Reactions of urocanic acid (UCA) methyl esters with singlet oxygen and 4-methyl-1,2,4-triazoline-3,5-dione (MTAD)

Article abstract of DOI:10.1016/j.tet.2006.07.108

Singlet oxygen adds to the imidazole ring of cis- and trans-methyl urocanate (MUC) to yield the corresponding 2,5-endoperoxides, which are modestly stable at low temperature but decompose upon warming to form complex reaction mixtures. MTAD, a singlet oxygen mimic, reacts with cis- and trans-MUC to yield stereospecific [4+2] reaction products involving the olefinic side chain and the C₄-C₅ double bond of the imidazole ring. trans-MUC forms a 1:2 MTAD adduct while the cis isomer yields only the 1:1 adduct at 25 °C.?The stereospecificity and absence of MeOH trapping adducts indicate that these reactions may not involve open or trappable dipolar intermediates.

Full text of DOI:10.1016/j.tet.2006.07.108

Tetrahedron 62 (2006) 10700–10708
Reactions of urocanic acid (UCA) methyl esters with singlet
oxygen and 4-methyl-1,2,4-triazoline-3,5-dione (MTAD)
*
Roberto Roa and Kevin E. O’Shea
Department of Chemistry and Biochemistry, Florida International University, Miami, FL 33199, United States
Received 12 June 2006; revised 30 July 2006; accepted 30 July 2006
Dedicated in memory of Professor Christopher S. Foote
Abstract—Singlet oxygen adds to the imidazole ring of cis- and trans-methyl urocanate (MUC) to yield the corresponding 2,5-endoper-
oxides, which are modestly stable at low temperature but decompose upon warming to form complex reaction mixtures. MTAD, a singlet
oxygen mimic, reacts with cis- and trans-MUC to yield stereospecific [4+2] reaction products involving the olefinic side chain and the
ꢀ
C
stereospecificity and absence of MeOH trapping adducts indicate that these reactions may not involve open or trappable dipolar intermediates.
4
–C
5
double bond of the imidazole ring. trans-MUC forms a 1:2 MTAD adduct while the cis isomer yields only the 1:1 adduct at 25 C. The
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
mutations, premature aging, DNA damage, and respiratory
problems. The oxidative products of UCA are proposed
5
3
-(1H-Imidazol-4(5)-yl)-2-propenoic acid commonly re-
to be responsible for UV-induced immune suppression. In
2002, Elton and Morrison confirmed that the irradiation of
UCA with monochromatic light at 351 nm produces singlet
ferred to as urocanic acid (UCA) is a metabolite of histidine
and is present in the stratum corneum of the human skin.
UCA, produced naturally as the trans isomer, accumulates in
the upper layers of the epidermis and isomerizes to c-UCA
upon absorption of UV light (Fig. 1). t-UCA was found
to absorb ultraviolet B light and thus considered to be a
natural sunscreen against harmful UV rays. This led to the
addition of UCA to sunscreens and skin lotions to help pre-
vent skin cancer and skin related diseases. However, subse-
quent studies indicated that t-UCA can act as an initiator of
immunosuppression, which may be critical to UV-induced
6
oxygen. They concluded that UCA is a better scavenger
than generator of singlet oxygen and may play an important
role as an antioxidant. Colorimetric tests (EM Quant per-
oxide strip) on the reaction products of singlet oxygen with
1
,2
6
UCA showed the presence of peroxides, but detailed prod-
uct studies have not been reported. UCA-singlet oxygen
products catalyze further photo-destruction of UCA and
UCA peroxides can be major contributors to photo-initiated
6
nicking of plasmid DNA.
3
,4
pathogenesis of skin cancer and other cutaneous diseases.
The reactions of singlet oxygen with imidazole ring contain-
ing systems, especially histidine and guanine analogs, have
received considerable attention. Early studies on the photo-
oxygenation of histidine demonstrated that complex reaction
mixtures are produced, i.e., the reaction of singlet oxygen
with N-benzoyl histidine leads to the formation of 17 prod-
It was recently reported that photoexcited t-UCA can lead to
the formation of reactive oxygen species, which are linked to
a number of diseases and disorders, i.e., enzyme inactivation,
O
7
ucts. Wasserman and Lipshutz conducted extensive studies
on the reactions of singlet oxygen with substituted imidaz-
oles, proposed the involvement of 2,5-endoperoxides, and
reported two main pathways of oxidation depending on the
OH
N
h
N
NH
NH
h
O
OH
8
trans-UCA
cis-UCA
substitution pattern of the imidazole ring. When the imida-
zole ring bears a hydrogen atom at the 2- and/or 5-position,
a carbonyl group is produced at that specific site. In the
absence of a proton at the 2- or 5-position the formation
of the dioxetane by [2+2] addition of singlet oxygen to the
C4–C5 bond in the imidazole appears to be operative. Ryang
and Foote later reported the direct observation of imidazole
Figure 1. Photoisomerization of UCA.
Keywords: Photooxygenation; Singlet oxygen; Endoperoxides; Urocanic
acid.
