Biochemical and Biophysical Research Communications 394 (2010) 279–284

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications

journal homepage: www.elsevier.com/locate/ybbrc

Enzymatic Kolbe–Schmitt reaction to form salicylic acid from phenol:

Enzymatic characterization and gene identification of a novel enzyme,
Trichosporon moniliiforme salicylic acid decarboxylase
Kohtaro Kirimura *, Hiroaki Gunji, Rumiko Wakayama, Takasumi Hattori, Yoshitaka Ishii
Department of Applied Chemistry, Faculty of Science and Engineering, Waseda University, 3-4-1 Ohkubo, Shinjuku-ku 169-8555, Tokyo, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Salicylic acid decarboxylase (Sdc) can produce salicylic acid from phenol; it was found in the yeast
Received 4 February 2010 Trichosporon moniliiforme WU-0401 and was for the first time enzymatically characterized, with the
Available online 25 February 2010 sdc gene heterologously expressed. Sdc catalyzed both reactions: decarboxylation of salicylic acid to
phenol and the carboxylation of phenol to form salicylic acid without any byproducts. Both reactions
Keywords: were detected without the addition of any cofactors and occurred even in the presence of oxygen,
Salicylic acid decarboxylase suggesting that this Sdc is reversible, nonoxidative, and oxygen insensitive. Therefore, it is readily
Reversible decarboxylase
applicable in the selective production of salicylic acid from phenol, the enzymatic Kolbe–Schmitt
Trichosporon moniliiforme
Enzymatic Kolbe–Schmitt reaction
reaction. The deduced amino acid sequence of the gene, sdc, encoding Sdc comprises 350 amino acid
Selective production residues corresponding to a 40-kDa protein. The recombinant Escherichia coli BL21(DE3) expressing sdc
converted phenol to salicylic acid with a 27% (mol/mol) yield at 30 °C for 9 h.
Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction
The Kolbe–Schmitt reaction is well known as the carboxylation
reaction of alkali metal phenoxides, producing hydroxybenzoic
acid under conditions of high temperature and pressure [1]. Sev-
eral phenolic compounds such as salicylic acid, c-resorcylic acid,
and 4-aminosalicylic acid were produced through this reaction in
industrial processes and used as intermediates for medicines, her-
bicides, and industrial chemicals [2]. However, such a production
also generates a large amount of byproducts, which then need to
be separated [1]. Recently, the biological application of the Kol-
be–Schmitt reaction has been focused because selective and eco-
logical carboxylation of phenolic compounds might be possible
under environmentally benign conditions. Until now, several
nonoxidative decarboxylases, because of their reversibility, have
been found to catalyze the carboxylation of aromatics into aro-
matic acids; these include 4-hydroxybenzate decarboxylase [3],
3,4-dihydroxybenzate decarboxylase [4,5], c-resorcylate decarbox-
ylase [6–8], vanillate/4-hydroxybenzoate decarboxylase [9], pyr-
rol-2-carboxylate decarboxylase [10], and indole-3-carboxylate
decarboxylase [11]. However, despite the fact that salicylic acid
is a well-known aromatic carboxylic acid used as a precursor for
acetylsalicylic acid and its wide use as a nonsteroidal anti-inflam-
matory drug, analgesic aspirin [2], there is no report on the enzyme
catalyzing the carboxylation of phenol to form salicylic acid.
We isolated Trichosporon moniliiforme WU-0401 as a salicylic
acid-degrading yeast [12]. In the course of this degradation path-
way, we found a salicylic acid decarboxylase that reversibly cata-
lyzes the regioselective carboxylation of phenol to form salicylic
acid (Fig. 1). In this paper, we report the characterization of this
novel reversible and nonoxidative salicylic acid decarboxylase,
Sdc, from T. moniliiforme WU-0401 and its purification, molecular
characterization, and gene identification. This is the first report
on reversible salicylic acid decarboxylase applicable to the selec-
tive production of salicylic acid from phenol, the enzymatic
Kolbe–Schmitt reaction.
2. Materials and methods
2.1. Bacterial strains, plasmids, and cultivation
T. moniliiforme WU-0401 [12] was used as a source for purifica-
tion of the reversible and nonoxidative salicylic acid decarboxylase
and cloning of the sdc gene. Strain WU-0401 was cultivated as
described previously [12].
2.2. Whole cell reaction

* Corresponding author. Fax: +81 3 3232 3889. Whole cell suspension was prepared as described previously
E-mail address: kkohtaro@waseda.jp (K. Kirimura). [12]. For measuring carboxylase activity, 50 ll of 0.5 M phenol

