Riboflavin derivatives for enhanced photodynamic activity against Leishmania parasites

Article abstract of DOI:10.1016/j.tet.2014.11.072

Riboflavin derivatives with various substituents (O-acyl, N-methyl, N-alkylcarboxyalkyl or N-alkyl(trialkyl)ammonium) have been prepared and spectroscopically characterized (absorption, emission and fluorescence quantum yields). Their quantum yields of photosensitized singlet molecular oxygen production (Φ_Δ 0.24-0.58) and octanol/water partition coefficients (P_ow ≤0.01-11) were measured. Preliminary studies indicate that all derivatives display higher phototoxicity against the human protozoan parasite Leishmania than the parent riboflavin, with negligible toxicity in the absence of light. Their photodynamic action shows a higher correlation with P;bsubesub& than with Φbsubesub&, opening up their potential application to cutaneous diseases treatment.

Full text of DOI:10.1016/j.tet.2014.11.072

Tetrahedron 71 (2015) 457e462
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Riboflavin derivatives for enhanced photodynamic activity against
Leishmania parasites
,
b
c
a
Alexandre Vieira Silva a , Almudena Lopez-Sanchez , Helena Couto Junqueira ,
ꢀ
ꢀ
Luis Rivas c , Mauricio S. Baptista , Guillermo Orellana
,
a,
b,
*
*
*
a
~
~
~
Institute of Chemistry, University of Sao Paulo, 748 Prof. Lineu Prestes Ave., Butanta, Sao Paulo, Brazil
b Department of Organic Chemistry, Universidad Complutense de Madrid, 28040 Madrid, Spain
c
ꢀ
Department of Physico-Chemical Biology, Centro de Investigaciones Biologicas (CSIC), 28040 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 October 2014
Received in revised form 28 November 2014
Accepted 30 November 2014
Available online 4 December 2014
Riboflavin derivatives with various substituents (O-acyl, N-methyl, N-alkylcarboxyalkyl or N-alkyl(trialkyl)
ammonium) have been prepared and spectroscopically characterized (absorption, emission and fluores-
cence quantum yields). Their quantum yields of photosensitized singlet molecular oxygen production (FD
0.24e0.58) and octanol/water partition coefficients (Pow ꢀ0.01e11) were measured. Preliminary studies
indicate that all derivatives display higher phototoxicity against the human protozoan parasite Leishmania
than the parent riboflavin, with negligible toxicity in the absence of light. Their photodynamic action
shows a higher correlation with Pow than with FD, opening up their potential application to cutaneous
diseases treatment.
Keywords:
Riboflavin derivatives
Photosensitizers
Singlet oxygen
Ó 2014 Elsevier Ltd. All rights reserved.
Photodynamic therapy
Leishmaniasis
1. Introduction
prosthetic groups of flavoenzymes.5 Being natural compounds,
these molecules display low toxicity in the dark but produce
Photodynamic therapy (PDT) is a technique with scarce or nil
invasion for the treatment of different pathogenic diseases and
cancers.1 PDT involves incorporation of a photoactive compound
(called photosensitizer, PS) into the target cells that upon irradia-
tion at an appropriate wavelength within its absorption spectrum,
forms an electronically excited triplet species.2 The latter may
generate radical species by electron transfer (‘type I’ photosensiti-
zation) or, in the presence of O2, generate the highly reactive singlet
molecular oxygen (1O2) molecule by energy transfer (‘type II’
photosensitization).3 Both PDT mechanisms cause cell damage
leading to necrosis and/or apoptosis.4 The essential features of a PS
for PDT have been established: (i) an efficient absorption of light,
fast intersystem crossing and large singlet oxygen production
quantum yield; (ii) a high affinity and selectivity for the target cells;
(iii) low toxicity in the dark; (iv) enough chemical stability, and (v)
a good solubility in water.
a strong phototoxicity after absorbing light in the UV and blue re-
gions.6 Flavins undergo efficient intersystem crossing in their
lowest electronic excited state, forming transient triplet species,7
which in the presence of molecular oxygen and/or other bio-
molecules, generate 1O2 and/or radical species.8 Capitalizing on
these features, natural and artificial flavin derivatives have been
used as PSs in different applications such as water treatment,9
photodegradation of aminoacids,10 light-induced death of tumor
cells,8b,11 photodynamic treatment of keratitis,12 and blood disin-
fection.13,14 However, structure-activity relationships are missing
for flavin photosensitizers.3
Leishmaniasis is a protozoal disease produced by infection with
species of the genus Leishmania, reported in 98 countries mostly
located in tropical and subtropical areas.15 Its impact on human
health is only superseded by malaria among human protozoal
maladies, with roughly two million new cases every year, ca. 75% of
them associated to cutaneous forms of the disease and hence
amenable to PDT treatment. Dyes such as phtalocyanines and
methylene blue,16 different porphyrin scaffold surrogates absorbing
at the blue region of the spectra,17 and the porphyrin precursor 5-
aminolevulinic acid (ALA),18 have been tested as PSs against
Leishmania parasites. Nevertheless, RF was only employed for the
clearance of the microorganism in blood banks.14
Riboflavin (RF, vitamin B2, an essential nutrient in humans) and
its natural derivatives, flavin mononucleotide (FMN) and flavin
adenine dinucleotide (FAD), are present in aerobic organisms as
* Corresponding authors. E-mail addresses: luis.rivas@cib.csic.es (L. Rivas),
baptista@iq.usp.br (M.S. Baptista), orellana@quim.ucm.es (G. Orellana).
http://dx.doi.org/10.1016/j.tet.2014.11.072
0040-4020/Ó 2014 Elsevier Ltd. All rights reserved.

458
A.V. Silva et al. / Tetrahedron 71 (2015) 457e462
In spite of the above mentioned uses of flavins as PS dyes, ap-
the sought cationic flavins were obtained in 10% and 68% yields,
respectively. Such disparate chemical yields are due to the strong
difference of steric hindrance at the reactive brominated carbon
atom. These novel RF derivatives display similar absorption and
emission spectra to the parent compound, but they are significantly
more soluble in water (see below). Moreover, the cationic RF de-
rivatives 1 and 2 were also soluble in organic solvents (chloroform,
acetone, ethanol, etc.). This amphiphilic character is expected to be
useful for their implementation as pharmaceuticals.
