Cleavage of α-dicarbonyl compounds by terpene hydroperoxide

Article abstract of DOI:10.1271/bbb.120378

The highly reactive α-dicarbonyl compounds, glyoxal, methylglyoxal (MGO), and 3-deoxyglucosone, react with the amino groups of proteins to form advanced glycation end-products (AGEs) which have been implicated in diabetic complications, aging, and Alzheimer's disease. We found that a test sample of terpinen-4-ol (T4) containing hydroperoxides showed cleaving activity toward an α-dicarbonyl compound, but that the freshly isolated pure sample did not. Prepared terpinen-4-ol hydroperoxide (T4-H) also efficiently cleaved the C-C bond of the α-dicarbonyl compounds via Baeyer-Villiger-like rearrangement and subsequent hydrolysis of an acid anhydride moiety in the rearranged product to give carboxylic acids. Other terpene hydroperoxides, as well as T4-H, showed significant cleaving activities, and all these hydroperoxides protected RNase A from the lowering of enzyme activity induced by MGO. The cleaving mechanism via Baeyer-Villiger-like rearrangement was confirmed by time-interval NMR measurements of the reaction mixture of the symmetrical α-dicarbonyl compound, diacetyl with T4-H.

Full text of DOI:10.1271/bbb.120378

Biosci. Biotechnol. Biochem., 76 (10), 1904–1908, 2012
Cleavage of ꢀ-Dicarbonyl Compounds by Terpene Hydroperoxide
y
Ryu-ichiro NAGAMATSU, Shinya MITSUHASHI, Kengo SHIGETOMI, and Makoto UBUKATA
Division of Applied Bioscience, Graduate School of Agriculture, Hokkaido University,
Kita-9, Nishi-9, Kita-ku, Sapporo 060-8589, Japan
Received May 16, 2012; Accepted June 22, 2012; Online Publication, October 7, 2012
[doi:10.1271/bbb.120378]
The highly reactive ꢀ-dicarbonyl compounds, glyoxal,
methylglyoxal (MGO), and 3-deoxyglucosone, react
with the amino groups of proteins to form advanced
glycation end-products (AGEs) which have been impli-
cated in diabetic complications, aging, and Alzheimer’s
disease. We found that a test sample of terpinen-4-ol
(T4) containing hydroperoxides showed cleaving activity
toward an ꢀ-dicarbonyl compound, but that the freshly
isolated pure sample did not. Prepared terpinen-4-ol
hydroperoxide (T4-H) also efficiently cleaved the C-C
bond of the ꢀ-dicarbonyl compounds via Baeyer-
Villiger-like rearrangement and subsequent hydrolysis
of an acid anhydride moiety in the rearranged product
to give carboxylic acids. Other terpene hydroperoxides,
as well as T4-H, showed significant cleaving activities,
and all these hydroperoxides protected RNase A from
the lowering of enzyme activity induced by MGO. The
cleaving mechanism via Baeyer-Villiger-like rearrange-
ment was confirmed by time-interval NMR measure-
ments of the reaction mixture of the symmetrical ꢀ-
dicarbonyl compound, diacetyl with T4-H.
major components of tea tree oil used as an aromather-
apy product. Although the pleiotropic biological activ-
ities of the essential oil have been reported, no logical
explanation at the molecular level for the therapeutic
effects has appeared in the scientific literature.13) We
were therefore interested in elucidating the mystery for
the effects of the essential oil and found one possible
answer. The proposed action mechanism involved the
allylic hydroperoxide formation of T4 by autoxidation,
the Baeyer-Villiger-like reaction of the ꢀ-dicarbonyl
compound with the resulting hydroperoxide, and hy-
drolysis of the acid anhydride moiety of the rearranged
product to give two carboxylic acid molecules.
Results and Discussion
Identification of the oxidative compounds from the
reaction mixture of 1-phenyl-1,2-propanedione (PPD)
and unpurified terpinen-4-ol
An unpurified test sample (unpurified T4 (1)), after
exposing (ꢀ)-terpinen-4-ol (>95% by GC) to air for one
month, showed moderate cleaving activity toward 1-
phenyl-1,2-propanedione (PPD) to give benzoic acid as
shown in Fig. 1A and C. To test the significance of the
double bond, trans-4-methyl-1-(1-methylethyl)cyclo-
hexanol, MMC (2) was prepared as indicated in Fig. 1B,
and its cleaving activity was compared with that of
unpurified T4. MMC lost the cleaving activity as shown
in Fig. 1C. We next isolated the three oxidative
derivatives of (ꢀ)-terpine-4-ol, 5-hydroxy-2-methyl-
5-(1-methylethyl)cyclohex-2-enone (3), (1R,2R,4R)-
1-methyl-4-(1-methylethyl)-1,2,4-cyclohexanetriol (4)
and (2Z)-3,7-dimethyl-6-oxooct-2-enal (5), from the
reaction mixture of unpurified T4 and PPD. The
presence of these oxidized compounds suggested that
allylic hydroperoxides of T4 in the active samples were
the true cleaving agent of the ꢀ-dicarbonyl compound,
and allylic hydroperoxide in the unpurified T4 sample
triggered the formation of the oxidative products during
the cleavage reaction of the ꢀ-dicarbonyl compound.
