Synthesis of Gallic Acid: Cu2+-Mediated Oxidation of 3-Dehydroshikimic Acid

Article abstract of DOI:10.1021/jo000335z

With the elaboration of high-yielding, high-titer syntheses of 3-dehydroshikimic acid from glucose using recombinant Escherichia coli, oxidation of this hydroaromatic becomes a potential route for synthesis of gallic acid. Conversion of 3-dehydroshikimic acid into gallic acid likely proceeds via initial enolization of an a-hydroxycarbonyl and oxidation of the resulting enediol. 3-Dehydroshikimate enolization in water was catalyzed by inorganic phosphate while Zn²⁺ was used to catalyze enolization in acetic acid. Enediol oxidation employed Cu²⁺ as either the stoichiometric oxidant or as a catalyst in the presence of a cooxidant. Gallic acid was produced in a yield of 36% when 3-dehydroshikimic acid in phosphate-buffered water reacted for 35 h with H₂O₂ and catalytic amounts of CuSO₄. 3-Dehydroshikimate-containing, phosphate-buffered culture supernatants reacted with stoichiometric amounts of CuCO₃Cu(OH)₂ and Cu_x(H_3-xPO₄)₂ to give gallic acid in yields of 51% in 5 h and 43% in 12 h, respectively. Solutions of 3-dehydroshikimic acid in acetic acid reacted with stoichiometric amounts of Cu(OAc)₂ to afford a 74% yield of gallic acid in 36 h. Acetic acid solutions of 3-dehydroshikimic acid could also be oxidized by air using catalytic quantities of Cu(OAc)₂. ZnO accelerated these oxidations leading to a 67% yield of gallic acid in 4 h when an acetic acid solution of 3-dehydroshikimic acid was reacted with O₂ and a catalytic amount of Cu(OAc)₂.

Full text of DOI:10.1021/jo000335z

6
904
J . Org. Chem. 2000, 65, 6904-6909
2
+
Syn th esis of Ga llic Acid : Cu -Med ia ted Oxid a tion of
3
-Deh yd r osh ik im ic Acid
†
,†,‡
Spiros Kambourakis and J . W. Frost*
Department of Chemistry and Department of Chemical Engineering, Michigan State University,
East Lansing, Michigan 48824
frostjw@argus.cem.msu.edu
Received March 8, 2000
With the elaboration of high-yielding, high-titer syntheses of 3-dehydroshikimic acid from glucose
using recombinant Escherichia coli, oxidation of this hydroaromatic becomes a potential route for
synthesis of gallic acid. Conversion of 3-dehydroshikimic acid into gallic acid likely proceeds via
initial enolization of an R-hydroxycarbonyl and oxidation of the resulting enediol. 3-Dehydroshiki-
2
+
mate enolization in water was catalyzed by inorganic phosphate while Zn was used to catalyze
enolization in acetic acid. Enediol oxidation employed Cu2 as either the stoichiometric oxidant or
as a catalyst in the presence of a cooxidant. Gallic acid was produced in a yield of 36% when
+
3
-dehydroshikimic acid in phosphate-buffered water reacted for 35 h with H
amounts of CuSO . 3-Dehydroshikimate-containing, phosphate-buffered culture supernatants
reacted with stoichiometric amounts of CuCO Cu(OH) and Cu to give gallic acid in
yields of 51% in 5 h and 43% in 12 h, respectively. Solutions of 3-dehydroshikimic acid in acetic
acid reacted with stoichiometric amounts of Cu(OAc) to afford a 74% yield of gallic acid in 36 h.
Acetic acid solutions of 3-dehydroshikimic acid could also be oxidized by air using catalytic quantities
of Cu(OAc) . ZnO accelerated these oxidations leading to a 67% yield of gallic acid in 4 h when an
acetic acid solution of 3-dehydroshikimic acid was reacted with O
Cu(OAc)
2 2
O and catalytic
4
3
2
x 4 2
(H3-xPO )
2
2
2
and a catalytic amount of
2
.
3
-Dehydroshikimic acid (DHS, Scheme 1) is a hy-
Sch em e 1
droaromatic intermediate in the common pathway of
aromatic amino acid biosynthesis and a key intermediate
1
2
in the biocatalytic synthesis of vanillin, catechol, and
3
adipic acid. Although DHS is typically present in nature
only in trace quantities, recombinant Escherichia coli
have recently been constructed that are capable of
synthesizing DHS from glucose in 30% yield and at
4
concentrations of 69 g/L. This drastic increase in the
) resulting from Cu2+-mediated oxidation of DHS would
1
availability of DHS provides an opportunity to expand
the range of molecules synthesized via DHS intermediacy
by integrating chemical catalysis with microbe-catalyzed
synthesis of DHS. A prime candidate for such a strategy
is gallic acid, which the pioneering studies of Haslam
demonstrated could be formed when DHS was reacted
with Fehling solution.5 Gallic acid is an important
starting material used in the synthesis of a wide range
of food additives and pharmaceuticals and is currently
obtained from hydrolysis of gallotannins isolated from
gall nuts and tara powder.6
be expected to rapidly enolize to form gallic acid. Initial
strategies focused on the use of inorganic phosphate to
catalyze Cu -mediated oxidation of DHS in aqueous
solutions. Attention then turned to oxidations of DHS in
AcOH solutions using either stoichiometric amounts of
2+
2+
2+
Cu as the oxidant or catalytic amounts of Cu with
serving as the oxidant. Nonoxidative Lewis acids such
O
2
2+
as Zn were discovered to accelerate the rate that acetic
acid solutions of DHS were oxidized by O in the presence
2
2+
of catalytic quantities of Cu . Beyond the utility of
increasing gallic acid availability by elaboration of a
hybrid biocatalytic/chemical synthesis from glucose, oxi-
dation of microbe-synthesized DHS provides a useful
opportunity for comparison of oxidations employing sto-
ichiometric versus catalytic quantities of a metal.