The N-methyl RF derivative 3 has been described already and its
1O2 quantum yield production was reported to be 20% higher than
that of riboflavin.21 For the sake of comparison, we also synthesized
it. Derivatives 4 and 5 have been prepared by Banekovich and
Matuszczak, and the fluorescence spectrum of the latter reported.22
However, none of the flavins 3e5 has been tested as PS against
human pathogens. The chemical structure of all synthesized flavins
was confirmed by NMR and HRMS (see Supplementary data).
plying natural flavins in PDT displays some drawbacks, mainly re-
lated to their low solubility in physiological medium and poor
photostability. Hence, we set out to synthesize riboflavin de-
rivatives with diverse features and test them against Leishmania
promastigotes.
2. Results and discussion
2.1. Synthesis
In order to improve the efficiency and the application of flavins
in photodynamic processes, we modified RF derivatives with di-
verse motifs. The genesis of the different derivatives synthesized in
this work from the parent RF is shown in Scheme 1.
2.2. Photophysical properties and n-octanol/water partition
The electronic absorption and emission spectra of RF derivatives
1e5, TARF and TPRF are all very similar to those of riboflavin (see
Supplementary data for typical spectra exemplified for derivative
2): two absorption maximums in the UV region at 267e273 nm and
327e358 nm, and an additional maximum in the vis region at
442e449 nm.23 Their fluorescence peak appears at 515e529 nm in
methanol. Due to the largely different solubility properties of the
flavins, the spectra had to be recorded sometimes in a different
solvent (Table 1). Similarly to RF, all the prepared derivatives are
also significantly fluorescent (emission quantum yields from 0.4 to
0.5). Derivatives 3 and 5 are somewhat less fluorescent due to the
higher intersystem crossing efficiency of the former (shows
a higher 1O2 production, see below) or specific hydrogen bonding to
the solvent of the carboxylic group of the latter.
Table 1
Photophysical data of riboflavin and its derivativesa
PS
l1/nm
l2/nm
l3/nm
lmF ax/nm
FF
(ε/Mꢁ1 cmꢁ1
)
(ε/Mꢁ1 cmꢁ1
)
(ε/Mꢁ1 cmꢁ1
)
RF
TARF
TPRF
1
2
3
4
5
267(33,000)
267(32,030)
345(9050)
351(9575)
340(7850)c
358(7230)
327(8240)e
353(9810)
357(9015)
351(8775)
442(10,850)
446(13,830)
529
525
512c
521
523
515
524
522
0.39b
0.46
0.49d
0.41
0.42
0.12
0.39
0.23
270(26,700)c
270(27,400)
271(29,620)e
272(38,050)
272(32,920)
273(31,865)
446(10,780)c
446(9485)
449(9865)e
447(12,800)
449(12,060)
445(10,940)
a
In methanol, except otherwise stated; l1
,
l2 and l3 (ꢂ1 nm) are the maximums
Scheme 1. Synthesis of riboflavin derivatives. a) DMF/BrCnH2nCH2N(CH3)þ3 Brꢁ/Cs2CO3/
KI/50 ꢃC/4 h; b) DMF/CH3I/Cs2CO3/40 ꢃC/24 h; c) (CH3CO)2O/DMAP/40 ꢃC/24 h; d)
DMF/CH3(CH2)14COCl/DMAP/50 ꢃC/24 h; e) DMF/BrCH2CO2C(CH3)3/K2CO3/KI/40 ꢃC/
20 h; f) CF3CO2H/CH2Cl2/50 ꢃC/5 h; yields of isolated compounds are given (full details
of the syntheses are provided in the Experimental section).
in the absorption spectra, ε (ꢂ2%) is their molar absorption coefficient, lmF ax (ꢂ1 nm)
is the fluorescence band maximum (lexc¼445 nm) and FF is the fluorescence
quantum yield (ꢂ10%).
b
Standard from Ref. 24.
In chloroform.
In acetone.
In water.
c
d
e
A first analog was obtained by acetylation of the four alkoxy
groups of RF with good yield. Using a similar strategy, tetra-O-
palmitoyl riboflavin (TPRF) was obtained with a significantly im-
proved yield (67%) compared to those of reported methodologies
(1e8% yield).19 Acyl derivatives of RF, more lipophilic and photo-
stable than the natural compound, are known to keep its 1O2
photosensitization and fluorescence properties.20
To turn the acetylated RF into an amphiphilic photosensitizer,
tetra-O-acetyl riboflavin was decorated with a tetraalkylammo-
nium group, yielding the hitherto unknown derivatives 1 and 2.
Their synthesis was performed by nucleophilic substitution on the
desired alkyl bromide with the flavin N3 atom using nucleophilic
catalysis by Ie. After ion-exchange gel permeation chromatography,
The quantum yield of photosensitized 1O2 production by ribo-
flavin and its derivatives (FD, Table 2) was measured in air-
equilibrated solution and calculated from the area under the 1O2
emission decay curve at 1270 nm (Fig. 1), using 1H-phenalen-1-one
as a reference photosensitizer (FD¼0.97 in methanol).25
The kinetic profiles of the investigated flavins are biphasic due
to the small contribution of the tail of their luminescence. Conse-
quently, only the area under the long-lived decay component, ex-
trapolated to time zero, was taken into account for the 1O2
production quantum yield measurements.

A.V. Silva et al. / Tetrahedron 71 (2015) 457e462
459
Table 2
Quantum yield of photosensitized production of singlet oxygen (FD) and n-octanol/
water distribution ratio (Pow) of riboflavin and its derivatives
a
b
PS dye
FD
Pow
Riboflavin
TARF
TPRF
1
2
3
4
5
0.50
0.46
0.35c
0.53
0.54
0.58d
0.52
0.24
0.08
1.28
10.80
ꢀ0.01
ꢀ0.01
2.28
3.64
0.05
a
FD ꢂ10%, in methanol.
b
Pow ꢂ5% in n-octanol/aqueous Hank’s buffer.
In chloroform.
c
d
0.61 in Ref. 21.