Key words: terpinen-4-ol;
hydroperoxide;
methyl-
glyoxal; advanced glycation end-products
(AGEs); RNase A
Advanced glycation end products (AGEs) are a group
of compounds formed via a non-enzymatic reaction
between reducing sugars and amine residues on proteins,
and have been implicated as the cause of diabetic
complication,1) aging,2,3) Alzheimer’s disease,4,5) and
arteriosclerosis.6) The major AGEs in vivo appear to be
formed as highly reactive intermediate carbonyl groups,
known as ꢀ-dicarbonyl compounds such as 3-deoxyglu-
cosone, glyoxal, and methylglyoxal (MGO).7,8) The first
cleaving agent of an ꢀ-dicarbonyl compound appearing
in the literature was N-phenacylthiazolium bromide
(PTB), and the mechanism for cleaving the C-C bond
has been proposed.9) 3-Phenacyl-4,5-dimethylthiazolium
chloride (ALT-711) was later introduced as a stable
analogue of PTB.10) These cleaving compounds might
be useful as therapeutic agents for diabetes complica-
tions and ointment ingredients for skin care.11,12) We
have found that the volatile oil from Citrus junos Tanaka
and certain terpenes could cleave 1-phenyl-1,2-propane-
dione (PPD), a model ꢀ-dicarbonyl compound like PTB,
whereas we had difficulty replicating the phenomenon.
To investigate this anomalous phenomenon, we focused
our attention first on terpinen-4-ol (T4, 1), one of the
Efficient cleavage of the ꢀ-dicarbonyl compound by
terpene hydroperoxides, and their protection of the
enzyme function of RNase A from methylglyoxal
This hypothesis was verified by two experiments;
pure T4 (1) freshly isolated from commercially available
(ꢀ)-terpinen-4-ol (>95% by GC) did not cleave the ꢀ-
dicarbonyl compound, while the hydroperoxide, T4-H
(6) prepared from T4 with singlet oxygen,14,15) demon-
y
To whom correspondence should be addressed. Tel/Fax: +81-11-706-3638; E-mail: m-ub@for.agr.hokudai.ac.jp

Terpene Hydroperoxide as an ꢀ-Dicarbonyl Breaker
1905
O
O
A
C
B
O
H2, Pd-C, EtOAc
Sample, 37 °C, 50 mM PB (pH 7.4)
containing 50% MeOH
OH
OH
T4 (1)
OH
Benzoic acid
PPD
MMC (2)
D
100
80
60
40
20
0
100
80
HOO
OH
OH
60
40
T4-H (6)
Purified T4 (1)
20
0
Control T4 (>95%) MMC
Control
PTB
T4
T4-H
Fig. 1. Cleaving Effect of Unpurified T4 (>95%) with Conversion of 1-Phenyl-1,2-propanedione (PPD) to Benzoic Acid.
A: The cleaving activity of the ꢀ-dicarbonyl compound by unpurified T4 was evaluated by a modified procedure of Vasan’s protocol.9) B:
trans-4-Methyl-1-(1-methylethyl)cyclohexanol (MMC) was prepared by hydrogenation of unpurified T4. C: MMC did not cleave PPD,
suggesting the importance of the double bond for the cleavage of ꢀ-diketone. D: Cleavage rate of PPD with purified T4 or terpine-4-ol
hydroperoxide (T4-H). Control, DMSO (vehicle); PTB, N-phenacylthiazolium bromide. Error bar: ꢂSD (n ¼ 3).
strated significant cleaving activity of the ꢀ-dicarbonyl
OOH
OOH
A
OOH
compound as shown in Fig. 1D. To prove the generality
of the cleaving effect of the terpene hydroperoxide and
its protective effect on glycation-induced oxidative
damage of the enzyme, we next prepared additional
hydroperoxides and evaluated the protective effects of
these hydroperoxides on ribonuclease A (RNase A) in
the presence of MGO,16,17) a major precursor of AGEs.
