5048
J. J. Maresh et al. / Tetrahedron Letters 55 (2014) 5047–5051
of mixing NaOCl and tyrosine at ice bath temperature followed
Results and discussion
by warming to 37 °C, was used to synthesize 4-HPAA (2a) for
enzymatic (S)-norcoclaurine synthesis.17 However, the aldehyde
product itself was not isolated.
Sodium hypochlorite oxidation of tyrosine to 4-HPAA
We found that slow addition of one equivalent of NaOCl solu-
tion at 60.2 mM of over 5–10 min with vigorous stirring gave
clean, quantitative conversion of 1.0 mM tyrosine to 4-HPAA in
10 mM phosphate buffer after 1–2 h at 37–45 °C (see Fig. 1).18 Cold
NaOCl addition9 had no noticeable effect on the overall reaction.
Control of reaction stoichiometry is crucial. Two equivalents of
NaOCl yielded significant concentrations of the corresponding
nitrile. This side product was identified by matching HPLC-UV, 1H
NMR, and GC–MS data to those of authentic 4-hydroxyphenyl-
acetonitrile. Therefore, we recommend measuring the active hypo-
chlorite concentration of the NaOCl solution (see Supplemental
material for details).19 Solid Ca(OCl)2 is not a suitable source of
hypochlorite for this reaction as it produced significant side prod-
ucts. The rate of reagent addition is also critical. Rapid addition of
NaOCl resulted in a yellow reaction mixture with several unidenti-
fied minor side products apparent by HPLC, including the corre-
sponding nitrile. Finally, the initial concentration of tyrosine was
also critical for minimization of side-products. Supplemental
Figure S2 shows the results of oxidation at six different concentra-
tions of tyrosine ranging from 1.0 to 8.0 mM. We observed that
concentrations >4.0 mM yielded significant side products.
We initially evaluated multiple approaches to the synthesis of
2a and halogenated variants 2b–d (numbering in Table 1). We
attempted oxidation of 4-(2-hydroxyethyl)phenol by pyridinium
dichromate4, Dess–Martin periodinane, Swern oxidation, and Pari-
kh–Doering oxidation.5 All methods gave poor yields and signifi-
cant side products in our hands.6 Parikh–Doering oxidation was
the only method to give yields >30%, but produced significant side
products that must be removed by column chromatography
(Supporting Table S1 and Fig. S1).
We next turned our attention to oxidative decarboxylation of
a
-amino acids. We selected sodium hypochlorite (NaOCl) over
chloramine-T, the most common reagent for this reaction,7
because it is inexpensive and the only byproducts are biologically
compatible salts, thus allowing the products to be used directly for
biological studies. In humans, the enzyme myeloperoxidase gener-
ates hypochlorite ion and is known to transform L-tyrosine (1a)
into 4-HPAA (2a) in the enzyme active site.8 The same transforma-
tion was observed in pH 7.0 phosphate buffer with hypochlorous
acid, but not with hypobromous acid.9
We found no report that describes the general application of
NaOCl oxidative decarboxylation in organic synthesis. The reaction
is prone to many side reactions such as nitrile formation,10 aro-
matic chlorination,11,12 and formation of unidentified products13
when pH, concentration, and rate of addition are not controlled.
The first high-yield procedure was reported by Langheld14 who
used an apparatus that mixed amino acids with NaOCl at ice-cold
temperature and then dropped the reaction into a tube of glass
wool with a jet of steam passing through to simultaneously cata-
lyze rapid decomposition of the amino acid while removing the
volatile products. The aldehydes were isolated as nitrophenyl-
hydrazone derivatives. Tryptophan and 5-methoxy-tryptophan
have been converted into 2-(indol-3-yl)acetaldehyde and 2-(5-
methoxy-indol-3-yl)acetaldehyde using a two-phase reaction with
benzene, requiring multiple additions of reagents and column
chromatography to remove side products.15,16 Hazen’s method9
Three intermediates were observed by HPLC-UV during the
course of the reaction (Fig. 1). Initial attempts to characterize these
species by GC–MS from extraction of the reaction mixture have
been inconclusive. All species present in solution have nearly
identical UV spectra. Intermediate 1 appears immediately after
the addition of NaOCl and maintains the greatest intensity
(Supplemental Fig. S3). An iminium intermediate species has been
previously isolated from oxidative decarboxylation of a-N-diphen-
ylglycine.20 By analogy, we propose that intermediate 1 is the
iminium ion that precedes hydrolysis in Scheme 2.13 A detailed
characterization of this reaction is in progress.
When the observed turnover of tyrosine to 4-HPAA (Fig. S3) was
fit to a two-step irreversible kinetic model, the rate constant for
reactant disappearance was 16.7 0.6 MÀ1 sÀ1 and rate constant
Table 1
HO
X
Y
X
Y
+
NH2
3
NaOCl
HO
Y
X
H2N
COOH
O
H
1a-f
2a-f
4a-f
X
Y
Amino
acid
Oxidation
product
Conversiona (isolated
yield)
Pictet–Spengler
product
Conversion in
phosphatea
Relative rate with
NCSc
(isolated yield)
OH
OH
OH
OH
H
H
Cl
Br
I
1a
1b
1c
1d
1e
2a
2b
2c
2d
2e
>99% (71%)
>99% (83%)
>99% (82%)
>99% (77%)
>99% (61%)
4a
4b
4c
4d
4e
>98% (81%)
>96% (83%)
>95% (82%)
>95% (70%)
>98% (58%)
100%
24 4%
40 8%
43 4%
185 16%
H
HN
1f
2f
61%b (N/A)
4f
>98% (38%)
N/A
H N
2
COOH
a
b
c
Estimated from HPLC peak area at 225 nm for the product and starting material peaks just before extraction.
For this reaction yield estimate, side product peaks were assumed to have the same extinction coefficient as the product.
Initial rates were measured at 1.0 mM concentration of both the aldehyde and dopamine substrates at 23 °C in 50 mM BES at pH 7.0. The reported rates are referenced to
the rate of 2a reaction. Error represents the standard deviation of three trials.