DHS possesses R-hydroxycarbonyl functionality (Scheme
2
+
1
), which is precedented to undergo selective, Cu -
7
,8
mediated oxidation. The R-dicarbonyl species (Scheme
†
Department of Chemistry.
Department of Chemical Engineering.
‡
(1) Li, K.; Frost, J . W. J . Am. Chem. Soc. 1998, 120, 10545.
(2) Draths, K. M.; Frost, J . W. J . Am. Chem. Soc. 1995, 117, 2395.
(3) Draths, K. M.; Frost, J . W. J . Am. Chem. Soc. 1994, 116, 399.
(4) Li, K.; Mikola, M. R.; Draths, K. M.; Worden, R. M.; Frost, J . W.
Resu lts
Isola tion of Micr obe-Syn th esized DHS. Two dif-
ferent methods were effective for isolation of DHS from
Biotechnol. Bioeng. 1999, 64, 61.
5) Haslam, E.; Haworth, R. D.; Knowles, P. F. J . Chem. Soc. 1961,
854.
6) Leston, G. In Kirk-Othmer Encyclopedia of Chemical Technology;
Kroschwitz, J . I., Howe-Grant, M., Eds.; Wiley: New York, 1996; Vol.
9, p 778.
(
1
(7) Organic Syntheses; Wiley: New York, 1963; Collect. Vol. 4, p
838.
(8) (a) Lewbart, M. L.; Mattox, V. R. J . Org. Chem. 1963, 28, 2001.
(b) Weiss, M.; Appel, M. J . Am. Chem. Soc. 1948, 70, 3666.
(
1
1
0.1021/jo000335z CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/16/2000

Synthesis of Gallic Acid
J . Org. Chem., Vol. 65, No. 21, 2000 6905
Ta ble 1. Aqu eou s, P h osp h a te-Ca ta lyzed Rea ction s of
Ta ble 2. Aqu eou s, P h osp h a te-Ca ta lyzed Oxid a tion of
a ,b
2+ a ,b
DHS
DHS Usin g Stoich iom etr ic Am ou n ts of Cu
yield (%)
yield (%)
oxidant Cu (equiv) GA PCA DHS PGL TCBA time (h)
Cu
time
entry
oxidant
Pi (M) (equiv) GA PCA DHS (h)
air
H2O2
H2O2
0
0
0.01
14
22
36
12
10
3
10
20
3
14
2
50
36
35
1
2
3
4
CuCO3Cu(OH)2 1.0
CuCO3Cu(OH)2 0.25
CuX(H3-XPO4)2
CuX(H3-XPO4)2
4
4
2.2
2.2
51
47
43
31
2
4
2
3
5
20
12
23
21
34
1.0
0.25
a
Abbreviations: GA, gallic acid; PCA, protocatechuic acid; DHS,
-dehydroshikimic acid; PGL, pyrogallol; TCBA, tricarballylic acid.
All reactions were run at 40 °C.
3
b
a
Abbreviations: GA, gallic acid; PCA, protocatechuic acid; DHS,
-dehydroshikimic acid. b All reactions were run at 50 °C.
3
Sch em e 2
Ta ble 3. Oxid a tion of DHS in AcOH Usin g Ca ta lytic a n d
Stoich iom etr ic Am ou n ts of Cu 2
+ a
yield (%)
Cu
time
T
entry
catalyst
Cu(OAc)2
(equiv) GA PCA DHS (h) (°C)
1
2
3
4
5
2.2
2.2
0.05
0.1
0.1
53
74 0.7
38
21 4.3
48 2.5
3
12
36
24
18
23
50
40
40
40
40
Cu(OAc)2
Cu(OAc)2/NH4NO3
Cu(OAc)2/H2O2
Cu(OAc)2/Air
2
5
34
30
a
Abbreviations: GA, gallic acid; PCA, protocatechuic acid; DHS,
-dehydroshikimic acid.
3
3 2
using CuCO showed absorbances indicative of
.Cu(OH)
phosphate rather than carbonate as the counteranion.