MGO is a highly reactive endogenous metabolite, and its
levels are elevated in diabetic patients.18,19) ꢀ-Terpineol
hydroperoxide (T-ol-H,a 3:1 mixture of 8a and 8b)
prepared from ꢀ-terpineol(T-ol, 7), and ꢀ-terpinyl
acetate hydroperoxide(T-ace-H, 10) prepared from ꢀ-
terpinyl acetate (T-ace, 9) showed significant cleaving
activity toward PPD, whereas highly purified T-ol and
T-ace did not cleave the ꢀ-dicarbonyl compound as in
the case of T4 (Fig. 2). PPD could be slightly cleaved by
atmospheric oxygen in the sealed reaction tube,9) while
purified T4, T-ol, and T-ace rather inhibited the cleavage
of PPD as shown in Figs. 1D and 2, probably because
these pure terpenes rapidly consumed oxygen to give 4
as a major oxidative derivative, while 3 and 5 as minor
oxidative derivatives, and the allylic hydroperoxide
were not present in sufficient concentrations to cleave
PPD. Terpene hydroperoxide in test sample was there-
fore the real active agent for cleaving the ꢀ-dicarbonyl
compound. All these hydroperoxides, T4-H, T-ol-H and
T-ace-H, protected ribonuclease A (RNase A) from
highly reactive MGO, their protective effects being
comparable with those of the positive controls, amino-
guanidine (AG) and metformin (Met), in the RNase
A-MGO assay as shown in Fig. 3.
OH
OH
OH
OAc
OAc
T-ace-H (10)
T-ol (7)
8a
8b
T-ace (9)
T-ol-H
100
80
60
40
20
0
B
Control
PTB
T-ol
T-ol-H
T-ace
T-ace-H
Fig. 2. Structures of Terpene Hydroperoxides and Their Effects on
the Cleavage of PPD.
A: Structures of ꢀ-terpineol (T-ol), ꢀ-terpineol hydroperoxide
(T-ol-H), ꢀ-terpinyl acetate (T-ace), and ꢀ-terpinyl acetate hydro-
peroxide (T-ace-H). B: Cleavage rate of PPD treated with purified
monoterpene or its hydroperoxide. Control, DMSO (vehicle); PTB,
N-phenacylthiazolium bromide. Error bar: ꢂSD (n ¼ 3).
was treated with 20 mM of T4-H (6) for 0–48 h at 30 ꢁC
in CDCl3. Changes in the NMR signals of the starting
materials (T4-H and diacetyl), reaction intermediate
(acetic anhydride), and cleavage product (acetic acid)
1
were monitored by H-NMR measurements. A decrease
in the R-OOH signal of T4-H at 8.16 ppm was observed
during the 8–48 h period as shown in Fig. 4A. Figure 4B
shows that the CH3 peak at 2.34 ppm of the diacetyl
gradually decreased via the signal being split into three
peaks at 8 h, suggesting that one ketone of the diacetyl
had been attacked by T4-H to form a detectable
intermediate until 8 h. Three peaks at around 2.34 ppm
appearing at 8 h were tentatively assigned to a diaster-
eomeric pair of methyl groups of a methyl ketone of the
intermediate in the Baeyer-Villiger reaction, and the
Cleaving mechanism of the ꢀ-dicarbonyl compound
1
verified by H-NMR analysis
The cleaving mechanism of the ꢀ-dicarbonyl com-
pound by terpene hydroperoxide could be explained by
Baeyer-Villiger-like rearrangement and subsequent hy-
drolysis,20) this hypothesis being reliably established by
1H-NMR experiments as indicated in Fig. 4. To detect
the reaction intermediates, 10 mM of diacetyl as a
symmetrical ꢀ-dicarbonyl compound in an NMR tube

1906
R. NAGAMATSU et al.
methyl group of the diacetyl. The CH3 peak at 2.23 ppm
of acetic anhydride (authentic acetic anhydride was
2.23 ppm) as the Baeyer-Villiger rearranged product
appeared after 24 h and gradually increased over 48 h.
The acetic anhydride was hydrolyzed to acetic acid by
residual H2O (1.59 ppm), and the CH3 peak at 2.10 ppm
of acetic acid (authentic acetic acid was 2.11 ppm)
appeared after 24 h and gradually increased over 48 h as
shown in Fig. 4C. These observations validated the
cleaving mechanism of the ꢀ-dicarbonyl compound
treated with terpene hydroperoxide.
Discussion on the biological activities of monoter-
penes and their hydroperoxides
100
80
60
40
20
We clarified the cleaving mechanism of the ꢀ-
dicarbonyl compound by terpene hydoroperoxide, al-
though terpene hydroperoxides have only previously
been recognized as skin sensitizers. d-Limonene itself
gives no significant skin sensitization, and some of the
oxidation products formed when d-limonene is exposed
to air have been shown to be potent contact allergens,
this being analogous to the oxidation products of ꢁ-
carene as another monoterpene. The hydroperoxides of
these compounds have been reported to be responsible
for the main allergenic effect of turpentine.21,22) We
confirmed that fresh T4 easily reacted with air at room
temperature for 30 d to give at least three hydroperoxy
derivatives (Rf 0.3, 0.5, and 0.69) which were detected
as yellow spots by a saturated aqueous KI solution, an
indicator of hydroperoxides, on a silica gel TLC plate
(EtOAc:hexane, 1:2). These spots, which can also be
indicated as characteristic white spots on the TLC image
0
RNase A
+
-
+
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
MGO (10mM)
Compound (10 mM)
AG Met T4 T4-H T-ol T-ol-H T-ace T-ace-H
-
Fig. 3. Protective Effects of Terpene Hydroperoxides, T4-H, T-ol-H,
and T-ace-H in the RNase A-MGO Assay.
