Discovery of a sesamin-metabolizing microorganism and a new enzyme

Article abstract of DOI:10.1073/pnas.1605050113

Sesamin is one of the major lignans found in sesame oil. Although some microbial metabolites of sesamin have been identified, sesamin-metabolic pathways remain uncharacterized at both the enzyme and gene levels. Here, we isolated microorganisms growing on sesamin as a sole-carbon source. One microorganism showing significant sesamin-degrading activity was identified as Sinomonas sp. no. 22. A sesamin-metabolizing enzyme named SesA was purified from this strain and characterized. SesA catalyzed methylene group transfer from sesamin or sesamin monocatechol to tetrahydrofolate (THF) with ring cleavage, yielding sesamin mono- or di-catechol and 5,10-methylenetetrahydrofolate. The kinetic parameters of SesA were determined to be as follows: K_m for sesamin = 0.032 ± 0.005 mM, V_max = 9.3 ± 0.4 (μmol?min^-1), and k_cat = 7.9 ± 0.3 s^-1 . Next, we investigated the substrate specificity. SesA also showed enzymatic activity toward (+)-episesamin, (-)-asarinin, sesaminol, (+)-sesamolin, and piperine. Growth studies with strain no. 22, and Western blot analysis revealed that SesA formation is inducible by sesamin. The deduced amino acid sequence of sesA exhibited weak overall sequence similarity to that of the protein family of glycine cleavage T-proteins (GcvTs), which catalyze glycine degradation in most bacteria, archaea, and all eukaryotes. Only SesA catalyzes C1 transfer to THF with ring cleavage reaction among GcvT family proteins. Moreover, SesA homolog genes are found in both Gram-positive and Gram-negative bacteria. Our findings provide new insights into microbial sesamin metabolism and the function of GcvT family proteins.

Full text of DOI:10.1073/pnas.1605050113

Discovery of a sesamin-metabolizing microorganism
and a new enzyme
Takuto Kumanoa,1, Etsuko Fujikia,1, Yoshiteru Hashimotoa, and Michihiko Kobayashia,2
aGraduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8572, Japan
Edited by Julian Davies, University of British Columbia, Vancouver, BC, Canada, and approved June 14, 2016 (received for review March 28, 2016)
Sesamin is one of the major lignans found in sesame oil. Although
some microbial metabolites of sesamin have been identified,
sesamin-metabolic pathways remain uncharacterized at both the
enzyme and gene levels. Here, we isolated microorganisms growing
on sesamin as a sole-carbon source. One microorganism showing
significant sesamin-degrading activity was identified as Sinomonas
sp. no. 22. A sesamin-metabolizing enzyme named SesA was purified
from this strain and characterized. SesA catalyzed methylene group
transfer from sesamin or sesamin monocatechol to tetrahydrofolate
(THF) with ring cleavage, yielding sesamin mono- or di-catechol and
5,10-methylenetetrahydrofolate. The kinetic parameters of SesA were
higher antioxidant activity than vitamin E (18); reduction of
cardiovascular disease risk (19); reduction of the risk of arte-
riosclerosis by decreasing the total cholesterol level (20); and
reduction of inflammation markers (21). However, neither ses-
amin-metabolizing enzymes nor their genes have been identified
in microorganisms, including intestinal bacteria and A. oryzae.
Here, we describe the isolation and identification of a sesamin-
catabolizing soil microorganism, Sinomonas sp. no. 22, together
with purification and characterization of a previously un-
known sesamin-modifying enzyme. This enzyme has the unique
catalytic ability to transfer the methylene group from sesamin to
tetrahydrofolate (THF) through ring cleavage (Fig. 1). In addition,
we clarify the biochemical properties of this enzyme and propose a
possible reaction mechanism. Our findings provide novel insights
into the catabolism of natural compounds and THF-dependent
metabolic pathways.
determined to be as follows: Km for sesamin = 0.032
0.005 mM,
Vmax = 9.3 0.4 (μmol·min−1·mg−1), and kcat = 7.9 0.3 s−1. Next,
we investigated the substrate specificity. SesA also showed enzymatic
activity toward (+)-episesamin, (−)-asarinin, sesaminol, (+)-sesamolin,
and piperine. Growth studies with strain no. 22, and Western blot
analysis revealed that SesA formation is inducible by sesamin. The
deduced amino acid sequence of sesA exhibited weak overall se-
quence similarity to that of the protein family of glycine cleavage
T-proteins (GcvTs), which catalyze glycine degradation in most
bacteria, archaea, and all eukaryotes. Only SesA catalyzes C1 trans-
fer to THF with ring cleavage reaction among GcvT family proteins.
Moreover, SesA homolog genes are found in both Gram-positive
and Gram-negative bacteria. Our findings provide new insights
into microbial sesamin metabolism and the function of GcvT
family proteins.
Results
Isolation and Identification of Sesamin-Metabolizing Bacteria. The
soil samples used for microbial screening were collected from
sesame gardens at the University of Tsukuba. At ∼1 mo from the
start of this study, by using the enrichment culture method de-
scribed in Materials and Methods, we isolated 40 microorganisms
that were able to grow on culture medium containing sesamin as
the sole carbon source. We selected one isolate, strain no. 22, for
further study. This strain showed 99% 16S rRNA gene sequence
similarity to Sinomonas atrocyanea DSM20127T. Morphological
and biochemical properties of strain no. 22 are shown in SI Ap-
pendix, Supplementary Data.
sesamin metabolism lignan tetrahydrofolate
|
|
|
e have been involved in studies of not only microbial
W
metabolism of man-made compounds (1–3) but also bio-
logically active natural compounds, such as curcumin (4). In this
study, we characterize the microbial metabolism of the lignan
sesamin.
