Cyclodextrin inclusion interferes with trolox oxygen radical scavenging capacity measurement

Article abstract of DOI:10.1524/zpch.2012.0233

The interference of cyclodextrin solubilization with the measurement of oxygen radical scavenging capacity was investigated. Cyclodextrin (CD) that can solubilize water-insoluble compounds has been used in the oxygen radical scavenging capacity assay called oxygen radical absorbance capacity (ORAC) method. A vitamin E analog, trolox (2-hydroxy-2,5,7,8-tetramethylchroman-2- carboxylic acid) has been employed as a standard compound in ORAC methods and the results were often expressed in the trolox equivalent unit. We found that trolox ORAC values measured with electron spin resonance-based ORAC method, were markedly dependent on the CD concentration, i.e., it decreased by 50% when [CD]/[Trolox]=100 was present. 2D ROESY NMR study of trolox/CD inclusion complex revealed that trolox resided within the CD cavity. The reactive phenoxyl group in trolox is shielded from the attack by the oxygen radical, suggesting that this hindrance was causal for the decrease in ORAC values. by Oldenbourg Wissenschaftsverlag, Mu?nchen.

Full text of DOI:10.1524/zpch.2012.0233

Z. Phys. Chem. 227 (2013) 49–55 / DOI 10.1524/zpch.2012.0233
© by Oldenbourg Wissenschaftsverlag, München
Cyclodextrin Inclusion Interferes with Trolox
Oxygen Radical Scavenging Capacity Measurement
By Yoshimi Sueishi1 ∗, Misa Ishikawa1, Masashi Hori1, and Naoya Inazumi2
,
1
Department of Chemistry, Faculty of Science, Okayama University, 3-1-1 Tsushima-naka, Kita-ku,
Okayama 700-8530, Japan
Technical Support Division, Graduate School of Science, Osaka University, 1-1 Machikaneyama,
2
Toyonaka, Osaka, 560-0043, Japan
(Received March 16, 2012; accepted in revised form April 27, 2012)
(Published online June 11, 2012)
Oxygen Radical Scavenging Capacity / Spin Trapping / Inclusion Complex /
Trolox / Cyclodextrin
The interference of cyclodextrin solubilization with the measurement of oxygen radical scavenging
capacity was investigated. Cyclodextrin (CD) that can solubilize water-insoluble compounds has
been used in the oxygen radical scavenging capacity assay called oxygen radical absorbance capacity
(ORAC) method. A vitamin E analog, trolox (2-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic
acid) has been employed as a standard compound in ORAC methods and the results were often
expressed in the trolox equivalent unit. We found that trolox ORAC values measured with electron
spin resonance-based ORAC method, were markedly dependent on the CD concentration, i.e., it
decreased by 50% when [CD]/[Trolox]= 100 was present. 2D ROESY NMR study of trolox/CD
inclusion complex revealed that trolox resided within the CD cavity. The reactive phenoxyl group
in trolox is shielded from the attack by the oxygen radical, suggesting that this hindrance was
causal for the decrease in ORAC values.
1. Introduction
Evaluations of antioxidant capacity of pure antioxidants and food extracts have been
performed by using the fluorescence-based oxygen radical absorbance capacity method
(ORAC-FL) [1–3]. In order to overcome some experimental disadvantages of ORAC-
FL, such as the necessity of time-course measurements and computer-aided analysis,
we have proposed a new method of ORAC evaluation using electron spin resonance
(ESR) spin trapping technique (ORAC-EPR or ORAC-ESR) [4,5]. The ORAC-ESR
method, in which the free radicals are directly quantified with ESR spin trapping, is
a simple and useful method for the radical scavenging rate determination of antiox-
idants. Because natural antioxidants are often lipophilic and water-insoluble, Huang
et al. adopted a cyclodextrin (CD) analog to solubilize lipophilic compounds [6–8]. CD
* Corresponding author. E-mail: ysueishi@okayama-u.ac.jp
Brought to you by | Tulane University
Authenticated
Download Date | 12/30/14 10:47 AM