MGO, methylglyoxal; AG, aminoguanidine (positive control 1);
Met, metformin (positive control 2). The remaining activity of
RNase A was determined as described in the Experimental section.
Error bar: ꢂSD (n ¼ 3).
A
B
8.16 ppm
2.34 ppm
2.23 ppm
Diacetyl (10mM)
T4-H (20mM)
8 h
24 h
48 h
8.4
8.2
8.0
2.5
2.4
2.3
2.2
C
1.59 ppm
2.10 ppm
Diacetyl (10mM)
T4-H (20mM)
8 h
24 h
48 h
2.1
2.0
1.6
1.5
Fig. 4. Time-Interval 1H-NMR Spectra of the Reaction Mixture of Diacetyl and T4-H.
Changes in signal intensity of (A) the hydroperoxide proton in T4-H (8.16 ppm), (B) the methyl protons in diacetyl (2.34 ppm) and acetic
anhydride (2.23 ppm) as the Baeyer-Villiger oxidation products, and (C) acetic acid (2.10 ppm) as the cleavage product and the proton of H2O
(1.59 ppm).

Terpene Hydroperoxide as an ꢀ-Dicarbonyl Breaker
1907
the resulting solution was diluted 5-fold with a combined solution of
equal parts of 100 mM sodium phosphate buffer (pH 7.4) and methanol.
The components were analyzed by HPLC (Kanto Chemical ODS
Mightysil RP-18 column, 4:6 ꢃ 250 mm (5 mm) and mobile phase of
a 35% methanol/0.1% TFA solution), and the yield of benzoic acid
was calculated by area integration (A254) and interpolation of a
calibration curve obtained by injecting standard amounts of pure
benzoic acid. The percentage of cleaving activity was calculated as
100 ꢃ [concentration (mM) of generated benzoic acid/10 mM].
before turning to blue spots on a green-yellow back-
ground by a phosphomolybdic acid solution, were
detected in an active T4 sample as the ꢀ-dicarbonyl
cleaving agent. Autoxidation of 1,2-dimethylcyclohex-
ene with triplet oxygen gives the hydroperoxide with an
unshifted double bond as the major product, while
singlet oxygen-mediated oxidation of the same cyclo-
hexene produces the hydroperoxide derivative having
exomethylene as the major product.15) It has recently
been reported that ꢀ-terpinene was rapidly degraded to
form allylic epoxides and p-cymene as the major
oxidation products.23) Australian tea tree oil containing
T4 as a major component has exhibited such activities as
antibacterial,24) antifungal,25) and antiviral,26) and has
been used topically to treat such conditions as acne, cold
sores, dandruff, onychomycosis, oral candidiasis, and
tinea pedis.27,28) Although studies on human skin
penetration of T4 have appeared,29,30) no molecular
mechanism for such biological activities has been
reported. The biological activities of monoterpenes
including T4 might therefore need to be reevaluated in
consideration of the effects caused by a trace amount of
hydroperoxy derivatives contained in the test samples.
All terpene hydroperoxides synthesized in this study
showed significant cleaving activities of ꢀ-dicarbonyl
compounds and converted potentially toxic MGO into
non-toxic or less toxic carboxylic acids to protect the
enzyme function. The present findings are important to
provide insight into the quality control of essential oils
for traditional remedies and aromatherapy, and might
contribute to the development of a novel class of
inhibitors of AGE formation and/or skin care products
whose mechanism is different from that of such
thiazolium-type inhibitors as PTB and ALT-711.
Synthesis of trans-4-methyl-1-(1-methylethyl)cyclohexanol (MMC, 2).
To 153 mg of Pd–C was added 2.31 g (15.0 mmol) of T4 (1) dissolved in
20 mL of EtOAc. The mixture was stirred in a hydrogen atmosphere at
room temperature for 24h. After filtering through a Celite pad, the
filtrate was evaporated in vacuo and subjected to silica gel column
chromatography (EtOAc:hexane, 1:9) to afford MMC (2, 1.19g, 51%)
as a syrup together with cis-4-methyl-1-(1-methylethyl)-cyclohexanol
(0.56g, 24%) as crystals.
Compound 2 (MMC). 1H-NMR (500 MHz, CDCl3): 0.91 (9H, d,
J ¼ 6:9 Hz, isopropyl-CH3 ꢃ 2, and 4-CH3), 1.03 (1H, s, OH), 1.23–
1.39 (5H, m), 1.51–1.59 (5H, m). 13C-NMR (125 MHz, CDCl3): 16.9
(isopropyl-CH3 ꢃ 2), 22.4 (4-CH3), 30.4 (3- and 5-CH2), 32.4 (4-CH),
33.7 (2- and 6-CH2), 38.6 (isopropyl-CH), 72.6 (COH). HR-FI-MS:
m=z [M]þ; calcd. for C10H20O, 156.1514; found, 156.1506.