0006-291X/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.bbrc.2010.02.154
280 K. Kirimura et al. / Biochemical and Biophysical Research Communications 394 (2010) 279–284
Selective
production
OH OH
CO2
COOH
CO2
Phenol Salicylic acid
Fig. 1. Reversible enzymatic conversion of salicylic acid and phenol.
solution was added to 1 ml of the whole cell suspension containing
150 mM KHCO3 (pH 9.6) in a 2-ml microcentrifuge tube. The reac-
tion was allowed to proceed for 24 h at 30 °C, and 50 ll of 12 M HCl
was added to stop the reaction.
2.3. Purification of salicylic acid decarboxylase
Cell-free extract was prepared as described by previously [12].
It was applied to a Toyopearl-DEAE 650S column (Tosoh, Tokyo,
Japan) equilibrated with 25 mM Tris–HCl buffer (pH 8.0) and
washed with 90 ml of the same buffer. Elution was carried out with
a continuous linear gradient of 0 to 1.0 M KCl in 150 ml of the same
buffer at a flow rate of 2 ml/min. The active fractions (0.14–0.15 M
KCl) showing decarboxylation of salicylic acid to form phenol were
collected and applied to a Q-Sepharose FF column (GE Healthcare,
Buckinghamshire, England) equilibrated with 25 mM Tris–HCl buf-
fer (pH 8.0) and washed with 90 ml of the same buffer. Elution was
carried out with a continuous linear gradient of 0–1.0 M KCl in
150 ml of the same buffer at a flow rate of 2 ml/min. The active
fractions (0.22–0.26 M KCl) were concentrated by ultrafiltration
and applied to a Superdex 200 HR 10/30 column (GE Healthcare)
equilibrated with 25 mM Tris–HCl buffer (pH 8.0) and washed with
90 ml of the same buffer. Elution was carried out with the same
buffer at a flow rate of 0.5 ml/min. The active fractions (fractions
containing proteins with a molecular mass of approximately
140 kDa) were collected for applying to a Bio-Scale CHT-1 column
(BIO-RAD, CA, USA) equilibrated with 10 mM K2HPO4–KH2PO4
buffer (pH 7.0) and washed with 90 ml of the same buffer. Elution
was carried out with a continuous linear gradient of 0–500 mM
K2HPO4–KH2PO4 in 150 ml of the same buffer at a flow rate of
2 ml/min. The active fractions (0.09–0.10 mM K2HPO4–KH2PO4)
were collected and then concentrated by ultrafiltration. Protein
assays, molecular mass and concentration, were carried out as
described previously [6]. The partial amino acid sequence of the
reversible salicylic acid decarboxylase was determined by Shimadzu
Biotech (Ibaraki, Japan).
2.4. Enzyme assays
For measurement of decarboxylase activity, the following stan-
dard reaction conditions were used: the mixture contained 10 lg
of protein and 30 mM salicylic acid in 100 mM MES–NaOH buffer
(pH 5.5) with a final volume of 1 ml in a 1.5-ml microcentrifuge
tube and incubation with shaking at 40 °C for 1 h. For measure-
ment of carboxylase activity, the following standard reaction con-
ditions were used: the mixture contained 100 lg of protein,
20 mM phenol, and 2.5 M KHCO3 in 100 mM K2HPO4–KH2PO4 buf-
fer (pH 7.0) with a final volume of 1 ml in a 1.5-ml microcentrifuge
tube and incubation with shaking at 30 °C for 1 h. Both the reac-
tions, decarboxylation and carboxylation, were stopped by heating
at 60 °C for 1 h. When substrates other than salicylic acid or phenol
No production
OH OH
COOH
COOH
m-Hydroxybenzoic acid p-Hydroxybenzoic acid
were used, the same standard assay conditions were used, unless
otherwise indicated. The amounts of salicylic acid, phenol, and
other substrates were determined by HPLC. One unit (U) of enzyme
activity is defined as the amount of enzyme catalyzing the forma-
tion of 1 lmol of product per minute.
2.5. Reverse transcriptase-polymerase chain reaction and 50 - and 30 -
RACE
Total RNA was isolated from powdered cells of strain WU-0401,
using Sepasol-RNA I Super (Nacalai tesque, Kyoto, Japan), followed
by Cloned DNase I (Takara Bio, Shiga, Japan) treatment with Cloned