*
Corresponding author. Tel.: +1 305 348 3968; fax: +1 305 348 3772;
e-mail: osheak@fiu.edu
9
2,5-endoperoxides as products of singlet oxygen.
0
040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.07.108

R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
10701
Singlet oxygenation of guanine and its nucleosides (guano-
sine) have received considerable attention because of the
implications on the function of DNA and its importance
during photodynamic therapy (PDT) used for the treatment
product mixtures. Given the unstable nature of the endo-
peroxides and the complexity of the reaction product mix-
tures we also studied the reactions of MUC with MTAD,
a singlet oxygen mimic. Surprisingly, MTAD and singlet
oxygen exhibit dramatically different reactivities toward
MUC.
1
0,11
of malignant tumors.
Because of the limited solubility
of guanine substrates, the instability of intermediates and
complexity of the product mixtures, the mechanism of pho-
tosensitized oxidation of guanine and its nucleosides was
1
1
unclear until recently. Foote and co-workers used a variety
of substituted guanosine adducts, isotopic labeling, and low
temperature NMR to suggest that singlet oxygenation of
guanosine involves a series of reactive intermediates, includ-
2. Results and discussion
2.1. Reactions of singlet oxygen with MUC
1
1b,c
ing endoperoxides, hydroperoxides, and dioxiranes.
The
Since the solubility of UCA in organic solvents is too low to
conduct detailed NMR studies, trans-UCA was converted
to the corresponding methyl, n-butyl, and tert-butyl esters,
using acid promoted esterification. t-MUC showed the best
solubility among the esters in the organic solvents com-
monly used for low temperature NMR. t-MUC is easily
isomerized to c-MUC by irradiation with 254 nm light under
reaction pathways are highly sensitive to solvent effects
and the substitution pattern, i.e., only the C-8 methylated
substrate reacts to form an observable endoperoxide, which
collapses to yield singlet oxygen and the starting substrate
(
Fig. 2). Kang and Foote conducted detailed product studies
on isotopically labeled imidazole ring systems and demon-
strated that the reaction pathways are dramatically influ-
enced by N-alkylation of the parent compound (Fig. 2).
1
5
argon-saturated conditions. The cis isomer was purified
using standard flash column chromatography.
1
2
The photooxidation of UCA may be a main pathway for pre-
mature aging, UV-induced immune suppression, and skin
For the typical experiment, MUC was dissolved in 0.5–1 mL
of the appropriate NMR solvent (CD Cl , acetone-d ,
CD CN, or CDCl ) along with Rose bengal, the singlet
3 3
2
2
6
1
3,14
cancer,
yet the mechanisms and products of photooxida-
tion have not been clearly established. While previous stud-
ies on the singlet oxygenation of imidazole ring systems
provide tremendous insight about the possible reaction
pathways of singlet oxygen with UCA, the presence of the
conjugated enone side chain in UCA may lead to different
singlet oxygen reactivities.
oxygen sensitizer. The solution was transferred to a 5 mm
NMR tube and placed in ice water in a windowed Dewar.
The sample was gently purged with oxygen and irradiated
using 200 W Hg–Xe lamp. Pyrex glass was employed to fil-
ter light ꢁ320 nm. Control experiments showed no conver-
sion of the cis- or trans-MUC in the absence of oxygen
(argon saturated) or by direct photolysis (in the absence of
Rose bengal). The photosensitized oxidation of t-MUC,
monitored by H NMR, was complete within 5 min and con-
tinued irradiation did not change the product distribution.
Because of the insoluble nature of UCA in organic solvents,
the methyl ester, MUC, was used as a model for UCA. MUC
is modestly soluble in the organic solvents required for low
temperature NMR studies. MUC is expected to exhibit the
same reactivity as UCA toward singlet oxygen. We report
herein singlet oxygenation of cis- and trans-MUC yields 2,5-
endoperoxides, which are modestly stable at low tempera-
ture but decompose at room temperature to afford complex
1
Likely pathways for the reaction of the MUC with singlet
oxygen are illustrated below (Fig. 3). They include [4+2]
cycloaddition to the imidazole ring to form a 2,5-endoper-
oxide 1, [4+2] across the double bond of the side chain
Guanosine
R = CH3
O
+
O
N
1
O2
N
1O
HN
N
2
HN
R
R
-
100 °C
N
H N
N
N
O
2
H N
2
Sugar
O
R = H
Complex
Product
Mixture
Sugar
Guanosine
1
5
N
15
N
H
H
13CO
2
R = H
+
Ph
15
Ph
Ph
15
Ph
13
Ph
N
N
1O
2
13
O
1
3
1
5
N
H
-
100 °C
Ph
N 15
Ph
Ph
15
O
R = CH3
N
R
R
Imidazole
1O
+
2
N 15
R
1
1
12
Figure 2. Singlet oxygenation products of guanosine and labeled imidazole.

