International Journal of
Molecular Sciences
Article
Analogs of Natural 3-Deoxyanthocyanins:
O-Glucosides of the 40 ,7-Dihydroxyflavylium Ion
and the Deep Influence of Glycosidation on Color
Nuno Basílio 1 , Sheiraz Al Bittar 2 , Nathalie Mora 2 , Olivier Dangles 2, * and Fernando Pina 1, *
1
2
*
LAQV-REQUIMTE, Department of Chemistry, Faculty of Science & Technology,
Universidade NOVA de Lisboa, 2829-516 Monte de Caparica, Portugal; [email protected]
Faculty of Sciences, Technology & Health, University of Avignon, INRA, UMR408, 84000 Avignon, France;
[email protected] (S.A.B.); [email protected] (N.M.)
Correspondence: [email protected] (O.D.); [email protected] (F.P.);
Tel.: +33-490-144-446 (O.D.); +351-212-948-355 (F.P.)
Academic Editors: Nuno Mateus and Iva Fernandes
Received: 23 August 2016; Accepted: 12 October 2016; Published: 20 October 2016
Abstract: 3-Deoxyanthocyanidins and their O-β-D-glucosides are natural pigments abundant in
black sorghum. O-glycosidation can perturb the acid-base properties of the chromophore and
lower its electron density with a large impact on the distribution of colored and colorless forms in
aqueous solution. In this work, the influence of O-glycosidation on color is systematically studied
from a series of 3-deoxyanthocyanin analogs. The pH- and light-dependent reversible reactions of
7-β-D-glucopyranosyloxy-40 -hydroxyflavylium (P3) and 40 -β-D-glucopyranosyloxy-7-hydroxyflavylium
(P5) were completely characterized in mildly acidic solution and compared with the parent aglycone
40 ,7-dihydroxyflavylium ion and the O-methylethers of P3 and P5. Except P5, the chalcone forms of
the pigments exhibit a high cis-trans isomerization barrier that allows a pseudo-equilibrium involving
all species except the trans-chalcone. At equilibrium, only the flavylium cation and trans-chalcone
are observed. With all pigments, the colored flavylium ion can be generated by irradiation of the
trans-chalcone (photochromism). Glycosidation of C7–OH accelerates hydration and strongly slows
down cis-trans isomerization with the pH dependence of the apparent isomerization rate constant
shifting from a bell-shaped curve to a sigmoid. The color of P5 is much more stable than that of its
regioisomer P3 in near-neutral conditions.
Keywords: 3-deoxyanthocyanidin; flavylium; glucoside; photochromism; multistate
1. Introduction
3-Deoxyanthocyanidins and their O-glycosides are natural pigments that are especially abundant
in black sorghum [1] but also occur in other plants such as ferns [2–4]. 3-deoxyanthocyanidins of
black sorghum are known to express a variety of health-promoting effects [5] and a better chemical
stability [6] than the much more common anthocyanins, an important advantage for their potential
development as natural food colorants.
3-Deoxyanthocyanidins are typically isolated (sometimes, chemically synthesized), and thus
represented, as the 40 ,5,7-trihydroxyflavylium ion (apigeninidin) and its derivatives with additional
OH and/or OMe groups (generally at C30 and C50 ). However, the flavylium ion (AH+ ) is stable only in
highly acidic medium (pH < 2) and those pigments actually constitute a multistate system of chemical
species, which are reversibly interconverted by external stimuli such as pH variations and light [7,8].
In the mildly acidic conditions typical of plant cell and food, the multistate system encompasses five
distinct colored or colorless species (Figure 1). Understanding the flavylium multistate is of utmost
Int. J. Mol. Sci. 2016, 17, 1751; doi:10.3390/ijms17101751
www.mdpi.com/journal/ijms
Int. J. Mol. Sci. 2016, 17, 1751
2 of 14
Int. J. Mol.for
Sci. 2016,
17, 1751 (3-deoxy) anthocyanin research in plant, food and humans. As extraction
2 of 14
importance
advanced
of natural 3-deoxyanthocyanins is typically time-consuming and low-yielding, investigations on
humans. As extraction of natural 3-deoxyanthocyanins is typically time-consuming and lowsimpler analogs easily accessible by chemical synthesis may be a relevant alternative, especially for
yielding, investigations on simpler analogs easily accessible by chemical synthesis may be a relevant
the systematic
of substituent
effects. Moreover,
synthetic
flavylium
ions
are also an
alternative, investigation
especially for the
systematic investigation
of substituent
effects.
Moreover,
synthetic
important
class
of
photochromic
compounds
with
several
applications,
such
as
models
for optical
flavylium ions are also an important class of photochromic compounds with several applications,
memories
ormodels
even for
basic
properties
of neurons
[9].properties of neurons [9].
such as
for investigating
optical memories
or even
for investigating
basic
Figure 1. The reaction network of 04′,7-dihydroxyflavylium ion (DHF). With DHF, proton loss gives
Figure
1. The reaction network of 4 ,7-dihydroxyflavylium ion (DHF). With DHF, proton loss gives
the 7-keto A tautomer (formation of the 4′-keto A tautomer requires glycosidation of C7–OH). The
the 7-keto A tautomer (formation of the 40 -keto A tautomer requires glycosidation of C7–OH).
corresponding rate constants for the direct and reverse reactions are noted kn and k−n, respectively
The corresponding rate constants for the direct and reverse reactions are noted kn and k−n , respectively
(n = a, h, t, i). Kn = kn/k−n.
(n = a, h, t, i). Kn = kn /k−n .
Hydroxyl groups at positions C4′ and C7 are common to all (3-deoxy)anthocyanins and critical
0 and C7 of
to
color
development
through the
corresponding
quinoidal bases (A). Thisand
is the
Hydroxyl
groups at positions
C4formation
arethe
common
to all (3-deoxy)anthocyanins
critical
purpose
of
this
work
to
show
that
glycosidation
of
the
4′,7-dihydroxyflavylium
ion
(DHP,
a
simple
to color development through the formation of the corresponding quinoidal bases (A). This is the
apigeninidin analog lacking the OH group at C5), besides0 increasing its water solubility, has a deep
purpose
of this work to show that glycosidation of the 4 ,7-dihydroxyflavylium ion (DHP, a simple
and contrasted impact on its reaction network depending on the glycosidation site. Thus, the rate and
apigeninidin analog lacking the OH group at C5), besides increasing its water solubility, has a deep
thermodynamic constants pertaining to the reaction networks of the two O-β-D-glucosides P3 and P5
and contrasted
impact on its reaction network depending on the glycosidation site. Thus, the rate and
of 4′,7-dihydroxyflavylium (DHF) have been assessed. The O-methylethers of P3 and P5, respectively,
thermodynamic
to the
of the
two
O-β-D-glucosides
noted P4 andconstants
P6, as wellpertaining
as DHF itself,
are reaction
included networks
for comparison
(see
structures
in Figure 2).P3 and P5
of 40 ,7-dihydroxyflavylium
(DHF)
have
been
assessed.
The
O-methylethers
of
P3
and P5,
respectively,
A convenient way to characterize the flavylium-based multistate of chemical species
is to
carry
notedout
P4aand
P6,pH
as jump
well as
areofincluded
for comparison
(see structures
in Figure
2).
direct
by DHF
a fast itself,
addition
a small volume
of strong base
to equilibrated
solutions.
The
first reaction
that
place upon
a direct pH jump multistate
is the formation
of quinoidal
base
A,carry
A
convenient
way
to takes
characterize
the flavylium-based
of chemical
species
is to
(1),jump
by proton
loss addition
from the of
most
acidic volume
OH group,
generally
C7–OH.
This reaction
is
out aEquation
direct pH
by a fast
a small
of strong
base
to equilibrated
solutions.
fast
and
requires
special
equipment,
such
as
temperature
jumps
or
flash
photolysis,
to
follow
the
The first reaction that takes place upon a direct pH jump is the formation of quinoidal base A,
corresponding kinetics [10,11]. The flavylium cation and the quinoidal base behave as a single species
Equation (1), by proton loss from the most acidic OH group, generally C7–OH. This reaction is
fast and requires special equipment, such as temperature jumps or flash photolysis, to follow the
Int. J. Mol. Sci. 2016, 17, 1751
Int. J. Mol. Sci. 2016, 17, 1751
3 of 14
3 of 14
Int.the
J. Mol. Sci. 2016, 17,kinetic
1751
3 of 14
in
Int. J.subsequent
Mol. Sci. 2016, 17, 1751 steps, which are by far much slower than proton transfer. Competing for
3 of 14
thecorresponding
disappearancekinetics
of the [10,11].
flavylium
cation
(and
consequently
the
quinoidal
base),
water
addition
The flavylium cation and the quinoidal base behave as a single species
Int.in
Mol.
Sci.
2016, 17, 1751
3 of 14for
thethe
subsequent
kinetic
steps,
which
areare
by far(2).
much
than
proton
transfer.
Competing
atJ. in
C2
(hydration)
gives
hemiketal
B, Equation
Onslower
the
other
hand,
Btransfer.
is transfer.
in equilibrium
with
subsequent
kinetic
steps,
which
much
slower
than
proton
Competing
for
in
the
subsequent
kinetic
steps,
which
are bybyfarfar
much
slower
than
proton
Competing
for the
the
disappearance
of
the
flavylium
cation
(and
consequently
the
quinoidal
base),
water
addition
cis-chalcone
Cc
with
an
interconverting
rate
falling
in
the
sub-second
timescale,
i.e.,
sufficiently
slow
the
disappearance
of
the
flavylium
cation
(and
consequently
the
quinoidal
base),
water
addition
of
the steps,
flavylium
cation
(and
the
quinoidal
base), water
addition
J.disappearance
Mol.
Sci. 2016, 17,
1751
3 for
of at
14 C2
inInt.
the
subsequent
kinetic
which
by
farconsequently
much
slower
than
proton
transfer.
Competing
at
C2
(hydration)
hemiketal
B,are
Equation
(2).(2).
On
the
other
hand,
B B
is is
in in
equilibrium
with
to
be
monitored
by gives
stopped-flow
UV-visible
spectroscopy,
Equation
(3).
at
C2
(hydration)
gives
hemiketal
B,
Equation
On
the
other
hand,
equilibrium
with
(hydration)
gives
hemiketal
B,
Equation
(2).