RNase Inhibitor (Takara Bio). Reverse transcriptase-polymerase
chain reaction (RT-PCR) was carried out using the ReverTra-Plus-™
kit (TOYOBO, Osaka, Japan). The primers used for RT-PCR were de-
signed from the partial amino acid sequences of purified enzyme:
50 -CTCTA CATCG CCCCC AA-30 was used as the sense primer and
50 -GTAGA TGGTG TAGCC GATGC C-30 , the anti-sense primer.
The primers used for 50 - and 30 -RACE were designed from the
amplified fragments thorough RT-PCR: 50 -GAGAT TCTCA ACCCG
TGCGG CAA-30 was the sense primer and 50 -TTGCC GCACG GGTTG
AGAAT CTC-30 , the anti-sense primer. Total RNA was converted
into mRNA by using a Poly(A)+ Isolation Kit (NIPPON GENE, Tokyo,
Japan), and the mRNA generated was used as a template for 50 - and
30 -RACE. Rapid amplification of 50 - and 30 -cDNA ends was per-
formed using a BD SMART™ RACE cDNA Amplification Kit (Takara
Bio).
2.6. Expression of sdc in Escherichia coli
Based on the nucleotide sequence of ORF1 encoding the salicylic
acid decarboxylase, the primers used for sdc amplification were as
follows: 50 -TTAAA ACACA TCCAT CCATA TGCG-30 [the NdeI restric-
tion site is underlined] was the sense primer and 50 -TTCAT TACTA
AGCTT CTAAG CCTCC GAG-30 [the HindIII restriction site is under-
lined], the anti-sense primer. The gene sdc was amplified by PCR
from the cDNA of strain WU-0401 and then inserted into pET21-
d(+) (Novagen, WI, USA) to generate pSDC. Recombinant E. coli
BL21 (DE3) carrying pSDC was cultivated at 30 °C for 18 h in
500-ml Erlenmeyer flasks containing 100 ml LB medium supple-
mented with 100 lg ampicillin/ml and 0.1 mM isopropyl-b-D-thio-
galactopyranoside with reciprocal shaking at 120 strokes/min.
2.7. Biosynthesis of salicylic acid by whole cell reaction using
recombinant E. coli cells
The recombinant E. coli cells cultivated were harvested by cen-
trifugation at 6000g for 10 min at 4 °C and washed twice with
50 mM K2HPO4–KH2PO4 buffer (pH 7.0). A reaction mixture
containing the washed cells (26 g dry-cells/l), 5 mM phenol, and
150 mM KHCO3 in 50 mM K2HPO4–KH2PO4 buffer (pH 7.0) in
 K. Kirimura et al. / Biochemical and Biophysical Research Communications 394 (2010) 279–284
1 ml in a 1.5-ml microcentrifuge tube was incubated at 30 °C for
16 h with reciprocal shaking at 120 strokes/min. The reaction
was stopped by adding 10 ll 12 M HCl; the mixture was then cen-
trifuged at 6000g for 10 min at 4 °C and the supernatant filtered
using a 0.2-lm PTFE membrane.
2.8. Nucleotide sequence accession number
The nucleotide sequence discussed in this paper is available
under the DDBJ under accession number DM040453 as the revers-
ible salicylic acid decarboxylasegene of strain WU-0401, sdc.
2.9. Analytical methods
Thin-layer chromatography (TLC) and high-performance liquid
chromatography (HPLC) were carried out as described previously
[12].
3. Results
3.1. Trichosporon moniliiforme WU-0401 converts phenol into
salicylic acid
TLC analysis and two-dimensional HMBC spectrum (Supple-
mentary Fig. 1) revealed that a whole cell reaction of strain
WU-0401 mixed with phenol and KHCO3 produced salicylic acid.
Whole cells of strain WU-0401 cultivated in LB medium did not
show salicylic acid production activity. In contrast, whole cells of
strain WU-0401 cultivated in LB medium supplemented with
20 mM salicylic acid showed the production activity (data not
shown). These results suggested that enzymes responsible for deg-
radation and production of salicylic acid might be inducible. A cell-
free extract of strain WU-0401 also showed salicylic acid produc-
tion from phenol with KHCO3 (data not shown).
In our previous study, salicylic acid degradation activity was
detected in strains closely related to the species of T. moniliiforme,
i.e., T. moniliiforme NBRC 1527, T. cutaneum NBRC 1198T, and T.