Isolation and structural determination of (5S)-5-hydroxy-2-methyl-
5-(1-methylethyl)cyclohex-2-enone (3), (1R,2R,4R)-1-methyl-4-(1-
methylethyl)-1,2,4-cyclohexanetriol (4), and (2Z)-3,7-dimethyl-6-
oxooct-2-enal (5). A 40-mL solution of 10 mM of PPD with 150 mM
T4 (1) in 50 mM PB (pH 7.4) containing 50% methanol was stirred for
72 h at 37 ꢁC. To the reaction mixture, 8 mL of 2 M HCl was added, and
the resulting solution was extracted three times with EtOAc. The
combined organic layer was dried over anhydrous Na2SO4 and
evaporated in vacuo to give a crude product. Silica gel column
chromatography (EtOAc: hexane, 1:2) of the crude product afforded
compound 4 (13.4 mg) and crude compounds 3 and 5. Compounds 3
(1.5 mg) and 5 (1.4 mg) were purified by further preparative silica gel
TLC (MeOH:CHCl3, 1:99, developed twice).
Compound 3. 1H-NMR (500 MHz, CDCl3): 0.97 (6H, d,
J ¼ 6:9 Hz, isopropyl-CH3), 1.75 (1H, sep, J ¼ 6:8 Hz, isopropyl-
CH), 1.82 (3H, d, J ¼ 1:4 Hz, 2-CH3), 2.41 (1H, dd, J ¼ 18:6, 5.4 Hz,
4-CH2), 2.52–2.60 (3H, m, 4- and 6-CH2), 6.62 (1H, m, 3-CH). 13C-
NMR (125 MHz, CDCl3): 15.5 (2-CH3), 16.6 (isopropyl-CH3), 16.8
(isopropyl-CH3), 35.7 (4-CH2), 37.4 (isopropyl-CH), 48.0 (6-CH2),
76.5 (5-C), 135.3 (2-C), 141.2 (3-CH). HR-FD-MS: m=z [M]þ; calcd.
for C10H16O2, 168.1151; found, 168.1160.
Experimental
Materials. (ꢀ)-Terpinen-4-ol and diacetyl were purchased from
Tokyo Kasei Co. (Tokyo, Japan). 1-Phenyl-1,2-propanedione (PPD),
thionine-acetate, molecular sieves catalyst support Na–Y zeolite,
methylglyoxal (MGO, 40% in H2O), RNase A from bovine pancreas
and transfer RNA (tRNA) from baker’s yeast (S. cerevisiae) were
purchased from Sigma Aldrich (St. Louis, MO, USA). Benzoic acid,
trifluoroacetic acid (TFA), (S)-ꢀ-terpineol, aminoguanidine, metfor-
min, perchloric acid and lanthanum nitrate hexahydrate were pur-
chased from Wako Co. (Osaka, Japan). PTB was purchased from Prime
Organics (Woburn, MA, USA). Thin-layer and silica gel column
chromatography were respectively performed with Silica Gel 60 F254
(Merck, Darmstadt, Germany) and Silica Gel 60N (spherical, neutral)
(Kanto Chemical, Tokyo, Japan). 1H, 13C, HH-COSY, HMBC,
HMQC, DIF-NOE and NOESY NMR spectra were measured with
an AMX-500 instrument (Bruker, Billeria, MA, USA) or JNM-EX270
instrument (Jeol, Tokyo, Japan). Chemical shifts are reported in ꢂ ppm,
using tetramethylsilane as an internal standard, and coupling constants
(J) are given in Hertz. Mass spectra were acquired with a JMS-
T100GCV instrument (Jeol, Tokyo, Japan). Some of the NMR and MS
data were measured at the GC-MS and NMR Laboratory of Faculty of
Agriculture at Hokkaido University. Optical rotation data were
determined with a P-2000 polarimeter (Jasco, Tokyo, Japan). Singlet
oxygen photo-oxygenation was performed for 1 h at 0 ꢁC with a UM-
102 high-pressure mercury lamp (Ushio, Tokyo, Japan).
27
Compound 4. ½ꢀꢄD ¼ ꢀ14:3 (c 1.00, EtOH). 1H-NMR (270 MHz,
DMSO-d6): 0.82 (6H, d, J ¼ 6:9 Hz, isopropyl-CH3 ꢃ 2), 1.10 (3H, s,
CH3), 1.20–1.52 (4H, m, CH2), 1.59–1.81 (3H, m, isopropyl-CH,
CH2), 3.30 (1H, m, 2-CH), 4.14 (1H, s, 4-OH), 4.55 (1H, s, 1-OH),
5.12 (1H, d, J ¼ 6:9 Hz, 2-OH). 13C-NMR (67.5 MHz, CD3OD): 17.1
(isopropyl-CH3), 17.2 (isopropyl-CH3), 27.1 (1-CH3), 30.3 (5-CH2),
30.4 (6-CH2), 34.8 (2-CH2), 39.0 (isopropyl-CH), 72.0 (1-C), 75.7
(4-C and 2-CH). HR-FD-MS: m=z [M]þ; calcd. for C10H20O3,
188.1412; found, 188.1423.