Fig. 2. Photobiocidal properties of riboflavin and its derivatives on Leishmania major
promastigotes. Parasites were incubated with the PS dye for 1 h prior to illumination
with blue light (30 min, 470-nm 9000-mcd 15ꢃ LED, 62 J mꢁ2 sꢁ1, see Supplementary
data). Black bar: concentration of the dye that inhibits promastigote viability by 50%
(EC50); gray bar: concentration of the dye that inhibits parasite proliferation by 50%
(LC50).
Under these conditions, all the tested flavins display higher PD
activity against L. major than that of the natural RF. Although more
detailed experiments are under way, we believe that the primary
reason for the enhanced effect lies on the improved photochemical
stability of the acetylated RF derivatives.21 Moreover, the tert-butyl
ester 4 gave the best PD results. As anticipated by its high Pow ratio,
we can speculate that the higher affinity of derivative 4 for the
parasite cell membrane is the most important factor in the light-
induced cell death. Being a natural compound or close relatives of
it, riboflavin and derivatives have shown very little toxicity in the
Fig. 1. Decay of the emission at 1270 nm of the 1O2 photogenerated by laser excitation
at 355 nm of derivative 2 in methanol solution (A355¼0.10).
dark (EC50 and LC50 >100 mM). Furthermore, we observed that their
Our results are in agreement to those reported for riboflavin
(0.51) and derivative 3 (0.61) in methanol.21 With the exception of
TPRF and derivative 5, the investigated flavins display similar FD
values in the 0.46e0.58 range, regardless the nature of the sub-
stituent at N3 or acetylation of the OH groups at the ribityl side
chain. For tetra-O-palmitoyl riboflavin and derivative 5, lower FD
values were found probably due to partial aggregation in solution
driven by the long alkyl chains of the former or the specific hy-
drogen bonding to solvent of the CO2H group of the latter. Table 2
also gathers the measured n-octanol/water distribution ratio
(Pow), evidencing that solubility in aqueous media is determined by
the polarity of the chemical substituents introduced in the ribo-
flavin core. Pow is often used to predict the distribution of a given PS
between the cell membrane and the external media as a conse-
quence of the physicochemical properties of the molecule. Never-
theless, this parameter is unable to predict how other more specific
interactions are influenced by the PS structure, amphiphilicity,
functional groups and electrical charge.26 Due to their net positive
charge, the novel RF derivatives 1 and 2 show by far the highest
affinity for water, while TPRF is the most hydrophobic PS assayed.
phototoxicity (EC50) increases with the PS lipophilicity. Thus, the
affinity of the latter for the parasite cell membrane seems to be
playing the leading role in the light-induced Leishmania killing,
suggesting that the damage occurs by short-lived reactive species
such as 1O2 (Type II photosensitization).28 Nevertheless, it must be
underlined that type I mechanism might also contribute to the
photocytotoxicity of the flavin derivatives 3 and 5 as FD never ex-
ceeds 0.6 and their FF are the lowest (Table 1).29 The fact that LC50 is
markedly larger than EC50 for most of the investigated flavins in-
dicates that the damage caused by the PS can be, in some extension,
repaired by the parasite, but the damage inflicted by flavin de-
rivative 4, the best PS so far, was largely irreversible.
3. Conclusion
Our results indicate that riboflavin can be successfully derivat-
ized with N-alkylammonium and acetyl groups to provide photo-
sensitizing dyes with better photobiocidal and solubility features
than the natural compound. Their absorption in the blue region,
high 1O2 generation efficiency and low dark toxicity make them
potentially useful for their application to PDT of cutaneous diseases
(such as leishmaniasis caused by L. major), blood and water
disinfection.
2.3. Photobiocidal properties
The in vitro photodynamic (PD) activity of riboflavin and its
derivatives (except TPRF for solubility reasons) was tested against
Leishmania major promastigotes under blue light illumination. The
parasites were seeded at a density of 20ꢄ106 cells mLꢁ1 (see
Supplementary data). Cytotoxicity measurements were performed
4. Experimental details
4.1. General
in the 0.1e40
m
M range after 470 nm LED irradiation for 30 min.
Unless stated otherwise, all reagents are commercially available
(SigmaeAldrich, Acros Organics or Alfa-Aesar) and used without
further purification. Solvents were HPLC grade. Thin-layer
Promastigotes death was evaluated by the MTT method,26 after
a post-illumination period of 4 and 72 h (Fig. 2).

460
A.V. Silva et al. / Tetrahedron 71 (2015) 457e462
chromatography (TLC) was performed on pre-coated aluminum foil
plates (silica gel 60 F254 Merck). Silica gel (0.035e0.070 mm, Acros)
was used for column chromatography. SP SephadexÔ C-25 (GE
Healthcare) was employed for ion exchange column
chromatography.
1H), 7.51 (s, 1H), 5.62 (br s, 1H), 5.42e5.32 (m, 2H), 4.87 (br s, 2H),
4.39 (dd, J1¼12; J2¼3 Hz, 1H), 4.13 (dd, J1¼12; J2¼6 Hz, 1H), 2.49 (s,
3H), 2.40e2.35 (m, 6H), 2.26e2.21 (m, 3H), 2.05e1.94 (m, 2H),
1.62e1.47 (m, 16H), 1.18 (br s, 80H), 0.81 (br t, J¼6, 20H). 13C NMR
(125 MHz, CDCl3) d: 173.3, 173.0, 172.5, 172.4, 159.3, 154.2, 150.7,
Infrared spectra were recorded on a PerkineElmer 1750 FTIR
spectrometer. Nuclear Magnetic Resonance (NMR) spectra were
recorded on a Bruker AVIII 700, Bruker AV 500, or Bruker DPX 300
instrument (UCM NMR Central Instrumentation Facilities). For 1H
NMR (operating at 700 MHz, 500 MHz or 300 MHz, respectively),
147.9, 136.8, 136.1, 134.6, 133.0, 131.4, 115.7, 70.4, 69.0, 63.1, 61.8,
44.9, 34.2, 34.1, 34.0, 33.7, 32.8, 31.9, 29.7 (br), 29.7, 29.6, 29.5, 29.5,
29.4, 29.4, 29.4, 29.4, 29.3, 29.3, 29.2, 29.1, 29.1, 29.1, 28.9, 25.7, 24.9,
24.8, 24.8, 24.3, 22.7, 21.4, 19.4, 14.1. IR (KBr) cmꢁ1: 3453, 2917,
2850, 1743, 1547, 1468, 1160, 722. HRMS [ESI(þ)]: Calculated for
(C81H140N4O10þH)þ, 1300.0648; measured, 1300.0597.
