Effect of phosphate buffer on the complexation and photochemical interaction of riboflavin and caffeine in aqueous solution: A kinetic study

Article abstract of DOI:10.1016/j.jphotochem.2013.09.007

A study of the photodegradation of 5 × 10^-5 M riboflavin (RF) in 0.2-1.0 M phosphate buffer in the presence and absence of 2.50 × 10^-4 M caffeine at pH 6.0-8.0 has been carried out. RF in phosphate buffer is photodegraded simultaneously by normal photolysis (photoreduction) and photoaddition reactions giving rise to lumichrome (LC) and cyclodehydroriboflavin (CDRF) as the main final products, respectively. RF and its photoproducts, formylmethylflavin (FMF), lumiflavin (LF), LC and CDRF in degraded solution have been determined by a specific multicomponent spectrophotometric method with an accuracy of ±5%. The apparent first-order rate constants for the photodegradation of RF and for the formation of LC and CDRF are 5.47-15.05 × 10^-3 min^-1, 1.06-8.30 × 10^-3 min^-1 and 4.31-8.05 × 10 ^-3 min^-1, respectively. An increase in phosphate concentration leads to an increase in the rate of formation of CDRF and alters the photodegradation of RF in favor of the photoaddition reaction. This photoaddition reaction is further enhanced in the presence of caffeine which results in a further decrease of the fluorescence of RF in phosphate buffer. Caffeine may facilitate the photoaddition reaction by suppression of the photoreduction pathway of RF.

Full text of DOI:10.1016/j.jphotochem.2013.09.007

Journal of Photochemistry and Photobiology A: Chemistry 273 (2014) 17–22
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Effect of phosphate buffer on the complexation and photochemical
interaction of riboflavin and caffeine in aqueous solution: A kinetic
study
Muhammad Ali Sheraza, Sadia Hafeez Kazia, Sofia Ahmeda, Tania Mirzaa, Iqbal Ahmada,
Maxim P. Evstigneevb,c,∗
a Institute of Pharmaceutical Sciences, Baqai Medical University, Toll Plaza, Super Highway, Gadap Road, Karachi 74600, Pakistan
b Department of Physics, Sevastopol National Technical University, Universitetskaya str. 33, Sevastopol 99053, Crimea, Ukraine
c Department of Biological and Chemical Sciences, Belgorod State University, Pobeda str. 85, Belgorod 308015, Russia
a r t i c l e i n f o
a b s t r a c t
A study of the photodegradation of 5 × 10−5 M riboflavin (RF) in 0.2–1.0 M phosphate buffer in the pres-
ence and absence of 2.50 × 10−4 M caffeine at pH 6.0–8.0 has been carried out. RF in phosphate buffer is
photodegraded simultaneously by normal photolysis (photoreduction) and photoaddition reactions giv-
ing rise to lumichrome (LC) and cyclodehydroriboflavin (CDRF) as the main final products, respectively.
RF and its photoproducts, formylmethylflavin (FMF), lumiflavin (LF), LC and CDRF in degraded solution
have been determined by a specific multicomponent spectrophotometric method with an accuracy of
5%. The apparent first-order rate constants for the photodegradation of RF and for the formation of LC
and CDRF are 5.47–15.05 × 10−3 min−1, 1.06–8.30 × 10−3 min−1 and 4.31–8.05 × 10−3 min−1, respectively.
An increase in phosphate concentration leads to an increase in the rate of formation of CDRF and alters
the photodegradation of RF in favor of the photoaddition reaction. This photoaddition reaction is further
enhanced in the presence of caffeine which results in a further decrease of the fluorescence of RF in phos-
phate buffer. Caffeine may facilitate the photoaddition reaction by suppression of the photoreduction
pathway of RF.
Article history:
Received 28 June 2013
Received in revised form
11 September 2013
Accepted 14 September 2013
Available online xxx
Keywords:
Riboflavin
Caffeine
Phosphate buffer
Photoaddition
Photoreduction
Kinetics
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
cyclization) reactions of RF [9–14]. These two major pathways of
the photodegradation of RF occur simultaneously depending on
Buffers are considered as an important component of chemical
and pharmaceutical systems to achieve optimum stability of a com-
pound or to perform a reaction under controlled condition of pH,
buffer concentration and ionic strength. In many cases, the buffers
affect the stability of pharmaceutical compounds in aqueous solu-
tion [1–6]. Several studies have been carried out to evaluate the
effect of buffer species on the photolysis of riboflavin (RF). These
studies include the catalytic effect of acetate [7] and phosphate
buffers [8–14] and the inhibitory effect of borate [15] and citrate
buffers [16]. It has been shown that the monovalent phosphate
ions (H2PO4–) exert a catalytic effect on the normal photolysis or
photoreduction (side-chain cleavage) of RF [11,17] and the diva-
lent phosphate ions (HPO42–) on the photoaddition (side chain
the pH, buffer concentration and light intensity and wavelengths
[9–14]. The rate of photoaddition reaction of RF at pH 7.0 is more
than twice of the photolysis reaction [11].
RF is known to form molecular complexes with caffeine [18–24]
which causes the stabilization of RF and thus inhibits its rate of
chemical [25] and photodegradation reactions [26,27] in aque-
ous solution. However, the previous studies deal only with the
individual effect of caffeine (CF) [27], or the phosphate [11–14]
on the photodegradation of RF. These studies do not provide
information on the combined effect of CF and phosphate on the
reaction.