asteroides NBRC 0173 [12]. Therefore, whole cells of these three
strains were prepared to examine whether salicylic acid would
be produced from phenol with KHCO3. The whole cells of three
strains also showed salicylic acid production activity (data not
shown). Therefore, salicylic acid decarboxylase reversibly catalyz-
ing the carboxylation of phenol, i.e., production of salicylic acid,
might be widely distributed in strains closely related to the species
of T. moniliiforme.
3.2. Purification of the salicylic acid decarboxylase
Based on the decarboxylase activity converting salicylic acid
into phenol, the salicylic acid decarboxylase was purified to homo-
geneity from cell-free extracts of strain WU-0401. The enzyme was
purified 20.8-fold with a yield of 5.5%, and the specific activity of
the purified enzyme for salicylic acid decarboxylation was
0.47 U/mg (Table 1). The purified enzyme produced a 40 kDa single
band on the gel by SDS–PAGE (Supplementary Fig. 2). The native
molecular mass of the enzyme was found to be 140 kDa by gel
Table 1
Purification of the reversible salicylic acid decarboxylase from T. moniliiforme WU-0401.
Step Total protein (mg) Specific activity (U/mg)
Cell-free extract 18.9 0.023
Toyopearl-DEAE 1.79 0.069
Q-Sepharose FF 0.80 0.13
Superdex 200 0.35 0.23
Bio-scale CHT-1 0.050 0.47
a
U, One unit (U) was defined as the amount of enzyme producing one micromole of phenol from salicylic acid per minute.
281
filtration (Supplementary Fig. 2), suggesting that the enzyme has
a homotetrameric structure.
3.3. Properties on decarboxylation of the salicylic acid decarboxylase
The optimal pH and temperature for decarboxylase activity
were 5.5 and 40 °C, respectively (Fig. 2). Decarboxylase activity
was stable up to 40 °C and retained 40% of its activity after the
treatment of heating at 50 °C for 1 h. As shown in Table 2, the
decarboxylase activity was inhibited by AgNO3, HgCl2, p-chloro-
mercuribenzoic acid as a sulfhydryl group inhibitor, and NiCl2.
Addition of EDTA, pyridoxal 50 -phosphate, avidin, biotin, NADH,
and NADPH to the reaction mixture had no or slight effect on the
enzyme activity. Hydroxylamine, a known inhibitor of pyridoxal
50 -phosphate-dependent decarboxylases, also had no effect. These
results suggest that this enzyme is a nonoxidative decarboxylase
that requires no cofactor such as pyridoxal 50 -phosphate, NADH,
and NADPH.
The enzyme catalyzed the decarboxylation of salicylic acid into
stoichiometric amounts of phenol with a specific activity of 0.47 U/
mg. The Km, Vmax, and kcat values at 40 °C and pH 5.5 were 1.08 mM,
1.22  102 mM/min, and 2.03  10 min1, respectively, for sali-
cylic acid. The enzyme also catalyzed the decarboxylation of
b-resorcylic acid (2,4-dihydroxybenzoic acid) into resorcinol (1,3-
dihydroxybenzene), c-resorcylic acid (2,6-dihydroxybenzoic acid)
into resorcinol, 2,3-dihydroxybenzoic acid into catechol (1,2-
dihydroxybenzene), and 4-aminosalicylic acid into 3-aminophenol.
The enzyme did not catalyze the decarboxylation of 3-hydroxyben-
zoic acid, 4-hydroxybenzoic acid, protocatechuic acid (3,4-dihy-
droxybenzoic acid), a-resorcylic acid (3,5-dihydroxybenzoic acid),
3-methylsalicylic acid, 4-methylsalicylic acid, and vanillic acid
(4-hydroxy-3-methoxybenzoic acid). These results indicate that
the substrate recognition of the enzyme for decarboxylation seems
to depend strictly on hydroxybenzoic acid with two neighboring
hydroxyl and carboxyl groups in the active center of the enzyme.
3.4. Properties on carboxylation of the reversible salicylic acid
decarboxylase
The purified enzyme also catalyzed the carboxylation of phenol
into salicylic acid, indicating that this enzyme is a reversible sali-
cylic acid decarboxylase. The carboxylase activity showed a sub-
strate saturation dependence from HCO 3 with an optimal HCO3