On
the
other
hand,
B
is
in
equilibrium
with
cis-chalcone
Cc
thecis-chalcone
disappearance
of the
flavylium cationrate
(and
consequently
the quinoidal
base),
water
addition
Cc
with
an
interconverting
falling
in
the
sub-second
timescale,
i.e.,
sufficiently
slow
cis-chalcone
Cc kinetic
with ansteps,
interconverting
rate
falling
inslower
the sub-second
timescale,
i.e.,Competing
sufficiently
slow
in
the
subsequent
which
are
by
far
much
than
proton
transfer.
for
with
an
interconverting
rate
falling
in
the
sub-second
timescale,
i.e.,
sufficiently
slow
to
be
monitored
at to
C2be(hydration)
gives
hemiketal UV-visible
B, Equation
(2). On the Equation
other hand,
B is in equilibrium with
monitored
by
stopped-flow
spectroscopy,
(3).(3).
to
be
monitored
by
stopped-flow
UV-visible
spectroscopy,
Equation
the
disappearance
of
the
flavylium
cation
(and
consequently
the
quinoidal
water addition
by
stopped-flow
UV-visible
spectroscopy,
Equation
(3).
cis-chalcone Cc with an interconverting rate falling in the sub-second timescale, base),
i.e., sufficiently
slow
at
C2
(hydration)
gives
hemiketal
B,
Equation
(2).
On
the
other
hand,
B
is
in
equilibrium
with
to be monitored by stopped-flow UV-visible spectroscopy, Equation (3).
cis-chalcone Cc with an interconverting rate falling in the sub-second timescale, i.e., sufficiently slow
to be monitored by stopped-flow UV-visible spectroscopy, Equation (3).
Figure 2. Pigments investigated in this work.
Figure 2. Pigments investigated in this work.
Pigments investigated
in this work.
AH+ Figure
+ H2O 2.
⇆Pigments
A + H3O+
Ka (acid-base)
(1)
Figure 2. Pigments investigated in this work.
+ ++
+ +++2H
AH
2O ⇆
AAB+A++H
3O
+H
AH
⇆⇆
AH
++H
O
AH
+22O
H2O
+H
H33O
3O+
(acid-base)
KKha K
(hydration)
Ka (acid-base)
a (acid-base)
(1)(1)(1)
(2)
Figure 2. Pigments investigated in this work.
+
+
+
AH
+++H
2O
⇆
3+
OH
AH+++AH
2H
22O
⇆+⇆BH
BB+
AH
2H
O
+ +B
⇆ 2A
Cc
33O
2H
O
+HH
3O+
+ + H2O ⇆ A + H3O++
+ + 2H
AHAH
2O
B + H3O
BB ⇆
Cc
Cc
Cc
⇆
Ct
B⇆
Cc
KaKK(acid-base)
(1)(3)
(2)(2)(2)
Kh (hydration)
th(tautomerization)
K(hydration)
h (hydration)
a (acid-base)
(1)
KK
hK
(2)
Kt (tautomerization)
(tautomerization)
(3) (3)
K(hydration)
it (isomerization)
K
t (tautomerization) (4) (3)
+ + 2H2O
Tautomerization is the
fastest
process
except
in very acidic medium
AHsecond
⇆CtCt
B + H3O+of the multistate,
h (hydration)
(2)
Ki (isomerization)
B Cc
⇆
Cc
KK
tK
(tautomerization)
(3)
Cc
⇆
i (isomerization)
(4)(4)(4)
Cc
⇆ Ct
Ki (isomerization)
where hydration could be even faster. Finally, cis-trans isomerization leads to the trans-chalcone,
Tautomerizationisisthe
thesecond
secondBfastest
fastest
of
multistate,
except
in very
acidic
medium
Tautomerization
process
ofthe
the
multistate,
except
in in
very
acidic
medium
⇆fastest
Ccinprocess
K
t (tautomerization)
(3)where
Equation
(4). This reaction
could
less
than
one
second
up
days
depending
on
the
Tautomerization
is the
second
process
of
the
multistate,
except
very
acidic
medium
Ccoccur
⇆
Ct
K
i to
(isomerization)
(4)
hydration
could
be
even
faster.
Finally,
cis-trans
isomerization
leads
to
the
trans-chalcone,
Equation
(4).
where
hydration
could
be
even
faster.
Finally,
cis-trans
isomerization
leads
to
the
trans-chalcone,
isomerization
barrier.could
Brouillard
andfaster.
Dubois
[10] discovered
a peculiar characteristic
of the
(3-deoxy)
where hydration
be even
Finally,
cis-trans isomerization
leads to the
trans-chalcone,
Tautomerization
is
the
second
fastest
process
of
the
multistate,
except
in
very
acidic
medium
This reaction
could
occur in
less Cc
than
one
second
up to
days
depending
on
the depending
isomerization
barrier.
Equation
(4).(4).
This
reaction
could
occur
less
than
one
second
upK
days
on(4)
thethe
⇆
Ctindoes
ito
(isomerization)
anthocyanin
multistate:
the quinoidal
base
not
undergo
water
addition
in
acidic
medium
and
Equation
This
reaction
could
occur
in
less
than
one
second
up
to
days
depending
on
where
hydration
could
be [10]
even
faster.
Finally,
cis-trans
isomerization
leads
to the
trans-chalcone,
Brouillard
and
Dubois
discovered
a peculiar
characteristic
of the (3-deoxy)
anthocyanin
multistate:
isomerization
barrier.
Brouillard
and
Dubois
[10]
discovered
a
peculiar
characteristic
of
the
(3-deoxy)
theisomerization
system
evolves
from
Equation
(1)
toDubois
Equation
onlymultistate,
by means
of flavylium
hydration.
Thus,
barrier.
Brouillard
and
[10](4)
discovered
a peculiar
characteristic
of the
(3-deoxy)
Tautomerization
is
the
second
fastest
of
the
except
inthe
very
acidic
medium
Equation
(4). This
reaction
could
occur
in process
less
than
one
second
upaddition
to days
depending
on the
the
quinoidal
base
does
not
undergo
water
addition
in acidic
medium
and
system
evolves
from
anthocyanin
multistate:
the
quinoidal
base
does
not
undergo
water
in
acidic
medium
and
theanthocyanin
quinoidal
base
can be
regarded
as a Finally,
kinetic
product
the flavylium
reaction
network.
multistate:
the
quinoidal
base does
not of
undergo
water addition
in
acidic
medium
and
where
hydration
could
be
even
faster.
cis-trans
isomerization
leads
to
the
trans-chalcone,
isomerization
barrier.
Brouillard
and
Dubois
[10]
discovered
a
peculiar
characteristic
of
the
(3-deoxy)
Equation
(1) to Equation
(4) only(1)
byto
means
of flavylium
hydration.
Thus,
the quinoidal
baseThus,
can be
thethe
system
evolves
from
Equation
Equation
(4)(4)
only
byby
means
of
flavylium
hydration.
With
natural
anthocyanins,
i.e.,
3-βD
-glycopyranosyloxyflavylium
ions,
isomerization
is
a
system
evolves
from
Equation
(1)
to
Equation
only
means
of
flavylium
hydration.
Thus,
Equation
(4).
reaction
could
occur
in
less
than
one
second
up to days
depending
the
anthocyanin
multistate:
the
quinoidal
base
does
not
undergo
water
addition
in acidic
mediumonand
regarded
asThis
a kinetic
product
of the
reaction
thethe
quinoidal
base
can
be
regarded
as
aflavylium
kinetic
product
ofnetwork.
the
flavylium
reaction
network.
minor
processbarrier.
and
the
trans-chalcone
typically
represents
less
than
10%
ofreaction
the colorless
forms [7].
quinoidal
base
can
be
regarded
as
a kinetic
product
of
the
flavylium
network.
isomerization
Brouillard
and
Dubois
[10]
discovered
a
peculiar
characteristic
of
the
(3-deoxy)
the system
evolves
from
Equation (1)
to3-βEquation
(4) only by means of flavylium
hydration.
Thus,
With
natural
anthocyanins,
i.e.,
DD-glycopyranosyloxyflavylium
ions,
isomerization
is a is
minor
With
anthocyanins,
3-β-glycopyranosyloxyflavylium
ions,
isomerization
a a
Indeed,
thenatural
trisubstituted
double
bond
of3-βanthocyanin
chalcones
has
no marked
preference
forand
the
With
natural
anthocyanins,
i.e.,
D-glycopyranosyloxyflavylium
ions,
isomerization
is
anthocyanin
multistate:
the
quinoidal
base
does
not
undergo
water
addition
in
acidic
medium
theminor
quinoidal
base
can
be regarded
as
a kinetic
product
ofless
thethan
flavylium
reaction
network.
process
and
the
trans-chalcone
typically
represents
10% of
theofcolorless
formsforms
[7]. Indeed,
process
and
the
trans-chalcone
typically
represents
less
than
10%
the
colorless
[7].
trans
vs. cis
configuration.
The situation
totally different
with
natural
3-deoxyanthocyanidins
and[7].
minor
process
and
theEquation
trans-chalcone
typically
represents
than
of thehydration.
colorless forms
the
system
evolves
from
(1)
to is
Equation
(4)
only has
by less
means
of 10%
flavylium
Thus,
With
natural
anthocyanins,
3-βD-glycopyranosyloxyflavylium
ions,
isomerization
is vs.
athecis
the
trisubstituted
double
bondi.e.,
ofbond
anthocyanin
chalcones
no marked
preference
for the trans
Indeed,
the
trisubstituted
double
of
anthocyanin
chalcones
has
no
marked
preference
for
theIndeed,
artificial
investigated
in
this
work.
Indeed,
with
such pigments,
the
trans-chalcone
isthe
theanalogs
trisubstituted
double
bond
of anthocyanin
chalcones
hasreaction
no marked
preference for
the
quinoidal
base
can
be
regarded
as
a
kinetic
product
of
the
flavylium
network.
minor
process
and
the
trans-chalcone
typically
represents
less
than
10%
of
the
colorless
forms
[7].
configuration.
The situation
is
totally different
with
natural
3-deoxyanthocyanidins
and the artificial
trans
vs.vs.
cisthe
configuration.
The
situation
is totally
different
with
natural
3-deoxyanthocyanidins
and
essentially
sole
colorless
form
at
equilibrium
in
mildly
acidic
solution.
trans
cis
configuration.
The
situation
is
totally
different
with
natural
3-deoxyanthocyanidins
With
natural
anthocyanins,
i.e., Indeed,
3-βD-glycopyranosyloxyflavylium
ions, preference
isomerization
is aand
Indeed,
the trisubstituted
double
bond
of
anthocyanin
chalcones
hasthe
no
marked
for the
analogs
investigated
in
this work.
with
such
pigments,
trans-chalcone
is essentially
thethe
artificial
analogs
investigated
in
this
work.
Indeed,
with
such
pigments,
the
trans-chalcone
isthe
The
flavylium
multistate
can
be
viewed
as
a
single
acid-base
equilibrium
involving
the
flavylium
artificialand
analogs
investigated in
this work.
Indeed,less
with
such
pigments,
the trans-chalcone
minor
process
theattrans-chalcone
typically
represents
than
10%
of the colorless
formsand
[7]. is
trans
vs.
cis
configuration.
The
situation
is
totally
different
with
natural
3-deoxyanthocyanidins
sole
colorless
form
equilibrium
in
mildly
acidic
solution.
essentially
theone
sole
colorless
form
at at
equilibrium
inbase
mildly
acidic
solution.
cation,
on
hand,
and
the
overall
conjugated
(CB),
constituted
the mixture
of quinoidal
essentially
the
sole
colorless
form
equilibrium
in
mildly
acidic
solution.