d
values are referenced to (CH3)4Si (0 ppm) in CDCl3 or D2O. For 13
C
NMR (operating at 175 MHz, 125 MHz or 75 MHz, respectively),
values are referenced to CDCl3 (77.0 ppm) in CDCl3, to CH3OD
d
4.2.3. Tetra-O-acetyl-N(3)-(ethyl)trimethylammonium
riboflavin
(49.0 ppm) in CD3OD, or to (CH3)4Si (0 ppm) in D2O. Chemical shifts
are given in ppm and coupling constants (J) are given in Hertz
(multiplicity of the signal: s¼singlet, d¼doublet, dd¼double dou-
blet, t¼triplet, dt¼double triplet, quart¼quartet, quint¼quintet,
dquint¼double quintet, m¼multiplet; br¼broad signal). High Res-
olution Mass Spectra (HRMS) analyses were performed on a LCMS
Bruker Daltonics MicroTOF with ESI ionization. UVeVIS absorption
spectra were recorded on a Varian Cary 3Bio spectrophotometer.
Steady-state fluorescence spectra were recorded on a Horiba Flu-
oromax4-TCSPC spectrofluorometer.
chloride (1). In a two-necked round-bottomed flask, tetra-O-acetyl
riboflavin (136 mg, 0.25 mmol), Cs2CO3 (90 mg, 0.275 mmol), po-
tassium iodide (41 mg, 0.25 mmol) and dry N,N-dimethylforma-
mide (4 mL) were mixed under argon and stirred at 0 ꢃC for 30 min.
Then, a solution of (2-bromoethyl)trimethylammonium hexa-
fluorophosphate (see Supplementary data) in dry DMF (1 mL) was
added and the stirring was continued for 4 h at 50 ꢃC. The reaction
mixture was left to reach room temperature and a suspension of SP-
Sephadex (1.5 g) in water (10 mL) was added. The resulting slurry
was stirred further for 1 h, placed into a glass column and eluted
with aqueous NaCl solution (0e0.1 M gradient). The orange-yellow
fractions eluted using 10ꢁ3 and 10ꢁ2 M NaCl were collected. Water
was removed under vacuum and the residue was extracted with
cold methanol (0 ꢃC) (2ꢄ5 mL). The methanol was evaporated in
vacuum and the new residue was extracted with chloroform
(3ꢄ5 mL). After chloroform was removed, 28.2 mg (0.042 mmol) of
the chloride derivative 1 was obtained. Yield: 10%. 1H NMR
4.2. Synthesis of riboflavin derivatives
4.2.1. Tetra-O-acetyl riboflavin (TARF). In a two-necked round-bot-
tomed flask, riboflavin (2.0 g, 5.3 mmol), acetic anhydride (20 mL),
and 4-dimethylaminopyridine (2.6 g, 21.3 mmol) were mixed and
stirred under argon for 24 h at 40 ꢃC. After this period, the solution
was left to cool to room temperature, diluted in chloroform (60 mL)
and followed by addition of saturated aqueous NH4Cl solution
(40 mL). After phase separation, the aqueous layer was extracted
with chloroform (2ꢄ60 mL). The combined organic layers were
dried over MgSO4, filtered off and the solvent removed under
vacuum. The residue was purified by column chromatography on
silica gel, using ethyl acetate as eluent to afford tetra-O-acetyl ri-
(300 MHz, CDCl3) d: 7.95 (s, 1H), 7.55 (s, 1H), 5.60e5.55 (m, 1H),
5.42e5.31 (m, 2H), 4.94 (br s, 2H), 4.47 (br t, J¼6, 2H), 4.40 (dd,
J1¼12; J2¼3 Hz, 1H), 4.16 (dd, J1¼12; J2¼5 Hz, 1H), 3.88 (br t, J¼6,
2H), 3.49 (s, 9H), 2.50 (s, 3H), 2.39 (s, 3H), 2.18 (s, 3H), 2.14 (s, 3H),
2.01 (s, 3H), 1.64 (s, 3H). 13C NMR (125 MHz, D2O)
d: 174.0, 173.2,
174.0,173.0,162.0,156.5,151.2,149.5,139.7,135.5,134.4,132.0,131.4,
116.6, 71.0, 70.1, 70.0, 62.5, 62.5, 53.7, 45.1, 36.1, 21.2, 20.8, 20.6, 20.4,
20.0, 19.0. IR (KBr) cmꢁ1: 1743, 1687, 1549, 1232, 1206, 1131, 1049,
802. HRMS [ESI(þ)]: Calculated for (C30H40N5O10)þ, 630.2775;
measured: 630.2782.
boflavin. Yield: 75%. 1H NMR (500 MHz, CDCl3)
d: 9.12 (br s, 1H),
7.92 (s, 1H), 7.51 (s, 1H), 5.59 (br d, J¼9 Hz, 1H), 5.40e5.39 (m, 1H),
5.35e5.32 (m, 1H), 4.85 (br s, 2H), 4.36 (dd, J1¼12; J2¼3 Hz,1H), 4.18
(dd, J1¼12; J2¼6 Hz, 1H), 2.50 (s, 3H), 2.37 (s, 3H), 2.21 (s, 3H), 2.15
(s, 3H), 2.01 (s, 3H), 1.68 (br s, 3H). 13C NMR (125 MHz, CDCl3)
d:
4.2.4. Tetra-O-acetyl-N(3)-(pentyl)trimethylammonium
riboflavin
170.6, 170.3, 169.8, 169.7, 159.3, 154.9, 150.6, 148.1, 137.0, 135.9,
134.6, 132.8, 131.1, 115.6, 70.4, 69.4, 68.9, 61.8, 21.4, 21.0, 20.8, 20.6,
20.3, 18.4. IR (KBr) cmꢁ1: 3036, 1749, 1662, 1538, 1212. HRMS
[ESI(þ)]: Calculated for (C25H28N4O10þH)þ, 545.1884; measured:
545.1881.
