tapia, C. Zhao, A. R. Akram, S.
Preciado, F. Albericio, M. Lee, A. Serrels, N. Kielland,
N. D. Read, R. Lavilla, M. Vendrell, Nat. Commun.
2016, 7, 10940.
Localization assay. The cells were harvested on coverslips.
Then the bodipys were added in a final concentration of 10
µM. 24h later the cells were washed twice with PBS.
Finally, 48 h after the treatment LisoTracker red DND-99
was dissolved in optiMEM to 50 nM final concentration.
The solution was incubated 5 min with the cells at room
temperature and the cells were then washed twice with
PBS. Then, the samples were studied in a Leica DMI3000
M inverted microscope (Leica, Wetzlar, Germany) at 800
exposure units. The bodibys signal was observed with the
green filter of the microscope (λex 460-500 nm; λem 515-
545 nm) and the lysotracker with the red filter (λex 540-580
nm; λem 615-680 nm). The images were modified for a
correct visualization with Fiji-Image J program.
[5] a) P. S. Deore, D. V. Soldatov, R. A. Manderville, Sci.
Rep. 2018, 8, 16874; b) M. H. Y. Cheng, H. Savoie, F.
Bryden, R. W. Boyle, Photochem. Photobiol. Sci. 2017,
16, 1260-1267; c) N. Zhao, T. M. Williams, Z. Zhou, F.
R. Fronczek, M. Sibrian-Vazquez, S. D. Jois, M. G. H.
Vicente, Bioconjugate Chem. 2017, 28, 1566−1579; d)
L. C. D. de Rezende, F. A. da Silva Emery, Orbital
Elec. J. Chem. 2013, 5, 62-83.
[6] a) A. A. Morgan, E. Rubenstein, PLOS One 2013, 8,
e53785; b) G. Wu, F. W. Bazer, R. C. Burghardt, G. A.
10
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
Johnson, S. Woo Kim, D. A. Knabe, P. Li, X. Li, J. R.
[15] a) A. Guerrero-Corella, J. Asenjo-Pascual, T.
Janardan Pawar, S. Díaz-Tendero, A. Martín-Sómer, C.
Villegas Gómez, J. L. Belmonte-Vázquez, D. E.
Ramírez-Ornelas, E. Peña-Cabrera, A. Fraile, D. Cruz
Cruz, J. Alemán, Chem. Sci. 2019, 10, 4346-4351; b) Y.
Liu, X. Lv, M. Hou, Y. Shi, W. Guo, Anal. Chem. 2015,
87, 11475-11483.
McKnight, M. Carey Satterfield, T. E. Spencer, Amino
Acids 2011, 40, 1053-1063; c) S Kaul, S. S. Sharma, I.
K. Mehta, Amino Acids 2008, 34, 315-320.
[7] a) Y. Che, G. R. Marshall, Biopolymers 2006, 81, 392-
406; b) S. M. Cowell, Y. S. Lee, J. P. Cain, V. J. Hruby,
Curr. Med. Chem. 2004, 11, 2785-2798; c) P. Karoyan,
S. Sagan, O. Lequin, J. Quancard, S. Lavielle, G.
Chassaing in: Targets in Heterocyclic Systems Vol. 8
(Eds.: O. A. Attanasi, D. Spinelli), RSC, Cambridge,
2004, pp 216-273; d) C. Toniolo, M. Crisma, F.
Formaggio, C. Peggion, Biopolymers (Pept. Sci.) 2001,
60, 396-419.
[16] We carried out the reaction at -30 ºC but the
conversion was very low and the ee value was similar
to the obtaining at 4 ºC.
[17] CCDC 1899217 (3k) contains the crystallographic
data. These data can be obtained free of charge at
www.ccdc.cam.ac.uk.
[8] X.-Y. Zheng, H.-L. Zhang, Q. Luo, J. Zhu, J. Biomed.
Biotechnol. 2011, 2011, Article ID 457079.
[18] a) All the optimized structures computed can be
visualized
and
download
from
M.