CF molecules may enter the human body from various food
sources in large amounts and, thereby, may influence RF function
in organism. Hence, the knowledge of photochemical interaction
of CF–RF is important for understanding their interaction in a
biological system. The knowledge of the set of factors altering
RF photodegradation in solution is also important in view of
the positive synergistic effect of blue light and RF with respect
to suppression of tumor cells growth in vivo [28] and the blue
∗
Corresponding author at: Department of Physics, Sevastopol National Technical
University, Universitetskaya str. 33, Sevastopol 99053, Crimea, Ukraine.
Tel.: +38 067 9210912; fax: +38 692 243 590.
E-mail address: max evstigneev@mail.ru (M.P. Evstigneev).
1010-6030/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jphotochem.2013.09.007

18
M.A. Sheraz et al. / Journal of Photochemistry and Photobiology A: Chemistry 273 (2014) 17–22
light-mediated DNA repair by photolysate enzymes containing
flavin adenine dinucleotide (FAD) as cofactor [29].
at 356 and 445 nm, the respective absorption maxima of LC and
LF [11]. The other three compounds (RF, FMF, CDRF), present in
the aqueous phase, are determined by a three-component assay
at 385, 410, 445 nm, corresponding to the absorption maxima of
FMF, CDRF and RF, respectively [11]. The absorption maxima of RF
and FMF (445 nm, pH 7.0) can be distinguished at pH 2.0 where
FMF exists in a protonated form (pKa 3.5) [34] and thus can be
assayed by this method. CMF is a minor product of the oxida-
tion of FMF [13,17,35] and has not been considered in the assay
scheme.
The present work is based on the evaluation of the effects of
CF and phosphate together on the photodegradation of RF and the
role of CF vis-a-vis phosphate in altering the rates of the reac-
tions. The study provides kinetic evidence to support the view that
CF is involved in modifying the role of phosphate to enhance the
photodegradation of RF. This would facilitate the understanding of
the interaction of CF–RF in phosphate buffer and its influence on
structural orientation to cause a particular change in the mode of
degradation reactions.
2.4. Spectral measurements
2. Materials and methods
The spectral and absorbance measurements on RF and its
photolyzed solutions were performed on a Shimadzu UV-1601
recording spectrophotometer using quartz cells of 10-mm path
length.
RF, LC, LF and CF were obtained from Sigma–Aldrich Chemi-
cals Co. (St. Louis, MD, USA). FMF, CMF and CDRF were prepared
by the previously described methods, respectively [30,31]. All
the reagents and solvents were of analytical grade or of the
purest form available from Merck & Co. (Whitestone Station, NJ,
USA).
2.5. Fluorescence measurements
The fluorescence of RF solutions was measured at room temper-
ature (25 ◦C) with a Spectromax 5 fluorimeter (molecular devices,
Sunnyvale, CA, USA) in the end point mode using 374 nm as the
excitation wavelength and 525 nm as the fluorescence wavelength
[36]. The fluorescence was measured in relative fluorescence units
using a pure 0.05 mM RF solution as the standard.
2.1. Photolysis of RF
The photolysis of 5 × 10−5 M RF solutions, containing
2.5 × 10−4 M CF was carried out at 25 1 ◦C in the presence
and absence of 0.2–1.0 M Na2HPO4 at pH 6.0–8.0. The pH of the
solution was adjusted with 1.0–5.0 M HCl/NaOH solution. The
solution (100 ml) was irradiated in a dark chamber using a Philips
125 W high pressure mercury vapor fluorescent lamp (emission
at 405 and 435 nm), fixed horizontally at a distance of 30 cm from
the center of the flask. The 435 nm band of the radiation source
overlaps the 445 nm absorption maximum of RF [11] while CF
(ꢀmax 273 nm) [32] does not absorb in the visible region. The
solution was stirred by bubbling a stream of air into the flask.
Samples were withdrawn at appropriate intervals for chromato-
graphic examination and spectrophotometric determination. The
photolysis of RF was also carried out under the same conditions in
the presence of 1 M phosphate and 0.625–2.50 × 10−4 M CF at pH
7.0.
2.6. Light intensity measurement
The measurement of the intensity of Philips HPL-N 125 W high
pressure mercury vapor fluorescent lamp was performed using
potassium ferrioxalate actinometry [37]. The value of the intensity
of the lamp was determined as 1.16 0.098 × 1017 quanta s−1
.
3. Result and discussion
3.1. Composition of photodegraded solution of RF
It is necessary to ascertain the nature of products formed in the
photodegraded solutions of RF in the presence of phosphate buffer
and CF at pH 6.0–8.0. TLC of the solutions using solvent system
(a) showed the presence of the compounds (Rf values in paren-
theses): undergraded RF (0.36), FMF (0.61), LC (0.67), LF (0.42)
and CMF (0.26), with their characteristic fluorescence emission
(mentioned in Section 2 under thin layer chromatography), and
solvent system (b): undergraded RF (0.37), LC (0.82) and CDRF
(0.46) (red color). These products have previously been identified
in the photodegradation of RF under similar conditions and arise
from the photoreduction and photoaddition pathways [9,17,27,38].