Significance
Lignans, including sesamin, are produced by a wide variety of
plants, but the microbial degradation of lignan has not been
identified biochemically. Here, we show that Sinomonas sp. no.
22 can catabolize sesamin as a sole-carbon source. We identi-
fied the sesamin-converting enzyme, SesA, from strain Sino-
monas sp. no. 22. SesA catalyzed methylene group transfer
from sesamin to tetrahydrofolate (THF). The resulting 5,10-CH2-
THF might find use as a C1-donor for bioprocesses. SesA gene
homologs were found in the genomes of both Gram-positive
and Gram-negative bacteria, suggesting that sesamin (lignan)
utilization is a widespread, but still unrecognized, function in
environments where lignans are produced and degraded.
Lignans (5) are plant-derived compounds consisting of dimers
of phenylpropane units (6). They are found in a wide variety of
plant-based foods. Whole-grain products, vegetables, fruits, nuts,
seeds, and beverages such as tea, coffee, and wine are dietary
sources of lignans. In Asian countries, sesame, which contains
lignans, is used traditionally as a food. A major lignan is sesamin,
which is a biologically active compound with antioxidative (7),
cholesterol-lowering (8), lipid-lowering (9), antihypertensive (10),
and antiinflammatory (11) properties.
In humans, sesamin is metabolized by CYP450 enzymes into
sesamin mono- and di-catechol in liver microsomes (12). Sesamin
monocatechol is metabolized further by UDP-glucuronosyltrans-
ferase (UGT) and O-methyl transferase (COMT) (13), and the
resulting glucuronides of sesamin metabolites are excreted in the
bile and urine (14).
In microorganisms, on the other hand, metabolism of sesamin
has been reported in few species. Aspergillus oryzae converts
sesamin to sesamin mono- and di-catechol (15) whereas in-
testinal bacteria convert sesamin to the so-called “mammalian
lignans” enterodiol and enterolactone (16). The activities of
these metabolites make them useful dietary substances. Sesamin
mono- and di-catechol show stronger antioxidant activity than
sesamin (17). Also, mammalian lignans show other properties:
Author contributions: T.K., E.F., and M.K. designed research; T.K. and E.F. performed
research; T.K., E.F., Y.H., and M.K. analyzed data; and T.K. and M.K. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Data deposition: The nucleotide sequence data reported in this paper have been depos-
ited in the DNA Data Bank of Japan (DDBJ) database [accession nos. LC101493 (sesA and
thf2), LC101494 (thf1), and LC101495 (16S rRNA gene)].
1T.K. and E.F. contributed equally to this work.
2To whom correspondence should be addressed. Email: kobay@agbi.tsukuba.ac.jp.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1605050113/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1605050113
PNAS Early Edition
|
1 of 6

Time-Dependent Sesamin Metabolism During Cultivation. The
growth curve of strain no. 22 is shown in SI Appendix, Fig. S7. In
liquid medium containing sesamin as a sole-carbon source (SI
Appendix, Fig. S7A), the specific activity of the enzyme increased
after 8 h cultivation, followed by a decrease in sesamin in the
medium and an increase in the protein compared with cell growth.
On the contrary, in medium containing glucose as sole-carbon
source, cells grew exponentially for 4 h, but sesamin-converting
activity was not observed during cultivation. Furthermore, in 2×
YT liquid medium, sesamin-converting activity was detected only
in the presence of added sesamin (SI Appendix, Fig. S7B).
A
B
Fig. 1. Schematic representation of the bond cleavage location in the
structure of substrates for SesA (A) and previously reported THF-dependent
methyl transferases (B). The cleavage site of the bond between atoms within
a substrate is indicated by a dashed line.
Western Blot Analysis. To confirm correlation of SesA expression
with sesamin metabolism, the induction of SesA in the presence
of sesamin was followed by Western blot analysis on cell-free
extracts in the following culture conditions: One was in 0.1% (wt/vol)
sesamin-supplemented 2× YT liquid medium, and the other was in
in 2× YT liquid medium. A cross-reacting band was observed
only in the cell-free extract of cells cultured with sesamin (SI Ap-
pendix, Fig. S7D). In addition, SesA enzyme activity (0.053 units/mg)
was observed only in cells grown in the presence of sesamin.
Structure Determination of the Sesamin Metabolites. We incubated
a cell-free extract of strain no. 22 with sesamin as a substrate.
Two products (compounds A and B) were isolated by ethyl ac-
etate extraction and preparative HPLC (Fig. 2).
High-resolution mass spectrometry (HRMS) analysis in the
negative mode revealed the molecular ion of compound A at m/z
341.1029 [M-H]−, which was in agreement with the calculated
mass of C19H17O6, and that of compound B at m/z 329.1022
[M-H]−, in agreement with the calculated mass of C18H17O6 (SI
Appendix, Fig. S1). The loss of 12 atomic mass units suggested
modification of the methylenedioxyphenyl group.
Cloning and Heterologous Expression of sesA and Biochemical
Properties of SesA. The 1.4-kb region of the SesA-coding gene
was inserted into an expression vector, and recombinant SesA
was produced in Escherichia coli and purified. The specific ac-
tivity of the recombinant SesA was approximately the same
(8.8 units/mg) as that of SesA purified from strain no. 22.
We examined the effects of temperature and pH on SesA
activity. The optimal reaction temperature and pH were below
40 °C and pH 7.5–8.5, respectively (SI Appendix, Fig. S8 A and B).
SesA was most stable under 30 °C and in the pH range of 5.5–10.0
(SI Appendix, Fig. S8 C and D).
The absorption spectrum of the purified SesA showed an ab-
sorbance maximum near 280 nm. No other absorption peak or
shoulder was observed at higher wavelengths (SI Appendix, Fig.
S9). These results suggest that no cofactor is bound to the pu-
rified enzyme. The CD spectrum is shown in SI Appendix, Fig.