Given anticipated problems with exchange of a carbonate
for a phosphate counteranion during recycling of the Cu
oxidant, heterogeneous oxidation of DHS in phosphate-
buffered (1 M, pH 6.6) culture supernatant using sto-
microbial culture supernatants. The first method relied
on liquid-liquid extraction while the second method was
based on ion-exchange interactions. For both isolation
methods, cells were removed by centrifugation, and the
resulting cell-free supernatant was acidified. Precipitated
protein was removed by a second centrifugation to afford
the culture supernatant from which DHS was isolated.
Continuous, liquid-liquid extraction using EtOAc re-
sulted in a 95% recovery of the DHS from the culture
supernatant. An 85% recovery of DHS was realized by
2
+
ichiometric concentrations of Cu
Similar yields of gallic acid and protocatechuic acid were
4 2
observed when DHS was oxidized using Cu (H3-xPO )
x 4 2
(H3-xPO ) was explored.
x
(
Table 2, entry 3) relative to DHS oxidation using CuCO
3
.-
Cu(OH)
2
(Table 2, entry 1). Reducing the high concentra-
tions of inorganic phosphate during heterogeneous oxi-
.
dation of DHS using stoichiometric amounts of CuCO
3
-
-
loading culture supernatant on AG1 × 8 (AcO ) anion-
Cu(OH) or Cu led to significantly slower
2
x 4 2
(H3-xPO )
exchange resin followed by elution with glacial AcOH.
Dehydration of the DHS to protocatechuic acid was not
observed, and since AcOH was the solvent employed for
many of the Cu2 -mediated oxidations, eluted DHS could
be directly oxidized.
reaction rates and significant concentrations of unreacted
DHS (Table 2, entry 2 versus entry 1; entry 4 versus
entry 3).
+
Cu 2+-Med ia ted DHS Oxid a tion s In AcOH. High
concentrations of inorganic phosphate were avoided when
DHS purified from culture supernatant was dissolved in
Aqu eou s, Cu +2-Med ia ted DHS Oxid a tion s Ca ta -
lyzed by In or ga n ic P h osp h a te. Reaction of DHS with
air in phosphate-buffered water leads (Table 1) to the
formation (Scheme 2) of gallic acid, protocatechuic acid,
pyrogallol, and tricarballylic acid.9 Reaction of DHS
AcOH/H
geneous reaction conditions using stoichiometric (2.2
equiv) amounts of Cu(OAc) (Table 3, entry 1). Water had
to be included with the AcOH reaction solvent in order
to dissolve the amounts of Cu(OAc) required for sto-
2
O (3/1, v/v) and oxidized at 50 °C under homo-
2
2 2
purified from culture supernatant with H O in phosphate-
2
buffered water gave (Table 1) higher yields of gallic acid
relative to protocatechuic acid and tricarballylic acid with
no observable formation of pyrogallol. Subsequently, it
was discovered (Table 1) that addition of catalytic
ichiometric oxidations. These reaction conditions were
apparently quite sensitive to temperature. Although the
rate of reaction was significantly slower, higher yields
of gallic acid and only trace concentrations of protocat-
echuic acid were observable when DHS dissolved in
amounts (1% mol/mol relative to DHS) of CuSO
4
to the
reactions of DHS with H in phosphate-buffered water
2
O
2
AcOH/H
Table 3, entry 2).
To determine whether catalytic amounts (0.1 equiv) of
Cu(OAc) could be employed during oxidation of DHS to
gallic acid, use of NH NO , H and air as oxidant were
2 2
O (3/1, v/v) was oxidized by Cu(OAc) at 40 °C
eliminated tricarballylic acid formation, increased the
yield of gallic acid, and decreased protocatechuic acid
formation.
(
2
Heterogeneous oxidation of DHS in phosphate-buffered
4
3
2 2
O
(
1.0 M, pH 6.5) culture supernatant using stoichiometric
examined (Table 3, entries 3-5). Higher yields of gallic
acid were obtained when air was the oxidant relative to
use of H O or NH NO as the oxidant. This difference
2 2 4 3
.
3 2
amounts of CuCO Cu(OH) led to faster reaction rates
and higher yields of gallic acid (Table 2, entry 1). IR
analysis of the copper salts recovered after DHS oxidation
in gallic acid yield prompted an appraisal of gallic acid
stability under oxidative conditions. Since the lowest
yields of gallic acid were obtained during Cu2 -catalyzed
(9) Richman, J . E.; Chang, Y.-C.; Kambourakis, S.; Draths, K. M.;
+
Almy, E.; Snell, K. D.; Strasburg, G. M.; Frost, J . W. J . Am. Chem.
Soc. 1996, 118, 11587.
2 2
oxidation using H O , this reaction was examined for

6
906 J . Org. Chem., Vol. 65, No. 21, 2000
Kambourakis and Frost
Ta ble 4. Zn 2 -Ca ta lyzed Oxid a tion of DHS in AcOH
+
+
2 a ,b
Usin g Ca ta lytic Am ou n ts of Cu
yield (%)
DHS
Cu
Zn
time
entry (M) cooxidant (equiv) (equiv) GA PCA DHS (h)
1
2
3
4
5
6
7
8
0.05
0.05
0.05
0.05
0.5
1
air
air
air
air
air
air
O2
O2
trace trace 90
14
16
23
23
7
7
6
4
0.5
0.5
0.1
0.1
0.1
0.1
0.2
1
80
4
5
0.1
0.1
0.05
0.05
0.05
0.05
73
71
70
46
64
67
5
3
14
3.5
3
20
11
8
1
1
9
a
Abbreviations: GA, gallic acid; PCA, protocatechuic acid; DHS,
3
-dehydroshikimic acid. b All reactions were run at 50 °C.