[9] O. M. Pulido, Mar. Drugs 2008, 6, 180-219.
http://dx.doi.org/10.19061/iochem-bd-8-6;
b)
Álvarez-Moreno, C. de Graaf, N. Lopez, F. Maseras, J.
M. Poblet, C. Bo, J. Chem. Inf. Model. 2015, 55, 95-
103.
[10] H. Shinozaki, M. Ishida, Y. Gotoh, S. Kwak in:
Amino Acids (Eds.: G. Lubec, G. A. Rosenthal),
Springer, Dordrecht, 1990, pp 281-293.
[19] Gaussian 09, Revision E.01, M. J. Frisch et al.,
[11] F. Gray, L. Amphlett, H. Bright, L. Chambers, A.
Cheasty, R. Fenwick, D. Haigh, D. Hartley, P. Howes,
R. Jarvest, F. Mirzai, F. Nerozzi, N. Parry, M. Slater, S.
Smith, P. Thommes, C. Wilkinson, E. Williams, 42nd
Meeting of the European Association for the Study of
Liver Diseases, Barcelona, Spain, April 11–15, 2007.
Gaussian, Inc., Wallingford CT, 2013.
[20] I. Fleming, Frontier Orbitals and Organic Chemical
Reactions, Wiley, London, 1978.
[21] For cycloaddition with maleimide, -nitrostyrene and
dimethyl fumarate, see: S. Cabrera, R. Gómez Arrayás,
J. C. Carretero, J. Am. Chem. Soc. 2005, 127, 16394-
16395. For cycloaddition with (E)-4-phenylbut-3-en-2-
one, see: b) J. Hernández-Toribio, R. Gómez Arrayás,
B. Martín-Matute, J. C Carretero, Org. Lett. 2009, 11,
393-396.
[12] a) G. Pandey, P. Banerjee, S. R. Gadre, Chem. Rev.
2006, 106, 4484-4517; b) J. Adrio, J. C. Carretero,
Chem. Commun. 2011, 47, 6784-6794; c) J. Adrio, J. C.
Carretero, Chem. Commun. 2014, 50, 12434-12446; d)
T. Hashimoto, K. Maruoka, Chem. Rev. 2015, 115,
5366-5412.
[22] The importance of pre-association complexes (PAC)
was studied in an aminocatalytic case, see: a) E. Arpa,
M. Frías, C. Alvarado, J. Alemán, S. Díaz-Tendero, J.
Mol. Cat. A. 2016, 423, 308-318; b) F.
Aguilar‑ Galindo, A. M. Tuñón, A. Fraile, J. Alemán,
S. Diaz‑ Tendero, Theor. Chem. Acc. 2019, 138, 59; c)
A. Martín-Somer, E. M. Arpa, S. Díaz-Tendero, J.
Alemán, Eur. J. Org. Chem. 2019, 574-581.
[13] a) C. Bhat, S. G. Tilve, RSC Adv. 2014, 4, 5405-5452;
b) M. I. Calaza, C. Cativiela, Eur. J. Org. Chem. 2008,
3427-3448.
[14] a) A. Pascual-Escudero, A. de Cózar, F. P. Cossío, J.
Adrio, J. C. Carretero, Angew. Chem. Int. Ed. 2016, 55,
15334-15338; b) K. Liu, Y. Xiong, Z.-F. Wang, H.-Y.
Taoa, C.-J. Wang, Chem. Commun. 2016, 52, 9458-
9461.
[23] V. Leen, P. Yuan, L. Wang, N. Boens, W. Dehaen,
Org. Lett. 2012, 14, 6150-6153.
11
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901465
Advanced Synthesis & Catalysis
FULL PAPER
Boron Dipyrromethene (BODIPY) as Electron-
Withdrawing Group in Asymmetric Copper-
Catalyzed [3+2] Cycloadditions for the Synthesis
of Pyrrolidine-Based Biological Sensors
Adv. Synth. Catal. Year, Volume, Page – Page
Thomas Rigotti, Juan Asenjo-Pascual, Ana Martín-
Somer, Paula Milán Rois, Marco Cordani, Sergio
Díaz-Tendero, Álvaro Somoza, Alberto Fraile,*
José Alemán*
12
This article is protected by copyright. All rights reserved.