Compound 5. 1H-NMR (270 MHz, CDCl3): 1.11 (6H, d,
J ¼ 6:9 Hz, isopropyl-CH3 ꢃ 2), 2.18 (3H, s, CH3), 2.45 (1H, sep,
J ¼ 6:9 Hz, 7-CH), 2.66 (2H, t, J ¼ 8:2 Hz, 4-CH2), 2.84 (2H, t,
J ¼ 8:2 Hz, 5-CH2), 5.90 (1H, d, J ¼ 7:6 Hz, 2-CH), 9.99 (1H, d,
J ¼ 7:6 Hz, 1-CHO). 13C-NMR (67.5 Hz, CDCl3): 21.4 (isopropyl-
CH3 ꢃ 2), 24.5 (3-CH3), 29.9 (4-CH2), 35.8 (5-CH2), 43.6 (7-CH),
125.2 (2-CH), 172.1 (3-C), 191.3 (1-CHO), 206.4 (6-C=O). HR-FI-
MS: m=z [M]þ; calcd. for C10H20O3, 168.1151; found, 168.1157.
Cleavage rate of the ꢀ-dicarbonyl compound. The cleavage rate of
PPD was determined by a previously described method9) with slight
modifications. Briefly, 1-mL solution of 10 mM of PPD with various
concentrations of the test compound in 50 mM PB (pH 7.4) containing
50% methanol was incubated for 24 h at 37 ꢁC while shaking at
300 rpm. To the reaction mixture, 200 mL of 2 M HCl was added, and
Preparation of (1R,3S)-3-hydroperoxy-1-(1-methylethyl)-4-methyl-
ene cyclohexanol (T4-H, 6), 2-(4-hydroperoxy-4-methylcyclohex-2-
ene-1-yl) propan-2-ol (8a), 2-(3-hydroperoxy-4-methylenecyclohexyl)-
propan-2-ol (8b, (T-ol-H, a 3:1 mixture of 8a and 8b), and 2-((1S4R)-4-
hydroperoxy-4-methylcyclohex-2-en-1-yl)propan-2-yl acetate (T-ace-
H, 10). T4-H (6), T-ol-H (8a and 8b), and T-ace-H (10) were prepared

1908
R. NAGAMATSU et al.
by the previously described method14,15) with slight modifications. The
preparation of 10 as an example of the general procedure started by
drying thionine-supported Na–Y zeolite (3.0 g) in a two-neck flask at
120 ꢁC for 1.5 h in vacuo. To the flask, 30 mL of distilled hexane and
30 mL of pyridine were added, and the mixture was stirred for 5 h in an
argon atmosphere. To the resulting mixture, 30.7 mg of (S)-ꢀ-terpinyl
acetate (9) prepared from (S)-ꢀ-terpineol (7) was added, and the mixture
was photooxygenated at 0 ꢁC for 1 h in a slow stream of oxygen. After
stirring for 12 h with 50 mL of CH3CN, the mixture was passed through
a Celite pad, and the filtrate was evaporated in vacuo and subjected to
silica gel column chromatography (CHCl3: Et2O, 16:1) to give 17.1 mg
of 10 as a white solid in a 48% yield.
Acknowledgments
The authors are grateful to Dr. N. Matsuura of
Okayama University of Science for helpful discussions,
and to Dr. H. Oguri of Hokkaido University for use of
the high-pressure mercury lamp.
References
1) Thomas MC, Baynes JW, Thorpe SR, and Cooper ME, Curr.
Drug Targets, 6, 453–474 (2005).
2) Vlassara H, Fuh H, Makita Z, Krungkrai S, Cerami A, and Bucala
R, Proc. Natl. Acad. Sci. USA, 289, 12043–12047 (1992).
3) Semba RD, Nicklett EJ, and Ferrucci L, J. Gerontol. A Biol. Sci.
Med. Sci., 65, 963–975 (2010).
T4-H (6). 1H-NMR (270 MHz, CDCl3): 0.93 (1H, d, J ¼ 7:6 Hz,
isopropyl-CH3), 1.25 (1H, s, OH), 1.39–1.53 (2H, m, 2-CH2-a, 6-CH2-
a), 1.61–1.71 (2H, m, isopropyl-CH, 2-CH2-b), 2.14 (1H, ddd,
J ¼ 12:8, 4.9, 2.6 Hz, 6-CH2-b), 2.27 (1H, ddd, J ¼ 13:5, 4.9, 2.9 Hz,
5-CH2-a), 2.44 (1H, dt, J ¼ 13:5, 4.2 Hz, 5-CH2-b), 4.76 (1H, dd,
J ¼ 11:5, 4.9 Hz, 3-CH), 4.86 (1H, d, J ¼ 1:3 Hz, olefinic-CH2-a), 4.96
(1H, d, J ¼ 1:3 Hz, olefinic-CH2-b), 8.37 (1H, s, OOH). HR-FD-MS:
m=z [M]þ; calcd. for C10H18O3, 186.1256; found, 186.1226.