The formation (FMF, LC, LF, CMF, CDRF) and loss (RF, FMF, CMF)
of these compounds was monitored by changes in the intensity
of their spots. The formation of LF in the reaction takes place at
pH 7.0–8.0.
2.2. Thin layer chromatography
The identification of the photodegradation products of RF
was carried out by thin-layer chromatography (TLC) using 250-
␮m silica gel G plates (Merck) and the solvent systems: (a)
1-butanol–acetic acid/water (40:10:50 (v/v), organic phase) [33],
and (b) chloroform–methanol (9:2 (v/v)) [9]. The detection of
RF and its photodegradation products was performed by their
characteristic fluorescence emission under UV (365 nm) excita-
tion (RF, FMF, CMF – yellow green; LC, sky blue) using a Uvitech
lamp (Cambridge, UK) or by visual examination (CDRF – red
color). The progress of the photolysis reactions was monitored
by observing the intensity variations of the spots of different
products.
3.2. Spectral characteristics of photodegraded solutions
2.3. Spectrophotometric assay
The spectral variations in the photodegraded solutions of RF in
the presence of CF [27] and the phosphate buffer [9,11,14] have
been studied. In the presence of CF, RF shows a loss of absorbance
at 374 and 445 nm and a slight increase around 385 nm indicat-
ing the formation of FMF with time in the aqueous phase (pH 2.0)
after chloroform extraction to remove LC and LF. The mixture of
these products in the aqueous solution (pH 4.5) exhibits a grad-
ual increase in absorbance around 356 and 445 nm, the respective
absorption maxima of LC and LF. The absorption spectrum of the
photodegraded solutions of RF in the presence of phosphate buffer
The assay of RF and its side-chain cleavage photoproducts (FMF,
LC, LF) and cycloaddition photoproduct (CDRF) was performed
by a five component spectrophotometric method of Ahmad et al.
[11]. This method has previously been used in a number of pho-
todegradation studies of RF in the presence of phosphate buffer
[11–14]. It involves the pre-adjustment of the photodegraded solu-
tions to pH 2.0 (0.2 M HCl/KCl buffer), followed by the extraction
of LC and LF with chloroform and determination of the chloroform
residue at pH 4.5 (0.2 M acetate buffer) by a two-component assay

M.A. Sheraz et al. / Journal of Photochemistry and Photobiology A: Chemistry 273 (2014) 17–22
19
Table 1
Apparent first-order rate constants (kobs) for the photodegradation of RF and for the formation of CDRF (k1) and LC (k2ꢀ) in the presence of CF (2.5 × 104 M) and phosphate
buffer (0.2–1.0 M) at different pH valuesa.
ꢀ
pH
6.0
Phosphate conc. (M)
kobs × 103 (min−1
)
SD
k1 × 103 (min−1
)
SD
k2ꢀ × 103 (min−1
)
SD
k1/k2
0.2
0.4
0.6
0.8
1.0
5.69 0.28
7.49 0.45
8.76 0.46
9.94 0.62
11.17 0.60
(5.14 0.21)
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
6.5
7.0
7.5
0.2
0.4
0.6
0.8
1.0
5.43 0.24
6.76 0.34
8.83 0.40
10.40 0.69
11.70 0.70
(6.15 0.29)
1.12 0.07
2.19 0.11
3.70 0.22
4.99 0.32
6.05 0.35
(1.35 0.05)
4.31 0.26
4.57 0.25
5.13 0.32
5.41 0.33
5.65 0.28
(4.80 0.21)
0.26
0.48
0.72
0.92
1.07
(0.28)
0.2
0.4
0.6
0.8
1.0
5.87 0.32
7.92 0.51
10.99 0.55
13.18 0.71
15.05 0.90
(8.13 0.43)
1.53 0.08
2.81 0.14
4.85 0.29
6.68 0.30
8.30 0.42
(3.23 0.19)
4.32 0.28
5.11 0.31
6.14 0.36
6.50 0.26
6.75 0.39
(4.90 0.28)
0.36
0.55
0.79
1.03
1.23
(0.66)
0.2
0.4
0.6
0.8
1.0
8.85 0.57
2.15 0.15
3.43 0.24
4.49 0.27
5.52 0.33
6.92 0.38
(1.88 0.12)
6.70 0.40
7.30 0.37
7.36 0.38
7.60 0.34
8.05 0.55
(3.93 0.26)
0.32
0.47
0.61
0.73
0.86
(0.48)
10.73 0.60
11.85 0.53
13.12 0.79
14.98 0.82
(5.81 0.33)
8.0
0.2
0.4
0.6
0.8
1.0
5.47 0.33
6.57 0.43
7.59 0.44
9.31 0.47
10.25 0.62
(4.88 0.32)
1.06 0.04
1.70 0.10
2.31 0.12
3.26 0.19
4.03 0.22
(1.32 0.07)
4.41 0.29
4.87 0.29
5.28 0.26
6.05 0.39
6.21 0.41
(3.56 0.15)
0.24
0.35
0.44
0.54
0.65
(0.37)
a
The values of the rate constants in parentheses are in the absence of CF in 1.0 M phosphate buffer.