10702
R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
H CO
3
O
O
OCH3
O
O
O
NH
O
N
OCH3
O
N
NH
1
2
NH
N
OCH3
O
OCH3
O
t-MUC
O
O
NH
O
O
NH
N
N
4
H
O
OCH3
OCH3
O
O
+
O
NH
H
N
5
6
NH
O
H
N
O
7
Figure 3. Likely pathways for the reaction of singlet oxygen with t-MUC.
and the C –C double bond of the imidazole ring to yield
4
Given the complex nature of the product mixtures observed
at room temperature and the fact that endoperoxides and
dioxetanes often decompose at room temperature the photo-
5
endoperoxide 2, and [2+2] cycloadditions to the side chain
to form dioxetane 3 or across the C –C double bond of the
ring to form 6. Dioxetanes generally collapse to the corre-
4
5
1
oxygenation and H NMR characterization were conducted
at low temperature. The NMR tube containing the reaction
sponding carbonyl compounds at room temperature.
ꢀ
solution was kept at ꢂ78 C throughout the photooxygena-
The reaction products of singlet oxygen and t-MUC at room
temperature exhibit a nearly continuous series of overlap-
tion (dry ice/acetone bath in a windowed Dewar). The
NMR tube was transferred to a pre-cooled NMR magnet
1
ꢀ
was gradually warmed within the NMR magnet and proton
ping H NMR peaks between 6 and 8 ppm. The complex
nature of the H NMR spectrum indicates a considerable
and the proton spectrum collected at ꢂ78 C. The solution
1
ꢀ
ences were observed in the spectra obtained at 0 and 25 C.
number of products formed at room temperature. The
[
spectra collected at ꢂ40, ꢂ20, 0, and 25 C. Dramatic differ-
ꢀ
The predominant peaks in the low temperature H NMR
2+2] addition to the side chain would result in the formation
1
1
of the dioxetane 3, which will collapse to 5 and 6. H NMR
and GC–MS analyses of the reaction solution show no
evidence of 5 or 6, thus indicating the [2+2] reaction does
not occur at the side chain of t-MUC under our experimental
spectrum are singlets at 8.02 and 4.90 ppm (1H/ea); a pair
of doublets at 6.80 (1H) and 6.51 ppm (1H) with
J¼16.1 Hz indicating the trans double bond is not oxidized;
and a methoxy resonance at 3.63 ppm. While solubility limi-
tations at low temperatures did not allow for acquisition of
a carbon NMR spectrum, the proton spectrum for the reac-
tion mixture at low temperature is consistent with 1 trans
based on the literature values for the endoperoxide of
1-methylimidazole (Fig. 4). There are no peaks indicating
1
conditions. The room temperature H NMR spectrum also
does not correspond to endoperoxide 1 in accordance with
1
2
chemical shifts reported for analogous compounds. To
test for the formation of endoperoxide 2 we subjected the
reaction mixture at low temperature to treatment with
cobalt(II) tetraphenylporphyrin, which is known to convert
1
6
ꢀ
endoperoxides to furans via loss of water. Subsequent
H NMR and GC–MS analyses showed no evidence
the presence of dioxetane products down to ꢂ78 C.
1
that the corresponding furan was formed. These results indi-
cate that singlet oxygen does not react at the side chain of
t-MUC.
Singlet oxygenation of c-MUC was also conducted at
low temperature. The H NMR of the product is shown
below (Fig. 5). The characteristic H NMR peaks are as
1
1

R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
10703
O
3.63
6
.80
CH3
6.60
5.94
H
H
H
5.46
O
2
.38
H
O
O
O
H C
3
N
H
O
HN
HN
O
O
6.51
O
H
O
H
CH3
3.87
H
N
H
.02
H
N
N
8
.12
H
8.19
8
6
.23
4.90
6
.38
1-methyl imidazole
1 trans
1 cis
1
Figure 4. Comparison of the H NMR chemical shifts for the 2,5-endoperoxides of 1-methyl imidazole with 1 trans and cis.
9
follows: 8.19 (1H) and 6.38 (1H) ppm, a pair of doublets at
6
cis double bond is not oxidized, and a singlet at 3.87 (3H)
corresponding to the methoxy resonance. The results are
consistent with the formation of the endoperoxide formed
by [4+2] addition of singlet oxygen to the imidazole ring.
2.2. Reactions of MTAD
.57 (1H) and 5.94 (1H) ppm with J¼11.6 Hz indicating the
Due to the instability of the endoperoxides and limited sol-
ubility of MUC we were unable to further characterize the
complex reaction mixtures produced from singlet oxygena-
tion. With this in mind we chose to study the reactions of
MUC with MTAD, a singlet oxygen mimic. MTAD, a power-
ful electrophile, undergoes the same type of reactions as
A number of 2,5-endoperoxides decompose via retro Diels–
Alder reaction to yield starting material and singlet oxygen.
Tetramethylethylene singlet oxygen trapping experiments
indicate the endoperoxides of MUC do not decompose to
2
0
singlet oxygen, namely [4+2], [2+2], and the ene reactions.
We expected analogous products for the reactions of MTAD
and singlet oxygen with MUC, but unlike the peroxide prod-
ucts formed from singlet oxygen, the expected [2+2] and
[4+2] products from MTAD (Fig. 6) should be stable at
room temperature and readily characterized using NMR.