S10. Qualitative analysis of metal content was performed by in-
ductively coupled plasma atomic emission spectroscopy analysis.
The enzyme contained 1.70 mol of phosphorus and 0.70 mol of
sulfur per mole of subunit. No other metal was detected within
the assay limits (SI Appendix, Supplementary Data 3).
1H NMR spectra of compounds A and B showed proton signals
corresponding to the tetrahydrofuran ring and 1,3,4-substituted
benzene moiety expressed (SI Appendix, Fig. S1). On the other
hand, the proton signal corresponding to H-10′ of sesamin was
absent in compound A (SI Appendix, Figs. S1 and S2). The proton
signals corresponding to H-10 and H-10′ of sesamin were not
observed for compound B (SI Appendix, Figs. S1 and S3).
Based on these observations, compounds A and B were iden-
tified as sesamin monocatechol (1 in Fig. 3) and sesamin di-
catechol (2 in Fig. 3), respectively (22).
Purification of the Sesamin-Metabolizing Enzyme. The sesamin-
metabolizing enzyme was purified 1.2-fold with a yield of 3.4%
(SI Appendix, Table S3). Considering the relationship between
the protein amount and the activity of each fraction obtained
on gel-filtration chromatography, the target enzyme was as the
∼50-kDa protein band observed on SDS/PAGE (SI Appendix,
Fig. S4 and Table S4). An N-terminal amino acid sequence of
this protein was determined to be TAEQAIN.
The gene for the sesamin-metabolizing enzyme, named SesA,
was identified from the draft genome sequence data for strain
no. 22 (see Identification of the Sesamin-Metabolizing Enzyme).
The primary structure of SesA suggested that it might require
THF as a cofactor. On the addition of THF to cell-free extracts,
the sesamin-metabolizing activity increased 350 times (Table 1
and SI Appendix, Table S3).
SesA was purified as a single band on SDS/PAGE (Table 1
and SI Appendix, Fig. S5) after ammonium sulfate precipitation
(35–55%) and column chromatography (TOYOPEARL Butyl
650M, Mimetic Orange 1 A6XL). The molecular mass of native
SesA was 150 kDa, (SI Appendix, Fig. S6). The molecular mass of
SesA was calculated to be 50,385 Da, which was consistent with
that of the purified enzyme determined by SDS/PAGE. These
results indicate that SesA consists of three identical subunits.
Stoichiometry. The stoichiometry of the SesA reaction was exam-
ined. The amounts of sesamin, sesamin monocatechol, sesamin
di-catechol, and folates were determined by HPLC and liquid
chromatography (LC)/MS/MS. After 30 min incubation, sesamin
mono- and di-catechol increased to 77 and 9.3 μM, respectively.
Coincidentally, sesamin decreased by 80 μM (SI Appendix, Fig. S11A).
We detected 5,10-CH2-THF in SesA reaction mixtures. After
30-min incubation, 5,10-CH2-THF increased to 85 μM (SI
75
sesamin
50
A
B
25
0
Identification of the Sesamin-Metabolizing Enzyme. We determined
the draft genome sequence for strain no. 22 using a next-generation
sequencer, Hiseq2500 (Illumina). From the sequence information,
we identified an ORF of 1,359 nucleotides, 21 of which corre-
sponded to the above N-terminal sequence (SI Appendix, Supple-
mentary Data 2). The deduced amino acid sequence identified a
putative folate-binding domain as in the glycine cleavage T-protein
(GcvT) (pfam01571).
0
5
10
15
Retention time (min)
Fig. 2. HPLC analysis of reaction products of sesamin. Chromatogram of the
reaction mixture after incubation of sesamin with a cell-free extract of strain
no. 22. The products are indicated as A and B.
2 of 6
|
www.pnas.org/cgi/doi/10.1073/pnas.1605050113
Kumano et al.

sesamin mono-catechol, di-catechol, and 5,10-CH2-THF were
formed stoichiometrically with the consumption of sesamin dur-
ing the enzymic reaction (Fig. 4A).
Specific
activity
(units/mg)
Relative
activity
Substrate
Product
10
O
O
O
H
O
Kinetic Analysis. Using varying concentrations of sesamin in the
presence of 1 mM THF, a typical hyperbolic curve of product
formation over substrate concentration was obtained (SI Appen-
dix, Fig. S11C), indicating that the reaction followed Michaelis–
Menten kinetics. Apparent steady-state kinetic constants were
estimated for sesamin by quantitative measurements of sesamin
monocatechol by HPLC analysis. Nonlinear regression analysis
revealed the following: Km = 0.032 0.005 mM, Vmax = 9.3
3
7’
H
8
1
4’
OH
O
O
1’
H
O
100%
8’
4
H
O
3’
7
10’
O
O
OH
sesamin
1
HO
O
H
OH
HO
H
O
OH
2
HO
O
H
O
O
H
O
O
O
HO
31%
88%
0.4 (μmol·min−1·mg−1), and kcat = 7.9 0.3 s−1
.
H
O
H
O
O
O
(+)-episesamin
3
Substrate Specificity. To examine the substrate specificity of SesA,
we investigated the compounds listed in Fig. 3. The structures of
the products were determined by LC/MS/MS and NMR analyses.
SesA catalyzed the demethylenation of (+)-episesamin, (−)-asarinin,
sesaminol, and (+)-sesamolin to yield monocatechols 3 (SI Appendix,
Figs. S12–S16), 4 (SI Appendix, Figs. S17–S22), 6 (SI Appendix, Figs.
S23–S25), and 8 and 9 (SI Appendix, Figs. S26–S28), and di-catechols
5 (SI Appendix, Figs. S17 and S29–S32), 7 (SI Appendix, Fig. S23),
and 10 (SI Appendix, Fig. S26). Moreover, piperine (derived
from black pepper) was also demethylated (SI Appendix, Figs.