chloride (2). In a two-necked round-bottomed flask, tetra-O-acetyl
riboflavin (272 mg, 0.5 mmol), Cs2CO3 (179.19 mg, 0.55 mmol),
potassium iodide (83 mg, 0.5 mmol) and dry DMF (8 mL) were
mixed under argon and stirred at 0 ꢃC for 30 min. Then a solution of
(5-bromopentyl)trimethylammonium
bromide
(159
mg,
0.55 mmol) in dry DMF (2 mL) was added and the stirring was
continued for 4 h at 50 ꢃC. The reaction mixture was left to reach
room temperature and a suspension of SP-Sephadex (3 g) in water
(15 mL) was added. The resulting mixture was slowly stirred fur-
ther for 1 h, then placed into a glass column and eluted with
aqueous NaCl solution (0e0.1 M gradient). The orange-yellow
fractions eluted with 10ꢁ3 and 10ꢁ2 M NaCl were collected. Water
was removed in vacuum and the solid residue was extracted with
cold methanol (0 ꢃC) (2ꢄ5 mL). The latter was evaporated under
vacuum and the solid residue was extracted with chloroform
(3ꢄ5 mL). After chloroform was removed, 276 mg (0.39 mmol) of
the chloride derivative was obtained with 90% purity. Derivative 2
was further purified by preparative HPLC (see Supplementary data)
to afford 241 mg (0.34 mmol) of product with purity higher than
4.2.2. Tetra-O-palmitoyl riboflavin (TPRF). In a two-necked round-
bottomed flask, riboflavin (250 mg, 0.66 mmol) and 4-
dimethylaminopyridine (322 mg, 2.64 mmol) were mixed and
stirred under argon in dry N,N-dimethylformamide (DMF, 4 mL) at
0
ꢃC. After 15 min, a solution of palmitoyl chloride (800
mL,
2.64 mmol) in dry DMF (4 mL) was added, and the mixture was
stirred for 12 h at 40 ꢃC. Then, an additional volume of palmitoyl
chloride solution (800 mL, 2.64 mmol) in dry DMF (4 mL) was in-
corporated to the reaction flask and the reaction stirred for 12 h.
After this period, the solution was left to cool down to room tem-
perature, diluted in dichloromethane (50 mL) and 30 mL of satu-
rated aqueous solution of NH4Cl were added. After phase
separation, the aqueous layer was extracted with chloroform
(2ꢄ30 mL). The combined organic layers were dried over MgSO4,
filtered off and the solvent was removed under vacuum. The resi-
due was purified by column chromatography on silica gel, using
hexaneeethyl acetate as eluent to afford tetra-O-palmitoyl ribo-
99% in 68% yield. 1H NMR (700 MHz, D2O)
d: 7.74 (s, 1H), 7.64 (s, 1H),
5.51e5.48 (m, 1H), 5.39 (t, J¼5 Hz,1H), 5.31e5.28 (m, 1H), 4.96 (br s,
2H), 4.36 (br d, J¼12,1H), 4.21 (dd, J1¼6; J2¼12 Hz,1H), 3.88 (dquint,
J1¼3; J2¼7 Hz, 2H), 3.18 (m, 2H), 2.96 (s, 9H), 2.43 (s, 3H), 2.29 (s,
3H), 2.08 (s, 3H), 2.05 (s, 3H), 1.90 (s, 3H), 1.73 (quint, J¼4, 2H), 1.61
flavin. Yield: 67%. 1H NMR (300 MHz, CDCl3)
d: 8.31 (s, 1H), 7.97 (s,

A.V. Silva et al. / Tetrahedron 71 (2015) 457e462
461
(quint, J¼8 Hz, 2H), 1.56 (s, 3H), 1.29 (quint, J¼8, 2H). 13C NMR
the solvent was removed under vacuum. The residue was purified
by column chromatography on silica gel using a mixture of ethyl
acetate and ethanol (gradient up to 50% of ethanol) as eluent to
(175 MHz, D2O) d: 173.5, 172.7, 172.6, 172.6, 161.2, 156.9,150.4, 148.8,
139.1,134.8,134.1,131.4, 130.9,116.1, 70.5, 69.7, 69.5, 66.4, 62.0, 52.8,
44.5, 41.5, 26.3, 228, 21.8, 20.8, 20.3, 20.1, 20.0, 19.6, 18.5. IR (KBr)
cmꢁ1: 1746, 1655, 1586, 1550, 1229, 1051. HRMS [ESI(þ)]: Calculated
for (C33H46N5O10)þ, 672.3245; measured: 672.3248.
afford derivative 5. Yield: 50%. 1H NMR (300 MHz, D2O)
d: 7.84 (s,
1H), 7.76 (s, 1H), 5.67e5.62 (m, 1H), 5.541 (br t, J¼6, 1H), 5.49e4.44
(m, 1H), 5.09 (br s, 2H), 4.53 (s, 2H), 4.52 (dd, J1¼12; J2¼3 Hz, 1H),
4.37 (dd, J1¼12; J2¼6 Hz, 1H), 2.57 (s, 3H), 2.43 (s, 3H), 2.24 (s, 3H),
4.2.5. Tetra-O-acetyl-N(3)-methyl riboflavin (3). In a two-necked
round-bottomed flask, tetra-O-acetyl riboflavin (109 mg,
0.2 mmol), Cs2CO3 (130.3 mg, 0.4 mmol) and dry DMF (5 mL) were
stirred under argon for 15 min at room temperature. Then, an ex-
cess of methyl iodide (0.5 mL) was added and the resulting solution
was stirred further for 24 h at 40 ꢃC. After this period of time, the
solution was diluted with chloroform (20 mL) and a saturated
aqueous solution of NH4Cl (20 mL) was added. After the phase
separation, the organic layer was washed with distilled water
(2ꢄ20 mL) and dried over MgSO4. The desiccant was filtered off and
the solvent was removed under vacuum. The residue was purified
by column chromatography on silica gel using a gradient of hex-
aneeethyl acetate as eluent to afford derivative 3. Yield: 51%. 1H
2.22 (s, 3H), 2.04 (s, 3H), 1.70 (s, 3H). 13C NMR (75 MHz, CD3OD)
d:
175.4, 172.3, 171.8, 171.5, 171.4, 161.7, 157.3, 151.0, 149.1, 138.4, 137.2,
136.0, 133.1, 132.7, 117.7, 71.6, 70.8, 70.7, 63.0, 45.7, 45.7, 21.3, 21.1,
20.7, 20.6, 20.4, 19.3. IR (KBr) cmꢁ1: 3447, 1746, 1653, 1586, 1549,
1378, 1231, 1052. HRMS [ESI(þ)]: Calculated for (C27H30N4O12þH)þ:
603.1939; measured: 603.1937.