Sch em e 4
F igu r e 1. Rate of reaction at 50 °C of DHS (24 mM) in AcOH
2
+
+2
with 10 mol % Cu and: ×, O
2
/50 mol% Zn ; 4, air/50 mol%
+
2
+2
+2
Zn ; +, air/50 mol% Mn ; ], air/50 mol% Mg ; 2, no added
Lewis acid.
Sch em e 3
products resulting from overoxidation of gallic acid. With
catalytic amounts of Cu(OAc)
2
and reaction conditions
similar to those employed for DHS oxidation, H
2
O
2
oxidation resulted in complete consumption of gallic acid
in 5 h and formation of a 20% yield of aconitic acid and
a 10% yield of hydroxyacetic acid (Scheme 3). Neither
aconitic acid or hydroxyacetic acid were formed when
oxidation of gallic acid by air using catalytic amounts of
of Zn2 equivalents could be reduced by 5-fold without
reducing the yield of gallic acid or increasing reaction
times (Table 4, entry 4 versus entry 3).
+
Further increases in the rate of Cu2+-mediated oxida-
tion by air in the presence of Zn2 were achieved by
increasing the concentration of DHS by 10-fold (Table 4,
entry 5). However, additional increases in the concentra-
tion of DHS resulted in a marked decline in the yield of
gallic acid and an increase in the yield of protocatechuic
acid (Table 4, entry 6). This reduction in the yield of
2
Cu(OAc) was attempted. Approximately 95% of the
+
starting gallic acid remained unreacted after 16 h.
+
2
Cu -Med ia ted DHS Oxid a tion s in AcOH Ca ta -
lyzed by Lew is Acid s. In an attempt to accelerate the
rate of DHS oxidation using catalytic amounts of Cu-
(
OAc)
to the reaction was examined. All oxidations were run
at 50 °C under 1 atm of either air or O with DHS
concentrations at 0.05 M. Relative to the DHS, 0.1 equiv
of Cu(OAc) and 0.5 equivalents of ZnO, MgO, and MnO
2
, the impact of adding nonoxidative Lewis acids
2
product gallic acid was reversed upon use of O
2
instead
of air (Table 4, entry 7). At 1 M concentrations of DHS
with use of O
2
, increasing the Zn2 concentration in-
+
2
were employed. At 19/1 (mol/mol), the ratio of gallic acid/
protocatechuic acid formed during DHS oxidation was
independent of the added Lewis acid. The rate of DHS
oxidation was significantly increased (Figure 1) by all of
the nonoxidative Lewis acids examined with the greatest
creased the rate of the reaction but did not significantly
increase the yield of gallic acid. (Table 4, entry 8 versus
entry 7).
Discu ssion
rate acceleration associated with the presence of Zn2+
An additional acceleration in the overall rate of DHS
.
Mech a n istic Con sid er a tion s. Oxidation of DHS to
form gallic acid is an example of a larger class of
oxidations involving conversion of R-hydroxycarbonyls to
oxidation (Figure 1) was observed when Zn2 oxidations
were run under O instead of air.
DHS was quite stable in AcOH solutions heated to 50
+
2
7
8
R-dicarbonyls using stoichiometric or catalytic amounts
of Cu2 . These oxidations likely begin with complexation
+
C in the absence of added Cu2+ or Zn . As opposed to
inorganic phosphate-buffered solutions of DHS (Table 1,
entry 1), 90% of the DHS in AcOH/H O (85:15, vol/vol)
2+
°
2
+
of the R-hydroxycarbonyl to form a Cu -ketol complex
(
Scheme 4).1
0,11
Subsequent enolization of the R-hydroxy-
2
carbonyl species to an enediol (Scheme 4) is the rate-
solutions exposed to air remained unreacted (Table 4,
entry 1) with formation of only trace amounts of gallic
determining step for the overall oxidation.1
1,12
This is
1
1
2
+
based on the report of a large kinetic isotope effect
and protocatechuic acids. Addition of Zn (0.5 equiva-
associated with Cu2 -mediated oxidation of a dueterium-
+
lents relative to DHS) in lieu of added Cu2+ led to an 80%
yield (Table 4, entry 2) of protocatechuic acid. A 73% yield
of gallic acid was realized (Table 4, entry 3) with Cu2
+
-
(10) Pedersen, K. J . Acta Chem. Scand. 1948, 2, 253.
(
(
11) Wiberg, K. B.; Nigh, W. G. J . Am. Chem. Soc. 1965, 87, 3849.