S33–S35). SesA activity was specific for the methylenedio-
xyphenyl and not the methoxyphenyl group.
O
O
H
O
O
H
OH
O
O
O
H
O
H
OH
O
O
(-)-asarinin
4
HO
O
H
OH
HO
H
O
OH
5
O
H
O
O
O
O
O
O
H
O
no products
O
O
diasesamin
O
H
O
O
H
OH
O
112%
H
H
O
O
OH
OH
OH
sesaminol
6
O
Mutational Analysis of SesA. To investigate the reaction mecha-
nism of SesA, we constructed a set of mutants with single-amino
acid substitutions. SesA was found to belong to a diverse family
of enzymes that include specific domains of dimethylglycine
oxidase (DMGO) (23), sarcosine oxidase (24), dimethylsulfo-
niopropionate demethylase (DmdA) (25), and demethylase. The
enzyme acts on lignin-degradation products such as syringate and
vanillin [DesA (26, 27) and LigM (28), respectively], as well as
GcvT (29).
Based on the amino acid sequence alignment of these enzymes,
we prepared three mutants of SesA: D95A, E189A, and Y221A.
Each of the mutant enzymes was expressed in E. coli and purified
by the method described in Materials and Methods (SI Appendix,
Fig. S36). Enzymatic activity was measured using the method as
that for wild-type SesA. E189A and Y221A exhibited no activity
at all. The activity of the D95A derivative was 40% compared
with that of the wild-type enzyme.
HO
H
OH
HO
H
O
OH
OH
7
O
O
O
H
O
H
O
O
OH
O
O
O
H
H
97%
O
O
O
OH
sesamolin
8
HO
O
H
HO
O
O
H
O
O
9
HO
O
H
OH
HO
O
H
O
OH
10
H3CO
HO
O
H
OH
H
O
no products
OCH3
pinoresinol
Discussion
N
N
In nature, many physiologically active compounds, such as fla-
vonoids, terpenoids, alkaloids, steroids, coumarins, glycosides,
and nucleosides, are produced by plants and microorganisms.
Unique pathways of microbial metabolism of these compounds
have been reported (30–34). Research on the microbial metab-
olism of a diversity of natural compounds can be expected to
reveal novel enzymes (4) and catalytic functions (34). We recently
reported a curcumin metabolic pathway in E. coli and identified a
novel curcumin/dihydrocurcumin reductase (4). Here, we isolated
sesamin-catabolizing microorganisms and determined the initial
steps of the sesamin-metabolic pathway at both protein and
gene levels.
OH
O
O
O
O
11%
OH
OH
piperine
11
O
O
O
O
O
O
H
O
O
H
O
HO
OH
O
OCH3
OCH3
curcumin*
samin*
piperonal*
HO
H3CO
HO
O
OH
O
H3CO
O
sesamol*
vanillin*
isovanillin*
Sesamin is a major lignan in sesame oil with characteristic
methylenedioxyphenyl groups. A methylenedioxyphenyl group
is present in some plant metabolites, such as berberin (isolated
from Berberis), piperonal (isolated from dill, vanilla, violet flowers,
and black pepper), and piperine (isolated from black pepper),
as well as sesamin. Also, drugs such as tadalafil and 3,4-methyl-
enedioxymethamphetamine (MDMA) have a methylenedioxyphenyl
group. In humans, the methylenedioxy bridges (O-C-O) of
methylenedioxyphenyl compounds are oxidized to catechols by
CYP450s (12, 35).
*no products were observed
Fig. 3. SesA activities for plant-derived methylenedioxyphenyl compounds
and their derivatives. The specific activities for each substance and structures
of reaction products are indicated.
Appendix, Fig. S11B); there was no change in the 5-CH3-THF
amount. THF could not be determined because of its instability
under our assay conditions. These results demonstrate that
Kumano et al.
PNAS Early Edition
|
3 of 6

Table 1. Purification of SesA
Step
Total protein, mg Total activity, units Specific activity, units/mg Yield, % Purification, fold
Cell-free extract
(NH4)2SO4
Hiprep Butyl FF
Mimetic Orange 1 A6XL
390
170
21
120
110
30
0.31
0.66
1.4
100
88
25
1
2.1
4.5
1.2
9.1
7.6
7.5
24
On the other hand, microbial enzymes that metabolize meth-
ylenedioxyphenyl bridges have not been reported although me-
tabolites have been identified in a few cases. For example,
sesamin is converted into sesamin monocatechol and sesamin
di-catechol by A. oryzae (15). In addition, cell-free extracts of
Pseudomonas fluorescens strain PM3 oxidize piperonylic acid into
protocatechuate and formic acid, requiring NADH or NADPH
as a cofactor (36). Recently, it was reported that intestinal bac-
teria convert sesamin into mammalian lignans (16).
In this study, we identified SesA, which converts sesamin into
sesamin mono- and di-catechol. The cleaved methylene group
of sesamin is transferred to THF, 5,10-CH2-THF being formed.
Thus, SesA is a THF-dependent sesamin/sesamin-monocatechol
methylenetransferase. In this reaction, the methylene group
from the methylenedioxy bridge is transferred without an apparent
change in redox state to THF. This mechanism is distinct from that
of CYP450 (12), which oxidatively removes the methylene group
of sesamin. Compared with the enzymatic activity of SesA and
CYP2C9, which contributes significantly to the metabolism of ses-
amin in the liver, the Km values of SesA and CYP2C9 were 32 μM
and 5.4 μM, respectively. On the other hand, the kcat value of SesA
for sesamin was 220 times higher than that of CYP2C9 (12).