4) Sasaki N, Fukatsu R, Tsuzuki K, Hayashi Y, Yoshida T, Fujii N,
Koike T, Wakayama I, Yanagihara R, Garruto R, Amano N, and
Makita Z, Am. J. Pathol., 153, 1149–1155 (1998).
5) Reddy VP, Zhu X, Perry G, and Smith MA, J. Alzheimers Dis.,
16, 763–774 (2009).
6) Forbes JM, Yee LTL, Thallas V, and Lassila M, Diabetes, 53,
1813–1823 (2004).
T-ol-H (8a). 1H-NMR (270 MHz, CDCl3): 1.18 (3H, s, 1- or 3-
CH3), 1.24 (3H, s, 1- or 3-CH3), 1.33 (3H, s, 30-CH3), 1.52–1.92 (2H,
m, 50-CH2-a and 60-CH2-a), 2.05–2.36 (3H, m, 50-CH2-b, 60-CH2-b,
and 10-CH), 5.69 (1H, ddd, J ¼ 10:2, 2.3, 1.7 Hz, 30-olefinic-CH), 6.08
(1H, ddd, J ¼ 10:2, 2.3, 1.0 Hz, 20-olefinic CH), 7.97 (1H, s, OOH).
13C-NMR (67.5 Hz, CDCl3), 20.4 (60-CH3), 24.8 (40-CH3), 26.1 (1- or
3-CH3), 28.1 (1- or 3-CH3), 47.2 (1-C), 72.7 (2-C), 78.5 (40-C), 130.3
(30-olefinic-CH), 134.1 (20-olefinic-CH). HR-FI-MS: m=z [M+H]þ;
calcd. for C10H18O3, 186.1256; found, 186.1252.
7) Brownlee M, Nature, 414, 813–820 (2001).
8) Thornalley JP, Endocrinol. Metab., 3, 149–166 (1996).
9) Vasan S, Zhang X, Kapurnitotu A, Bernhagen J, Teichberg S,
Basgen J, Wagle D, Shih D, Terlecky I, Bucala R, Cerami A,
Egan J, and Ulrich P, Nature, 382, 275–278 (1996).
10) Wolffenbuttel BHR, Boulanger CM, Crijns FRL, Huijberts
MSP, Poitevin P, Swennen GNM, Vasan S, Egan JJ, Ulrich P,
Cerami A, and Levy BI, Proc. Natl. Acad. Sci. USA, 95, 4630–
4634 (1998).
T-ol-H (8b). 1H-NMR (270 MHz, CDCl3): 1.14 (3H, s, 1- or 3-
CH3), 1.19 (3H, s, 1- or 3-CH3), 1.38–1.92 (7H, m, 10-CH, 20-, 50-, and
60-CH2), 4.53 (1H, t, J ¼ 2:9 Hz, 30-CH), 4.98 (2H, br d, J ¼ 0:7 Hz),
8.74 (1H, s, OOH). 13C-NMR (67.5 MHz, CDCl3): 26.1 (20-CH2), 27.5
(1- or 3-CH3), 28.4 (1- or 3-CH3), 30.4 (60-CH2), 31.2 (50-CH2), 42.0
(10-CH), 72.7 (2-C), 85.2 (37-CH), 113.8 (40-olefinic-CH2), 145.2
(40-olefinic-CH). FI-MS: m=z 187 ½M þ Hꢄþ.
11) Goh SY and Cooper ME, J. Clin. Endocrinol. Metab., 93, 1143–
1152 (2008).
12) Vasan S, Foiles P, and Founds H, Arch. Biochem. Biophys., 419,
89–96 (2003).
13) Adorjan B and Buchbauer G, Flavour Fragr. J., 25, 407–426
(2010).
14) Stratakis M, Sofikiti N, Baskakis C, and Raptis C, Tetrahedron
Lett., 45, 5433–5436 (2004).