ꢀ
singlet oxygen upon warming from ꢂ78 C to room temper-
ature. The formation of the endoperoxides can occur by con-
1
7
certed or non-concerted processes. To test for the possible
involvement of dipolar intermediates, 3 equiv of methanol
was added to the solvent prior to running the reaction. Under
these conditions the formation of methanol adducts indicates
t-MUC was dissolved in 0.5–1 mL of CDCl in an NMR tube
3
and 1 equiv of MTAD was added. Upon addition of MTAD
to the t-MUC solution, the color immediately turns from red
to deep purple, but quickly fades to a very light pink. Positive
1
8,19
1
the involvement of dipolar intermediates.
The H NMR
spectra of the reaction run in the presence of MeOH did not
indicate the formation of MeOH adducts or any change in
+
mode APCI-MS established the molecular ion, MH ¼266,
ꢀ
the product distribution at ꢂ78 and 0 C. While the presence
indicating the addition of 1 equiv of MTAD. The proton
NMR spectrum of the product shows a clean set of three trip-
lets at 5.75, 5.49, and 5.24 ppm (1H each), with coupling
constants w3.4 Hz, as well as singlets at 7.45 (1H), 3.05
(3H), and 3.75 ppm (3H). The spectrum does not contain
peaks in the olefinic region with a characteristic trans cou-
pling constant. The absence of trans olefinic peaks rules out
the presence of 8 and 9. The diazetidine product, 11 would
exhibit a pair of doubletsw5.0–6.0 ppm with characteristic
of MeOH adducts would clearly indicate the involvement of
dipolar intermediates, under our experimental conditions,
where MeOH (3 equiv) was added to a relatively non-polar
solvent a dipolar intermediate may be too short lived to be
trapped. There was also no significant change in the product
distribution using CD Cl , acetone-d , CDCl , or CD CN as
2
2
6
3
3
solvents, suggesting that the polar intermediates may not
involve in the product forming reaction pathways.
H
H
6
7
O
8
O
5
9
HN1
O
CH3
4
O
H
3
2
N
H
J =11.6Hz
SM
SM
SM
SM
H(6)
H(2)
H(7)
H(4)
PPM
8.0
7.6
7.2
6.8
6.4
6.0
5.6
1
Figure 5. H NMR spectrum of the products from reaction of singlet oxygen of c-MUC at 0 C in CDCl
ꢀ
3
.

10704
R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
O
OCH3
H C
3
O
N
O
O
O
N
N
N
NH
N
CH3
OCH3
N
N
NH
N
O
MTAD
MTAD
OCH3
8
[
2+2]
9
[4+2]
O
NH
t-MUC
N
O
MTAD
MTAD
OCH3
O
H3C
[4+2]
[2+2]
N
O
N
N
N
O
CH3
O
NH
O
N
N
N
O
CH3
1
1
N
NH
10
Figure 6. Likely products from the reactions of MTAD with t-MUC.
2
0c
ꢀ
trans coupling constants as well as singlets corresponding
to the imidazole ring hydrogens. The H NMR spectrum
does not possess peaks consistent with diazetidines. The
The reaction of c-MUC with MTAD was run at ꢂ30 C and
exhibits the same color changes as the reaction of t-MUC.
The proton NMR spectrum of the product contains three
doublets of doublets at 5.93, 5.88, and 5.29 ppm (1H each)
as shown in Figure 8. The spectrum indicates a single prod-
uct and is consistent with, 10 trans, the [4+2] cycloaddition
of MTAD across the double bond of the side chain and the
C –C double bond of the imidazole ring.
1
[
4+2] addition of MTAD across the olefinic side chain and
the C –C double bond of the imidazole ring yields 10.
4
5
While analogous [4+2] reactions of MTAD with styrenes
have been reported, the reaction product of MUC is
not expected to lead to the observed set of triplets in the
2
1,22
4
5
1
H NMR spectrum. The H-shift product presented below
Fig. 7) would lead to the re-aromatization of the imidazole
1
(
ring system possessing –CHH –CH – set protons and may
The H NMR spectra of the MTAD products from the
cis- and trans-MUC indicate each isomer leads to a single
diastereomer (likely enantiotopic pairs). The observation
of three clean triplets in the proton NMR spectrum of the
t-MUC product and three doublets of doublets from
c-MUC product is rationalized in terms of long range and
different coupling constants due to geometrical differences.
The results are in accordance with the Woodward–Hoffmann
rules, with MTAD reacting in a disrotatory fashion to yield
the cis or trans product depending on the stereochemistry
of the starting material. One might expect 10 would be sus-
ceptible to a retro Diels–Alder transformation. However,
0
00
exhibit three triplets in the H NMR spectrum.
1
To test the presence of such a product, low temperature ex-
periments were monitored by H NMR at ꢂ30 C. No rear-
1
ꢀ
rangement of the product was observed upon warming to
ꢀ
at room temperature. Carbon, DEPT, and H– C HETCOR
2
5 C. The initial product, however, does slowly degrade
1
13
ꢀ
NMR spectra were obtained at ꢂ30 C. The DEPT spectra
indicate that the reaction product does not contain any meth-
ylene groups ruling out the presence of the H-shifted prod-
uct. Based on the NMR data summarized below (Figs. 8
and 9), we propose that the observed reaction product is 10.