(pH 7.0) show the loss of RF around 445 nm and the formation of LC
and CDRF around 356 and 410 nm, respectively [14]. The absorp-
tion spectra of RF solution (aqueous phase), photodegraded in the
presence and absence of CF in phosphate buffer (pH 7.0) exhibit
similar spectral variations indicating the loss of RF and the for-
mation of FMF and CDRF by different pathways. It is interesting
to note that the 410/445 nm (CDRF/RF) ratio in the presence and
absence of CF in 1 M phosphate buffer on photodegradation in 2 h
changes from 0.63 to 3.90 and 0.63 to 2.70, respectively (Fig. 1).
This indicates a two-fold increase in the absorbance ratio leading
to the formation of CDRF in the presence of CF. A three-component
analysis of the spectral data in the presence and absence of CF and
their kinetic treatment in terms of the rate constants (Table 1) indi-
cates the effect of CF on the rates of the reactions. The magnitude
of spectral variations depends on the concentration of phosphate
buffer which alters the normal photolysis (photoreduction) path-
way (leading to the formation of FMF, LC and LF by cleavage of
the ribityl side-chain) in favor of the photoaddition pathway (lead-
ing to the formation of CDRF by cyclophotoaddition of the ribityl
side-chain).
the photodegradation of RF in the presence of phosphate buffer and
applied it to the kinetic studies of the reaction [11–14]. Ahmad and
Rapson [39] had previously reported a similar spectrophotometric
method for the assay of RF, FMF, LC and LF in the photodegraded
solutions of RF. This method has extensively been applied to study
the kinetics of normal photolysis of RF in aqueous solution in the
absence [15–17] and presence of CF [27].
The assay method used in this study [11] takes into
consideration the simultaneous determination of RF and its pho-
todegradation products of normal photolysis (FMF, LC and LF) and
photoaddition (CDRF). This method would throw light on the rel-
ative composition of the photodegradation product of the two
reaction pathways and has previously been used for similar studies
in the absence of CF [11–14]. CF (ꢀmax 273 nm) does not inter-
fere in the region of wavelengths used for the assay of RF and its
photoproducts [27].
3.4. Kinetics of photodegradation reactions
The photodegradation of RF takes place by several pathways
including the photoreduction, photoaddition and photodealkyla-
tion [10,14,40–42]. In the presence of divalent phosphate ions
(HPO42–), RF undergoes simultaneous photoreduction and pho-
toaddition reactions in aqueous solution to form LC and CDRF,
respectively, as the main final photoproducts [9,11–14]. The rates
of these reactions depend on the condition such as pH, buffer con-
centration and light intensity. The photodegradation of RF in the
presence of phosphate buffer can be expressed by the following
reactions:
3.3. Assay of RF and photoproducts
The photodegraded solutions of RF represent a complex mix-
ture containing RF, FMF, LC, LF and CDRF in a variable composition
and need a specific, convenient and reliable method for the assay
of RF and its photodegradation products. In a previous study [9],
no quantitative and kinetic study of these reactions has been
carried out. Ahmad et al. [11] developed a multicomponent spec-
trophotometric method for the assay of RF, FMF, LC, LF and CDRF in

20
M.A. Sheraz et al. / Journal of Photochemistry and Photobiology A: Chemistry 273 (2014) 17–22
the overall first-order rate constant for the photodegradation of
RF. The mathematical treatment of differential equations for the
reactant and the products, the determination of the values of k1
ꢀ
and k2 (considering diꢀrect formation of LC from RF) and the deter-
ꢀ
–
2–
mination of k3 and k4 for the H2PO4 and HPO4 ions catalyzed
reactions has been dealtꢀin an earlier communication [5,11,14]. The
values of kobs, k1 and k2 in the presence of CF along with the val-
ues in the absence of CF ꢀin 1.0 M phosphate buffer are reported in
Table 1. The values of k4 have been determined from the plots of
kobs versus H2PO4–/Total phosphate concentration. This value in
the presence of CF (2.29 × 10−2 M−1 min−1) is nearly twice of the
value (1.39 × 10−2 M−1 min−1) in the absence of CF in 1.0 M phos-
phate buffer. This indicates that CF accelerates the photoaddition
ꢀ
pathway of RF. This is further substantiated by the values of k1/k2
in the presence and absence of CF as discussed under Section 3.6.
3.5. Effect of phosphate buffer
Previous studies of the effect of phosphate buffer on the pho-
todegradation of RF have indicated the simultaneous formation
of CDRF and LC by different pathways. The rates of these reac-
tions depend on the concentration of phosphate species present
at a particular pH [11–14]. In the present study the photodegra-
dation reactions of RF in phosphate buffer have been carried out
in the presence or absence of CF. A consideration of the reaction
in 1.0 M phosphate buffer alone at pH 7.0 shows that the ratio
of the first-order rate constants for the formation of CDRF (pho-
toaddition) and LC (photoreduction) (k1/k2ꢀ) is 0.66, i.e., one-third
of the photodegradation reaction leads to the formation of CDRF
and two-third to the formation of LC. This is the same pattern as
observed in the earlier studies of the effect of phosphate buffer on
the photodegradation of RF [11,12,14]. This ratio has been found to
increase with an increase in buffer concentration in the presence of
2.5 × 10−4 M CF (0.36–1.23) indicating an enhancement in the rate
of photodegradation pathway of RF in favor of the photoaddition
reaction (Table 1).