Indeed,
thethe
trisubstituted
double
bond
ofwork.
anthocyanin
chalcones
has
no by
marked
preference
for
the
the
artificial
analogs
investigated
in
this
Indeed,
with
suchequilibrium
pigments,
theinvolving
trans-chalcone
is
The
flavylium
multistate
can
be
viewed
as
a single
acid-base
equilibrium
the
flavylium
The
flavylium
multistate
can
be
viewed
as
a
single
acid-base
involving
thethe
flavylium
base,
hemiketal
and
both
chalcones,
Equation
(5):
The
flavylium
multistate
can
be
viewed
as
a
single
acid-base
equilibrium
involving
flavylium
trans
vs.
cis
configuration.
The
situation
is
totally
different
with
natural
3-deoxyanthocyanidins
and
essentially
sole
colorless
form
at
equilibrium
in mildly
acidic
solution.
cation,
on
the
one
hand,
and
the
overall
conjugated
base
(CB),
constituted
by
the
mixture
of quinoidal
cation,
onthe
thethe
one
hand,
and
thethe
overall
conjugated
base
(CB),
constituted
byby
thethe
mixture
of of
quinoidal
on
one
hand,
and
overall
conjugated
base
(CB),
constituted
quinoidal
the cation,
artificial
analogs
investigated
in
this
work.
Indeed,
with
such
pigments,
the mixture
trans-chalcone
is
+ + H
+
AH
2O
⇆
CB
+
H
3O
K’
a
The
flavylium
multistate
can
be
viewed
as
a
single
acid-base
equilibrium
involving
the
flavylium
base,
hemiketal
and
both
chalcones,
Equation
(5):
base,
hemiketal
and
both
chalcones,
Equation
(5):(5):
base,
hemiketal
and
both
chalcones,
Equation
essentially
sole
colorless
form
at equilibrium
mildly
solution.
(5)
cation,
on thethe
one
hand,
and the
overall
conjugatedin
base
(CB),acidic
constituted
by the mixture of quinoidal
+ ++
+
AH
H
2Oa⇆
CB
+
H
3O+ + equilibrium
K’
a
The
flavylium
multistate
can
be
viewed
as
single
acid-base
involving
the
flavylium
[CB]=[A]+[B]+[Cc]+[Ct]
+
+3O
H3 O
K 0K’
AH ++(5):
HH
2O
⇆
CBCB
+H
base, hemiketal and both chalcones, Equation
a a
2O
(5)(5) (5)
cation, on the one hand, and the overall[CB
conjugated
base
(CB),
constituted
by the mixture of quinoidal
=
A
+
B
+
Cc
+
Ct
]
[
]
[
]
[
]
[
]
+
+
AH
+ H2O ⇆
CB + H3O
K’a
[CB]=[A]+[B]+[Cc]+[Ct]
base, hemiketal and both chalcones,
KEquation
'a [CB]=[A]+[B]+[Cc]+[Ct]
= Ka +(5):
Kh + Kh Kt + Kh Kt Ki
(6)
(5)
0
K+a = Ka + Kh + Kh Kt+ + Kh Kt Ki
(6)
AH + H2O ⇆ CB + H3O
K’a
[CB]=[A]+[B]+[Cc]+[Ct]
K 'K
=
K
+
K
+
K
K
+
K
K
K
(6)
a ' =aK +hK +hK t K +hK t Ki K
(5) (6)
a
a
h
h t
h t i
K[CB]=[A]+[B]+[Cc]+[Ct]
'a = Ka + Kh + Kh Kt + Kh Kt Ki
K 'a = Ka + Kh + Kh Kt + Kh Kt Ki
(6)
(6)
Int. J. Mol. Sci. 2016, 17, 1751
4 of 14
2. Results and Discussion
2.1. Investigation of the 7-β-D-Glucopyranosyloxy-40 -hydroxyflavylium Ion (P3)
Natural (3-deoxy)anthocyanins and some related pigments such as P3 (see below) possess a high
cis-trans isomerization barrier. This fact has a dramatic influence on the kinetics of the multistate.
In particular, it is possible to define a pseudo-equilibrium, a long-life transient state where all species
except the trans-chalcone are equilibrated. Analogous with the equilibrium, the pseudo-equilibrium
can be defined by Equations (7) and (8):
AH+ + H2 O CBˆ + H3 O+
[CBˆ] = [A] + [B] + [Cc]
K ˆa
(7)
Kˆa = Ka + Kh + Kh Kt
(8)
Owing to the high cis-trans isomerization barrier, three distinct kinetic processes after a direct pH
jump can be distinguished: (i) formation of the quinoidal base (not observed by stopped-flow);
(ii) hydration followed by tautomerization, where the rate-limiting step is hydration; and (iii)
cis-trans isomerization.
The three successive kinetic steps, all very well separated in time, are given by Equation (9)
(too fast, not investigated in this work), Equations (10) and (11) [12]:
k1st = ka + k −a [H+ ]
(9)
k2nd =
[H+ ]
1
kh +
k [H+ ]
1 + Kt −h
[ H ] + Ka
(10)
k3th =
Kh Kt
k i + k −i
[ H+ ] + Ka + Kh + Kh Kt
(11)
+
The alternative to direct pH jumps consists of adding a small volume of strong acid to equilibrated
solutions of CB (at moderately acidic pH values) and is defined as reverse pH jumps. If the final pH
is sufficiently low, dehydration becomes faster than tautomerization and four distinct steps can be
observed: (i) conversion of the quinoidal base into flavylium cation during the mixing time of the
stopped flow; (ii) dehydration of the hemiketal into flavylium cation, Equation (12); (iii) a slower
step due to the conversion of the cis-chalcone into flavylium cation via the hemiketal, Equation (13);
(iv) in a much slower process conversion of the trans-chalcone into flavylium cation, Equation (11).
When the reverse pH jumps are carried out from pseudo-equilibrium, the kinetics steps (ii) and (iii) are
better defined because the concentration of hemiketal and cis-chalcone is higher. This is especially
important with the pigments investigated in this work as the trans-chalcone is practically the only CB
component at full equilibrium. In Equation (13), the term kt was not considered because dehydration
is much faster and there is no reversibility. In other words, as soon as the hemiketal is formed from the
cis-chalcone, it immediately gives the flavylium cation:
k1st reverse =
[H+ ]
k h + k −h [ H+ ]
[ H+ ] + Ka
−
+
OH
k2nd reverse = k −t + kH
−t [H ] + k −t [OH ]
(12)
(13)
OH
The terms kH
−t and k −t account for the acid and base catalysis of cis-chalcone cyclization [13].
2.1.1. pH Jumps
The absorption spectra of P3 after direct pH jumps (taken 10 ms after mixing, base formed during
mixing time of solutions, ca. 6 ms) are shown in Figure 3a as a function of the final pH. The spectra
Int. J. Mol. Sci. 2016, 17, 1751
Int. J. Mol. Sci. 2016, 17, 1751
5 of 14
5 of 14
2.1.1. pH Jumps
2.1.1. pH Jumps
Int. J. Mol. Sci. 2016, 17, 1751
5 of 14
The absorption spectra of P3 after direct pH jumps (taken 10 ms after mixing, base formed during
The absorption spectra of P3 after direct pH jumps (taken 10 ms after mixing, base formed during
mixing time of solutions, ca. 6 ms) are shown in Figure 3a as a function of the final pH. The spectra
mixing time of solutions, ca. 6 ms) are shown in Figure 3a as a function of the final pH. The spectra
are compatible with an acid-base
acid-base equilibrium, Equation (1), with
with pK
pKa = 5.4. This value is consistent
are compatible with an acid-base 0equilibrium, Equation (1), with pKaa = 5.4. This value is consistent
it is
higher
than
thethe
pKapK
of 4.0
with the pK
pKaa of
of5.5
5.5reported
reportedfor
for4′-hydroxyflavylium.
4 -hydroxyflavylium.However,
However,
it much
is much
higher
than
of
with the pKa of 5.5 reported for 4′-hydroxyflavylium.
However, it is much higher than the pKa ofa 4.0
0
estimated
for for
thethe
corresponding
aglycone
(4′,7-dihydroxyflavylium),
4.0 estimated
corresponding
aglycone
(4 ,7-dihydroxyflavylium),meaning
meaningthat
thatglycosidation
glycosidation at
estimated for the corresponding aglycone (4′,7-dihydroxyflavylium), meaning that glycosidation at
C7–OH cancels the most acidic OH group.
C7–OH cancels the most acidic OH group.
1.2
1.2
503 nm
503 nm
452 nm
452 nm
a
1
1 A
1 A
0.8
0.8
0.5
A
A
0.5
0
2
0
2
0.5
0.5
0.7
A
A
4
4
6
pH 6
pH
8
0
0.3
0
0.4
0.4
8
3
3
obs
0.7
0.3
4
4
-1
k =0.038 s -1
kobs =0.038 s
1.1
A(458
A(458
nm)nm)
1
1.1
pK =5.4
pKa =5.4
hydration
1.5
log(k
log(k
) )
hydration
1.5
1.5 1.5
20
60
80
100
(s)
60
Time (s)
40
80
100
20 Time
40
Reverse pH jumps
Reverse pH jumps
2
2
1
1
0
0
Direct pH jumps
Direct pH jumps
-1
-1
0
0250
250
350
350
450
Wavelength 450
(nm)
Wavelength (nm)
0
0 250
250
550
550
(a)
(a)
350
450
-2
-2 0
0
550
550
350
450
Wavelength
(nm)
Wavelength (nm)
(b)
(b)
1
1
2
2
3
3
(c)
(c)
4
4
pH
5
5
6
6
7
7
8
8
pH
Figure
3. (a)
(a) pH-Dependenceofofthe
the UV-visible
spectrum
of P3
taken
10 after
ms after
a direct
pH jump
Figure
spectrum
of P3
10 ms
a direct
pH jump
from
Figure 3.
3. (a)pH-Dependence
pH-Dependence of theUV-visible
UV-visible
spectrum
of taken
P3 taken
10 ms after
a direct
pH jump
from
pH
1.0
to
higher
pH
values.
On
this
timescale,
the
flavylium
cation
and
the
quinoidal
base
pH
1.0
to
higher
pH
values.
On
this
timescale,
the
flavylium
cation
and
the
quinoidal
base
equilibrate
from pH 1.0 to higher pH values. On this timescale, the flavylium cation and the quinoidal base
= 5.4;
and (b)
spectralupon
variations
upon
direct
pH pH
jump
1.0 to 5.5
equilibrate
pKa (b)
with
pKa = with
5.4; and
spectral
variations
a direct
pH aajump
from
1.0from
to 5.5pH
followed
by
and (b)
spectral variations
upon
direct
pH jump
from
pH
1.0 to 5.5
equilibrate
with
pKa = 5.4;
−
1
followed
by
stopped
flow.