12) Singh, S. V.; Saxena, O. C.; Singh, M. P. J . Am. Chem. Soc.
mediated reaction of DHS (0.05 M) with air in the
presence of Zn2 (0.5 equiv relative to DHS). The number
+
1970, 92, 537.

Synthesis of Gallic Acid
J . Org. Chem., Vol. 65, No. 21, 2000 6907
labeled ketol, and the first-order dependence12 on base
concentration during base-catalyzed oxidation of R-hy-
droxycarbonyls using Cu2 . R-Hydroxycarbonyl enoliza-
tion thus provides a convenient basis for mechanistic
appraisal of the methods employed to oxidize DHS.
In aqueous solutions, the rate of oxidation of DHS
Sch em e 5
+
2
+
using either stoichiometric amounts of Cu or catalytic
amounts of Cu2 with O
+
functioning as the stoichiometric
2
oxidant depends on the concentration of inorganic phos-
phate. This suggests that inorganic phosphate is cata-
lyzing rate-determining enolization of DHS. Formation
of gallic acid when DHS is dissolved in oxygenated,
phosphate-buffered water has previously been demon-
strated to be a general base-catalyzed process with
9
inorganic phosphate functioning as the base. Phosphate-
catalyzed oxidation of DHS also occurs when H
2 2
O is
2
an 85:15 (v/v) ratio of AcOH/H O. Inhibition of oxidation
employed as the oxidant. Upon addition of even catalytic
has also been observed upon addition of water to steroidal
quantities of Cu2 to the oxidation of DHS using H
+
O ,
2 2
R-hydroxycarbonyls.8 The formation of Cu dimers in
phosphate-buffered, aqueous solutions is uncertain given
the expected disruption of Cu2 dimerization by water
and inorganic phosphate.
a
2+
gallic acid becomes the predominant product. This switch
in the ratio of products is consistent with the Cu2
complexing with the ketol prior to enolization (Scheme
+
+
4
). Resulting stabilization of the enediol as it is formed
Syn th etic Con sid er a tion s. Our experience using
would likely direct the reaction coordinate away from
dehydration (Scheme 4) and attendant formation of
protocatechuic acid.
Cu2 -containing Fehling solution as reported by Haslam
+
5
to oxidize DHS indicated that gallic acid was produced
in low yield concurrent with formation of high levels of
protocatechuic acid contamination. Nonetheless, Haslam’s
In acetic acid solutions containing water, DHS oxida-
2
+
tions employ either a stoichiometric quantity of Cu or
5
a catalytic quantity of Cu2+ in combination with a
observations prompted us to add catalytic quantities of
CuSO
resulting improvements in gallic acid synthesis (Table
) when H was used as the oxidant led to our
4
to phosphate-buffered solutions of DHS. The
nonoxidative Lewis acid catalyst. Use of a stoichiometric
amount of Cu2+ enables the oxidation of DHS to take
1
2 2
O
advantage of this cation’s ability to catalyze enolization
2
+
of the R-hydroxycarbonyl.10 The pronounced catalytic
evaluation of using Cu as both a catalytic and stoichio-
metric oxidant. Observations that inorganic phosphate
in aqueous solutions and ZnO in AcOH solutions could
be used to catalyze the oxidation of DHS to gallic acid
introduces two new variables that can be manipulated
during the use of Cu2 for oxidation of R-hydroxy-
carbonyls.
2
+
roles played by nonoxidative Lewis acids such as Zn
during oxidation of DHS using catalytic amounts of Cu2+
has not previously been reported. Enolization catalyzed
2
+
2+
by Zn may reflect formation of a ternary Cu -enediol-
+
2
+
2+
Zn complex where Zn binds to the diene portion of
2
+
2+
an initially formed Cu -enediol complex. Zn -diene
complexes involving both σ- and π-type interactions are
known to form.13 It is also possible that Zn may initially
complex with the R-hydroxycarbonyl and catalyze eno-
lization resulting in formation of a Zn2+-enediol complex.
To avoid formation of uncomplexed enediol and dehydra-
tion resulting in formation of protocatechuic acid (Scheme
Cu2+-based oxidation of DHS in aqueous solution using
inorganic phosphate catalysis minimizes the number of
manipulations that are required between microbe-
catalyzed synthesis of DHS from glucose and oxidation
of this hydroaromatic. DHS does not need to be isolated
from the culture supernatant. Heterogeneous oxidation
2+
+
2
4
Zn
), Cu would likely need to bind prior to dissociation of
3 2 x 4 2
conditions using CuCO Cu(OH) or Cu (H3-xPO ) gave
2
+
.
comparable yields of gallic acid and similar levels of
protocatechuic acid contamination. Catalytic amounts of
Cu2 could also be employed for oxidation of DHS in
Cu(OAc)
with carboxylic acids with the dimeric complex (Scheme
2
forms both monomeric and dimeric complexes
+
1
4
5
) known to dominate in acetic acid solutions. This
phosphate-buffered, aqueous solution. However, O
not be used as the stoichiometric oxidant. The need to
use H and the resulting modest yields of gallic acid
can likely be attributed to overoxidation of gallic acid by
. Oxidation of DHS in aqueous solution and culture
2
could
2
+
dimeric complex may play an important role in Cu
-
mediated oxidations in AcOH solutions. Displacement of
one of the acetic acid ligands by DHS (Scheme 5) would
allow the two Cu2 ions required for oxidation of the
enediol to the R-dicarbonyl to be in close proximity to one
another and the functionality undergoing oxidation.