Since the amount of LF formed at pH 7.0–8.0 is less than 5%
of the reaction mixture at 50% photodegradation of RF, it could
be omitted in the above scheme. LC is formed through FMF as
an intermediate in this reaction, it could be assumed that CDRF
and LC are formed by parallel first-order reactions. Therefore, the
ꢀ
values of k1 and k2 can be determined from the values of kobs
,
3.6. Effect of CF and phosphate buffer
An increase in CF concentration at a fixed concentration of phos-
phate buffer (1.0 M) also resulted in an increase in the ratios of
ꢀ
k1/k2 (0.70–1.23) (Table 2), suggesting that CF in the presence
of phosphate buffer influences the photoaddition pathway. This
could be due to the suppression of the photoreduction pathway
as observed in a previous study [27]. In the presence of CF, the
ꢀ
k1/k2 ratio is twice (1.23) compared to that observed in the pres-
ence of phosphate buffer alone (0.66) at pH 7.0. Thus, in addition
to the suppression of the photoreduction pathway of RF, CF may be
involved in some interaction with phosphate buffer to accelerate
the photoaddition reaction. CF forms molecular complexes with RF
(see Section 1) and thus inhibits its rate of chemical [25] and pho-
todegradation reactions [26,27] in aqueous solutions. It has been
suggested that CF forms a hetero association complex with flavin
mononucleotide (FMN) as a result of vertical stacking interaction
between the aromatic chromophores of the two molecules [24,43].
In view of the formation of such an association complex between
CF and RF and the fact that an increase in CF concentration would
cause greater suppression of the photoreduction pathway [27], an
acceleration in the photoaddition pathway is expected as evident
from the values of k1 and k2ꢀ. It is interesting to know that the val-
ꢀ
ues of k1/k2 in the presence of CF are higher than these obtained
in the absence of CF. Therefore, CF appears to exert an influence in
determining the magnitude of the two degradation pathways of RF.
The observed ability of CF molecule to discriminate between
different degradation pathways of RF, found in the present work
and linked with the stacking-type hetero-complex formation, is
Fig. 1. UV and visible absorption spectra of photodegraded solutions of RF at pH 7.0
(1.0 M phosphate buffer). (a) In the presence of CF and (b) in the absence of CF at 0,
30, 60, 90 and 120 min.

M.A. Sheraz et al. / Journal of Photochemistry and Photobiology A: Chemistry 273 (2014) 17–22
21
Table 2
Apparent first-order rate constants (kobs) for the photodegradation of RF and for the formation of CDRF (k1) and LC (k2ꢀ) in the presence of CF (0.625–2.50 × 104 M) in phosphate
buffer (1.0 M) at pH 7.0.
ꢀ
Caffeine concentration (M × 104)
kobs × 103 (min−1
)
SD
k1 × 103 (min−1
)
SD
k2ꢀ × 103 (min−1
)
SD
k1/k2
0.625
1.250
1.875
2.500
8.42 0.51
11.11 0.67
12.86 0.75
15.05 0.90
3.47 0.19
5.10 0.33
6.53 0.39
8.30 0.42
4.95 0.35
6.01 0.36
6.33 0.41
6.75 0.39
0.70
0.85
1.03
1.23
Table 3
very similar to well-known regulatory action of CF with respect to
aromatic drugs binding with DNA [44,45] and antibiotic–vitamin
interaction [46], which is also considered as a consequence of the
hetero-association between CF and other molecular species in solu-
tion. The likely reason for the influence of the hetero-complex
formation on the kinetics of RF photodegradation is the formation
of low dielectric environment within the volume of hetero-complex
as a consequence of water exclusion, altering the energetic barri-
ers for chemical transformation of RF. This effect, being specific to
stacking type of interactions, is also known to result in a variety of
biochemical effects such as, for instance, the lowering of pKa of cer-
tain atomic groups within the complexes [47], changing the drugs’
pharmacokinetics [48] and solubility [49].
Fluorescence intensity of 5 × 10−5 M RF solution at pH 7.0 in the presence of CF and
phosphate buffera.
Caffeine concentration
Phosphate
concentration (M)
Relative fluorescence
intensity at 525 nm
(M × 104)
0.00
1.25
2.50
0.00
1.25
2.50
0.005
0.005
0.005
1.00
100.0
100.0
100.0
81.1
1.00
78.0
1.00
74.6
a
Measurements were made at ambient temperature (∼25 ◦C) under aerobic con-
ditions.
the presence of 1.0 M phosphate buffer as observed in a previous
study [12]. The addition of 2.50 × 10−4 M CF to this solution further
decreases the fluorescence of RF to the extent of about 6%. Since CF
alone does not affect the fluorescence of RF and its presence in RF
solutions containing phosphate buffer leads to a further decrease in
the fluorescence intensity, some interaction between CF and phos-
phate buffer may be stipulated. The fact that the first-order rate
constants for the formation of CDRF are faster in phosphate buffer
containing CF compared to those of the phosphate buffer alone
(Table 1) and the values increase with an increase in phosphate con-
centration suggests that some interaction takes place between CF
and phosphate ions to facilitate the reaction (photoaddition path-
way). The loss of fluorescence of RF in the presence of phosphate
buffer alone indicates an interaction between RF and phosphate
ions (Table 3). This interaction could result from a RF–HPO42– com-
plex in the excited singlet state (1RF) as suggested by Schuman Jorns
et al. [9], to promote the photoaddition pathway. A further decrease
in RF fluorescence in phosphate buffer in the presence of CF (6%)
suggests that CF plays a role in the enhancement of the reaction
leading to the photoaddition pathway.