Mono-exponential
curve-fitting
gives
rate
constant
k
obs = 0.038 s−1; and
stopped
Mono-exponential
curve-fitting gives
rate constant
kobsrate
= 0.038
s ; and
apparent
followedflow.
by stopped
flow. Mono-exponential
curve-fitting
gives
constant
kobs (c)
= 0.038
s−1; rate
and
(c)
apparent
rate constants
of hydration
for different
pHpH
values:
(●)(#)
direct
pH jumps,
() reverse
pH
constants
of hydration
for different
pH values:
( ) direct
jumps,
reverse
pH jumps.
(c) apparent
rate constants
of hydration
for different
pH values:
(●) direct
pH jumps,
() Fitting
reversewas
pH
−1
−
1
3
−
1
jumps.
Fitting
was
achieved
for
k
h = 0.08
s
;
kh−=1 9 × 103 M−1·s−1; pKa set at 5.4, Kt set at 0.6.
achieved
for khwas
= 0.08
s ; k-hfor
= 9kh×= 10
·sh = ;9pK
5.4,
Kt aset
3 Mat
−1·s
−1; pK
a set
jumps. Fitting
achieved
0.08M
s−1; k× 10
setatat0.6.
5.4, Kt set at 0.6.
The Kt value was calculated from the amplitudes of the curves recorded after a reverse pH jump
The K
Ktt value was calculated from the amplitudes of the curves recorded after a reverse pH jump
The
from pH = 4.65 to pH = 1.9 (Figure 4). The initial absorbance is due to the mixture of flavylium ion
pH == 4.65 to pH
pH == 1.9 (Figure 4). The initial absorbance is due to the mixture
mixture of
of flavylium
flavylium ion
ion
from pH
and quinoidal base present at pseudo-equilibrium at pH = 4.65. The first step corresponds to fast
quinoidal base
base present
present at
at pseudo-equilibrium
pseudo-equilibrium at
at pH
pH == 4.65.
4.65. The first step corresponds to fast
fast
and quinoidal
dehydration at pH = 1.9 (apparent rate constant = 110 s−−1−11). The second step corresponds to the slower
dehydration at
atpH
pH== 1.9
1.9 (apparent
(apparentrate
rateconstant
constant==110
110ss ).). The
dehydration
The second
second step
step corresponds
corresponds to the slower
tautomerization. From the amplitudes of both steps, the Kt value can be estimated: Kt = 0.6.
tautomerization. From the amplitudes
amplitudes of
of both
both steps,
steps, the
the K
Ktt value can be estimated: Ktt == 0.6.
tautomerization.
0.6.
(452
nm)
AA
(452
nm)
0.6
0.6
Cc
Cc
K =0.6
Kt =0.6
+
0.2
0.2
0
00
0
B
B
t
0.4
0.4
AH ++ A
AH + A
1
1
2
2
Time (s)
Time (s)
3
3
4
4
Figure 4. Reverse pH jump from a pseudo-equilibrated solution of P3 at pH = 4.65 (final pH = 1.9).
Figure 4.
4. Reverse
Reverse pH
pH jump
jump from
from aa pseudo-equilibrated
pseudo-equilibrated solution
solution of
P3 at
at pH
pH =
= 4.65
4.65 (final
(final pH
pH =
= 1.9).
Figure
of P3
1.9).
Two kinetic steps are observed, the first one ascribed to dehydration, kobs = 110 s−1−1, and the slowest to
−
1
Twokinetic
kineticsteps
stepsare
areobserved,
observed,the
thefirst
firstone
oneascribed
ascribedtotodehydration,
dehydration,
k
obs==110
110
s
,
and
the
slowest
to
Two
k
s
,
and
the
slowest
to
obs
ring closure, k−t = 0.6 s−1−1.
ring closure,
closure, kk−−tt==0.6
ring
0.6ss−. 1 .
The time-dependence of the UV-visible spectra after a direct pH jump is shown in Figure 3b for
The time-dependence of the UV-visible spectra after a direct pH jump is shown in Figure 3b for
The
theflavylium
UV-visible
spectra
after
a directbase
pH jump
shown
in Figure
3b for a
a final
pHtime-dependence
= 5.5. At this pH,ofthe
cation
and
quinoidal
are inisnear
equal
concentration.
a final pH = 5.5. At this pH, the flavylium cation and quinoidal base are in near equal concentration.
final pH
= 5.5.vanish
At thisaccording
pH, the flavylium
cation and
quinoidal
are in rate
near constant
equal concentration.
Both
species
to a first-order
kinetics
withbase
apparent
= 0.038 s−1−1
Both species vanish according to a first-order kinetics with apparent rate constant = 0.038 −
s
Both species
a first-order
kinetics with
apparent
rate constant
= 0.038 and
s 1
(half-life
≈ 18vanish
s). Theaccording
spectral to
variations
are compatible
with
the formation
of hemiketal
(half-life ≈ 18 s). The spectral variations are compatible with the formation of hemiketal and
(half-life
≈ 18
s). expense
The spectral
variations
are compatible with the formation of hemiketal and
cis-chalcone
at the
of the colored
forms.
cis-chalcone at the expense of the colored forms.
cis-chalcone at the expense of the colored forms.
Int. J. Mol. Sci. 2016, 17, 1751
6 of 14
The Sci.
apparent
Int. J. Mol.
2016, 17, rate
1751
Int. J. Mol. Sci. 2016, 17, 1751
constant for the first step of the reverse pH jumps from pseudo-equilibrium
6 of 14
6 of 14
reverse
+
follows Equation (12) and was used to estimate k−h as k1st
≈ k− h [ H ] for P3 at pH < 2 (Figure 3c).
The apparent
rate constant
for the second
step of the
direct
pH jumps
follows
Equation
(10) and was
The
of the
the
reverse
pH jumps
jumps
from
pseudo-equilibrium
The apparent
apparent rate
rate constant
constant for
for the
the first
first step
step of
reverse
pH
from
pseudo-equilibrium
−1; k−h = 9 × 103 M−1·s
−1. Thus, Kh =
−6 M, pKh = 5.05.
reverse
+ k
used
to
estimate
k
h
.
One
has:
k
h
=
0.08
s
h
/k
−h
=
8.9
×
10
reverse
+
follows
(12) and
and was
was used
usedto
toestimate
estimatekk−−hh as
≈≈k−kh−[hH[ H] ]for
follows Equation
Equation (12)
as kk1st
forP3
P3atatpH
pH<<22(Figure
(Figure3c).
3c).
1st
After a direct pH jump, a pseudo-equilibrium
is established before formation of the
The apparent rate constant for the second step of the direct pH jumps follows Equation (10) and was
trans-chalcone. The pH dependence −1of
the UV-visible
spectrum
at pseudo-equilibrium
can
be
1 ·s−1 . Thus,
6 M,
3M
−1.−Thus,
−6 8.9 pK
One has:
has: kkhh==0.08
0.08ss−; 1k;−hk=−h9 =
9×
10−13·sM
K−h
kh /k
× h10
used to estimate khh. One
× 10
Kh = kh/k
× −10
=−5.05.
h = 8.9
h = M,
fitted according to a single proton transfer with thermodynamic constant K^a (Figure 5a). One obtains:
pKh =
5.05. a direct pH jump, a pseudo-equilibrium is established before formation of the
After
pK^a = 4.7.
After a direct
pHpH
jump,
a pseudo-equilibrium
is established
beforeatformation
of the trans-chalcone.
trans-chalcone.
The
dependence
of the UV-visible
spectrum
pseudo-equilibrium
can be
^a (Figure
The pH
dependence
of
the
UV-visible
spectrum
at
pseudo-equilibrium
can
be
fitted
according
to a
fitted
according
to
a
single
proton
transfer
with
thermodynamic
constant
K
5a).
One
obtains:
2
1.6
ˆ
ˆ
1.6
^
single
proton transfer with thermodynamic
constant K a (Figure 5a). One obtains:
pK a = 4.7.
pK
a = 4.7.
pK' =2.3
^
2
a
2
1.5
1.6
A
pK =4.7
0.8
1.2
0.8
A
a
A
1.5
1
0.8
0
0.8
0.4
2
1
^
pK' =2.3
a
A (452 nm)
1.6
1.2
a
A (452 nm)
A (452 nm) A (452 nm)
pK =4.7
1
0
0
2
2
4
6
8
1
0.5
pH
4
6
4
6
pH
A
0
0
2
pH
0
2
4
6
8
pH
0.4
0
250
350
450
550
Wavelength (nm)
0.5
0
200
250
350
450
400
500
Wavelength (nm)
(a)
0
300
(b)
0
200
550
300
400
500
Wavelength
(nm)UV-visible spectrum of P3; (a)Wavelength
Figure 5. pH-Dependence
of the
recorded(nm)
at pseudo-equilibrium
(all species except the trans-chalcone);
and (b) recorded at full equilibrium
(all species including
(a)
(b)
the trans-chalcone).
Figure
5. pH-Dependence
at pseudo-equilibrium
pseudo-equilibrium
Figure 5.
pH-Dependence of
of the
the UV-visible
UV-visible spectrum
spectrum of
of P3;
P3; (a)
(a) recorded
recorded at
(all
species except
except the
thetrans-chalcone);
trans-chalcone);and
and(b)
(b)recorded
recorded
equilibrium
(all
species
including
(all
species
at at
fullfull
equilibrium
species
including
the(third
Finally,
the spectral
variations of P3
to reach
full equilibrium
from(all
pseudo-equilibrium
the
trans-chalcone).
trans-chalcone).
step of
direct pH jumps, Figure 6) express the formation of the trans-chalcone at the expense of the
other species. The pH-dependence of the corresponding first-order rate constant can be fitted with
Finally, the spectral variations of P3 to reach full equilibrium from pseudo-equilibrium (third
−4.7), thus yielding:
Equation
(11)the
(K^spectral
a set at 10
KhKtkifull
= 2.9equilibrium
× 10−8 M·s−1;from
k−i = 5.9
× 10−6 s−1. Therefore,
we
Finally,
variations
of P3 to reach
pseudo-equilibrium
(third
step of direct pH jumps, Figure 6) express the formation of the trans-chalcone at the expense of the
−3
−1
× 10
s , KFigure
i = 935.6) express the formation of the trans-chalcone at the expense of the
deduce:
ki = 5.5pH
step of direct
jumps,
other species. The pH-dependence of the corresponding first-order rate constant can be fitted with
theThe
set
of
thermodynamic
the pK’
a value rate
can constant
be calculated
to be with
2.3,
otherFrom
species.
pH-dependence
of theconstants,
corresponding
first-order
can be fitted
Equation (11) (K^aˆ set at 10−4.7−),4.7thus yielding: KhKtki = 2.9 × 10−8 M·s−1;−k8−i = 5.9
× 10−6 s−1. Therefore,
we
−1 ;pH-dependence
−of
6 sthe
−1 .