Homology searches of the protein database demonstrated that
SesA exhibits low similarity (∼20%) to GcvT, which is involved
in the glycine cleavage system (GCS) together with P-protein,
H-protein, and L-protein. In general, GCS is found in most bac-
teria, archaea, and the mitochondria of all eukaryotes and plays
critical roles in both glycine degradation and one-carbon metab-
olism (37).
equatorial positions, when the stereochemistry of the tetrahy-
drofuran ring is 8S, 8′S.
Also, the activity of SesA was affected by substrate size. Pip-
erine was an active substrate whereas small methylene dioxy-
phenyl compounds such as samin were inert substrates (Fig. 3).
The crystal structure of GcvT of Thermotoga maritima shows
that aspartic acid (D96), glutamic acid (E195), and tyrosine
(Y100) are hydrogen-bonded to THF (29). These hydrogen bonds
are also observed in the structures of DMGO, sarcosine oxidase,
and DmdA; sequence alignment of SesA, LigM, DesA, GcvT,
DmdA, and DMGO demonstrated that D95 and E189 of SesA
corresponded to D96 and E195 of GcvT (SI Appendix, Fig. S38).
Tyrosine is not conserved at the corresponding position in the
protein sequences of GcvT family proteins. Strictly speaking, Y100
of GcvT, Y660 of DMGO, and Y206 of DmdA are hydrogen-
bonded to THF. In this alignment, Y221 of SesA corresponded
to Y247 and Y242 of LigM and DesA, respectively. The E189A
and Y221A mutants of SesA were inactive. Considering the
reported crystal structures of DmdA (25) and GcvT (29), this
O
O
thf1
sesA thf2
A
B
H
O
O
H
O
O
Sesamin
THF
1 kb
Thf1
10-CHO-THF
SesA
5,10-CH2-THF
Thf2
O
O
H
The folate-binding domain of GcvT (pfam01571) is conserved
in SesA and also in DMGO, sarcosine oxidase, DmdA, DesA,
and LigM. These GcvT family proteins exhibit weak sequence
identities and are classified into separate clades from one an-
other (SI Appendix, Fig. S37).
SesA is distinct as follows: (i) SesA is a homo-trimer that
forms no complex with any other proteins; (ii) the substrates of
SesA are aromatic compounds; and (iii) SesA catalyzes a ring
cleavage reaction to transfer the methylene group to THF (Fig. 1).
In particular, point iii is unique to SesA among the GcvT family
proteins.
Nucleotides
Vitamins
Amino acids
OH
O
Folate-dependent
one carbon pool
H
O
OH
Sesamin mono-catechol
Thf1
10-CHO-THF
THF
SesA
5,10-CH2-THF
Thf2
HO
O
H
OH
HO
H
O
OH
Sesamin di-catechol
SesA, LigM, and DesA all metabolize aromatic compounds
although the sequence similarities of SesA to LigM and DesA
are only 22% and 26%, respectively. On the other hand, SesA
showed distinct enzymatic activities compared with LigM and
DesA. LigM and DesA transfer the “methyl” group to THF,
yielding “5-CH3-THF.” On the contrary, SesA transferred the
“methylene” group of sesamin to THF, giving “5,10-CH2-THF.”
Although we investigated each of the compounds listed in Fig. 3,
we found that SesA, does not transfer the methyl group to THF.
The enzymatic activity of SesA was influenced by the stereo-
chemistry of the methylenedioxyphenyl group (Fig. 3). Dia-
sesamin and (+)-episesamin monocatechol are inert substrates
(Fig. 3), suggesting that SesA catalyzes only the demethylenation
of an equatorial methylenedioxyphenyl group, when the stereo-
chemistry of the tetrahydrofuran ring is 8R, 8′R. On the other
hand, (−)-asarinin is converted into (−)-asarinin di-catechol,
which suggests that SesA is able to catalyze the demethylenation
of methylenedioxyphenyl groups that are in both axial and
TCA cycle
R'
R'
R'
C
HO
H
HO
O
R
B
H
B
BH
O
OH
O
O
B'
R
HB'
R
H
B'
N
CH2
N
O
N
O
O
H
N
N
N5
R
N
H
HN
H2N
HN
H2N
HN
H2N
N10
HN
H2N
N
H
H
N
H
N
N
N
N
H
N
N
H
H
(i)
(ii)
(iii)
(iv)
Fig. 4. Sesamin metabolic pathway and proposed reaction mechanism.
(A) Proposed sesamin metabolic pathway in strain no. 22. (B) DNA sequence
of region encoding SesA and its flanking region. The thfl gene exhibits 60%
amino acid sequence identity with that of Mycobacterium bovis (UniProtKB
accession code P0A5T7), and the thf2 gene exhibits 69% amino acid se-
quence identity with that of Clavibacter michiganensis (UniProtKB accession
code B0RD2). (C) Proposed reaction mechanism for SesA. In C, “B” represents
a base.
4 of 6
|
www.pnas.org/cgi/doi/10.1073/pnas.1605050113
Kumano et al.

finding suggests that E189 and Y221 of SesA would be hydrogen-
bonded to THF.
5,10-CH2-THF produced by the SesA reaction would be me-
tabolized as follows. Analysis of the gene annotation of strain no.
22 indicated that strain no. 22 has a putative formyltetrahydrofolate
deformylase gene (thf1), and a putative bifunctional 5,10-CH2-THF
dehydrogenase/cyclohydrolase gene (thf2) upstream and down-
stream of the sesA gene, respectively (Fig. 4B). Thf2 could
convert 5,10-CH2-THF into 10-formyltetrahydrofolate (10-CHO-
THF), which is a substrate for the biosynthesis of purine. Also,
Thf1 could convert 10-CHO-THF into THF. These findings sug-
gest that strain in sesamin no. 22 is important physiologically.
Some SesA homologs, which we found by Blast searches, form
a gene cluster with folate-metabolizing enzyme genes (SI Ap-
pendix, Fig. S39). Most were derived from Actinobacteria, but
this gene set was also observed in the Gram-negative bacterium,
Bradyrhizobium japonicum WSM2793. These findings suggest
that THF-dependent C1 transferases are distributed in various
microorganisms.