Increasing the concentration of water would be expected
to disrupt formation of the dimeric copper complex.8a This
may account for the observed decrease in DHS oxidation
rates as the percentage of water was increased beyond
2 2
O
+
2 2
H O
supernatant using either stoichiometric or catalytic
amounts of Cu2 required high concentrations (1 M) of
inorganic phosphate. Such high-salt reaction conditions
would be problematic from a disposal standpoint if they
were employed on a large scale.
+
Cu2+-based oxidation of DHS to gallic acid in AcOH
solution required the use of either continuous liquid-
liquid extraction or an anion-exchange resin for isolation
of DHS from culture supernatants. Although this in-
creases the number of manipulations required for syn-
thesis of gallic acid from microbe-synthesized DHS, high-
salt reaction conditions are avoided and the yields of
gallic acid using Cu2 -based oxidations in AcOH are
(
13) (a) Budzelaar, P. H. M.; Boersma, J .; van der Kerk, G. J . M.;
Spek, A. L.; Duisenberg, A. J . M. J . Organomet. Chem. 1985, 281, 123.
b) Fischer, B.; Wijkens, P.; Boersma, J .; van Koten, G.; Smeets, W. J .
J .; Spek, A. L.; Budzelaar, P. H. M. J . Organomet. Chem. 1989, 376,
(
2
23.
(14) (a) van Niekerk, J . N.; Schoening, F. R. L. Acta Crystallogr.
1
1
953, 6, 227. (b) Kato, M.; J onassen, H. B.; Fanning, J . C. Chem. Rev.
964, 64, 99.
+

6
908 J . Org. Chem., Vol. 65, No. 21, 2000
Kambourakis and Frost
higher than Cu2+-based oxidations run in aqueous envi-
ronments. An additional advantage of using AcOH for
an ice bath led to the precipitation of DHS as an off-white
powder (10.5 g, 68%).
Resin -Ba sed Isola tion of DHS fr om Cu ltu r e Su p er n a -
ta n t. KL3/pKL4.66A was cultured under fed-batch fermentor
conditions according to ref 4. Fermentation broth was centri-
fuged (3000g for 10 min) to remove cells and the resulting
culture supernatant (900 mL) containing DHS (17 g, 0.11 M)
acidified to pH 2with addition of concentrated HCl. Precipi-
tated protein was removed by centrifugation (13000g for 10
min). The resulting black-colored solution was adjusted to pH
DHS oxidation is that catalytic quantities of Cu(OAc)
can be used with O serving as the cooxidant. Irrespective
of whether stoichiometric amounts of Cu(OAc) or cata-
lytic amounts of Cu(OAc) were used in the presence of
ZnO, the yields of gallic acid resulting from oxidation of
DHS in AcOH solution were approximately the same.
Oxidation of DHS using stoichiometric amounts of
2
2
2
2
5
with addition of NaOH and then applied to a column
Cu(OAc)
echuic acid byproduct formation relative to DHS oxida-
tion using O catalyzed by Cu(OAc) and ZnO.
Oxidation of DHS catalyzed by Cu(OAc) and ZnO is
consonant with the trend toward development of
Cu-catalyzed oxidations of alcohols using O as the
oxidant.15 Oxidation of DHS by O
using catalytic (5 mol
) amounts of Cu(OAc) required significant amounts
10-20 mol %) of ZnO in order to achieve sizable rate
2
did give significantly lower levels of protocat-
-
containing 400 mL of AG-1 × 8 (AcO ). After washing the
column with H O (600 mL) and MeOH (600 mL), decolorized
DHS (14.5 g, 85%) was selectively eluted with glacial AcOH
500 mL). The acetate form of the resin was regenerated by
eluting it with NaOH (0.1 N) followed by elution with HOAc/
O (v/v, 1/1).
Cu SO -Ca ta lyzed Oxid a tion of DHS by H
p h a te-Bu ffer ed H O. DHS (1 g, 5.88 mmol), CuSO
0.0147 g, 0.0588 mmol), NaH PO ‚H O (4.14 g, 30 mmol), and
Na HPO (4.26 g, 30 mmol) were dissolved in 49 mL of H O.