concentration above 2.5 M (Fig. 3). The optimal substrate concen-
tration of phenol and optimal temperature for enzyme activity
were 30 mM and 30 °C, respectively (Fig. 3). Carboxylase activity
was stable up to 30 °C and retained 60% of its activity after heating
at 40 °C for 1 h.
Effects of various metal ions and chemical reagents on carbox-
ylase activities were investigated (Table 2). Carboxylase activity
was inhibited by HgCl2, NiCl2, CuCl2, AgNO3, p-chloromercuriben-
zoic acid as a sulfhydryl group inhibitor, CaCl2, and diethyl pyro-
carbonate as a histidine residue-specific inhibitor. Enzyme
activity was activated by FeCl2.
The purified enzyme catalyzed the regioselective carboxylation
of phenol into stoichiometric amounts of salicylic acid. The Km,
Total activity (U)a Purification (fold) Yield (%)
0.43 1.00 100
0.12 3.05 29
0.11 5.89 25
0.082 10.4 19
0.023 20.8 5.5
282 K. Kirimura et al. / Biochemical and Biophysical Research Communications 394 (2010) 279–284

A
Table 2
100
Effects of various compounds on decarboxylase and carboxylase activities of the
reversible salicylic acid decarboxylase.

Compound Relative activity (%)

Relative activity (%)

Decarboxylation Carboxylation
None 100 100
50 NaCl 97 80
KCl 92 81
MgCl2 106 90
MnCl2 105 83
CaCl2 84 60
HgCl2 0 6
ZnCl2 111 84
0 CuCl2 86 30
3 4 5 6 7 8 9 10 11 FeCl2 93 149
pH NiCl2 54 15
AgNO3 0 40
EDTA 95 104
B 100 N-Ethylmaleimide 94 72
Iodoacetamide 97 82
p-Chloromercuribenzoic acid 18 51
Relative activity (%)

Diethyl pyrocarbonate 79 63
Pyridoxal 50 -phosphate 88 93
Avidin 95 103
Biotin 91 82
50 NADH 90 86
NADPH 83 82
Hydroxylamine 91 77

The carboxylation and decarboxylation reactions were performed under the stan-
dard conditions supplemented with the compounds tested at 1 mM except for p-
chloromercuribenzoic acid (0.2 mM) and avidin (2 U/ml).
0
0 10 20 30 40 50 60 70 80
Temperature (°C)
agents and/or anaerobic conditions during the handling [3–
5,10,11]. We examined the effects of O2 on the decarboxylase
C 100
and carboxylase activity (Supplementary Fig. 3). There was no ef-
fect on the activity when the conditions were substituted with
an Ar atmosphere, as compared with the standard condition in a
Relative activity (%)

shaking reaction without the exchange of the gas in the head space,
in other words without removal of O2. These results indicate that
50 the purified enzyme is insensitive to O2 for reactions.

3.6. The reversible salicylic acid decarboxylase gene

Two partial amino acid sequences of the purified enzyme were

0 Based on these amino acid sequences and codon usage of T. monil-
0 10 20 30 40 50 60 70 80 90
Temperature (°C)
Fig. 2. Effects of pH and temperature on decarboxylase activity. (A) Effects of pH on
activity. The following buffers (100 mM) were used: open diamond, citrate–NaOH
(pH 4.0–6.0); open square, MES–NaOH (pH 5.5–6.5); open triangle, K2HPO4–KH2PO4
(pH 6.5–8.0); open circle, Tris–HCl (pH 7.0–9.0); closed diamond, H3BO3–NaOH (pH
8.0–10.0). (B) Effects of temperature on activity. (C) Effects of temperature on
stability. After the purified enzyme was preincubated at the indicated temperature
for 1 h, the remaining activity was measured under standard conditions. The
remaining activity is expressed relative to the activity, taken as 100%, when the
reaction mixture was not preincubated.
Vmax, and kcat values at 30 °C were 1.23  102 mM, 1.23  10–
2
mM/min, and 6.64  10 min1, respectively, for phenol. The en-
zyme also catalyzed the carboxylation of resorcinol into b- and
c-resorcylic acid, catechol into 2,3-dihydroxybenzoic acid, and
3-aminophenol into 4-aminosalicylic acid. The enzyme did not
catalyze the carboxylation of cresol.
3.5. Effects of oxygen on the reversible salicylic acid decarboxylase
Almost all of the already-known reversible aromatic decarbox-
ylases were sensitive to O2 and required the addition of reducing
determined to be VKAELYIAPN and VGIGYTIYLIY, respectively.
iiforme, a complete 1035-bp open reading frame (ORF1) was deter-
mined by RT-PCR, and 50 - and 30 -RACE. The molecular size and the
partial amino acid sequence of the protein encoded by ORF1 were
identical to those of purified enzyme from strain WU-0401.
The ORF1 was amplified by PCR and subcloning in pET21-d(+) to
generate a plasmid pSDC. A cell-free extract of E. coli BL21(DE3)
harboring pSDC showed both decarboxylase activity toward sali-
cylic acid and carboxylase activity toward phenol; ORF1 was then
designated as sdc encoding the reversible salicylic acid decarboxyl-
ase of strain WU-0401. Salicylic acid production activity of the cell-
free extract of E. coli BL21(DE3) harboring pSDC was 0.08 U/mg and
four times higher than that of the original strain WU-0401. How-
ever, overexpression of the sdc gene in E. coli was clearly detected
on SDS–PAGE.
The deduced amino acid sequence of sdc shows 50% and 40%
identities to the 2,3-dihydroxybenzoic acid decarboxylase (DHBD,
AP007151) of Aspergillus oryzae [14] and a c-resorcylic acid decar-
boxylase (Rdc, AB185333) of Rhizobium radiobacter WU-0108 [6],
respectively. On the other hand, as to the primary structure of
reversible and nonoxidative aromatic decarboxylase, 4-hydroxy-
benzoate decarboxylase (Ohb1, AAD50377) from Clostridium
hydroxybenzoicum JW/Z-1T was also reported [3,5], but the
 K. Kirimura et al. / Biochemical and Biophysical Research Communications 394 (2010) 279–284 283