The reaction mechanisms of GcvT, DmdA, and DMGO were
proposed based on 3D structural and mutational analyses (23,
25, 38). (i) In GcvT of E. coli, the electron relay from D97 (or
D96) to N113 through the hydrogen bond could make N113 act
as a base to deprotonate the protonated amino group of the
aminomethyllipoyllysine arm in the substrate, followed by the
cleavage of the C–S bond of the arm after migration of a proton
from the protonated R223 to the substrate, yielding a reactive
iminium intermediate (38). The iminium intermediate reacts
with the N5 atom of THF to form ammonia and an iminium ion
including the N5 atom of THF. Next, D97 deprotonates the N10
atom of 5-CH2-THF, and this deprotonated N10 attacks the
iminium ion including N5, to form 5,10-CH2-THF. (ii) In DmdA,
methyl transfer is suggested to be coupled with proton transfer
that is initiated by a base and mediated by a water molecule in
the active site. In this reaction, as a nucleophile, Sp3 hybridized-N5
of THF attacks the CH3 group on the sulfonium ion of the sub-
strate, to yield 3-(methylthio)propionic acid (25). Therefore, the
proton donor is not required in this reaction. (iii) In DGMO, THF
attacks the iminium ion of the substrate via the nucleophilic N10
atom, with concomitant deprotonation by D552, followed by the
formation of sarcosine and 5,10- CH2-THF through intramolecular
rearrangement of the covalent intermediate formed between THF
and the iminium intermediate (i.e., N5 of THF attacks on the
covalent intermediate with concomitant deprotonation of N5 of
THF by the nascent sarcosine) (23).
Materials and Methods
Bacterial Strains, Plasmids, Primers, and Additional Methods. For bacterial
strains, plasmids, and primers, see SI Appendix, Tables S1 and S2.
For chemicals, HPLC and LC/MS/MS analyses, structure determination,
purification of the sesamin-metabolizing enzyme from Sinomonas sp. no. 22,
the draft genome sequence of Sinomonas sp. no. 22, cloning and heterol-
ogous expression of sesA, determination of the molecular mass of SesA, time
courses of cell growth and enzymatic activity, Western blot analysis, mea-
surement of folate, temperature dependency and stability, pH dependency
and stability, substrate specificity, circular dichroism analysis, and site-directed
mutagenesis, see SI Appendix, Supplementary Methods.
In our study, on the other hand, the activity of D95A was
found to be 40% less compared with that of the wild-type enzyme.
Considering this finding and the proposed reaction mechanisms of
other GcvT family enzymes, direct nucleophilic attack on sesamin
by N5 of THF would initiate the reaction, as seen in the case of
DmdA. However, the reaction mechanism of SesA is not the same
as other members of the GcvT family. In the SesA reaction, the
methylenedioxy groups (O–C–O) are cleaved to yield OH groups
of catechol moieties; proton donation is required to cleave the
O–C–O bond, which is not present in substrates of other GcvT
family enzymes. Therefore, we propose a possible reaction
mechanism, in which proton donors (indicated by BH+ and B′H+
in Fig. 4C) are involved. In previously reported reaction mech-
anisms of GcvT family enzymes, proton donation does not occur
except in GcvT. In GcvT, R223 is predicted to donate a proton to
the substrate for cleavage of the C–S bond. According to the
amino acid alignments of these proteins, an arginine residue,
which corresponds to R223 in GcvT, is not conserved in SesA,
LigM, and DesA. We predict that the amino acid residues act as
proton donors that provide the methylenedioxy bridges of ses-
amin with a proton. Considering the crystal structures of GcvT,
DmdA, and DMGO in the GcvT family proteins, candidates of
proton donor residues in the predicted active site of SesA are as
follows: R81, H82, R100, R179, and H225 (SI Appendix, Fig.
S38). BH+ and B′H+ provide protons and become :B and :B′,
respectively, in sesamin. In the proposed reaction (Fig. 4C), the
ring closure to yield 5,10-CH2-THF is predicted to be initiated by
a base (:B′). Then, the :B′ should accept the proton on the N10
atom of an iminium ion, including the N5 atom of 5-CH2-THF in
the last step (step iii). The SesA reaction is different from that of
other GcvT family enzymes in that SesA requires proton donors
for the reaction. To identify the amino acid residues B and B′,
studies on the crystal structure and site-directed mutagenesis
studies of SesA are required.
Isolation of Sesamin-Metabolizing Microorganisms. Sesamin-metabolizing
microorganisms were isolated from soil in the University of Tsukuba and sesame
gardens by the following enrichment method. Step 1 was as follows: 1 g of
collected soil was added to 10 mL of sesamin medium, which consisted of 0.1%
(wt/vol) sesamin, 1% (wt/vol) (NH4)2SO4, 0.05% (wt/vol) KH2PO4, 0.05% (wt/vol)
K2HPO4, 0.05% (wt/vol) MgSO4·7H2O, 0.0005% (wt/vol) FeSO4·7H2O, and
10% (vol/vol) tap water, adjusted to pH 7.0 with NaOH, followed by incubation
at 28 °C or 37 °C for 3 d. Step 2 was as follows: 2% (vol/vol) of the cultivated
medium was added to the same fresh medium, followed by incubation at
28 °C or 37 °C for 3 d. Step 2 was repeated three times.
After enrichment, the culture broth was spread on sesamin sole-carbon
agar plates, which contained 1.5% (wt/vol) agar in addition to the above
sesamin sole-carbon medium, and colonies that grew on these plates on 1 wk
incubation at 28 °C were isolated.
Each of the isolated strains was inoculated into a test tube containing
10 mL of sesamin sole-carbon medium, followed by incubation at 28 °C for 2 d.