50
Y. Sueishi et al.
Fig. 1. Structures of antioxidant trolox and modified β-cyclodextrin (DM-β-CD).
molecule is a cyclic oligomer of glucose, having a cage-shaped void space, wherein
a lipophilic molecule is included and solubilized [9,10]. However, possible interference
between the ORAC assay and the solubilizer is not well understood.
In most ORAC-like methods, a vitamin E analog, 2-hydroxy-2,5,7,8-tetramethyl-
chroman-2-carboxylic acid (trolox, Fig. 1) has been employed as a standard antioxidant
of the radical scavenging capacity. Thus, ORAC values are often expressed in the trolox
equivalent unit. Fortunately, trolox is slightly water soluble and its ORAC value can
be evaluated in the absence or presence of solubilizers. When a cyclodextrin analog,
DM-β-CD (2,6-di-O-methylated β-cyclodextrin, Fig. 1) was added to the trolox solu-
tion, ORAC values markedly decreased as a function of increasing CD concentration. In
order to study the cause of this decrease, trolox/DM-β-CD complex was characterized
with 2D ROESY NMR. The structure has lead us to the conclusion that trolox’s reac-
tive site was effectively included in the CD cavity, blocking the approach of the oxygen
radical, which resulted in the decrease of the scavenging rate.
2. Experimental
2.1 Materials
Chemical structures of the compounds used in this study are shown in Figs. 1 and 2.
Trolox and DM-β-CD were purchased from Tokyo Chemical Industry Co. (Tokyo,
Japan) and Nakalai Tesque (Kyoto, Japan), respectively. The spin-trap DMPO (5,5-
dimethyl-pyrroline N-oxide) was obtained also from Tokyo Chemical Industry Co.
(Tokyo, Japan). As a source of oxygen radical (alkoxyl radical (RO·)), 2,2ꢀ-azobis(2-
amidino-propane) dihydrochloride (AAPH) was purchased from Wako Pure Chemicals
(Osaka, Japan) (Fig. 2).
2.2 ESR and NMR measurements
A competitive reaction method between the spin trap DMPO and trolox was employed
to quantify the radical scavenging capacity (ORAC values) [4,5]. A phosphate buffer
(0.1 M (M = mol dm−3), pH = 7.4) was used as a solvent and the RO· radical was gen-
erated with in situ UV irradiation (1 s irradiation time with a RUF-203S 200 W mercury
arc (Radical Research Inc.)). The ESR signals of RO· radical adducts with DMPO were
recorded in a JEOL FA200 X-band spectrometer (Akishima, Japan). The sample solu-
tion temperature was controlled at 298 0.1 K.
Brought to you by | Tulane University
Authenticated
Download Date | 12/30/14 10:47 AM

Cyclodextrin Inclusion Interferes Trolox Oxygen Radical Scavenging Capacity
51
Fig. 2. Structures of spin trap (DMPO) and radical source (AAPH).
1
In order to elucidate the structure of the DM-β-CD inclusion complex, H-NMR
spectra were obtained in D2O with a Varian Mercury 300 NMR spectrometer
(300 MHz) (Palo Alto CA, USA) at room temperature. Chemical shifts were reported
as δ values relative to HOD (δ 4.79) as an internal standard [11]. A 2D ROESY NMR
experiment was performed at 600 MHz in D2O on a Varian VNS600 NMR spectrometer
at 303 K. The mixing time for the ROESY NMR experiments was set at 200 ms.
2.3 Scavenging capacity measurement
In the present system, trolox and DMPO compete to scavenge AAPH-derived oxygen
radical (RO· radical). Thus, the scavenging capacity of trolox was obtained relative to
that of DMPO [4,5]. The reaction that we monitored in excess of CD is illustrated as
follows [12]:
hv
AAPH −→RO·
DMPO : CD+RO· −→ DMPO-OR : CD
scavenging rate const: kDMPO
scavenging rate const: kTrolox
Trolox : CD+RO· −→ Product (ESR silent)
,
where DMPO : CD and Trolox : CD denote CD-inclusion complexes of DMPO and
trolox, respectively. A simple formulation for the radical scavenging calculation can be
derived from the above reaction scheme and has been reported elsewhere [4,5]. The
resultant equation is:
I0 − I
kTrolox [Trolox]0
kDMPO [DMPO]0
=
,
(1)
I
where I and I0 are ESR signal heights for the radical adduct in the presence and ab-
sence of trolox, respectively, and the [ ]0 symbol denotes the initial concentrations of
DMPO or trolox. The measurement of ESR signal intensities by selecting several trolox
concentrations would give a linear plot of (I0 − I)/I against [Trolox]0/[DMPO]0 like
Fig. 3c. The slope of such a plot yielded kTrolox/kDMPO, which is equivalent to the radical
scavenging rate constant for trolox relative to that of DMPO.
3. Results
3.1 ESR spectrum of spin-trapped oxygen radical
Figure 3a shows the ESR spectra of RO· radical spin-trapped with DMPO in a UV-
irradiated AAPH solution in the absence of DM-β-CD. The hyperfine splitting constant
Brought to you by | Tulane University
Authenticated
Download Date | 12/30/14 10:47 AM

