56
3 Dopaquinone Conversion and Related Reactions
we employed density functional theory-based first principles calculation, with
parameters described in Sect. 3.2.
Section 3.3 describes the simulated results for the competition between cyclization
and thiol binding of dopaquinone and RD-quinone. As a model of thiols, the simplest
structure methane thiolate ion (SCH 3
− ) was chosen. We compared the binding energy
of SCH 3
− to dopaquinone and RD-quinone before and after the cyclization. As
the cyclized structure of dopaquinone and RD-quinone, we used dopachrome and
RD-cyclic quinone, respectively. As a result, dopaquinone case showed an unstable
SCH 3
− -bound state after cyclization, while RD-quinone could bind SCH 3
− even
after cyclization.
Section 3.4 describes computational studies on the initial cyclization process
for various o-quinones. As an elementary step of the initial cyclization process,
we considered the C6–N or C6–O cyclic bond formation. For the cyclization of oquinoneamines, we chose dopaminequinone and its analogs, namely α-carboxylated,
N-methylated, and N-formylated derivatives. Besides, the methylene- (–CH 2 −)
inserted analogs in their side chain were also considered. Furthermore, as an example
of cyclization with hydroxyl group, RD-quinone cyclization was investigated. As a
result, the α-carboxylated and N-methylated dopaminequinone showed decreased
activation barriers for the C6–N cyclic bond formation. In contrast, introduction of
N-formyl group resulted in a remarkable increase in the activation barrier. Using
the methylene-inserted structures, the increased hydrocarbon side chains showed
slightly decreased activation barriers. In the case of RD-quinone, the C6–O cyclic
bond formation was not possible from the electroneutral structure, indicating the
necessity of hydroxyl deprotonation as the initial step.
Section 3.5 describes a mechanistic investigation on binding of cysteine with
dopaquinone. We found that cysteine thiolate could form a bound state on C5, C2,
C6, C3–C4 bridge, and C1 of dopaquinone. Especially, we identified C3–C4 bridge
site, which was not experimentally found, and the binding on C3–C4 bridge showed
the highest binding energy among all the sites, including C5 and C2. Furthermore,
the binding energy at C2 was not higher than that at C6. Therefore, the reported
preference of C2 over C6 for the thiol binding site cannot be explained by the energetic stability. From these results, we proposed the binding mechanism, in which
cysteine approaches C3–C4 bridge, and then migrates to C5 or C2, followed by
proton rearrangements to give 5-S-cysteinyldopa or 2-S-cysteinyldopa, respectively.
3.2 Calculation Methods for Simulating Dopaquinone
Conversion
As in the previous chapter, we conducted density functional theory-based first principles calculation [23, 24] with the Becke’s three-parameter hybrid functional [25]
combined with the Lee-Yang-Parr correlation functionals (B3LYP) [26]. Calculations were carried out with 6-31 ++G(d,p) basis set using the Gaussian09 computational package [27]. The atomic charges were estimated by the natural atomic orbital
3 Dopaquinone Conversion and Related Reactions
we employed density functional theory-based first principles calculation, with
parameters described in Sect. 3.2.
Section 3.3 describes the simulated results for the competition between cyclization
and thiol binding of dopaquinone and RD-quinone. As a model of thiols, the simplest
structure methane thiolate ion (SCH 3
− ) was chosen. We compared the binding energy
of SCH 3
− to dopaquinone and RD-quinone before and after the cyclization. As
the cyclized structure of dopaquinone and RD-quinone, we used dopachrome and
RD-cyclic quinone, respectively. As a result, dopaquinone case showed an unstable
SCH 3
− -bound state after cyclization, while RD-quinone could bind SCH 3
− even
after cyclization.
Section 3.4 describes computational studies on the initial cyclization process
for various o-quinones. As an elementary step of the initial cyclization process,
we considered the C6–N or C6–O cyclic bond formation. For the cyclization of oquinoneamines, we chose dopaminequinone and its analogs, namely α-carboxylated,
N-methylated, and N-formylated derivatives. Besides, the methylene- (–CH 2 −)
inserted analogs in their side chain were also considered. Furthermore, as an example
of cyclization with hydroxyl group, RD-quinone cyclization was investigated. As a
result, the α-carboxylated and N-methylated dopaminequinone showed decreased
activation barriers for the C6–N cyclic bond formation. In contrast, introduction of
N-formyl group resulted in a remarkable increase in the activation barrier. Using
the methylene-inserted structures, the increased hydrocarbon side chains showed
slightly decreased activation barriers. In the case of RD-quinone, the C6–O cyclic
bond formation was not possible from the electroneutral structure, indicating the
necessity of hydroxyl deprotonation as the initial step.
Section 3.5 describes a mechanistic investigation on binding of cysteine with
dopaquinone. We found that cysteine thiolate could form a bound state on C5, C2,
C6, C3–C4 bridge, and C1 of dopaquinone. Especially, we identified C3–C4 bridge
site, which was not experimentally found, and the binding on C3–C4 bridge showed
the highest binding energy among all the sites, including C5 and C2. Furthermore,
the binding energy at C2 was not higher than that at C6. Therefore, the reported
preference of C2 over C6 for the thiol binding site cannot be explained by the energetic stability. From these results, we proposed the binding mechanism, in which
cysteine approaches C3–C4 bridge, and then migrates to C5 or C2, followed by
proton rearrangements to give 5-S-cysteinyldopa or 2-S-cysteinyldopa, respectively.
3.2 Calculation Methods for Simulating Dopaquinone
Conversion
As in the previous chapter, we conducted density functional theory-based first principles calculation [23, 24] with the Becke’s three-parameter hybrid functional [25]
combined with the Lee-Yang-Parr correlation functionals (B3LYP) [26]. Calculations were carried out with 6-31 ++G(d,p) basis set using the Gaussian09 computational package [27]. The atomic charges were estimated by the natural atomic orbital