Equation (6),
is −1in
with the
deduced
from
the
(11)which
(K a −3
set
at perfect
10
),agreement
thus yielding:
Kh Kvalue
k
=
2.9
×
10
M
·
s
k
=
5.9
×
10
t i
−
i
deduce: ki = 5.5 × 10 s , Ki = 935. −3 −1
UV-visiblewe
spectrum
on10fullys equilibrated
Therefore,
deduce:recorded
ki = 5.5 ×
, Ki = 935. solutions (Figure 5b).
From the set of thermodynamic
constants,
the pK’a value can be calculated to be 2.3,
Equation (6), 2which is in perfect agreement with the0.002
value deduced from the pH-dependence of the
UV-visible spectrum recorded on fully equilibrated solutions (Figure 5b).
-1
(s )
1.2
(s )
-1
obs
k
obs
k
(s )
-1
(s )
-1
obs
0
0.6 0
50
100
150
200
250
0.001
200
250
obs
A
1.5
1
0.002
0.6
1.2
k
2
1.5
Time (min)
1
0.5
k
A
0
0
50
100
150
0.001
Time (min)
0.5
0
0
250
350
450
550
2
3
4
Wavelength (nm)
(a)
0
250
350
0
450
5
6
7
5
6
7
pH
550
2
3
(b)
4
−5
(nm)
pHfrom pH = 1.0 to pH = 3.3;
Figure
6. (a)
spectralWavelength
variations
of P3
P3 (5.4
(5.4×
× 10
Figure 6.
(a) spectral
variations
of
10−5M)
M)after
afteraadirect
directpH
pHjump
jump
from pH = 1.0 to pH = 3.3;
−4
−1
××10
(b)(a)
plot
of of
thethe
apparent
rate
constant
asasaafunction
of
kkobs ==3.1
−s4 s;−and
1 ; and
(b)
3.1
10
(b)
plot
apparent
rate
constant
function
ofpH;
pH; curve-fitting
curve-fitting was
was
obs
−1; k−
^6
−4.7.
8
−
−
1
ˆ
−
4.7
achieved
with
Equation
(11):
K
a set
hKtki = 3.4 × 10−8 −
M
s
−i1= 5.6 × 10−6 s−1; K
at
10
achieved
with
Equation
(11): Khof
Kt kP3
3.4 ××10
10−5 M)Mafter
s ; ak−direct
× 10
; K pH
a set
i =(5.4
i = 5.6pH
Figure
6. (a)
spectral
variations
jumpsfrom
= at
1.010to pH. = 3.3;
−4
−1
kobs = 3.1 × 10 s ; and (b) plot of the apparent rate constant as a function of pH; curve-fitting was
achieved with Equation (11): KhKtki = 3.4 × 10−8 M s−1; k−i = 5.6 × 10−6 s−1; K^a set at 10−4.7.
Int. J. Mol. Sci. 2016, 17, 1751
7 of 14
From the set of thermodynamic constants, the pK’a value can be calculated to be 2.3, Equation (6),
which is in perfect agreement with the value deduced from the pH-dependence of the UV-visible
Int. J. Mol. Sci.
2016, 17, 1751
7 of 14
spectrum
recorded
on fully equilibrated solutions (Figure 5b).
2.1.2.
2.1.2. Photochemistry
Photochemistry
In
of flavylium
multistates
for which
the trans-chalcone
is the major
at equilibrium,
In the
thecase
case
of flavylium
multistates
for which
the trans-chalcone
is species
the major
species at
itequilibrium,
is generally itpossible
to observe
photochromism,
i.e., color changesi.e.,
upon
lightchanges
absorption
bylight
this
is generally
possible
to observe photochromism,
color
upon
species
[7]. by
Upon
irradiation,
light irradiation,
absorption light
by the
trans-chalcone
P3 leads to its
absorption
this continuous
species [7]. Upon
continuous
absorption
by theof
trans-chalcone
of
disappearance
with concomitant
the flavylium
cation
(Figure
S1). Then,
system
P3 leads to its disappearance
withappearance
concomitantofappearance
of the
flavylium
cation
(Figurethe
S1).
Then,
reverts
backreverts
thermally
in this
way ainphotochromic
system.
the system
backdefining
thermally
defining
this way a photochromic
system.
Irradiation of the
the trans-chalcone
trans-chalcone with a light
light pulse
pulse is
is another
another useful
useful way
way to
to shift
shift the
the multistate
multistate
Irradiation
system from
from equilibrium
equilibrium (and
(and follow
follow the
the respective
respective relaxation
relaxation process)
process) to
to obtain
obtain kinetic
kinetic information
information
system
(Figure 7a).
7a). Immediately
into
cis-chalcone
in
(Figure
Immediately after
afterthe
theflash,
flash,aafraction
fractionofoftrans-chalcone
trans-chalconeisisconverted
converted
into
cis-chalcone
a few
nanoseconds
[14].[14].
Because
of theof
cis-trans
isomerization
barrier barrier
of P3, the
of the transin
a few
nanoseconds
Because
the cis-trans
isomerization
ofrecovery
P3, the recovery
of
chalcone
is not immediately
observed after
the flash
at the
pH flash
= 2.8 and
the=system
only forward
the
trans-chalcone
is not immediately
observed
after
at pH
2.8 andevolves
the system
evolves
to theforward
photochromic
products, i.e., products,
the flavylium
cation.
The dark
reactions
(afterreactions
the flash)(after
are thus
only
to the photochromic
i.e., the
flavylium
cation.
The dark
the
tautomerization
followed by dehydration.
At pH
= 2.8, hydration
is faster
tautomerization
and
flash)
are thus tautomerization
followed by
dehydration.
At pH
= 2.8, than
hydration
is faster than
Equation (13) applies.
In Figure(13)
7a up,
the raising
of the
Cc→AH+
tautomerization
and Equation
applies.
In Figure
7avisible
up, theabsorption,
raising of featuring
the visiblethe
absorption,
conversion,
is →
observed.
At a wavelength
the chalcones
(Figureabsorb
7a down),
featuring
the Cc
AH+ conversion,
is observed.where
At a wavelength
whereabsorb
the chalcones
(Figurethe
7a
instantaneous
bleaching after
the flash
anflash
indication
of Ct→Ccofconversion
(Ct having
higher
down),
the instantaneous
bleaching
afteristhe
is an indication
Ct→Cc conversion
(Ctahaving
absorption
coefficient
than Cc),than
followed
by a slower
consistentconsistent
with the Cc→AH
amolar
higher
molar absorption
coefficient
Cc), followed
by adecrease
slower decrease
with the+
conversion
(at 365 nm,(atCc365
absorbs
slightly
more
thanmore
AH+).than AH+ ).
Cc
→AH+ conversion
nm, Cc
absorbs
slightly
(a)
(b)
Figure 7.7.(a)
(a)Transient
Transient
absorption
upon
a flash
light applied
flash applied
to an equilibrated
P3
Figure
absorption
upon
a light
to an equilibrated
solution ofsolution
P3 at pHof
= 2.8.
−
1
obs = 0.7 s−1; A: flavylium formation;
at
pH
=
2.8.
A
mono-exponential
curve-fittings
gives:
k
A mono-exponential curve-fittings gives: kobs = 0.7 s ; A: flavylium formation; B: trans-chalcone
B: trans-chalcone
(b) plot
of constant
the apparent
rate constant
as a function of pH.
consumption;
andconsumption;
(b) plot of theand
apparent
rate
as a function
of pH.
The pH-dependence of the apparent rate constant of the flash photolysis experiments
The pH-dependence of the apparent rate constant of the flash photolysis experiments (Figure 7b)
(Figure 7b) clearly shows two distinct regimes [15]. At low pH, hydration is faster and the kinetics is
clearly shows two distinct regimes [15]. At low pH, hydration is faster and the kinetics is controlled by
controlled by tautomerization (Equation (13)). The shape of the branch is compatible with acid and
tautomerization (Equation
(13)). The shape of the branch is compatible with acid and base catalysis
base catalysis with k−tH and k−tOH respectively equal to 20 M−1·s−1 and 1 × 1010 M−1·s−1. At higher pH,
with k− t H and k− t OH respectively equal to 20 M−1 ·s−1 and 1 × 1010 M−1 ·s−1 . At higher pH, hydration
hydration becomes slower and Equation (10) applies.
becomes slower and Equation (10) applies.
2.2. Investigation
Investigation of
of the
the 44′-Glucopyranosyloxy-7-hydroxyflavylium
ion(P5)
(P5)
0 -Glucopyranosyloxy-7-hydroxyflavylium ion
2.2.
Unlike P3,
P3, P5
P5 does
does not
not display
display aa cis-trans
barrier. With
With such
such pigments,
pigments, after
after aa direct
direct
Unlike
cis-trans isomerization
isomerization barrier.
pH jump,
jump, only
only two
two separated
separated steps
steps can
can be
be detected:
detected: (i)
(i) aa very
very fast
fast proton
proton transfer
transfer to
to form
form the
the base;
base;
pH
and (ii) a slow kinetic process yielding the trans-chalcone. Equation (14) can be deduced from the
following assumptions [8]:
AH+ and A on the one hand, B and Cc on the other hand, are in fast equilibrium,
B and Cc taken collectively are in steady state,
Int. J. Mol. Sci. 2016, 17, 1751
8 of 14
and (ii) a slow kinetic process yielding the trans-chalcone. Equation (14) can be deduced from the
Int. J. Mol. Sci. 2016, 17, 1751
8 of 14
following assumptions [8]:
AH+ and A on the one hand, B and Cc
[Hon+ ]the other hand, are +in fast equilibrium,
B and Cc taken collectively are in steady
state,K h K t ki + k− i [H ]
+
kbell =
[H ] + Ka
(14)
K
[H+ ] +
Kh K
kti ki +t k −i [H+ ]
[H+[H
]+Ka ] +
k
[H+ ] + k−i hkKt
−h
kbell =
(14)
A plot of the apparent rate constant of Ct formation as a function of pH is a bell-shaped curve
+
]
A plottoofk−ithe
apparent
rate constant
of Ct
formation
as a function
of pH is aand
bell-shaped
at low
pH values
(reverse pH
jumps,
rate-limiting
isomerization)
to k h [Hcurve
that tends
++
[H ]
a
that tends to k at low pH values (reverse pH jumps, rate-limiting isomerization) and to k [H ] + Kat
−i
h [H+ ]+Ka
at high
values
(direct
jumps,
rate-limiting
hydration).
At intermediate
pH values
(maximal
high
pHpH
values
(direct
pHpH
jumps,
rate-limiting
hydration).
At intermediate
pH values
(maximal
rate
rate
of
Ct
formation),
there
is
no
rate-determining
step
and
isomerization
and
hydration
both
of Ct formation), there is no rate-determining step and isomerization and hydration both contribute
to
contribute
to kinetics.
the observed kinetics.
the
observed
2.2.1. pH Jumps
The pH-dependence of the UV-visible spectrum of P5 obtained 10 ms after direct pH jumps
at the
the equilibrium
equilibrium (Figure
(Figure 8b),
8b), are
are compatible
compatible with
with the
the initial
initial instantaneous
instantaneous
(Figure 8a), as well as at
formation of the quinoidal base and the final
final accumulation
accumulation of
of the
the trans-chalcone.
trans-chalcone.