A 100 B 100

Relative activity (%)

Relative activity (%)

50 50

0 0
0 1 2 3 0 10 20 30 40 50 60 70 80 90
KHCO3 (M) Phenol (mM)

C 100 D 100

Relative activity (%)

Relative activity (%)

50 50

0 0
0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 80 0 10 20 30 40 50 60 70 80 90
Temperature (°C) Temperature (°C)
Fig. 3. Effects of substrate concentration and temperature on carboxylase activity. (A) Effects of KHCO3 concentration on activity. (B) Effects of phenol concentration on
activity. (C) Effects of temperature on activity. (D) Effects of temperature on stability. After the purified enzyme was preincubated at the indicated temperature for 1 h, the
remaining activity was measured under standard conditions. The remaining activity is expressed relative to the activity, taken as 100%, when the reaction mixture was not
preincubated.

deduced amino acid sequence of sdc shows no homology to that of

Ohb1. 40

3.7. Selective production of salicylic acid, enzymatic Kolbe–Schmitt

Concentration (mM)

reaction toward phenol, by whole cell reaction 30

To optimize the conditions for the whole cell reaction using the
recombinant E. coli cells harboring pSDC, we examined the reaction
20
conditions by changing the standard temperature, KHCO3 concen-
tration, and reaction time (details not shown). The optimal condi-
10
tions were 30 °C, a phenol concentration of 20 mM, KHCO3 at 3 M
(saturated concentration), and a reaction time of 16 h. The time
course of salicylic acid production by the whole cell reaction using 0
the recombinant E. coli cells harboring pSDC is shown in Fig. 4. The 0 5 10 15 20 25
concentration of salicylic acid reached its maximum, 10.6 mM,
Time (h)
after 9 h and then remained constant. Under these conditions, sal-
icylic acid was the only product in the reaction mixture. These re- Fig. 4. Salicylic acid production by whole cell reaction using recombinant E. coli
sults indicate that recombinant E. coli cells expressing sdc must be BL21 (DE3)/pSDC under optimal conditions. Symbols: closed circle, salicylic acid;
applicable as efficient and convenient biocatalysts to the selective open triangle, phenol.