Cells were harvested by centrifugation (4,000 × g, 10 min, 4 °C) and, after
washing twice with 10 mM potassium phosphate buffer (KPB) (pH 7.0), were
resuspended in 200 μL of the same buffer. Then, the cells were disrupted by
sonication, and the cell debris was removed by centrifugation (27,000 × g,
10 min, 4 °C) to prepare a cell-free extract. Two hundred microliters of the
reaction mixture comprised 10 μL of 100 mM KPB (pH 7.0), 10 μL of 10 mM
sesamin (in DMSO), 100 μL of the cell-free extract, and milliQ water. After
incubation at 28 °C for 16 h, the reaction was stopped by adding 100 μL of
acetonitrile. The reaction samples were analyzed by HPLC and LC/MS.
Enzyme Assay. Measurement of enzyme activity was performed as follows.
One hundred microliters of the reaction mixture [1 μL of 0.73 mg/mL SesA,
5 μL of 1 M Tris·HCl (pH 8.0), 3 μL of 10 mM substrate (in DMSO), 10 μL of
10 mM THF, and 2 μL of Tween 80 were used]. THF was dissolved in 50 mM
Tris·HCl (pH 9.0), 1% 2-mercaptoethanol, and 2% (wt/vol) ascorbate. One
unit of sesamin-converting activity was defined as the amount of enzyme
required to catalyze the formation of 1 μmol of sesamin monocatechol per
minute. Specific activity is expressed as units per milligram of protein.
The reaction was initiated by adding the enzyme, followed by incubation
at 28 °C for an appropriate time. After incubation, the reaction was stopped
by adding 100 μL of acetonitrile.
For determination of the kinetic parameters for the demethylenation of
sesamin, 100 μL of the reaction mixture consists of 1 μL of 0.0731 mg/mL
SesA, 5 μL of 1 M Tris·HCl (pH 8.0), 10 μL of 10 mM THF, 2 μL of Tween 80, and
from 0.05 mM to 0.3 mM sesamin. The reactions were initiated by the ad-
dition of SesA, followed by incubation at 28 °C, and then termination at 1, 3,
5, 7, and 10 min by the addition of 50 μL of acetonitrile. The experiments
At the beginning of this study, we isolated strain no. 22 by
enrichment culture using sesamin as a sole-carbon source. In the
growth experiment, the enzymatic activity of SesA was found to
increase just before the onset of cell growth (SI Appendix, Fig. S7).
These experiments and Western-blot analysis revealed that SesA
formation was induced by sesamin in both media. Moreover,
Kumano et al.
PNAS Early Edition
|
5 of 6

were carried out in duplicate independently. The kcat values were calculated
using a Mr of 50,385 for SesA.
numbers LC101493 for sesA and thf2, LC101494 for thf1, and LC101495 for
the 16S rRNA gene.
Nucleotide Sequence Accession Numbers. The nucleotide sequence data
ACKNOWLEDGMENTS. We thank Dr. Kentaro Shiraki (University of Tsukuba)
reported in this paper appear in the DDBJ/GenBank database under accession
for help with circular dichroism spectra analysis.
1. Kobayashi M, Shimizu S (1998) Metalloenzyme nitrile hydratase: Structure, regula-
tion, and application to biotechnology. Nat Biotechnol 16(8):733–736.
2. Komeda H, Hori Y, Kobayashi M, Shimizu S (1996) Transcriptional regulation of the
Rhodococcus rhodochrous J1 nitA gene encoding a nitrilase. Proc Natl Acad Sci USA
93(20):10572–10577.
21. Hallund J, Tetens I, Bügel S, Tholstrup T, Bruun J-M (2008) The effect of a lignan
complex isolated from flaxseed on inflammation markers in healthy postmenopausal
women. Nutr Metab Cardiovasc Dis 18(7):497–502.
22. Nakai M, et al. (2003) Novel antioxidative metabolites in rat liver with ingested ses-
amin. J Agric Food Chem 51(6):1666–1670.
23. Leys D, Basran J, Scrutton N-S (2003) Channelling and formation of ‘active’ formal-
dehyde in dimethylglycine oxidase. EMBO J 22(16):4038–4048.
24. Ida K, Moriguchi T, Suzuki H (2005) Crystal structure of heterotetrameric sarcosine
oxidase from Corynebacterium sp. U-96. Biochem Biophys Res Commun 333(2):
359–366.
25. Schuller D-J, Reisch C-R, Moran M-A, Whitman W-B, Lanzilotta W-N (2012) Structures
of dimethylsulfoniopropionate-dependent demethylase from the marine organism
Pelagabacter ubique. Protein Sci 21(2):289–298.
3. Nomura J, et al. (2013) Crystal structure of aldoxime dehydratase and its catalytic
mechanism involved in carbon-nitrogen triple-bond synthesis. Proc Natl Acad Sci USA
110(8):2810–2815.
4. Hassaninasab A, Hashimoto Y, Tomita-Yokotani K, Kobayashi M (2011) Discovery of
the curcumin metabolic pathway involving a unique enzyme in an intestinal micro-
organism. Proc Natl Acad Sci USA 108(16):6615–6620.
5. Hearon W-M, MacGregor W-S (1955) The naturally occurring lignans. Chem Rev 55(5):
957–1067.
26. Nishikawa S, et al. (1998) Cloning and sequencing of the Sphingomonas (Pseudo-
monas) paucimobilis gene essential for the O demethylation of vanillate and sy-
ringate. Appl Environ Microbiol 64(3):836–842.
27. Masai E, et al. (2004) A novel tetrahydrofolate-dependent O-demethylase gene is
essential for growth of Sphingomonas paucimobilis SYK-6 with syringate. J Bacteriol
186(9):2757–2765.