3.7. Effect of pH
The present study of the photodegradation of RF has been car-
ried out at pH 6.0–8.0. The rate of HPO4 ion catalyzed reaction
2–
is highest around pH 7 as observed earlier [9,11,14]. An essential
requirement for the intramolecular photoaddition of RF is that it
occurs at pH value >6 [9]. This is in accordance with the name
assigned to the reaction as the anion-catalyzed neutral photochem-
istry by the discoverers [9] and has been found in this study. The
rate of reaction increases with pH up to a maximum (pH 7) and then
decreases as observed in a previous study [14]. The same pattern
of reaction is observed in the presence of CF, however, the value of
kobs is nearly twice than that obtained in the absence of CF (Fig. 2).
This shows that CF plays a significant role in the photodegradation
of RF in phosphate buffer. This would further be explained in the
next section.
3.8. Fluorescence studies
Fluorescence studies on RF solutions in the presence of CF and
phosphate buffer have been carried out to observe their effect on
the intensity of fluorescence emission (Table 3). RF exhibits intense
yellow green fluorescence in aqueous solution [41,50,51]. The flu-
orescence of a 5 × 10−5 M RF solution (pH 7.0) is not affected by
1.25–2.50 × 10−4 M CF. However, it is decreased by about 19% in
In addition to the formation of a molecular complex between RF
and CF to inhibit the degradation reaction, there is a possibility of
the suppression of the photoreduction pathway by deactivation of
the excited triplet state (3RF). CF could quench 3RF by an electron
transfer in phosphate buffer and thus facilitate the photoaddition
ꢀ
pathway. This is substantiated by a relatively little increase in k2
with an increase in CF concentration compared to that of k1 in the
overall photodegradation (kobs) of RF (Table 2). CF may influence the
other pathways of photodegradation of RF. It has been suggested
that the formation of LC also takes place by the participation of 1RF
in the intramolecular photoreduction [38] and the intramolecular
photodealkylation reactions [40,42] of flavins. However, the pre-
dominant reaction in phosphate buffer involves the photoaddition
of RF giving rise to CDRF. Further structural studies are required to
ascertain the exact role of CF in the kinetics of degradation of RF.
4. Conclusion
The photodegradation of riboflavin in phosphate buffer at pH
6.0–8.0 occurs by simultaneous photoreduction (side chain cleav-
age) and photoaddition (side chain cyclization) pathways. The
rates of these reactions depend on the concentration of phos-
phate buffer. An increase in phosphate concentration alters the
Fig. 2. Plots of kobs versus pH for the photodegradation of RF in 1.0 M phosphate
buffer solutions in the presence (—᭹—) and absence (—ꢀ—) of CF (2.50 × 104 M).

22
M.A. Sheraz et al. / Journal of Photochemistry and Photobiology A: Chemistry 273 (2014) 17–22
photodegradation of riboflavin in favor of the photoaddition path-
way and the highest rate of the reaction occurs around pH 7.
[23] J.C.M. Tsibris, D.B. McCormick, L.D. Wright, Donor–acceptor complexes relating
to the intramolecular association of the riboflavin and adenosine moieties of
flavin adenine dinucleotide, Biochemistry 4 (1965) 504–510.
–
2–
Both H2PO4 and HPO4 ions catalyze the photodegradation of
RF by photoreduction (LC) and photoaddition (CDRF) pathways,
respectively. The addition of caffeine to riboflavin solutions in phos-
phate buffer has been found to influence the photodegradation of
riboflavin by inhibiting the photoreduction pathway and enhanc-
ing the photoaddition pathway. Caffeine could interact with 3RF
in phosphate to cause suppression of the photoreduction pathway
and thus facilitate the formation of CDRF as observed during these
studies, and/or directly interact with RF within the RF–CF stacking-
type hetero-complex altering the energetic barriers for chemical
transformation of RF.
[24] M.P. Evstigneev, A.O. Rozvadovskaya, A.A. Hernandez Santiago, Y.V. Mukhina,
K.A. Veselkov, O.V. Rogova, D.B. Davies, A.N. Veselkov, A 1H NMR study of the
association of caffeine with flavin mononucleotide in aqueous solution, Russ.
J. Phys. Chem. 79 (2005) 573–578.
[25] D.E. Guttman, Complex formation influence on reaction rate, I. Effect of caf-
feine on riboflavin base-catalyzed degradation rate, J. Pharm. Sci. 51 (1962)
1162–1166.
[26] Y. Sato, M. Yokoo, S. Takahashi, T. Takahashi, Biphasic photolysis of riboflavin
with a low-intensity light source, Chem. Pharm. Bull. 30 (1982) 1803–1810.
[27] I. Ahmad, S. Ahmed, M.A. Sheraz, M. Aminuddin, F.H.M. Vaid, Effect of caffeine
complexation on the photolysis of riboflavin in aqueous solution: a kinetic
study, Chem. Pharm. Bull. 57 (2009) 1363–1370.
[28] M. Ohara, T. Fujikura, H. Fujiwara, Augmentation of the inhibitory effect of blue
light on the growth of B16 melanoma cells by riboflavin, Int. J. Oncol. 22 (2003)
1291–1295.