After adjusting the solution to pH 6.6, 11 mL of a 2% solution
of H (6.40 mmol) was added and the reaction stirred at 40
C for 35 h. The reaction solution was then adjusted to pH 2.5
by addition of H SO
2
2
2
2
(
2
H
2
2 2
O
in P h os-
‚H
2
4
2
4
2
O
%
2
(
2
4
2
(
2
4
2
accelerations. Nonetheless, this constitutes a sizable
reduction in total metal requirement relative to oxidation
of DHS using stoichiometric amounts (220 mol %) of
2
O
2
°
Cu(OAc)
2
. Recovery and recycling of metals also needs
, the Cu1 was oxidized by H
and the resulting Cu(OH) precipitated upon basicifica-
2
4
, extracted with EtOAc (4 × 50 mL), and
to be considered. After DHS oxidation using stoichiomet-
concentrated to a solid consisting of gallic acid (36%) and
protocatechuic acid (3%). Chromatography on octadecyl-func-
+
ric amounts of Cu(OAc)
2
2 2
O
2
tionalized silica gel (H O/MeOH, 9:1, v/v adjusted to pH 2.5
2
with AcOH) provided gallic acid (0.31 g, 31%), which was free
of protocatechuic acid contamination.
tion. Dissolving the filtered precipitate in AcOH, concen-
tration, and drying resulted in the recovery of 93% of the
.Cu (OH)
Oxid a tion of DHS by Cu CO
Bu ffer ed H
3
2
in P h osp h a te-
initially used Cu(OAc)
2
. A straightforward method for
2
O. E. coli AB2834/pMF63A was cultured in shake
1
+
high-yielding recovery and recycling of both Cu and
flasks as described in ref 2, and cells were removed from the
culture broth by centrifugation (3000g for 10 min) to provide
a culture supernatant (1 L) containing DHS (7.9 g, 46 mmol).
2
+
Zn was not identified for oxidation of DHS catalyzed
by Cu(OAc) and ZnO. As a result, the 10-fold lowering
in metal equivalents required to oxidize DHS using O
as the oxidant relative to DHS oxidation using stoichio-
metric amounts of Cu(OAc) will not be fully realized
2
KH
2
PO
4
‚H
2 2 4
O (68 g, 500 mmol) and K HPO (87 g, 500 mmol)
2
were then added and the culture supernatant adjusted to pH
.
6
.5. Addition of CuCO
a heterogeneous solution, which was vigorously stirred at 50
C for 5 h. After filtration to remove insoluble copper salts,
the solution was acidified to pH 2.5 with addition of concen-
trated H SO and extracted with EtOAc. Concentration of the
organic layer gave a red solid, which was dissolved in 150 mL
of H O/MeOH (9:1, v/v) and filtered through octadecyl-func-
3 2
Cu(OH) (20.3 g, 92 mmol) resulted in
2
until a successful procedure for total metal recovery and
recycling is elaborated.
°
2
4
Exp er im en ta l Section
2
1
Gen er a l Ch em istr y. H NMR spectra were recorded on a
tionalized silica gel. Concentration of the filtrate to 30 mL
resulted in formation of a precipitate. Filtration and drying
afforded 4.1 g of an off-white solid consisting of gallic acid
(51%) and protocatechuic acid (2%).
3
00 MHz spectrometer. Chemical shifts were reported in parts
per million (ppm) downfield from internal sodium 3-(trimeth-
ylsilyl)propionate-2,2,3,3-d O was the
4
(TSP, δ ) 0.00) when D
2
solvent. TSP was purchased from Lancaster. Octadecyl-func-
tionalized silica gel was purchased from Aldrich and activated
Oxid a tion of DHS by Cu
Bu ffer ed H O. E. coli AB2834/pMF63A was cultured in shake
flasks as described in ref 2 and cells removed from the culture
broth by centrifugation. KH PO (13.6 g) was added to a
x 4 2
(H3-xP O ) in P h osp h a te-
2
-
prior to use by elution with MeOH. AG-1 × 8 (AcO ) anion-
+
exchange resin and Dowex 50 (H ) cation-exchange resin were
2
4
purchased from Bio-Rad.
portion (100 mL) of this culture supernatant containing DHS
(0.79 g, 4.6 mmol) and the solution adjusted to pH 6.6 by
Liqu id -Liqu id Extr a ction of DHS fr om Cu ltu r e Su -
p er n a ta n t. KL3/pKL4.79B was cultured under fed-batch
fermentor conditions according to ref 4. Fermentation broth
was centrifuged (3000g for 10 min) to remove cells and the
resulting culture supernatant (500 mL) containing DHS (15.5
g, 0.18 M) acidified to pH 2 with addition of concentrated HCl.
Precipitated protein was removed by centrifugation (13000g
for 10 min). The resulting black solution was then stirred in a
continuous liquid-liquid extraction apparatus at a rate to
create a translucent colloidal suspension of EtOAc in the
aqueous culture supernatant while allowing the colloidal
suspension to separate into a clear organic phase at the top of
the extraction cylinder. DHS-containing EtOAc was replaced
with fresh EtOAc (400 mL) at 2, 5, 8 and 12 h. After the 2 L
of DHS-contaning EtOAc was filtered through 10-15 g of
Darco G-60 (100 mesh) activated charcoal and dried over
addition of NaOH. The Cu
by dissolving CuSO (1.6 g, 10 mmol in 50 mL of H
by addition of NaHPO (2.2 g, 0.016 mol). The blue-colored
x
(H3-xPO
4
)
2
oxidant was prepared
4
2
O followed
4
precipitate that formed was recovered by filtration and then
added to the reaction solution. This heterogeneous solution
was stirred under N
remove the insoluble copper salts, the reaction solution was
acidified to pH 2.2 with addition of H SO and extracted with
2
at 50 °C for 12 h. After filtration to
2
4
EtOAc (3 × 60 mL). Concentration of the organic layer afforded
0.35 g of a lightly yellow-colored solid consisting of gallic acid
(43%) and protocatechuic acid (2%).