ꢀ
upon heating to 50 C they decompose yielding complex
mixtures, which do not include MUC.
O
Addition of MTAD to a solution containing 10 cis results in
its disappearance with the clean formation of a secondary
product. The H NMR spectrum of the secondary product
has the following characteristics: a singlet at 7.93 ppm
O
H' H''
OCH3
H
H''
1
O
H
O
O
N
HN
N
N
N
(
1H), a pair of doublets at 6.01 and 5.37 ppm (1H each, J¼
H shift
N
O
N
N
CH3
N
1.6 Hz), singlets at 3.80, 3.10, and 3.03 ppm (3H each). The
H and C NMR spectra indicate a second equivalent of
N
H'
CH3
1
13
H
O
MTAD reacts via an ene reaction to form a stable 2:1 adduct.
The reaction of MTAD with b,b-dimethyl-p-methoxy
Figure 7. Potential H-shifted product of 10 cis.

R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
CH3
10705
O
6
7
H
O
6
7
CH3
H
O
H
H
O
O
O
N
N
N
N
N
N
N
N
J74 = 1.0 Hz
J67 = 7.7 Hz
J46 = 3.4 Hz
N
H
CH3
H
4
N
H
J74 = J67 = J46 = 3.4 Hz
CH3
H
O
O
4
10 cis
1
0 trans
4
4
6
7
7
6
5
.8
5.6
5.4
5.2
6.0
5.9
5.4
5.2
1
Figure 8. Expansion of H NMR spectral regions for 10 cis and trans.
styrene yields an analogous 2:1 [4+2]/ene adduct, but via
trappable dipolar intermediates. Addition of MTAD to
induced by the trans geometry. It is also possible that the
geometry of the allylic proton in 10 trans is not properly
aligned (co-planar with the p system) for the ene reaction
to occur.
2
1
the 10 trans did not result in the formation of a 2:1 adduct
ꢀ
at room temperature. While warming the solution to 50 C
promotes the formation of the 2:1 adduct as a minor product,
the mixture readily decomposes.
To probe the involvement of dipolar intermediates in the
reaction of MTAD with MUC, 3 equiv of MeOH was
added to the solvent prior to adding MTAD to MUC.
2
3,24
Based on these results we propose the following mechanism
Fig. 10) for the reactions of MTAD with MUC. The addition
1
(
Detailed analysis of the H NMR spectrum from the MeOH
trapping studies did not indicate the formation of any meth-
anol adducts under our experimental conditions from either
of the first equivalent of MTAD to trans- and cis-MUC
occurs via a suprafacial disrotatory [4+2] cycloaddition
such that the relative stereochemistries of the homo-allylic
protons are cis and trans, respectively. The cis homo-allylic
protons in 10 cis are susceptible to the ene reaction with
MTAD to form a 2:1 [4+2]/ene adduct, where as the ene re-
action of MTAD with 10 trans is inhibited by steric factors
ꢀ
the cis or the trans isomers at ꢂ30 and 0 C. While the forma-
tion of MeOH adducts is strong evidence for the involvement
of dipolar intermediates the absence of such adducts suggest
that dipolar intermediates are not involved or too short lived
to be trapped.
1
3
1
Figure 9. Summary of the C, DEPT, and H– C HETCOR NMR spectra for 10 cis.
13

10706
R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
O
state leading to addition of MTAD to the imidazole ring
O
O
will exhibit significant steric interactions between the
N-methyl group on MTAD and the lone pair of the pyridine
nitrogen atom of the imidazole ring. The [4+2] addition of
MTAD involving the enone side chain, the observed reaction
pathway, does not possess such steric interactions. Since
singlet oxygen is not subject to such secondary orbital inter-
actions or steric constraints a different reaction pathway is
observed compared to MTAD.
O
N
N
N
CO CH
+
2
3
N
CH3
N
N
N
+
N
CH3
N
H
N
H
O
O
t-MUC
4 + 2]
c-MUC
[
[4 + 2]
Disrotatory
Disrotatory
H3C
3. Conclusions
H C
O
3
N
O
O
O
N
Singlet oxygen and MTAD exhibit dramatically different
reactivities toward cis- and trans-MUC. Singlet oxygen
undergoes [4+2] cycloaddition to the imidazole ring in
MUC, while MTAD undergoes [4+2] stereospecific cyclo-
addition involving the olefinic side chain and C –C double
O
N
O
O
H
O
N
H
N
O
N
O
N
N
N
N
N
N
CH3
N
N
N
4
5
CH3
H
H
N
H
bond of the imidazole ring. The observed different reactiv-
ities may be due to secondary orbital interactions of MTAD
during the cycloaddition and/or steric factors not presented
by singlet oxygen. t-MUC forms a 1:2 MTAD [4+2]/ene
adduct while the cis isomer only yields the [4+2] adduct at
room temperature. While stereospecificity and absence of
MeOH trapping adducts indicate the reactions of MTAD
with MUC do not involve open or trappable dipolar inter-
mediates, the vivid short lived color changes during the pri-
mary stages of the reactions indicate a charge transfer
complex may be involved in the reaction process. The endo-
peroxides formed from the reaction of singlet oxygen with
MUC are modestly stable at low temperature and decompose
upon warming to form complex reaction mixtures. Studies
are underway to characterize the products formed from the
collapse of the MUC endoperoxides in an attempt to better
understand the antioxidant role of UCA.