[29] C.W.M. Kay, A. Bacher, M. Fischer, G. Richter, E. Schleicher, S. Weber,
Blue light-initiated DNA repair by photolyase, in: E. Silva, A.M. Edwards
(Eds.), Flavins–Photochemistry and Photobiology, Royal Society of Chemistry,
Cambridge, UK, 2006, pp. 151–182.
[30] C. Fukumachi, Y. Sakurai, Vitamin B2 photolysis, V. The photolytic formation
of 6,7-dimethylflavin-9-acetic acid ester from riboflavin, Vitamins (Kyoto) 7
(1954) 939–943.
References
[1] J.T. Carstensen, Kinetic pH profiles, in: J.T. Carstensen, C.T. Rhodes (Eds.), Drug
Stability Principles and Practices, 3rd ed., Marcel Dekker, New York, USA, 2000,
pp. 73–84.
[2] P.J. Sinko, Martin’s Physical Pharmacy and Pharmaceutical Sciences, 5th
ed., Lippincott Williams
416–420.
&
Wilkins, Baltimore, USA, 2006, pp. 406–407,
[31] H.H. Fall, H.G. Petering, Metabolic inhibitors, 1,6,7-dimethyl-9-
formylmethylisoalloxazine, 6,7-dimethyl-9-(2-hydroxyethyl)-isoalloxazine
and derivatives, J. Am. Chem. Soc. 78 (1956) 377–381.
[32] A.C. Moffat, M.D. Osselton, B. Widdop, Clarke’s Analysis of Drugs and Poisons,
4th ed., Pharmaceutical Press, London, UK, 2011, pp. 1028–1029.
[33] G.E. Treadwell, W.L. Cairns, D.E. Metzler, Photochemical degradation of flavins,
V. Chromatographic studies of the products of photolysis of riboflavin, J. Chro-
matogr. 35 (1968) 376–388.
[34] C.H. Suelter, D.E. Metzler, The oxidation of a reduced pyridine nucleotide analog
by flavins, Biochim. Biophys. Acta 44 (1960) 23–33.
[35] I. Ahmad, T. Mirza, K. Iqbal, S. Ahmed, M.A. Sheraz, F.H.M. Vaid, Effect of pH,
buffer, and viscosity on the photolysis of formylmethylflavin: a kinetic study,
Aust. J. Chem. 66 (2013) 579–585.
[36] P.S. Song, D.E. Metzler, Photochemical degradation of flavins, IV. Studies of the
anaerobic photolysis of riboflavin, Photochem. Photobiol. 6 (1967) 691–709.
[37] C.G. Hatchard, C.A. Parker, A new sensitive chemical actinometer, II. Potassium
ferrioxalate as a standard chemical actinometer, Proc. R. Soc. (Lond.) A 235
(1956) 518–536.
[38] W.L. Cairns, D.E. Metzler, Photochemical degradation of flavins, VI. A new pho-
toproduct and its use in studying photolytic mechanism, J. Am. Chem. Soc. 93
(1971) 2772–2777.
[39] I. Ahmad, H.D.C. Rapson, Multicomponent spectrophotometric assay of
riboflavin and photoproducts, J. Pharm. Biomed. Anal. 8 (1990) 217–223.
[40] P.S. Song, Chemistry of flavins in their excited states, in: H. Kamin (Ed.), Flavins
and Flavoprotein, University Park Press, Baltimore, USA, 1971, pp. 37–61.
[41] P.F. Heelis, The photophysical and photochemical properties of flavins (isoal-
loxazines), Chem. Soc. Rev. 11 (1982) 15–39.
[42] P.F. Heelis, The photochemistry of flavins, in: F. Muller (Ed.), Chemistry and
Biochemistry of Flavoenzymes, vol. 1, CRC Press, Boca Raton, FL, USA, 1991, pp.
171–193.
[43] P.A. Bolotin, S.F. Baranovsky, D.N. Chernyshev, M.P. Evstigneev, Spectro-
photometric study of the solution interactions between riboflavin, sodium
salicylate and caffeine, Int. J. Phys. Sci. 2 (2007) 68–72.
[44] E. Bedner, L. Du, F. Traganos, Z. Darzynkiewicz, Caffeine dissociates complexes
between DNA and intercalating drugs: application for bleaching fluorochrome-
stained cells for their subsequent restaining and analysis by laser scanning
cytometry, Cytometry 43 (2001) 38–45.
[45] M.P. Evstigneev, A.O. Lantushenko, V.P. Evstigneev, Yu.V. Mykhina, D.B. Davies,
Quantitation of the molecular mechanisms of biological synergism in a mixture
of DNA-acting aromatic drugs, Biophys. Chem. 132 (2008) 148–158.
[46] M.P. Evstigneev, V.P. Evstigneev, D.B. Davies, NMR investigation of the effect of
caffeine on the hetero-association of an anticancer drug with a vitamin, Chem.
Phys. Lett. 432 (2006) 248–251.
[47] N.A. Corfù, H. Sigel, Acid–base properties of nucleosides and nucleotides as a
function of concentration, Eur. J. Biochem. 199 (1991) 659–669.
[48] V.C. Okore, A.C. Osuji, A model study on caffeine–paracetamol interactions, Acta
Pharm. 51 (2001) 139–145.