28. Abe T, Masai E, Miyauchi K, Katayama Y, Fukuda M (2005) A tetrahydrofolate-dependent
O-demethylase, LigM, is crucial for catabolism of vanillate and syringate in Sphingomonas
paucimobilis SYK-6. J Bacteriol 187(6):2030–2037.
29. Lee H-H, Kim D-J, Ahn H-J, Ha J-Y, Suh S-W (2004) Crystal structure of T-protein of the
glycine cleavage system: Cofactor binding, insights into H-protein recognition, and
molecular basis for understanding nonketotic hyperglycinemia. J Biol Chem 279(48):
50514–50523.
30. Hiratsuka T, et al. (2008) An alternative menaquinone biosynthetic pathway oper-
ating in microorganisms. Science 321(5896):1670–1673.
31. Ahmad M, et al. (2011) Identification of DypB from Rhodococcus jostii RHA1 as a
lignin peroxidase. Biochemistry 50(23):5096–5107.
32. Van der Geize R, et al. (2007) A gene cluster encoding cholesterol catabolism in a soil
actinomycete provides insight into Mycobacterium tuberculosis survival in macro-
phages. Proc Natl Acad Sci USA 104(6):1947–1952.
33. Marín L, Miguélez E-M, Villar C-J, Lombó F (2015) Bioavailability of dietary poly-
phenols and gut microbiota metabolism: antimicrobial properties. BioMed Res Int
2015:905215.
6. Umezawa T (2003) Diversity in lignan biosynthesis. Phytochem Rev 2(3):371–390.
7. Fukuda Y, Nagata M, Osawa T, Namiki M (1986) Chemical aspects of the antioxidative
activity of roasted sesame seed oil, and the effect of using the oil for frying. Agric Biol
Chem 50(4):857–862.
8. Hirose N, et al. (1991) Inhibition of cholesterol absorption and synthesis in rats by
sesamin. J Lipid Res 32(4):629–638.
9. Tsuruoka N, Kidokoro A, Matsumoto I, Abe K, Kiso Y (2005) Modulating effect of
sesamin, a functional lignan in sesame seeds, on the transcription levels of lipid- and
alcohol-metabolizing enzymes in rat liver: A DNA microarray study. Biosci Biotechnol
Biochem 69(1):179–188.
10. Miyawaki T, et al. (2009) Antihypertensive effects of sesamin in humans. J Nutr Sci
Vitaminol (Tokyo) 55(1):87–91.
11. Utsunomiya T, Chavali S-R, Zhong W-W, Forse R-A (2000) Effects of sesamin-supple-
mented dietary fat emulsions on the ex vivo production of lipopolysaccharide-
induced prostanoids and tumor necrosis factor alpha in rats. Am J Clin Nutr 72(3):
804–808.
12. Yasuda K, Ikushiro S, Kamakura M, Ohta M, Sakaki T (2010) Metabolism of sesamin by
cytochrome P450 in human liver microsomes. Drug Metab Dispos 38(12):2117–2123.
13. Sakaki T, et al. (2012) Metabolism of sesamin by CYPs and UGTs in human liver. J Food
Drug Anal 20(1):376–379.
14. Moazzami A-A, Andersson R-E, Kamal-Eldin A (2007) Quantitative NMR analysis of a
sesamin catechol metabolite in human urine. J Nutr 137(4):940–944.
15. Miyake Y, et al. (2005) Antioxidative catechol lignans converted from sesamin and
sesaminol triglucoside by culturing with Aspergillus. J Agric Food Chem 53(1):22–27.
16. Liu Z, Saarinen N-M, Thompson L-U (2006) Sesamin is one of the major precursors of
mammalian lignans in sesame seed (Sesamum indicum) as observed in vitro and in
rats. J Nutr 136(4):906–912.
17. Kiso Y (2004) Antioxidative roles of sesamin, a functional lignan in sesame seed, and
it’s effect on lipid- and alcohol-metabolism in the liver: A DNA microarray study.
Biofactors 21(1-4):191–196.
34. Nishiyama T, Hashimoto Y, Kusakabe H, Kumano T, Kobayashi M (2014) Natural low-
molecular mass organic compounds with oxidase activity as organocatalysts. Proc Natl
Acad Sci USA 111(48):17152–17157.
35. Tucker G-T, et al. (1994) The demethylenation of methylenedioxymethamphetamine
(“ecstasy”) by debrisoquine hydroxylase (CYP2D6). Biochem Pharmacol 47(7):
1151–1156.
36. Buswell J-A, Cain R-B (1973) Microbial degradation of piperonylic acid by Pseudo-
monas fluorescens. FEBS Lett 29(3):297–300.
18. Prasad
K (2000) Antioxidant activity of secoisolariciresinol diglucoside-derived
37. Hasse D, et al. (2013) Structure of the homodimeric glycine decarboxylase P-protein
from Synechocystis sp. PCC 6803 suggests a mechanism for redox regulation. J Biol
Chem 288(49):35333–35345.
metabolites, secoisolariciresinol, enterodiol, and enterolactone. Int J Angiol 9(4):
220–225.
19. Peterson J, et al. (2010) Dietary lignans: Physiology and potential for cardiovascular
disease risk reduction. Nutr Rev 68(10):571–603.
20. Cunnane S-C, et al. (1995) Nutritional attributes of traditional flaxseed in healthy
young adults. Am J Clin Nutr 61(1):62–68.
38. Okamura-Ikeda K, et al. (2010) Crystal structure of aminomethyltransferase in com-
plex with dihydrolipoyl-H-protein of the glycine cleavage system: Implications for
recognition of lipoyl protein substrate, disease-related mutations, and reaction
mechanism. J Biol Chem 285(24):18684–18692.
6 of 6
|
www.pnas.org/cgi/doi/10.1073/pnas.1605050113
Kumano et al.