H
O
O
+
enantiomer
+
enantiomer
ene Rxn
RT
ene Rxn
X RT
O
H C
3
N
NH O
N
NOT
OBSERVED
N
O
O
H
H
O
N
N
N
CH3
N
H
O
Figure 10. Proposed mechanism for the reaction of MTAD with cis- and
trans-MUC.
4
. Experimental
Singlet oxygen and MTAD exhibit different reactivities to-
ward MUC. One might expected that the imidazole ring sys-
tem is more electron rich than the diene, which includes the
enone side chain and C –C double bond of the imidazole
ring. If the reaction was controlled primarily by electronic
factors, addition to the imidazole ring may be predominant
as was observed with singlet oxygen. On the other hand,
the observed [4+2] addition of MTAD involves the enone
side chain. The observed different reactivities may be due
to steric factors induced during secondary orbital interac-
tions of MTAD with MUC (Fig. 11). The endo transition
4
.1. General
4
5
Acetone-d , CDCl , CD Cl , CD CN, t-UCA, Rose bengal,
3
6
3
2
2
cobalt(II) tetraphenylporphyrin, and MTAD were purchased
from Aldrich and used as received. TLC analyses were per-
formed on 0.2 mm thick plates pre-coated with fluorescent
silica (Whatman), using short wavelength UV light to visu-
alize the spots. H and C NMR spectra were recorded on
a Bruker ARX-400 spectrometer equipped with a variable
temperature probe. Chemical shifts values are in parts per
million based on the chemical shift of the solvent.
1
13
CO2CH3
1
3
C NMR peak multiplicity was determined by DEPTexperi-
ments. MS measurements were made using a direct probe on
a Thermoquest Navigator Mass Spectrometer.
N
CO2CH3
H
N
H
H
CH3
N
C
N
H
O
4.2. Reactions
N
H
O
N
4
.2.1. Synthesis of trans-methyl, n-butyl, and tert-butyl
O
N
1
° bonding
urocanate. trans-UCA (1.0 g, 7.2 mmol) and 11 mL of
10% v/v solution of concentrated sulfuric acid in the appro-
priate alcohol were placed into a 50 mL round bottom flask.
The solution was then heated to reflux forw20 h. The resul-
tant solution was allowed to cool to room temperature and
18 mL of the corresponding alcohol was added. The pH of
O
N
N
interactions
N
2° bonding
interactions
endo T.S. predicted for addition
to imidazole ring
endo T.S. predicted for addition
involving side chain
Figure 11. Possible transitions states for [4+2] additions of MTAD to
t-MUC.

R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
10707
the solution was adjusted to 8–8.5 by adding solid anhydrous
sodium carbonate. The basic solution was then gravity fil-
tered and the solvent was removed under vacuum. The crude
product was extracted with hot ethyl acetate. Evaporation of
the solvent gave the desired ester in 90–95% yield. TLC
analysis and column chromatography were performed using
5.88 (d of d, J¼3.4 and 1.1 Hz, 1H), 5.29 (d of d, J¼7.7 and
1.1 Hz, 1H), 3.80 (s, 3H), 3.13 (s, 3H). 100 MHz (CD CN):
3
d (ppm) 25.3, 54.1, 55.4, 73.3, 102.1, 131.3, 156.8, 157.1,
162.7, 164.6. Positive mode APCI-MS of the 10 cis and trans
yielded a molecular ion (MH : 266) indicating 1:1 adducts.
+
a 4:1 mixture of CHCl /EtOH. trans-Methyl urocanate:
3
1:2 t-MUC/MTAD adduct 7.93 (s, 1H), 6.01 (d, J¼1.6 Hz,
1H), 5.37 (d, J¼1.6 Hz, 1H), 3.80 (s, 3H), 3.10 (s, 3H), 3.03
1
4
00 MHz H NMR (CD CN): d (ppm) 7.63 (s, 1H), 7.57
3
(
d, J¼15.7 Hz, 1H), 7.33 (s, 1H), 6.41 (d, J¼15.7 Hz, 1H),
(s, 3H). 100 MHz (CD CN): d (ppm) 24.8, 25.1, 50.1, 53.2,
3
1
3
3
1
4
.69 (s, 3H); 100 MHz C NMR: 167.3, 138.0, 136.4,
34.6, 123.8, 113.4, 51.3; trans n-butyl urocanate:
60.0, 123.1, 132.4, 139.9, 153.5, 155.1, 155.4, 157.3, 168.4.
1
00 MHz H NMR (acetone-d ): d (ppm) 7.65 (s, 1H), 7.53
6
Acknowledgements
(
4
d, J¼15.6 Hz, 1H), 7.30 (s, 1H), 6.39 (d, J¼15.6 Hz, 1H),
.12 (t, 2H), 1.55 (m, 2H), 1.30 (m, 2H), 0.92 (t, 3H); trans
R.R. was supported by NIH/NIGMS R25GM061347. We
thank Yali Hsu for his assistant with the NMR experiments.