4.3. Determination of the singlet oxygen production quan-
tum yields
Singlet molecular oxygen (1O2) photogenerated by the riboflavin
derivatives was quantified by monitoring its characteristic emission
in the NIR upon excitation of the samples (in air-equilibrated
methanol solutions, except for tetra-O-palmitoyl riboflavin, which
was performed in chloroform for solubility reasons) with the third
harmonic (355 nm, 2.2 mJ/pulse) of a Nd:YAG laser (Minilite II,
Continuum, CA). The 1O2 lifetime and photogeneration quantum
yields were determined by the kinetic analysis of its phosphores-
cence at 1270 nm with an Edinburgh Instruments (UK) LP-900 laser
kinetic spectrometer equipped with a NIR Hamamatsu H10330-45
PMT detector. Between sample and detector, in front of the emis-
sion monochromator, a 1270 nm wide band-pass interference filter
(Roithner-laser, Austria) was placed to isolate the 1O2 phospho-
rescence by suppressing or minimizing detection of the scattered
laser light and the sensitizer emission. Transient signals captured at
20 MS/s with a digital storage oscilloscope (Tektronix TDS 340A)
were analyzed with the Origin 8.0 software (OriginLab Corp.,
Northampton, MA). The measured lifetime for 1O2 under those
conditions was always in agreement with its typical decay in
NMR (500 MHz, CDCl3) d: 8.05 (s, 1H), 7.54 (s, 1H), 5.70e5.66 (m,
1H), 5.47 (br t, J¼5 Hz, 1H), 5.43e5.40 (m, 1H), 4.91 (br s, 2H), 4.44
(dd, J1¼12; J2¼3 Hz, 1H), 4.25 (dd, J1¼12; J2¼6 Hz, 1H), 3.50 (s, 3H),
2.56 (s, 3H), 2.44 (s, 3H), 2.30 (s, 3H), 2.22 (s, 3H), 2.08 (s, 3H), 1.74
(br s, 3H). 13C NMR (125 MHz, CDCl3)
d:170.6, 170.3, 169.9, 169.7,
159.9, 155.3, 149.1, 147.4, 136.5, 135.7, 134.7, 133.0, 131.2, 115.3, 70.5,
69.4, 69.1, 61.9, 44.5, 28.7, 21.4, 21.0, 20.8, 20.7, 20.3, 19.4. IR (KBr)
cmꢁ1: 3473,1748,1550,1372,1222,1048. HRMS [ESI(þ)]: Calculated
for (C26H30N4O10þNa)þ, 581.1860; measured: 581.1861.
4.2.6. Tetra-O-acetyl-N(3)-tert-butoxycarbonylmethyl
riboflavin
(4). In a two-necked round-bottomed flask, a mixture of tetra-O-
acetyl riboflavin (54.4 mg, 0.1 mmol), K2CO3 (16.6 mg, 0.12 mmol),
a catalytic amount of potassium iodide, and dry DMF (1 mL) was
stirred under argon at room temperature for 30 min. Then, a solution
of t-butyl-2-bromoacetate (0.1 mL, 0.69 mmol) in dry DMF (1 mL)
was added slowly and the stirring was continued for 20 h at 40 ꢃC.
The reaction mixture was diluted with dichloromethane (5 mL) and
the organic phase was washed with saturated aqueous solution of
NaHCO3 (5 mL), then water (5 mL), and then brine (5 mL). The or-
ganic extract was dried over MgSO4, filtered off and the solvent was
removed under vacuum. The residue was purified by column chro-
matography on silica gel using a mixture of dichloromethane and
ethyl acetate (1:1 v/v) as eluent to afford tetra-O-acetyl-N(3)-tert-
butoxycarbonylmethyl riboflavin. Yield: 85%. 1H NMR (300 MHz,
methanol (10 ms) and in chloroform solution (200 m
s).30 The mea-
surements were performed with all samples having the same ab-
sorbance (0.10) at 355 nm and the intensity of the 1O2 luminescence
(spectra integral) obtained for each flavin derivative (average of 5
determinations) was compared with phenalenone (perinaph-
thenone) reference standard for singlet oxygen quantum yields
measurements (FD¼0.97ꢂ0.02 in methanol and in chloroform).24
4.4. Measurement of the n-octanol/water partition
CDCl3)
d
: 7.96 (s, 1H), 7.94 (s, 1H), 5.58 (br d, J¼9 Hz, 1H), 5.40e5.31
(m, 2H), 4.80 (br s, 2H), 4.66 (s, 2H), 4.36 (dd, J1¼12; J2¼3 Hz, 1H),
4.18 (dd, J1¼12; J2¼6 Hz, 1H), 2.49 (s, 3H), 2.37 (s, 3H), 2.23 (s, 3H),
2.15 (s, 3H), 2.00 (s, 3H),1.69 (s, 3H),1.40 (br s, 9H).13C NMR (75 MHz,
The n-octanol/water distribution ratio (Pow) was measured fol-
lowing the procedure of Kessel et al.31 The photosensitizers were
dissolved separately in octanol-saturated Hanks aqueous buffer
(Hanks balanced salt solution, HBSS, made of 137 mM NaCl, 5.3 mM
KCl, 4.2 mM NaHCO3, 0.4 mM KH2PO4, 0.4 mM Na2HPO4, pH 7.2), at
CDCl3) d: 170.6, 170.3, 169.9, 169.7, 166.8, 159.4, 154.4, 149.5, 147.8,
136.7, 135.5, 134.7, 133.0, 131.2, 115.4, 82.2, 70.4, 69.5, 68.9, 61.8, 44.6,
43.5, 28.0, 21.4, 21.0, 20.8, 20.7, 20.3, 19.4. IR (KBr) cmꢁ1: 2980, 1749,
1669, 1587, 1550, 1371, 1225, 1155, 1048, 936, 854. HRMS [ESI(þ)]:
Calculated for (C31H38N4O12þNa)þ, 681.2384; measured: 681.2383.