1
1
pK =3.65
pK' =2.5
a
0.3
1
A (450 nm)
A
A
0.4
a
0.8
A (450 nm)
0.8
3
pH
5
7
0.4
0
0
2
4
6
pH
0
0
300
400
500
600
700
300
400
500
Wavelength (nm)
Wavelength (nm)
(a)
(b)
600
700
−5 M) recorded: (a) immediately after a direct pH jump; and
Figure 8.
8. UV-visible
1010
−5 M) recorded: (a) immediately after a direct pH jump;
Figure
UV-visible spectra
spectraof
ofP5
P5(4(4××
(b)
at
equilibrium.
and (b) at equilibrium.
The corresponding time-dependence of the spectrum is shown in Figure 9a for a direct pH jump
The corresponding time-dependence of the spectrum is shown in Figure 9a for a direct pH jump to
to 5.8. Plotting the apparent first-order rate constant as a function of pH leads to the expected
5.8. Plotting the apparent first-order rate constant as a function of pH leads to the expected bell-shaped
bell-shaped curve (Figure 9b), as observed for other flavylium ions bearing a hydroxyl group at C7.
curve (Figure 9b), as observed for other flavylium ions bearing a hydroxyl group at C7.
2.2.2. Photochemistry
2.2.2. Photochemistry
With pigments having no cis-trans isomerization barrier, the rate of flavylium appearance after
With pigments having no cis-trans isomerization barrier, the rate of flavylium appearance after the
the flash is controlled by hydration, Equation (15), except at very low pH where hydration becomes
flash is controlled by hydration, Equation (15), except at very low pH where hydration becomes faster
faster than tautomerization, Equation (16). Control by hydration assumes B and Cc in equilibrium
than tautomerization, Equation (16). Control by hydration assumes B and Cc in equilibrium (taking
(taking into account a contribution of diffusion-controlled base catalysis of tautomerization).
into account a contribution of diffusion-controlled base catalysis of tautomerization). Equations (15)
Equations (15) and (16) respectively correspond to Equations (10) and (13) used for P3 (slow
isomerization) but include a term representing isomerization, which cannot be neglected for P5:
kflash (hydration) =
[H + ]
1
Kt
kh +
k− h [H + ] +
ki
+
[H ] + Ka
1 + Kt
1 + Kt
(15)
Int. J. Mol. Sci. 2016, 17, 1751
9 of 14
and (16) respectively correspond to Equations (10) and (13) used for P3 (slow isomerization) but
include a term representing isomerization, which cannot be neglected for P5:
Int. J. Mol. Sci. 2016, 17, 1751
kflash (hydration) =
[H+ ]
1
Kt
kh +
k −h [ H+ ] +
ki
+
1 + Kt
1 + Kt
[ H ] + Ka
9 of 14
(15)
+
−
OH ] −
k flashk(tautomer ization) = k i +=k −kit +
+ kk−−Htt [H
] + k+ OH [OH
+ kH
−t [H −] t+ k −t [OH ]
flash (tautomer ization)
(16) (16)
0.004
0.8
0.6
-1
-1
=5.9x10 s
(s )
-4
obs
k
obs
A (483 nm)
k
0.1
0
50
100
150
0.002
Time (mim)
A
0
0
250
300
350
400
450
500
550
600
1
2
Wavelength (nm)
(a)
3
4
5
6
7
pH
(b)
Figure
9. (a)
of theofUV-visible
spectrum
of P5 after
directa pH
jump
to 5.8;1 to
Figure
9. time-dependence
(a) time-dependence
the UV-visible
spectrum
of P5aafter
direct
pHfrom
jump1 from
and5.8;
(b) and
pH-dependence
of the apparent
first-orderfirst-order
rate constant
flavylium
and its
(b) pH-dependence
of the apparent
rateofconstant
of consumption
flavylium consumption
−4 s−1; −
−1
fitting
according
to
Equation
(14):
K
hKtki = 8.5 × 10−7 M·s−1; k−i = 3.0 × 10
k
i1kt/k−h = 1.6 × 10−5 M−s
−
7
and its fitting according to Equation (14): Kh Kt ki = 8.5 × 10
M·s ; k−i = 3.0 × 10 4 s−1 ;
−1 Ka set at−10
−3.7
(kh =k 0.053
kt /k s =), 1.6
× 10 5 M·.s−1 (k = 0.053 s−1 ), Ka set at 10−3.7 .
i
−h
h
The flash photolysis experiments at pH = 4.2 are reported in Figure S2a. After the flash, Ct
The flash photolysis experiments at pH = 4.2 are reported in Figure S2a. After the flash,
(λmax = 360 nm) is instantaneously converted into Cc (lower molar absorption coefficient at this
Ct (λmax = 360 nm) is instantaneously converted into Cc (lower molar absorption coefficient at
wavelength),
and then rapidly regenerated due to the lack of cis-trans isomerization barrier, in
this wavelength), and then rapidly regenerated due to the lack of cis-trans isomerization barrier,
competition with flavylium formation (λmax = 450 nm).
in competition with flavylium formation (λmax = 450 nm).
The pH dependence of the corresponding apparent first-order rate constants shows the
The pH dependence of the corresponding apparent first-order rate constants shows the change
change from hydration regime (higher pH, Equation (15)) to tautomerization regime (lower pH,
from hydration regime (higher pH, Equation (15)) to tautomerization regime−1(lower pH, Equation (16)).
3
Equation (16)). The curve-fittings (Figure S2b) were achieved for −
kh = 0.053 s , k−h/(1 + Kt) =39 × −10
1 · s−1 ;
The
curve-fittings (Figure
S2b) were achieved
for kh−1 =−10.053 s 1 , k−h /(1
+
K−1t ) = 9 × 10 M
-1
−1
−1
−1
10
−1
−3.7
M s ; Ktki/(1 + Kt) = 0.15−s1 , k−t + ki = 1.35 s−1; H = 25 M−·s
;
= 2 × 10 M
a set to 10
in−3.7
10 ·s−1(K−
1 (K set to )10
K ki /(1 + Kt ) = 0.15 s , k−t + ki = 1.35 s ; k −t = 25 M 1 ·s−1 ; kOH
)
a
−t = 2 × 10 M ·s
goodt agreement
with the data from the bell-shaped curve.
in good agreement with the data from the bell-shaped curve.
The quantum yield, i.e., the percentage of incident light energy used for the Ct→Cc conversion,
The quantum yield, i.e., the percentage of incident light energy used for the Ct→Cc conversion, is
is a useful parameter for a complete estimation of the rate and equilibrium constants [13]. Its apparent
a useful parameter for a complete estimation of the rate and equilibrium constants [13]. Its apparent
value φ can be estimated from the percentage of color generated after illumination. It thus reflects
value φ can be estimated from the percentage of color generated after illumination. It thus reflects
the pH-dependent contribution of the forward process (Cc→B→AH+ + A).+ When hydration is the
the pH-dependent contribution of the forward process (Cc→B→AH + A). When hydration is
rate-determining step (tautomerization equilibrium achieved), the apparent rate constants for the
the rate-determining step (tautomerization equilibrium achieved), the apparent rate constants for
forward and backward reactions are, respectively, the first and the last parts of Equation (15).
the forward and backward reactions +are, respectively, the first and the last parts of Equation (15).
Neglecting the contribution of kh[H+]/([H
] + Ka), the pH-dependence of the quantum yield is given
Neglecting the contribution of kh [H+ ]/([H+ ] + Ka ), the pH-dependence of the quantum yield is given
by Equation (17):
by Equation (17):
kforward
k forward
[H + ] [H+ ]
(17)
ΦΦ==ΦΦ0 0
= =
Kt k i
kbackward + [HK+t ]k+
(17)
kkforward
+ k+
forward
i
backward
k −h
[H ] +
k−h
Analogously, when the rate-determining step is tautomerization, the pH-dependence of the
quantum yield is easily deduced from Equation (16) and follows Equation (18):
Φ = Φ0
−
k− t + k−Ht [H + ] + k−OH
t [OH ]
−
k− t + k−Ht [H + ] + k−OH
t [OH ] + ki
(18)
From the curve-fitting of the φ vs. pH plots according to Equations (17) and (18) (Figure 10), the
Int. J. Mol. Sci. 2016, 17, 1751
10 of 14
Analogously, when the rate-determining step is tautomerization, the pH-dependence of the
Int. J. Mol. quantum
Sci. 2016, 17,
1751
yield
is easily deduced from Equation (16) and follows Equation (18):
i
+
−
k −t + k H
+kOH
2.3. Substituent Effects on the Flavylium Multistate
−t [ H
−t [OH
10 of 14
i
Φ = Φ0
i
i
H [H + + kOH [OH − + k
k
+
k
−t reproduced
i
−twith the corresponding
−t
were
(18)
The investigations of P3 and P5
O-methylethers P4
and P6 and with the parent 4′,7-dihydroxyflavylium ion (DHF) itself. Although DHF has already
From the curve-fitting of the φ vs. pH plots according to Equations (17) and (18) (Figure 10), the ki
been studied
[7],
the complementary
analysis
of theof quantum
yield (deduced from the flash
value can
be estimated,
thereby allowing
the calculation
the other constants.
photolysis experiments) conducted in this work permitted a more accurate determination of kt and
2.3. Substituent Effects on the Flavylium Multistate
k−t. The complete
set of thermodynamic and kinetic constants characterizing the multistates of the
The
of P3 1.
and
P5 energy
were reproduced
with theofcorresponding
O-methylethers
five pigments
is investigations
reported in Table
The
level diagrams
P3 and P5 are
also represented in
0 ,7-dihydroxyflavylium ion (DHF) itself. Although DHF has
P4
and
P6
and
with
the
parent
4
Figure 11.
already been studied [7], the complementary analysis of the quantum yield (deduced from the flash
photolysis experiments) conducted in this work permitted a more accurate determination of kt and
Table
1. Complete sets of thermodynamic and kinetic rate constants characterizing the multistate of
k−t . The complete set of thermodynamic and kinetic constants characterizing the multistates of the
the pigments
investigated
work.
five pigments
is reportedininthis
Table
1. The energy level diagrams of P3 and P5 are also represented in
Figure 11.
^
Compound
pK’a
pK a
pKa
Kh/M *
Kt *
−6
DHF Table 1.
0.4 of
3.05
± 0.10sets of thermodynamic
–
4.0 rate
± 0.1
1.4 × 10
Complete
and kinetic
constants characterizing
the multistate
−6
the
pigments
investigated
in
this
work.