[3] S. Yoshioka, V.J. Stella, Stability of Drugs and Dosage Forms, Kluwer Aca-
demic/Plenum Publishers, New York, USA, 2000, pp. 97–99.
[4] J.K. Guillory, R.I. Poust, Chemical kinetics and drug stability, in: G.S. Banker, C.T.
Rhodes (Eds.), Modern Pharmaceutics, 4th ed., Marcel Dekker, New York, USA,
2002, pp. 158–159.
[5] A.T. Florence, D. Attwood, Physicochemical Principles of Pharmacy, 4th ed.,
Pharmaceutical Press, London, UK, 2006, pp. 113–118.
[6] L. Lachman, P. DeLuca, M.J. Akers, Kinetic principles and stability testing, in: L.
Lachman, H.A. Lieberman, J.L. Kanig (Eds.), The Theory and Practice of Industrial
Pharmacy, 3rd ed., Lea & Febiger, Philadelphia, USA, 1986, pp. 768–769.
[7] M. Halwer, The photochemistry of riboflavin and related compound, J. Am.
Chem. Soc. 73 (1951) 4870–4874.
[8] B. Holmstrom, Mechanism of the photoreduction of riboflavin, Arkiv. Kemi. 22
(1964) 329–346.
[9] M. Schuman Jorns, G. Schollnhammer, P. Hemmerich, Intramolecular addition
of the riboflavin side chain, anion-catalysed neutral photochemistry, Eur. J.
Biochem. 57 (1975) 35–48.
[10] I. Ahmad, F.H.M. Vaid, Photochemistry of flavins in aqueous and organic
solvents, in: E. Silva, A.M. Edwards (Eds.), Flavins Photochemistry
and Photobiology, Royal Society of Chemistry, Cambridge, UK, 2006,
pp. 13–40.
[11] I. Ahmad, Q. Fasihullah, F.H.M. Vaid, A study of simultaneous photolysis and
photoaddition reactions of riboflavin in aqueous solution, J. Photochem. Pho-
tobiol. B: Biol. 75 (2004) 13–20.
[12] I. Ahmad, Q. Fasihullah, F.H.M. Vaid, Effect of phosphate buffer on photodegra-
dation reactions of riboflavin in aqueous solution, J. Photochem. Photobiol. B:
Biol. 78 (2005) 229–234.
[13] I. Ahmad, Q. Fasihullah, F.H.M. Vaid, Effect of light intensity and wavelengths
on photodegradation reactions of riboflavin in aqueous solution, J. Photochem.
Photobiol. B: Biol. 82 (2006) 21–27.
[14] I. Ahmad, S. Ahmed, M.A. Sheraz, F.H.M. Vaid, I.A. Ansari, Effect of divalent
anions on photodegradation kinetics and pathways of riboflavin in aqueous
solution, Int. J. Pharm. 390 (2010) 174–182.
[15] I. Ahmad, S. Ahmed, M.A. Sheraz, F.H.M. Vaid, Effect of borate buffer on the
photolysis of riboflavin in aqueous solution, J. Photochem. Photobiol. B: Biol.
93 (2008) 82–87.
[16] I. Ahmad, S. Ahmed, M.A. Sheraz, S.H. Kazi, T. Mirza, M. Aminuddin, Stabiliz-
ing effect of citrate buffer on the photolysis of riboflavin in aqueous solution,
Results Pharma. Sci. 1 (2011) 11–15.
[17] I. Ahmad, Q. Fasihullah, A. Noor, I.A. Ansari, Q.N.M. Ali, Photolysis of riboflavin
in aqueous solution: a kinetic study, Int. J. Pharm. 280 (2004) 199–208.
[18] J.E. Wilson, Studies on the electronic nature of flavin-indole and flavin-purine
complexes, Biochemistry 5 (1966) 1351–1359.
[19] E. DeRitter, Vitamins in pharmaceutical formulations, J. Pharm. Sci. 71 (1982)
1073–1096.
[20] R.S. Rivlin, Riboflavin (vitamin B2), in: J. Zempleni, R.B. Rucker, D.B. McCormick,
J.W. Suttie (Eds.), Handbook of Vitamins, 4th ed., Taylor & Francis, CRC Press,
Boca Raton, FL, USA, 2007, pp. 233–251.
[49] M.P. Evstigneev, V.P. Evstigneev, A.A. Hernandez Santiago, D.B. Davies, Effect
of a mixture of caffeine and nicotinamide on the solubility of vitamin (B2) in
aqueous solution, Eur. J. Pharm. Sci. 28 (2006) 59–66.
[21] D.E. Guttman, M.Y. Athalye, Solubilization of riboflavin by complex formation
with caffeine, theophylline and dimethyluracil, J. Am. Pharm. Assoc. Sci. Ed. 49
(1960) 687–691.
[22] G.R. Penzer, G.K. Radda, The chemistry and biological functions of isoalloxazines
(Flavins), Q. Rev. 21 (1967) 43–65.
[50] British Pharmacopoeia, Monograph on Riboflavin, Her Majesty’s Stationary
Office, London, UK, 2013, Electronic version.
[51] P. Drossler, W. Holzer, A. Penzkofer, P. Hegemann, pH dependence of the
absorption and emission behaviour of riboflavin in aqueous solution, Chem.
Phys. 282 (2002) 429–439.