production of salicylic acid from phenol, enzymatic Kolbe–Schmitt

reaction toward phenol.
4. Discussion
As to the salicylate decarboxylating enzyme, salicylate hydrox-
ylase (salicylate 1-monooxygenase, EC 1.14.13.1), which catalyzes
oxidative decarboxylation of salicylate into catechol, was reported
[13]. However, there has been no report concerning nonoxidative
type decarboxylase reacting toward salicylic acid to form phenol.
In this paper, we report that for the first time nonoxidative sali-
cylic acid decarboxylase, Sdc, was purified from strain WU-
0401, and that it catalyzed reversibly the conversion of phenol
to salicylic acid. As to the reversible and nonoxidative decar-
boxylase, Ohb1 and 3,4-hydroxybenzoate decarboxylase from
C. hydroxybenzoicum JW/Z-1T [3–5], vanillate/4-hydroxybenzoate
284 K. Kirimura et al. / Biochemical and Biophysical Research Communications 394 (2010) 279–284
decarboxylase from Bacillus subtilis ATCC 6051 [9], pyrrole-2-car-
boxylate decarboxylase from Bacillus megaterium PYR2910 [10],
indole-3-carboxylate decarboxylase from Arthrobacter nicotianae
FI1612 [11], and c-resorcylic acid decarboxylase from R. radiob-
acter WU-0108, Rhizobium sp. MTP-10005, and Agrobacterium
tumefaciens IAM12408 have been previously reported [6–8].
These already-known reversible decarboxylases, except c-resor-
cylic acid decarboxylase and vanillate/4-hydroxybenzoate decar-
boxylase, were sensitive to O2, but Sdc was insensitive to O2.
Moreover, under both conditions in the presence and absence of
O2, the decarboxylase and carboxylase activities of Sdc were not
affected, suggesting that the handling of the recombinant E. coli
cells expressing sdc and the enzyme would be readily applicable
to practical salicylic acid production.
Several genes encoding the nonoxidative aromatic decarboxy-
lases have thus far been cloned and characterized [5–7,9,14–20].
Based on the searches using Conserved Domain Database (CDD)
and Clusters of Orthologous Groups (COG), Sdc and orthologous
proteins such as 2,3-dihydroxybenzoic acid decarboxylase (DHBD)
of A. oryzae [14] and c-resorcylic acid decarboxylase Rdc of R.
radiobacter WU-0108 [6] show homologies to the predicted
metal-dependent hydrolase of the TIM-barrel fold protein
(COG2159) or amidohydrolase (pfam04909). The reversible and
nonoxidative 4-dihydroxynbenzoate decarboxylase (Ohb1) from
C. hydroxybenzoicum belongs to the 3-octaprenyl-4-hydroxybenzo-
ate carboxy-lyase (UbiD) family (pfam01977) and contains the
conserved sequence EGP[F/Y][G/V][D/E]XXGXY of the UbiD family
[20] (Supplementary Fig. 4). The nonoxidative 4,5-dihydoxyphtha-
late decarboxylase (Pht5) from Pseudomonas putida [21] and PhtD
from Comamonas testosteroni [16] belong to the TauA family
(COG0715). The other nonoxidative aromatic decarboxylase such
as ferulic acid decarboxylase (Fdc) from Bacillus pumilus [19], p-
coumaric acid decarboxylase (PdcC) from Lactobacillus plantarum
[15], and phenolic acid decarboxylase (PadA) Bacillus sp. BP-7
[18] belong to the PA decarbox family (pfam05870). Sdc, the
orthologous proteins such as DHBD, and Rdc show no homology
to the decarboxylases of the UbiD, TauA, and PA decarboxylase
family. Moreover, they contain no sequence similar to EGP[F/
Y][G/V][D/E]XXGXY, and their subunit-molecular mass is approxi-
mately 35 kDa, a value different from that of other nonoxidative
aromatics decarboxylases, which are approximately 20 and
55 kDa. The subunit-molecular mass of already-known reversible
decarboxylases such as Ohb1, pyrrole-2-carboxylate decarboxyl-
ase, and indole-3-carboxylate decarboxylase was approximately
55 kDa. Vanillate/4-hydroxybenzoate decarboxylase consists of 3
subunits with molecular masses of 22.5, 53.0, and 8.6 kDa. These
results indicate that Sdc is a novel reversible and nonoxidative aro-
matic decarboxylase.
In conclusion, we succeeded in the molecular characterization
of a novel reversible and nonoxidative salicylic acid decarboxylase
from T. moniliiforme WU-0401 by purification, characterization and
gene identification of Sdc. A recombinant E. coli expressing sdc con-
verted 40 mM phenol to 10.6 mM salicylic acid with a 27% (mol/
mol) yield at 30 °C for 9 h. Therefore, we consider that the method
described here is a significant model process for selective and eco-
logical production by carboxylation of phenol to form salicylic acid,
that is, by the enzymatic Kolbe–Schmitt reaction.
Acknowledgments
This study was supported in part by the Global-COE Program
‘‘Center for Practical Chemical Wisdom” from the Ministry of