P3
0.60
2.30 ± 0.05
4.70 ± 0.05
5.40 ± 0.05
8.9 × 10
−6
P4
2.15 ± 0.05
4.80 ±pK0.05
–
8.4 × 10K *
1.0
ˆ
Compound
pK’a
pKa
Kh /M *
Ki *
t
a
−6
P5
2.50
±
0.05
–
3.70
±
0.05
4.3
×
10
0.33
−
6
DHF
3.05 ± 0.10
–
4.0 ± 0.1
0.4
1.4 × 10
1.4 × 103
−
6
−6
P32.60 ± 0.05
2.30 ± 0.05 5.20
4.70
0.05
5.40 ± 0.05
935
8.9 × 10 3.6 × 100.60
P6
±±0.05
–
0.68
P4
2.15 ± 0.05
4.80 ± 0.05
–
1.0
841
8.4 × 10−6
Compound P5 kh/s−12.50
* ± 0.05 k−h/M−1–·s−1 *
kt0.05
/s−1 * 4.3 × 10−6k−t/s−1 0.33
*
ki2.0
/s−1× *103
3.70 ±
−6
P6
5.20 ± 0.05
–
1.0 × 103
DHF
0.22 3.6 × 10 0.55 0.68
0.26
0.022.60 ± 0.05 1.3
× 104
−
1
−
1
−
1
−
1
−
1
−
1
Compound
k /s *
k−h /M ·s *
kt /s *
k−t /s *
k /s *
k /s−1 *
P3
0.36
0.60i
5.5 ×−i10−3−4
0.08 h
9 × 103 4
DHF
0.02
0.22
0.55
0.26
1.3 × 10
1.8 × 10
4
P4
0.09 0.08
1.19××10
0.75
× ×1010−3−6
P3
0.360.75
0.60
103
5.5 × 10−3 4.75.9
4
−
3
P4
0.75
0.75
104
4.7 × 10
5.6 × 10−6
P5
0.05 0.09
1.31.1
××10
0.25
0.75
0.60
P5
0.05
0.25
0.75
0.60
1.3 × 104
3.0 × 10−4
4
4
P6
0.500.50
0.75
10
7.8 × 10−5
P6
0.05 0.05
1.41.4××10
0.75 0.08
0.08
* Estimated
= 10%;
values obtained
at room
temperature.
* Estimated
error error
= 10%;
allallvalues
obtained
at room
temperature.
Ki *
1.4 × 103
935
841
2.0 × 103
1.0 × 103
k−i/s−1 *
1.8 × 10−4
5.9 × 10−6
5.6 × 10−6
3.0 × 10−4
7.8 × 10−5
0.04
Φ
0.02
0
1
2
3
4
5
6
7
pH
Figure 10.
Quantum
yield ofyield
P5 as
function
of pH.ofFitting
was was
achieved
with
Equations
Figure
10. Quantum
ofaP5
as a function
pH. Fitting
achieved
with
Equations(17)
(17)and (18)
−5 M s−1 ; k K /(1 +
−1 ; k /(1 + K3) = 9
3 M−1 ·s−1 ;
−5
−1
−1
−1
−1
and
(18)
for
k
k
/k
=
1.6
×
10
K
)
=
0.15
s
×
10
t
i t M
−h s ; kiKt/(1 + Kt) = i0.15
for kikt/k−h = 1.6 × 10
s ; tk−h/(1 + Kt) =−h9 × 10 t M ·s ;
= 25 M−1·s−1;
−1 ·s−1 ; kOH = 2 × 1010 M−1 ·s−1 .
kH
=
25
M
t 10 M−1·s−1.
−t
= 2 ×−10
0
1
2
3
4
5
6
7
pH
Figure 10. Quantum yield of P5 as a function of pH. Fitting was achieved with Equations (17) and (18)
for kikt/k−h = 1.6 × 10−5 M s−1; kiKt/(1 + Kt) = 0.15 s−1; k−h/(1 + Kt) = 9 × 103 M−1·s−1;
= 25 M−1·s−1;
Int. J. Mol. Sci. 2016,
17,
1751
11 of 14
= 2 × 1010 M−1·s−1.
(a)
(b)
Figure
state.
Figure11.
11.Energy
Energylevel
leveldiagrams
diagramsofofP3P3(a)(a)and
andP5P5(b).
(b).*: *:Excited
Excited
state.
Int. J. Mol. Sci. 2016, 17, 1751
11 of 14
The major consequence of methylation is obviously to cancel the acid-base
acid-base equilibrium
(Equation (1)). While
While the
the influence
influence of methylation
methylation on the hydration and tautomerization steps is
marginal,
chalcone
isomerization
is very
significant.
Interestingly,
only DHF
marginal, its
itsimpact
impacton
onthe
thecis-trans
cis-trans
chalcone
isomerization
is very
significant.
Interestingly,
only
and
P5,
i.e.,
the
two
pigments
with
an
OH
group
at
C7,
display
a
low
isomerization
barrier,
which
DHF and P5, i.e., the two pigments with an OH group at C7, display a low isomerization barrier,
prevents
the establishment
of the pseudo-equilibrium
(Equation (7)).
Glycosidation
or methylation or
of
which prevents
the establishment
of the pseudo-equilibrium
(Equation
(7)). Glycosidation
C7–OH
(P3,
P4,
P6)
raises
the
barrier
and
makes
the
cis-trans
chalcone
isomerization
a
slow
kinetically
methylation of C7–OH (P3, P4, P6) raises the barrier and makes the cis-trans chalcone isomerization
distinct
process, asdistinct
observed
with as
natural
anthocyanins.
However,
the influence
of glycosidation
is
a slow kinetically
process,
observed
with natural
anthocyanins.
However,
the influence of
much
more pronounced.
Forpronounced.
instance, there
a ca. 100-fold
difference
between
the ki values
of the
glycosidation
is much more
Forisinstance,
there is
a ca. 100-fold
difference
between
two
regioisomers
P3regioisomers
and P5, and glycosidation
of C7–OH
(P3 vs.of
DHP)
lowers
a factor
ca. 50,
vs.
ki values
of the two
P3 and P5, and
glycosidation
C7–OH
(P3 kvs.
lowers
ki by
i byDHP)
0
only
7.5 ca.
for 50,
methylation
(P5for
vs.methylation
P6). By contrast,
of C4glycosidation
-OH (P5 vs. DHP)
moderately
a factor
vs. only 7.5
(P5 vs.glycosidation
P6). By contrast,
of C4′-OH
(P5 vs.
accelerates
isomerization
(a factor
of ca. 2). (a factor of ca. 2).
DHP) moderately
accelerates
isomerization
As a powerful electron-donating group at C7 is expected to weaken the C=C
C=C bond of chalcones
through electron-delocalization
(Figure 12),
12), it may be concluded that the Glc moiety of P3 and P4
electron-delocalization (Figure
(through the combination of the electron-withdrawing effects of its O-atoms) largely quenches the
electron-donating
capacity of the O-atom at C7. Although much less pronounced, this effect is also
electron-donating capacity
significant with flavylium ions. Indeed, water addition to P3 and
and P4
P4 is
is four
four times
times as
asfast
fastas
aswith
withDHP.
DHP.
It is thus likely that lowering the electron-donating capacity of the substituent at C7 increases the
fraction of positive charge at C2 and consequently
consequently the
the hydration
hydration rate
rate constant.
constant.
Figure 12.
12. Electron
from the
the substituent
substituent at
at C7
C7 (R
(R7 == H,
Me, Glc) and its accelerating effect on
Figure
Electron donation
donation from
7 H, Me, Glc) and its accelerating effect on
cis-trans
chalcone
isomerization.
cis-trans chalcone isomerization.
The apparent rate constant of color loss for P3 increases with pH to reach a plateau value at
The apparent rate constant of color loss for P3 increases with pH to reach a plateau value at pH 6–7
pH 6–7 (Figure 6b). This trend reflects the rate-limiting isomerization and the higher fraction of
(Figure 6b). This trend reflects the rate-limiting isomerization and the higher fraction of cis-chalcone
cis-chalcone when the pH increases. By contrast, above pH = 4, the apparent rate constant of color
when the pH increases. By contrast, above pH = 4, the apparent rate constant of color loss for P5
loss for P5 decreases with pH (Figure 9b). In the absence of a high cis-trans isomerization barrier,
hydration becomes the rate-limiting step and the slower fading now reflects the lower fraction of
flavylium ion when the pH increases. These subtle changes make the P5 color much more stable than
the P3 color in mildly acidic to neutral conditions (at pH 6, the half-life of color is ca. 40 min for P5
vs. ca. 7 min for P3). By contrast, the coloring potential of P5 in such conditions is moderated by the
Int. J. Mol. Sci. 2016, 17, 1751
12 of 14
decreases with pH (Figure 9b). In the absence of a high cis-trans isomerization barrier, hydration
becomes the rate-limiting step and the slower fading now reflects the lower fraction of flavylium ion
when the pH increases. These subtle changes make the P5 color much more stable than the P3 color in
mildly acidic to neutral conditions (at pH 6, the half-life of color is ca. 40 min for P5 vs. ca. 7 min for
P3). By contrast, the coloring potential of P5 in such conditions is moderated by the quinoidal base
(proton loss at C7–OH) having ca. half the molar absorption coefficient of the corresponding flavylium
at λmax (Figure 8a), while, for P3, both flavylium and quinoidal base (proton loss at C40 –OH) have
roughly the same capacity for light absorption at λmax (Figure 3a).
2.4. Substituent Effects on the Performance of the Photochromic System
Glycosidation and/or methylation of DHF also have interesting implications on the photochemical
properties of the corresponding pigments. The primary photochemical event is trans-cis isomerization,
and the cis-chalcone thus formed decays to products, i.e., the colored flavylium cation or quinoidal
base
(depending
on1751
the pH, forward route) and the colorless trans-chalcone (backward route). At
Int.
J. Mol.
Sci. 2016, 17,
12 near
of 14
neutral pH, the cis-chalcone of P5 (low isomerization barrier) reverts back almost quantitatively to the
trans-chalcone
in a few
seconds without
forming appreciable
amounts
quinoidal
base (Figure
13a).
reverts
back almost
quantitatively
to the trans-chalcone
in a few
secondsofwithout
forming
appreciable
By contrast,
with P4 (high
barrier),13a).
the competing
reaction
is much
slower, and,
after the
amounts
of quinoidal
base (Figure
By contrast,backward
with P4 (high
barrier),
the competing
backward
flash, theisflavylium
ion accumulates
over
1 min,
then theion
system
reverts back
in less
reaction
much slower,
and, after the
flash,
theand
flavylium
accumulates
overto1 equilibrium
min, and then
the
than 1 hreverts
(Figureback
13b).toAs
P6 exhibitsin
a lower
barrier
P4 (k
(P6)/k
(P4)
=
17),
the
visible
absorbance
system
equilibrium
less than
1 h than
(Figure
13b).