a 5 mM concentration and the initial fluorescence intensity (Fi) of
each solution was determined. Then 1 mL of this flavin solution was
mixed with an equal volume of buffer-saturated n-octanol
(ChromasolvÒ for HPLC, SigmaeAldrich) and shaken vigorously for
2 min. The phases were separated by centrifugation. The final
fluorescence intensity (Ff) of the aqueous solution was measured
and the concentration rate was determined by the following
equation: Pow¼(FiꢁFf)/Fi
4.2.7. Tetra-O-acetyl-N(3)carboxymethyl riboflavin (5). In a two-
necked round-bottomed flask containing a solution of tetra-O-
acetyl-N(3)-tert-butoxycarbonylmethyl
riboflavin
(47
mg,
0.07 mmol) in dichloromethane (1 mL), trifluoroacetic acid (0.2 mL)
was added slowly under argon at 0 ꢃC. The solution was heated to
50 ꢃC and stirred for 5 h. The reaction mixture was then poured into
ice-water mixture, and a saturated aqueous solution of NaHCO3 was
added until pH 5. After the phase separation, the water layer was
washed with dichloromethane (3ꢄ2 mL) and the combined organic
extracts were washed with brine (3 mL) and then with water
(3 mL). After drying over MgSO4, the desiccant was filtered off and
4.5. PDT assays and cytotoxicity measurements
L. major promastigotes (Friendlin strain) were harvested at late
exponential phase, washed twice in Hanks buffer supplemented
with 10 mM
D-glucose, resuspended in the same medium and
transferred into a 96 microwell plate (Falcon Labs) to a final volume

462
A.V. Silva et al. / Tetrahedron 71 (2015) 457e462
L.; Yu, X.; Wu, W.; Zhao, J. Org. Lett. 2012, 14, 2594e2597; (d) Moussaron, A.;
of 120
m
L per well (final density of parasites: 20ꢄ106 cells mLꢁ1).
Arnoux, P.; Vanderesse, R.; Sibille, E.; Chaimbault, P.; Frochot, C. Tetrahedron 2013,
69, 10116e10122; (e) Ortiz, M. J.; Agarrabeitia, A. R.; Duran-Sampedro, G.;
Promastigotes were incubated for 1 h with the corresponding
concentration of the PS dye, followed by illumination with the
~
Banuelos Prieto, J.; Arbeloa Lopez, T.; Massad, W. A.; Montejano, H. A.; García,
custom-made 96-blue LED array (470 nm, 60 J mꢁ2 ꢁ1). After the
s
N. A.; Lopez Arbeloa, I. Tetrahedron 2012, 68, 1153e1162; (f) Yukruk, F.; Dogan,
A. L.; Canpinar, H.; Guc, D.; Akkaya, E. U. Org. Lett. 2005, 7, 2885e2887.
3. Wainwright, M. Photosensitisers in Biomedicine; Wiley-Blackwell: Chichester,
UK, 2009.
illumination, a 20
a replica plate containing 180
mL aliquot of each well was transferred into
mL of complete growth medium and
4. Oleinick, N. L.; Morris, R. L.; Belichenko, I. Photochem. Photobiol. Sci. 2002, 1,
1e21.
allowed to proliferate for 72 h in order to determine the PDT effect
on the parasite proliferation (LC50, see Supplementary data). To the
remaining parasites, MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide, Sigma) at a final concentration of
0.5 mg mLꢁ1 was added and reduction of this substrate to insoluble
formazan was allowed to proceed for 1 h to determine inhibition of
the reductase responsible for MTT reduction as the parameter to
represent parasite viability (EC50). Then, the resulting formazan
was dissolved in 1% SDS and measured in a BioRad 680 microplate
reader fitted with a cut-off filter at 595 nm, and referenced to the
control parasites. Proliferation of the parasites was measured using
the same MTT protocol. Table S1 (see Supplementary data) collects
the EC50 (enzyme concentration 50) and LC50 (lethal concentration
50) cytotoxicity parameters that indicate the concentration of the
dye that inhibits MTT by 50%, and represent the short- and long-
term effects, respectively.27 These parameters were calculated us-
ing the statistics module of the SSPS SigmaPlot software, v. 11.0.
Similar experiments were carried out without illumination to as-
sess the intrinsic toxicity of the PS assayed. Samples were made in
triplicate and the experiments repeated at least three times.
5. Massey, V. Biochem. Soc. Trans. 2000, 28, 283e296.
€
6. Drossler, P.; Holzer, W.; Penzkofer, A.; Hegemann, P. Chem. Phys. 2002, 282,
429e439.
7. Islam, S. D. M.; Penzkofer, A.; Hegemann, P. Chem. Phys. 2003, 291, 97e114.
8. (a) Siroska, E.; Khmelinskii, I.; Komasa, A.; Koput, J.; Ferreira, L. F. V.; Herance, J.
R.; Bourdelande, J. L.; Williams, S. L.; Worrall, D. R.; Insinska-Rak, M.; Sikorski,
M. Chem. Phys. 2005, 314, 239e247; (b) Edwards, A. M.; Silva, E.; Jofre, B.;
Becker, M. I.; De Ioannes, A. E. J. Photochem. Photobiol., B 1994, 24, 179e186.
9. (a) Pajaresa, A.; Bregliania, M.; Naterab, J.; Criadob, S.; Miskoskib, A.; Escalada,
J. P.; Garcíab, N. A. J. Photochem. Photobiol., A 2011, 219, 84e89; (b) Richard, A.;
Larson, R. A.; Stackhouse, P. L.; Crowley, T. O. Environ. Sci. Technol. 1992, 26,
1792e1798.
10. Remucal, C. K.; McNeill, K. Environ. Sci. Technol. 2011, 45, 5230e5237.
~
11. Munoz, M. A.; Pacheco, A.; Becker, M. I.; Silva, E.; Ebensperger, R.; Garcia, A. M.;
De Ioannes, A. E.; Edwards, A. M. J. Photochem. Photobiol., B 2011, 103, 57e67.
ꢀ
12. Szentmary, N.; Goebels, S.; Bischoff, M.; Seitz, B. Ophthalmologe 2012, 109,
165e170.