Education, Culture, Sports, Science and Technology (MEXT), Japan
and by Waseda University Grant for Special Research Projects.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bbrc.2010.02.154.
References
[1] A.S. Lindsey, H. Jeskey, The Kolbe–Schmitt reaction, Chem. Rev. 57 (1957) 583–
620.
[2] D. Jeffreys, Aspirin: The Remarkable Story of a Wonder Drug, Bloomsbury,
London, United Kingdom, 2004.
[3] Z. He, J. Wiegel, Purification and characterization of an oxygen-sensitive
reversible 4-dihydroxybenzoate decarboxylase from Clostridium
hydroxybenzoicum, Eur. J. Biochem. 229 (1995) 77–82.
[4] Z. He, J. Wiegel, Purification and characterization of an oxygen-sensitive,
reversible 3,4-dihydroxybenzoate decarboxylase from Clostridium
hydroxybenzoicum, J. Bacteriol. 178 (1996) 3539–3543.
[5] J. Huang, Z. He, J. Wiegel, Cloning, characterization, and expression of a novel
gene encoding a reversible 4-dihydroxybenzoate decarboxylase from
Clostridium hydroxybenzoicum, J. Bacteriol. 181 (1999) 5119–5122.
[6] Y. Ishii, Y. Narimatsu, Y. Iwasaki, N. Arai, K. Kino, K. Kirimura, Reversible and
nonoxidative c-resorcylic acid decarboxylase: characterization and gene
cloning of a novel enzyme catalyzing carboxylation of resorcinol, 1,3-
dihydroxybenzene, from Rhizobium radiobacter, Biochem. Biophys. Res.
Commun. 324 (2004) 611–620.
[7] M. Yoshida, N. Fukuhara, T. Oikawa, Thermophilic, reversible c-resorcylate
decarboxylase from Rhizobium sp. strain MTP-10005: purification, molecular
characterization, and expression, J. Bacteriol. 186 (2004) 6855–6863.
[8] T. Yoshida, Y. Hayakawa, T. Matsui, T. Nagasawa, Purification and
characterization of 2,6-dihydroxybenzoate decarboxylase reversibly
catalyzing nonoxidative decarboxylation, Arch. Microbiol. 181 (2004) 391–
397.
[9] B. Lupa, D. Lyon, L.N. Shaw, M. Sieprawska-Lupa, J. Wiegel, Properties of the
reversible nonoxidative vanillate/4-hydroxybenzoate decarboxylase from
Bacillus subtilis, Can. J. Microbiol. 54 (2008) 75–81.
[10] H. Omura, M. Wieser, T. Nagasawa, Pyrrole-2-carboxylate decarboxylase from
Bacillus megaterium PYR2910, an organic-acid-requiring enzyme, Eur. J.
Biochem. 253 (1998) 480–484.
[11] T. Yoshida, K. Fujita, T. Nagasawa, Novel reversible indole-3-carboxylate
decarboxylase catalyzing nonoxidative decarboxylation, Biosci. Biotechnol.
Biochem. 66 (2002) 2388–2394.
[12] Y. Iwasaki, H. Gunji, K. Kino, T. Hattori, Y. Ishii, K. Kirimura, Novel Metabolic
Pathway for Salicylate Biodegradation via Phenol in Yeast Trichosporon
moniliiforme, Biodegradation, in press. doi:10.1007/s10532-009-9323-6.
[13] J.R. Lee, K.R. Min, Y.C. Kim, C.K. Kim, J.Y. Lim, H.J. Yoon, K.H. Min, K.S. Lee, Y.
Kim, Cloning of salicylate hydroxylase gene and catechol 2,3-dioxygenase gene
and sequencing of an intergenic sequence between the two genes of
Pseudomonas putida KF715, Biochem. Biophys. Res. Commun. 211 (1995)
382–388.
[14] R. Santha, N.A. Rao, C.S. Vaidyanathan, Identification of the active-site peptide
of 2,3-dihydroxybenzoic acid decarboxylase from Aspergillus oryzae, Biochem.
Biophys. Acta 1293 (1996) 191–200.
[15] J.F. Cavin, L. Barthelmebs, C. Divies, Molecular characterization of an inducible
p-coumaric acid decarboxylase from Lactobacillus plantarum: gene cloning,
transcriptional analysis, overexpression in Escherichia coli, purification, and
characterization, Appl. Environ. Microbiol. 63 (1997) 1939–1944.
[16] J. Lee, T. Omori, T. Kodama, Identification of the metabolic intermediates of
phthalate by Tn5 mutants of Pseudomonas testosteroni and analysis of the 4,5-
dihydroxyphthalate decarboxylase gene, J. Ferment. Bioeng. 77 (1994) 583–
590.
[17] X. Peng, E. Masai, H. Kitayama, K. Harada, Y. Katayama, M. Fukuda,
Characterization of the 5-carboxyvanillate decarboxylase gene and its role in
lignin-related biphenyl catabolism in Sphingomonas paucimobilis SYK-6, Appl.
Environ. Microbiol. 68 (2002) 4407–4415.
[18] N. Prim, F.I. Pastor, P. Diaz, Biochemical studies on cloned Bacillus sp. BP-7
phenolic acid decarboxylase PadA, Appl. Microbiol. Biotechnol. 63 (2003) 51–
56.
[19] A. Zago, G. Degrassi, C.V. Bruschi, Cloning, sequencing, and expression in
Escherichia coli of the Bacillus pumilus gene for ferulic acid decarboxylase, Appl.
Environ. Microbiol. 61 (1995) 4484–4486.
[20] H. Zhang, G.T. Javor, Identification of the ubiD gene on the Escherichia coli
chromosome, J. Bacteriol. 182 (2000) 243–6246.
[21] Y. Nomura, M. Nakagawa, N. Ogawa, S. Harashima, Y. Oshima, Genes in PHT
plasmid encoding the initial degradation pathway of phthalate in Pseudomonas
putida, J. Ferment. Bioeng. 74 (1992) 333–344.