52
Y. Sueishi et al.
Fig. 3. ESR spectra of RO· radical adduct of DMPO obtained after the UV-photolysis of phosphate buffer
containing AAPH (5.0 mM) and DMPO (1.0 mM) in the absence of DM-β-CD: (a) [trolox] = 0, and
(b) [trolox] = 8.7×10−2 mM. Horizontal broken lines in the spectra demonstrate the change in ESR sig-
nal height due to the scavenging activity by trolox. (c) A plot of (I0-I)/I against [Trolox]0/[DMPO]0 in
the absence of DM-β-CD using Eq. (1). The slope is equal to trolox’s relative scavenging rate constant
k
Trolox/kDMPO.
(hfsc) was obtained after the analysis with a computer spectrum simulation, resulting in
AN (nitrogen nucleus hfsc) = 1.44 mT and AH (hydrogen nucleus hfsc) = 1.52 mT.
3.2 Oxygen radical scavenging capacity of trolox
The intensity of the ESR signal was decreased by the presence of trolox, such as shown
in Figs. 3a and 3b, indicating that part of the RO· radical was scavenged by trolox.
The intensity decrease was quantified and the relative scavenging rate constants were
calculated, using Eq. (1). In the absence of DM-β-CD, the ratio of scavenging rate con-
stant (kTrolox/kDMPO) was 43.6 1.4. As the concentration of DM-β-CD was increased,
k
Trolox/kDMPO monotonically decreased and leveled off when [DM-β-CD]0/[Trolox]0
was more than 100 (Fig. 4).
3.3 NMR spectra of trolox/DM-β-CD complex
We utilized 2D ROESY NMR technique to determine the disposition of trolox in
the DM-β-CD complex. Figure 5a shows the ROESY NMR spectrum of trolox
in D2O in the presence of DM-β-CD. The proton NMR peaks in trolox were as-
signed with reference to the NMR study of vitamin E by Baker and Myers [13].
The cross peaks of the H-3,5 protons in DM-β-CD were detected with C(2,7,8)-
CH3 protons as enclosed in the ellipse in Fig. 5. The cross peaks of the inner
H-3,5 protons of DM-β-CD with the C(3,4)-H and C(5)-CH3 protons did not ap-
pear, indicating that the moieties of C(2), C(7), and C(8) are encapsulated in the
CD cavity, in addition, C(6) in the chromane ring also showed strong interac-
tion with peripheral methyl groups in DM-β-CD. Such observations are consistent
Brought to you by | Tulane University
Authenticated
Download Date | 12/30/14 10:47 AM

Cyclodextrin Inclusion Interferes Trolox Oxygen Radical Scavenging Capacity
53
Fig. 4. The relative rate constant (kTrolox/kDMPO
8.7×10−2 mM.
)
plotted vs. [DM-β-CD]0/[Trolox]0: [Trolox]0 =
Fig. 5. (a) 2D ROESY NMR spectra (200 ms mixing time) at 303 K in D2O solution: [trolox] = 1.0 mM
and [DM-β-CD] = 10 mM. (b) A plausible structure of the inclusion complex of trolox with DM-β-CD
based on NMR results. (c) Changes in chemical shifts of C(2)-CH3 proton in trolox at various DM-β-CD
concentrations.
with the molecular modeling calculation for the trolox/methyl-β-CD inclusion com-
plex [14].
Based on the 2D ROESY NMR results of trolox and DM-β-CD in the complex,
the structure of the complex can be speculated. The schematic structure in Fig. 5b illus-
Brought to you by | Tulane University
Authenticated
Download Date | 12/30/14 10:47 AM

54
Y. Sueishi et al.
trates that trolox molecule resides in the CD cavity side ways. It should be emphasized
that the phenoxyl group in trolox is completely embedded in the entrance portion of CD
cavity.
3.4 NMR chemical shift of trolox/CD complexation
Association constant K for trolox : CD inclusion complex can be calculated based on
the induced chemical shift in trolox NMR spectrum. Figure 5c shows the chemical
shifts for the C(2)-CH3 protons of trolox as a function of DM-β-CD concentration. The
changes in the induced chemical shift leveled off at [DM-β-CD]/[Trolox] larger than
100, indicating that all trolox molecules formed inclusion complex.
The induced chemical shifts (Δδ)T in NMR signals in the 1 : 1 complex formation
equilibrium such as:
Trolox+DM-β-CD K(ꢀtrolox) Complex ,
is given by the following equation [15]:
ꢀ
ꢂ
ꢁ
Δδsat
Δδ =
β =
β − β2 −4[Trolox]0[DM-β-CD]0
,
(2)
(3)
2[Trolox]0
1
+[Trolox]0 +[DM-β-CD]0 ,
K(trolox)
where the [ ]0 symbol denotes the initial concentration. K(trolox) is an inclusion
constant of trolox with DM-β-CD and Δδsat is the saturated value of induced chem-
ical shifts. Figure 5c shows the saturation curve for the induced chemical shifts.
Using Δδsat obtained from Fig. 5c, a nonlinear least-square fit of the induced shifts to
Eq. (2) enabled us to determine K(trolox) as 98 5 M−1, which is much smaller than
those for DM-β-CD complexes of water-insoluble lipophilic antioxidants; for example,
K = 8670 and 23800 M−1 for catechin and epigallocatechin gallate, respectively [16].
4. Discussion
The present experiment demonstrated that upon addition of the solubilizer DM-β-CD,
trolox’s radical scavenging capacity decreased. In the presence of excess DM-β-CD
([DM-β-CD]0/[Trolox]0 = 100), it decreased by nearly 50% (Fig. 4) and leveled off
above this concentration ratio, indicating that all trolox molecules had formed inclusion
complexes above this ratio. The objective of this study was to elucidate the cause for
this decrease. NMR structure of trolox/DM-β-CD complex revealed that whole trolox
molecule was embedded within the wall of methoxy groups at the entrance of the CD
cavity. NMR results also indicated that the phenoxyl group in trolox strongly interacts
with CD molecule (Fig. 5b), suggesting that trolox is encapsulated in CD so that the
phenoxyl group is not exposed to the bulk solution phase. Phenoxyl group is highly re-
active to oxygen radical including RO· radical; therefore, the encapsulation of phenoxyl
group has made it difficult for RO· radical to contact and react with trolox. This means
Brought to you by | Tulane University
Authenticated
Download Date | 12/30/14 10:47 AM