1
tert-butyl urocanate: 400 MHz H NMR (acetone-d6):
d (ppm) 7.64 (s, 1H), 7.55 (d, J¼15.7 Hz, 1H), 7.32 (s,
1
H), 6.40 (d, J¼15.7 Hz, 1H), 1.38 (s, 9H).
References and notes
4
.2.2. Synthesis of cis-MUC. trans-MUC (0.760 g, 5 mmol)
was mixed with 250 mL of dichloromethane and purged
with argon for 2 min. The solution was prepared in a sealed
quartz reaction vessel and irradiated in a Rayonet photo-
chemical reactor equipped with sixteen 254 nm emitting
1. Morrison, H.; Avnir, D.; Bernasconi, C.; Fagan, G. Photochem.
Photobiol. 1980, 32, 711–714.
2. Baden, H. P.; Pathak, M. A. J. Invest. Dermatol. 1967, 48, 11–17.
3. Morrison, H. Photodermatology 1985, 2, 158–165.
4. Norval, M.; Simpson, T. J. Photochem. Photobiol. 1989, 50,
267.
5. Haralampus-Grynaviski, N.; Ransom, C.; Ye, T.; Rozanowska,
M.; Wrona, M.; Sarna, T.; Simon, J. D. J. Am. Chem. Soc. 2002,
124, 3461–3468.
ꢀ
1
bulbs for w4 h at 34 C. Analysis by H NMR indicates
a trans–cis ratio of 20:80. TLC analysis and column were
1
performed using CHCl /EtOH in a ratio 9:1. H NMR indi-
3
cates the purified sample was of high purity with no traces
of the trans isomer. 400 MHz H NMR (CD Cl ): d (ppm)
1
2
2
7
.70 (s, 1H), 7.34 (s, 1H), 6.88 (d, J¼12.5 Hz, 1H), 5.70
6. Elton, L. M.; Morrison, H. Photochem. Photobiol. 2002, 75,
565–569.
7. Tomita, M.; Irie, M.; Ukita, T. Tetrahedron Lett. 1968, 4933–
1
3
(d, J¼12.5 Hz, 1H), 3.78 (s, 3H); 100 MHz C NMR:
d (ppm) 168.7, 137.4, 136.7, 131.2, 127.4, 110.7, 51.7.
4
949.
4
.2.3. Reactions of cis- and trans-MUC with singlet oxy-
8. Wasserman, H. H.; Lipshutz, B. H. Reactions of Singlet
Oxygen with Heterocyclic Systems. In Organic Chemistry.
A Series of Monographs; Academic: New York, NY, 1979;
pp 430–506.
9. Ryang, H. S.; Foote, C. S. J. Am. Chem. Soc. 1979, 101, 6683–
6687.
10. Ravanat, J.-L.; Berger, M.; Bernad, F.; Langlois, R.; Ouellet,
R.; van Lier, J. E.; Cadet, J. Photochem. Photobiol. 1992, 55,
809–814 and references within.
11. (a) Ye, Y.; Muller, J. G.; Luo, W.; Mayne, C. L.; Shallop, A. J.;
Jones, R. A.; Burrows, C. J. J. Am. Chem. Soc. 2003, 125,
13926–13927; (b) Kang, P.; Foote, C. S. J. Am. Chem. Soc.
2002, 124, 4865–4873; (c) Sheu, C.; Kang, P.; Khan, S.;
Foote, C. S. J. Am. Chem. Soc. 2002, 124, 3905–3913.
12. (a) Kang, P.; Foote, C. S. Tetrahedron Lett. 2000, 9623–9626;
(b) Kang, P.; Foote, C. S. J. Am. Chem. Soc. 2002, 124, 9629–
9638.
13. (a) Kammeyer, A.; Eggelte, T. A.; Overmars, H.; Bootsma, A.;
Bos, J. D.; Teunissen, M. B. M. Biochim. Biophys. Acta, Gen.
Subj. 2001, 1526, 277–285; (b) Ramu, A.; Mehta, M. M.;
Leaseburg, T.; Aleksic, A. Cancer Chemother. Pharmacol.
2001, 47, 338–346; (c) Morrison, H.; Deibel, R. M.
Photochem. Photobiol. 1988, 48, 153–156.
gen. MUC (2 mg, 13 mmol) was mixed with 1 mL of deuter-
ated solvent (CD Cl , acetone-d , CD CN, or CDCl ) in an
2
2
6
3
3
NMR tube. Rose bengal (10 mM) was used as the sensitizer.