Oxid a tion of DHS by Cu (OAc)
DHS (5.4 g, 31.4 mmol) and Cu(OAc)
400 mL of AcOH/H O (3:1, v/v) was stirred at 40 °C under N
2
in AcOH. A solution of
2
(13.5 g, 67.5 mmol) in
2
2
for 36 h. After filtration to remove the insoluble copper salts,
the reaction solution was concentrated almost to dryness and
MgSO
4
, the yellow-colored solution was concentrated to a
volume of 50-60 mL. Chilling this concentrated solution in
2
then dissolved in H O (300 mL). This aqueous solution was
extracted with EtOAc (4 × 100 mL) and the organic layer
concentrated. The resulting brown solid consisting of gallic acid
(15) (a) Semmelhack, M. F.; Schmid, C. R.; Cort e´ z, D. A.; Chou, C.
(74%) and protocatechuic acid (0.7%) was dissolved in H
2
O/
S. J . Am. Chem. Soc. 1984, 106, 3374. (b) Mark o´ , I. E.; Giles, P. R.;
Tsukazaki, M.; Brown, S. M.; Urch, C. J . Science 1996, 274, 2044.
MeOH (9:1, v/v) and decolorized by filtration through octade-

Synthesis of Gallic Acid
J . Org. Chem., Vol. 65, No. 21, 2000 6909
plished by dissolving the gallic acid in 300 mL H O/MeOH (9:
1, v/v) followed by filtration through octadecyl-functionalized
silica gel. Upon concentration of the decolorized solution to
30 mL a solid formed which was filtered and dried to give 5.1
(49%) of gallic acid as a white solid.
cyl-functionalized silica gel. During concentration of the
filtrate, gallic acid precipitated and was recovered as an off-
white solid (3.3 g, 62%). Recovery of copper salts and recycling
of these salts back to Cu(OAc) began with combining the
2
filtered copper salts with the aqueous layer after removal of
gallic acid by EtOAc extraction. This mixture was reacted
2
Zn 2+, Mg2+, Mn 2+ Acceler a tion of Cu 2+-Ca ta lyzed DHS
overnight with 10 mL of 30% H
Addition of solid NaOH to give a solution of pH 13 resulted in
precipitation of Cu(OH) , which was recovered by filtration.
Stirring the Cu(OH) in AcOH for 2 h led to a homogeneous
solution that after concentration and drying gave 12.5 g of Cu-
OAc) (93%) indistinguishable by IR from the Cu(OAc)
initially used to oxidize DHS in the culture supernatant.
Zn O, Cu (OAc) -Ca ta lyzed Oxid a tion of DHS by O
AcOH. Oxygen was bubbled through a stirred solution of DHS
2 2
O at room temperature.
Oxid a tion s. DHS (0.1 g, 0.58 mmol) and Cu(OAc)
2
(0.012 g,
0
.058 mmol) were dissolved in 24 mL of AcOH/H O (85:15, v/v)
2
2
containing ZnO (0.23 g, 0.29 mmol), MgO (0.012 g, 0.23 mmol),
or MnO (0.29 g, 0.29 mmol). Reaction progress at 50 °C under
either air or O was determined by withdrawing 0.1 mL
2
2
(
2
2
aliquots from the reaction solution at timed intervals, diluting
with AcOH, and measuring the absorbance at 298 nm for each
time interval. To verify that the absorbance at 298 nm
corresponded to the concentration of gallic and protocatechuic
acids, 1 mL aliquots of the reaction solution at various timed
intervals were filtered through a pipet containing 0.3 mL of
2
2
in
(
(
°
10.5 g, 61.0 mmol), Cu(OAc)
0.995 g, 12.2 mmol) in 61 mL of AcOH/H
C for 4 h. Gallic acid (67%) and protocatechuic acid (3%) were
2
(0.61 g, 3.05 mmol), and ZnO
2
O (85:15, v/v) at 50
+
Dowex 50 (H ) to remove metals. These solutions were
the only products formed. The reaction solution was concen-
trated to 30 mL, chilled in an ice bath, and the black solid
that formed collected by filtration. This solid was dissolved in
2
concentrated, dissolved in D O, a known amount of TSP added,
and the concentration of gallic acid and protocatechuic acids
1
determined by H NMR.
2
H O (250 mL) and the resulting solution extracted with EtOAc
(
4 × 150 mL). The combined organic layers were concentrated
to 200 mL and chilled in an ice bath. Petroleum ether (200
mL) was then added. A precipitate formed which was filtered
and dried affording 5.5 g (53%) of gallic acid as a slightly
yellow, off-white solid. Further decolorization was accom-
Ack n ow led gm en t. Research was supported by a
contract from the Environmental Protection Agency.
J O000335Z