13. (a) Marschner, S.; Goodrich, R. Transfus. Med. Hemother. 2011, 38, 8e18; (b)
Wainwright, M. Chem. Soc. Rev. 2002, 31, 128e136.
14. (a) Cardo, L. J.; Rentas, F. J.; Ketchum, L.; Salata, J.; Harman, R.; Melvin, W.;
Weina, P. J.; Mendez, J.; Reddy, H.; Goodrich, R. Vox Sang. 2006, 90, 85e91; (b)
Goodrich, R. P.; Edrich, R. A.; Li, J.; Seghatchian, J. Transfus. Apher. Sci. 2006, 35,
5e17.
15. Alvar, J.; Velez, I. D.; Bern, C.; Herrero, M.; Desjeux, P.; Cano, J.; Jannin, J.; den
Boer, M. PLoS One 2012, 7, e35671 (12 pp.).
16. Baptista, M. S.; Wainwright, M. Braz. J. Med. Biol. Res. 2011, 44, 1e10.
17. Hooker, J. D.; Nguyen, V. H.; Taylor, V. M.; Cedeno, D. L.; Lash, T. D.; Jones, M. A.;
Acknowledgements
ꢀ
Robledo, S. M.; Velez, I. D. Photochem. Photobiol. 2012, 88, 194e200.
18. Enk, C. D.; Fritsch, C.; Jonas, F.; Nasereddin, A.; Ingber, A.; Jaffe, C. L.; Ruzicka, T.
The authors gratefully thank the financial support from Spanish
MINECO (CTQ2012-37573-C02-01) to G.O., FIS (PI12-02706) and
FEDER-RETICS (RICET) RD12/0018/0007 to L. R., and FAPESP (grants
12/50680-5 and 13/07937-8) to M. S. B.; A. V. S. was a recipient of
FAPESP and Airbus Military/Universidad San Pablo-CEU doctoral
fellowships in Brazil and Spain, respectively.
Arch. Dermatol. 2003, 139, 432e434.
~
19. (a) Edwards, A. M.; Saldano, A.; Bueno, C.; Silva, E.; Alegría, S. Bol. Soc. Chil.
Quím. 2000, 45, 647e656; (b) Ogasawara, F.; Wang, Y.; Bobbett, D. Anal. Chem.
1992, 64, 1637e1642.
~
20. Edwards, A. M.; Bueno, C.; Saldano, A.; Silva, E.; Kassab, K.; Polo, L.; Jori, G. J.
Photochem. Photobiol., B 1999, 48, 36e41.
21. Insinska-Rak, M.; Sikorska, E.; Bourdelande, J. L.; Khmelinskii, I. V.; Pruka1a, W.;
Dobek, K.; Karolczak, J.; Machado, I. F.; Ferreira, L. F. V.; Dulewicz, E.; Komasa,
A.; Worrall, D. R.; Kubicki, M.; Sikorski, M. J. Photochem. Photobiol., A 2007, 186,
14e23.
Supplementary data
22. Banekovich, C.; Matuszczak, B. Tetrahedron Lett. 2005, 46, 5053e5056.
23. An additional maximum for RF at 223 nm (in water) occurs below the meth-
anol cut-off wavelength: Koziol, J. Photochem. Photobiol. 1966, 5, 41e54.
Spectroscopic and HRMS data of all the flavin derivatives, HPLC
procedures, chemical actinometry and photodynamic results.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tet.2014.11.072.
€
24. Drossler, P.; Holzer, W.; Penzkofer, A.; Hegemann, P. Chem. Phys. 2003, 286,
409e420.
25. Schmidt, R.; Tanielian, C.; Dunsbach, R.; Wolff, C. J. Photochem. Photobiol., A
1994, 79, 11e17.
26. Engelmann, F. M.; Mayer, I.; Gabrielli, D. S.; Toma, H. E.; Kowaltowski, A. J.;
Araki, K.; Baptista, M. S. J. Bioenerg. Biomembr. 2007, 39, 175e185.
27. Luque-Ortega, J. R.; Rivas, L. Methods Mol. Biol. 2010, 618, 393e420.
28. (a) Moan, J. Lasers Med. Sci. 1986, 1, 5e12; (b) Fabris, C.; Valduga, G.; Miotto, G.;
Borsetto, L.; Jori, G.; Garbisa, S.; Reddi, E. Cancer Res. 2001, 61, 7495e7500; (c)
Ghesquiere, J.; Le Gac, S.; Marcelis, L.; Moucheron, C.; Kirsch-De Mesmaeker, A.
Curr. Top. Med. Chem. 2012, 12, 185e196.
29. Abdel-Shafi, A.; Wilkinson, F. J. Phys. Chem. A 2000, 104, 5747e5757.
30. Machadoa, A. E. H.; Miranda, J. A.; Oliveira-Campos, A. M. F.; Severino, D.;
Nicodemc, D. E. J. Photochem. Photobiol., A 2001, 146, 75e81.
31. Kessel, D. Photochem. Photobiol. 1989, 49, 447e452.
References and notes
1. (a) Bonnet, R. Chemical Aspects of Photodynamic Therapy; Gordon and Breach
Science: Amsterdan, The Netherlands, 2000; (b) Capella, M. A. M.; Capella, L. S. J.
Biomed. Sci. 2003, 10, 361e366; (c) Advances in Photodynamic Therapy: Basic,
Translational and Clinical; Hamblin, M. R., Mroz, P., Eds.; Artech House: Norwood,
MA, 2008; (d) Pedraz-Munoz, J.; Díez-Caballero, N. In Dermatological Treatments;
Conde-Taboada, A., Ed.; Bentham Science: Sharjah, UAE, 2012; pp 464e486; (e)
~
Bown, S. G. Philos. Trans. R. Soc., A 2013, 371, 20120371 (16 pp.).
2. (a) Peck, E. M.; Collins, C. G.; Smith, B. D. Org. Lett. 2013, 15, 2762e2765; (b)
Bastos, M. M.; Gomes, A. T. P. C.; Neves, M. G. P. M. S.; Silva, A. M. S.; Santos-Filho,
O. A.; Boechat, N.; Cavaleiro, J. A. S. Eur. J. Org. Chem. 2013, 1485e1493; (c) Huang,