A 200 W Hg–Xe lamp was used to irradiate the sample. A
water filter was placed in front of the lamp as a heat filter
and Pyrex glass was employed to filter light ꢁ320 nm. The
NMR tube containing the sample was placed inside a win-
dowed Dewar flask and oxygen was gently bubbled through
1
the solution during irradiation. H NMR spectra were taken
at 5-min intervals at the same temperature of irradiation. The
1
peaks in the H NMR spectrum at low temperature assigned
to endoperoxide, 1 trans are as follows: 400 MHz (CDCl3):
d (ppm) 8.02 (s, 1H), 6.80 (d, J¼16.1 Hz, 1H), 6.51 (d,
1
J¼16.1 Hz, 1H), 4.90 (s, 1H), 3.63 (s, 3H). The H NMR
signals for endoperoxide 1 cis are as follows: 400 MHz
(
CDCl ): d (ppm) 8.19 (s, 1H), 6.57 (d, J¼11.6 Hz, 1H),
3
6
.38 (s, 1H), 5.94 (d, J¼11.6 Hz, 1H), 3.87 (s, 3H).
4
.2.4. Reactions of cis- and trans-MUC with MTAD. MUC
2 mg, 13.1 mmol) was dissolved in 1 mL of deuterated
(
chloroform. The solution was then reacted with 1 equiv
of MTAD (1.5 mg, 13.1 mmol). The reactions of MTAD
ꢀ
were carried out at temperatures down to ꢂ30 C in
14. (a) Cadet, J.; Douki, T.; Gasparutto, D.; Ravanat, J. L. Mutat.
Res. 2003, 531, 5–35; (b) Ravanat, J. L.; Douki, T.; Cadet, J.
J. Photochem. Photobiol. B: Biol. 2001, 63, 88–102.
15. Mohammad, T.; Morrison, H. Org. Prep. Proced. Int. 2000, 32,
581–584.
a 5 mm NMR tube. NMR characterizations were run at the
same temperature as the reaction. A summary of the chem-
ical shift information for the major reaction products is pro-
vided below 400 MHz (CDCl ): d (ppm) 1:1 adduct, 10 cis;
3
7.45 (s, 1H), 5.75 (t, J¼3.4 Hz, 1H), 5.49 (t, J¼3.4 Hz, 1H),
5.24 (t, J¼3.4 Hz, 1H), 3.75 (s, 3H), 3.05 (s, 3H); 1:1 adduct,
10 trans; 7.51 (s, 1H), 5.93 (d of d, J¼7.7 and 3.4 Hz, 1H),
16. O’Shea, K. E.; Foote, C. S. J. Org. Chem. 1988, 54, 3475–3477
and references within.
17. O’Shea, K. E.; Foote, C. S. J. Am. Chem. Soc. 1988, 110, 7167.

10708
R. Roa, K. E. O’Shea / Tetrahedron 62 (2006) 10700–10708
1
8. Early examples include (a) Jefford, C. W.; Rimaubt, C. G. J. Am.
21. Stratakis, M.; Hatzimarinaki, M.; Froudakis, G. E.;
Orfanopoulos, M. J. Org. Chem. 2001, 66, 3682–3687.
22. Lai, Y.-C.; Mallakpour, S. E.; Butler, G. B.; Palenik, G. L.
J. Org. Chem. 1985, 50, 4378–4381.
23. The role and involvement of dipolar intermediates during
the reactions of triazolinedione have received considerable
attention recently, see: (a) Singleton, D. A.; Hang, C. J. Am.
Chem. Soc. 1999, 121, 11885–11893; (b) Roubelakis, M. M.;
Vougioukalakis, G. C.; Angelis, Y. S.; Orfanopoulos, M. Org.
Lett. 2006, 8, 39–42; (c) Kim, D. K.; O’Shea, K. E. J. Am.
Chem. Soc. 2003, 126, 7000–7001; (d) Vassilikogiannakis,
G.; Stratakis, M.; Orfanopoulos, M. Org. Lett. 2000, 2, 2245–
2248.
Chem. Soc. 1978, 100, 6437–6438; (b) Asveld, E. W. H.;
Kellogg, R. M. J. Am. Chem. Soc. 1980, 102, 3644–3648.
9. The involvement of dipolar intermediates in singlet oxygen re-
actions has recently been questioned. See: (a) Singleton, D. A.;
Hang, C.; Szymansky, M. J.; Meyer, M. P.; Leach, A. G.;
Kuwata, K. T.; Chen, J. S.; Greer, A.; Foote, C. S.; Houk,
K. N. J. Am. Chem. Soc. 2003, 125, 1319; (b) Singleton,
D. A.; Hang, C.; Szymansky, M. J.; Greenwald, E. E. J. Am.
Chem. Soc. 2003, 125, 1176.
1
2
0. (a) Smonou, I.; Orfanopoulos, M.; Foote, C. S. Tetrahedron
Lett. 1988, 29, 2769–2772; (b) Orfanopoulos, M.; Foote,
C. S.; Smonou, I. Tetrahedron Lett. 1987, 28, 15–18; (c)
Jensen, F.; Foote, C. S. J. Am. Chem. Soc. 1987, 109, 6376–
24. For an excellent review of the mechanisms of triazolinedione
ene reactions, see: Vougioukalakis, G. C.; Orfanopoulos, M.
Synlett 2005, 713–731.
6385; (d) Clennan, E. L.; Earlywine, A. D. J. Am. Chem.
Soc. 1987, 109, 7104–7110.