As
P6
exhibits
a
lower
barrier
than
i
i
is maximized
only
25 the
s after
the flash
(Figureis13c),
and thenonly
the system
returns
to equilibrium
inand
less
P4
(ki(P6)/ki(P4)
= 17),
visible
absorbance
maximized
25 s after
the flash
(Figure 13c),
than the
5 min.
Finally,
P3 and
display similar
However,
irradiation
then
system
returns
to P4
equilibrium
in lessisomerization
than 5 min. barriers.
Finally, P3
and P4while
display
similar
of P4 rapidly barriers.
forms the
flavylium
ion,irradiation
irradiationofofP4P3
more forms
slowlythe
leads
to the ion,
quinoidal
base
isomerization
However,
while
rapidly
flavylium
irradiation
+
+
+
+
(cf.P3
the
[H ]/([H
+ Ka ) to
term
Equationbase
(10)).
With
a maximal
of quinoidal
base
is
of
more
slowly]leads
thein
quinoidal
(cf.
the P3,
[H ]/([H
] + Ka)accumulation
term in Equation
(10)). With
P3,
5 min after the
flash (Figure
aobserved
maximalca.
accumulation
of quinoidal
base13d).
is observed ca. 5 min after the flash (Figure 13d).
(a)
(b)
(c)
(d)
Figure
Figure 13.
13. Transient
Transient absorption
absorption upon
upon aa light
light flash
flash applied
applied to
to an
an equilibrated
equilibrated solution
solution of
of (a)
(a) P5
P5 at
at
pH
=
6.05;
(b)
P4
at
pH
=
6.06;
(c)
P6
at
pH
=
5.9;
and
(d)
P3
at
pH
=
6.5.
pH = 6.05; (b) P4 at pH = 6.06; (c) P6 at pH = 5.9; and (d) P3 at pH = 6.5.
In conclusion, the O-glucosides of 4′,7-dihydroxyflavylium and their O-methylethers are simple,
water-soluble and easily accessible analogs of natural 3-deoxyanthocyanins. This work demonstrates
that glycosidation of critical OH groups exerts a subtle influence on the reaction network of flavylium
ions and so also on the resulting color and photochromic properties, thus confirming the high
versatility of the flavylium multistate.
3. Material and Methods
Int. J. Mol. Sci. 2016, 17, 1751
13 of 14
In conclusion, the O-glucosides of 40 ,7-dihydroxyflavylium and their O-methylethers are simple,
water-soluble and easily accessible analogs of natural 3-deoxyanthocyanins. This work demonstrates
that glycosidation of critical OH groups exerts a subtle influence on the reaction network of flavylium
ions and so also on the resulting color and photochromic properties, thus confirming the high versatility
of the flavylium multistate.
3. Material and Methods
3.1. Chemicals
All pigments were chemically synthesized according to already published procedures and
completely characterized by NMR and HPLC-DAD-MS analysis [16].
3.2. Analyses
Stock solutions of flavylium cations were prepared in 0.1 M HCl. The pH of solutions (measured
with a Crison basic 20+ pH meter, Barcelona, Spain) was adjusted by addition of HCl, NaOH and/or
pH 8.5 universal buffer (33 mM sodium citrate, 33 mM sodium phosphate, and 57 mM sodium borate).
Direct pH jumps were carried out by addition of 1 mL of flavylium cation at pH = 1 to a cuvette
(1 cm optical pathlength) containing 1 mL of 0.1 M NaOH and 1 mL of universal buffer previously
adjusted at the desired pH with concentrated HCl. Reverse pH jumps were carried out through the
addition of HCl to solutions equilibrated or pseudo-equilibrated (i.e., before significant formation of
trans-chalcone) at slightly acidic or neutral pH values.
UV/Vis absorption spectra were recorded on a Varian Cary 100 Bio or 5000 spectrophotometer
(Palo Alto, CA, USA). The stopped-flow experiments were conducted on an Applied Photophysics
SX20 stopped-flow spectrometer equipped with a photodiode array detector (Leatherhead, UK).
Direct pH jumps investigated by stopped-flow were carried out by charging one syringe (6 mL) with
flavylium cation solution at pH = 1 and the other syringe with 4 mL of universal buffer at the desired
pH and 2 mL of 0.3 M NaOH. The analyzed solutions were recovered and the pH measured with a
glass electrode.
Flash photolysis experiments were performed on a Varian Cary 5000 spectrophotometer with a
Harrick fiber-mate (Pleasantville, NY, USA) attached to a Ocean Optics 4-way cuvette holder (Dunedin,
FL, USA) to perform light excitation perpendicular to the analyzing beam (sample compartment
protected from daylight by black cardboard and black tape). As a pulsed white light source,
a commercially available Achiever 630AF camera flash (Hong Kong, China), placed in close contact
with the sample holder, was used (time resolution of ca. 0.05 s) [17]. Experiments under continuous
irradiation were conducted using a Xenon lamp (excitation band isolated with a monochromator)
(Osram, München, Germany). The solutions were irradiated in a quartz cuvette (path length = 1 cm)
with magnetic stirring. The absorption spectra were registered before irradiation and at selected time
points of irradiation period. The incident light intensity was measured by ferrioxalate actinometry [18].
All curve-fitting procedures were carried out using the solver program from Microsoft Excel 2016
(Redmond, WA, USA).
Supplementary Materials: Supplementary materials can be found at www.mdpi.com/1422-0067/17/10/1751/s1.
Acknowledgments: This work was supported by the Associated Laboratory for Sustainable Chemistry, Clean
Processes and Technologies—Laboratório Associado para a Química Verde (LAQV), which is financed by national
funds from Fundação para a Ciência e Tecnologia / Ministério da Ciência, Tecnologia e Ensino Superior
(UID/QUI/50006/2013) and co-financed by the European Regional Development Funds under the PT2020
Partnership Agreement (POCI-01-0145-FEDER-007265). Nuno Basílio gratefully acknowledges the post-doctorate
grant from the FCT/MEC (SFRH/BPD/84805/2012).
Author Contributions: Nuno Basílio and Fernando Pina conceived and designed the experiments; Nuno Basílio
performed the experiments; Nuno Basílio and Fernando Pina analyzed the data; Sheiraz Al Bittar, Nathalie Mora
and Olivier Dangles contributed the pigments; Olivier Dangles and Fernando Pina wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
Int. J. Mol. Sci. 2016, 17, 1751
14 of 14
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Kayodé, A.P.P.; Nout, M.J.R.; Linnemann, A.R.; Hounhouigan, J.D.; Berghofer, E.; Siebenhandl-Ehn, S.
Uncommonly high levels of 3-deoxyanthocyanidins and antioxidant capacity in the leaf sheaths of dye
sorghum. J. Agric. Food Chem. 2011, 59, 1178–1184. [CrossRef] [PubMed]
Swinny, E.E. A novel acetylated 3-deoxyanthocyanidin laminaribioside from the fern Blechnum novae-zelandiae.
Z. Naturforsch. C 2001, 56, 177–180. [CrossRef] [PubMed]
Iwashina, T.; Kitajima, J.; Matsumoto, S. Luteolinidin 5-O-glucoside from Azolla as a stable taxonomic marker.
Bull. Natl. Mus. Nat. Sci. Ser. B 2010, 36, 61–64.
Paula, J.T.; Paviani, L.C.; Foglio, M.A.; Sousa, I.M.O.; Duarte, G.H.B.; Jorge, M.P.; Eberlin, M.N.; Cabral, F.A.
Extraction of anthocyanins and luteolin from Arrabidaea chica by sequential extraction in fixed bed using
supercritical CO2 , ethanol and water as solvents. J. Supercrit. Fluids 2014, 86, 100–107. [CrossRef]
Taylor, J.R.N.; Belton, P.S.; Beta, T.; Duodu, K.G. Increasing the utilisation of sorghum, millets and
pseudocereals: Developments in the science of their phenolic phytochemicals, biofortification and protein
functionality. J. Cereal Sci. 2014, 59, 257–275. [CrossRef]
Yang, L.; Dykes, L.; Awika, J.M. Thermal stability of 3-deoxyanthocyanidin pigments. Food Chem. 2014, 160,
246–254. [CrossRef] [PubMed]
Pina, F.; Melo, M.J.; Laia, C.A.T.; Parola, A.J.; Lima, J.C. Chemistry and applications of flavylium compounds:
A handful of colours. Chem. Soc. Rev. 2012, 41, 869–908. [CrossRef] [PubMed]
Pina, F. Chemical applications of anthocyanins and related compounds. A source of bioinspiration. J. Agric.
Food Chem. 2014, 62, 6885–6897. [CrossRef] [PubMed]
Pina, F.; Melo, M.J.; Maestri, M.; Passaniti, P.; Balzani, V. Artificial chemical systems capable of mimicking
some elementary properties of neurons. J. Am. Chem. Soc. 2000, 122, 4496–4498. [CrossRef]
Brouillard, R.; Dubois, J.-E. Mechanism of the structural transformations of anthocyanins in acidic media.
J. Am. Chem. Soc. 1977, 99, 1359–1364. [CrossRef]
Basílio, N.; Laia, C.A.T.; Pina, F. Excited-state proton transfer in confined medium. 4-Methyl-7hydroxyflavylium and β-naphthol incorporated in cucurbit [7] uril. J. Phys. Chem. B. 2015, 119, 2749–2757.
[CrossRef] [PubMed]
Pina, F.; Oliveira, J.; de Freitas, V. Anthocyanins and derivatives are more than flavylium cations. Tetrahedron
2015, 71, 3107–3114. [CrossRef]
Basílio, N.; Fernandes, A.; de Freitas, V.; Gago, S.; Pina, F. Effect of β-cyclodextrin on the chemistry of
30 ,40 ,7-trihydroxyflavylium. New J. Chem. 2013, 37, 3166–3173. [CrossRef]
Leydet, Y.; Batat, P.; Jonusauskas, G.; Denisov, S.; Lima, J.C.; Parola, A.J.; McClenaghan, N.D.; Pina, F. Impact
of water on the cis-trans photoisomerization of hydroxychalcones. J. Phys. Chem. 2013, 117, 4167–4173.
[CrossRef] [PubMed]
Pina, F. Anthocyanins and related compounds. Detecting the change of regime between rate control by
hydration or by tautomerization. Dyes Pigment. 2014, 102, 308–314. [CrossRef]
Al Bittar, S.; Mora, N.; Loonis, M.; Dangles, O. A simple synthesis of 3-deoxyanthocyanidins and their
O-glucosides. Tetrahedron 2016, 72, 4294–4302. [CrossRef]
Maestri, M.; Ballardini, R.; Pina, F.; Melo, M.J. An easy and cheap flash spectroscopy experiment.
J. Chem. Educ. 1997, 74, 1314–1316. [CrossRef]
Hatchard, C.G.; Parker, C.A. A new sensitive chemical actinometer. II. Potassium ferrioxalate as a standard
chemical actinometer. Proc. R. Soc. Lond. Ser. A 1956, 235, 518–536. [CrossRef]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).