Cyclodextrin Inclusion Interferes Trolox Oxygen Radical Scavenging Capacity
55
that CD in this case is not a mere solubilizer but a reagent that could participate and
interfere with the free radical scavenging reaction.
Under in vivo conditions, natural proteins and lipids could play a role as solubiliz-
ers; therefore, the ORAC values (scavenging rates) of solubilized antioxidants in the
laboratory settings are thought to be not too far from in vivo situations. However, it may
be problematic if trolox is used as a standard in the scavenging assay of CD-solubilized
compounds, because trolox ORAC values are CD concentration-dependent. Therefore,
one should use trolox ORAC value in the absence of the CD solubilizer as a standard.
Acknowledgement
We thank Dr. Yashige Kotake for helpful discussion and critical reading of the
manuscript.
References
1. A. N. Glazer, Methods Enzymol. 186 (1990) 161.
2. G. Cao, H. M. Alessio, and R. G. Cutler, Free Radical Bio. Med. 14 (1993) 303.
3. T. P. Whitehead, G. H. G. Thorpe, and S. R. J. Maxwell, Anal. Chim. Acta 266 (1992) 265.
4. S. Kohri, H. Fujii, S. Oowada, N. Endo, Y. Sueishi, M. Kusakabe, M. Shimmei, and Y. Kotake,
Anal. Biochem. 386 (2009) 167.
5. N. Endo, S. Oowada, Y. Sueishi, M. Shimmei, K. Makino, H. Fujii, and Y. Kotake, J. Clin.
Biochem. Nutr. 45 (2009) 193.
6. D. Huang, B. Ou, M. Hampsch-Woodill, J. A. Flanagan, and E. K. J. Deemer, J. Agr. Food
Chem. 50 (2002) 1815.
7. Y. M. A. Naguib, S. P. Hari, R. Passwater Jr., and D. Huang, J. Nutr. Sci. Vitaminol. 49 (2003)
217.
8. D. V. Bangalore, W. McGlynn, and D. D. Scott, J. Agr. Food Chem. 53 (2005) 1878.
9. K. Frömming and J. Szejtli, Cyclodextrins in Pharmacy, Kluwer Acad. Publ., Dordrecht
(1993).
10. K. Uekama and I. Irie, Cyclodextrins and Their Industrial Uses, D. Duchene (Ed.), Edition de
Stante, Paris (1987).
11. H. E. Gottlieb, V. Kotlyar, and A. Nudelman, J. Org. Chem. 62 (1997) 7512.
12. Y. Sueishi, A. Miyata, D. Yoshioka, M. Kamibayashi, and Y. Kotake, J. Incl. Phenom. Macro-
cycl. Chem. 66 (2010) 357.
13. J. Baker and C. Myers, Pharm. Res. 8 (1991) 763.
14. S. Sapino, M. Trotta, G. Ermondi, G. Caron, R. Cavalli, and M. E. Carlotti, J. Incl. Phenom.
Macrocycl. Chem. 62 (2008) 179.
15. C. Binkowski, F. Hapiot, V. Lequart, P. Martin, and E. Monflier, Org. Biomol. Chem. 3 (2005)
1129.
16. C. Folch-Cano, C. Jullian, H. Speisky, and C. Olea-Azar, Food Res. Int. 43 (2010) 2039.
Brought to you by | Tulane University
Authenticated
Download Date | 12/30/14 10:47 AM