T-ace-H (10). 1H-NMR (270 MHz, CDCl3): 1.34 (3H, s, 40-CH3),
1.39 and 1.44 (each 3H, each s, 1- and 3-CH3), 1.54–1.62 (2H, m,
50-CH2-a, 60-CH2-a), 1.98 (1H, m, 50-CH2-b), 2.00 (3H, s, –OCOCH3),
2.17–2.23 (1H, m, 60-CH2-b), 2.77–2.84 (1H, m, 10-CH), 5.65 (1H, dt,
J ¼ 10:2, 2.1, 1.8 Hz, 30-olefinic-CH), 5.92 (1H, dd, J ¼ 10:2, 1.9 Hz,
20-olefinic-CH), 7.48 (1H, s, 40-OOH). 13C-NMR (67.5 MHz, CDCl3):
22.1 (6-CH2), 22.5 (OCOCH3), 23.0 (40-CH3), 23.7 (1- or 3-CH3), 24.8
(1- or 3-CH3), 31.6 (5-CH2), 44.3 (10-CH), 78.6 (2-C), 84.8 (40-C),
129.9 (30-olefinic-CH), 133.7 (20-olefinic-CH). HR-FD-MS: m=z
½M þ Hꢄþ; calcd. for C12H20O4, 228.1362, found 228.1346.
15) Shailaja J, Sivagaru J, Robbins RJ, Ramamurty V, Snoj RB, and
Chandrasekhar J, Tetrahedron, 56, 6927–6943 (2000).
16) Bergmeyer HU, ‘‘Methods of Enzymatic Analysis’’ third ed.,
Academic Press, New York, pp. 136–139 (1984).
17) Li D, Shinya M, and Ubukata M, Pharmaceut. Biol., Early
Online: 1–5 (2012).
18) Lu J, Randell E, Han Y, Adeli K, Krahn J, and Meng QH, Clin.
Biochem., 44, 307–311 (2011).
19) Beisswenger PJ, Howell SK, Touchette AD, Lal S, and
Szwergold BS, Diabetes, 48, 198–202 (1999).
Assay for protective effects of terpene hydroperoxides on RNase A in
the presence of methylglyoxal (MGO). The enzyme activity of RNase A
was determined by the previously described method16,17) with slight
modifications. Briefly, 200 mL of the reaction mixture in a PCR tube
containing 1 mg mLꢀ1 RNase A and 10 mM MGO with or without the
test compound dissolved in 2 mL of EtOH in a 20 mM phosphate buffer
(pH 7.2) was incubated for 24 h at 37 ꢁC. After 200-fold dilution with a
5% glycerol solution, the mixture was transferred into a 1.5-mL
Eppendorf tube to be incubated with 0.1 mg mLꢀ1 of tRNA in a
60 mM phosphate buffer (pH 8.0) with a total volume of 250 mL. The
enzyme reaction was carried out by incubating this reaction mixture at
37 ꢁC for 20 min while shaking (300 rpm), and then quenched with
250 mL of a stop solution (0.86 g of La(NO3)3 and 7.1 mL of HClO4 in
100 mL of H2O). The resulting solution was centrifuged at 13,000 rpm
for 10 min at 4 ꢁC, and the absorbance at 260 nm was measured by a
GeneQuant pro RNA/DNA calculator spectrometer (Cambridge,
20) Yan J, Travis BR, and Borhan B, J. Org. Chem., 69, 9299–9302
(2004).
21) Nilsson U, Bergh M, Shao LP, and Karlberg AT, Chromatog-
raphia, 42, 199–205 (1996).
22) Matura M, Goosens A, Bordalo O, Garcia-Bravo B, Magnusson
K, Wrangsjo K, and Karlberg AT, Contact Dermatitis, 49, 15–
21 (2003).
23) Rudback J, Bergstrom MA, Borje A, Nilsson U, and Karlberg
¨
¨
¨
AT, Chem. Res. Toxicol., 25, 713–721 (2012).
24) Carson CF and Riley TV, J. Appl. Bacteriol., 78, 264–269 (1995).
25) Hammer KA, Carson CF, and Riley TV, J. Antimicrob.
Chemother., 42, 591–595 (1998).
26) Carson CF, Ashton L, Dry L, Smith DW, and Riley TV,
J. Antimicrob. Chemother., 48, 450–451 (2001).
27) Seyed TA, Qureshi ZA, Ali SM, Ahned S, and Ahmed SA,
Trop. Med. Int. Health, 4, 284–287 (1999).
þ
England). The remaining activity (%) ¼ ½ðAs ꢀ AbÞ ꢀ ðAc ꢀ AbÞꢄ=
28) Bassett IB, Pannowitz DL, and Barnetson RS, Med. J. Aust.,
153, 455–458 (1990).
ꢀ
þ
ꢀ
½ðAc ꢀ AbÞ ꢀ ðAc ꢀ AbÞꢄ ꢃ 100, where Ac is the absorbance of the
þ
control (RNase A and tRNA solution), Ac is the absorbance of the
29) Cross SE, Russell M, Southwell I, and Roberts MS, Eur. J.
Pharm. Biopharm., 69, 214–222 (2006).
AGEs control (RNase A, MGO and tRNA solution), Ab is the
absorbance of a blank (MGO and tRNA solution), and As is the
absorbance of the sample (RNase A, MGO, sample, and tRNA solution).
30) Reichling J, Landvatter U, Wagner H, Kostka KH, and Schaefer
UF, Eur. J. Pharm. Biopharm., 64, 222–228 (2008).