In vitro demethoxycarbonylation of various chlorophyll analogs by a BciC enzyme

Article abstract of DOI:10.1007/s11120-018-0573-1

Unique light-harvesting antennas in the green sulfur bacterium Chlorobaculum tepidum, called chlorosomes, consist of self-aggregates of bacteriochlorophyll (BChl) c. In the biosynthesis of BChl c, BciC demethoxycarbonylase removes the C13²-methoxycarbonyl group to facilitate the self-aggregation of BChl c. We previously reported the in vitro BciC-enzymatic reactions and discussed the function of this enzyme in the biosynthesis of BChl c. This study aims to examine the substrate specificity of BciC in detail using several semi-synthetic (bacterio)chlorophyll derivatives. The results indicate that the substrate specificity of BciC is measurably affected by structural changes on the A/B rings including the bacteriochlorin π-systems. Moreover, BciC showed its activity on a Zn-chelated chlorophyll derivative. On the contrary, BciC recognized structural modifications on the D/E rings, including porphyrin pigments, which resulted in the significant decrease in the enzymatic activity. The utilization of BciC provides mild conditions that may be useful for the in vitro preparation of various chemically (un)stable chlorophyllous pigments.

Full text of DOI:10.1007/s11120-018-0573-1

Photosynthesis Research
https://doi.org/10.1007/s11120-018-0573-1
ORIGINAL ARTICLE
In vitro demethoxycarbonylation of various chlorophyll analogs
by a BciC enzyme
1
2
1
Misato Teramura · Jiro Harada · Hitoshi Tamiaki
Received: 6 April 2018 / Accepted: 11 August 2018
©
Springer Nature B.V. 2018
Abstract
Unique light-harvesting antennas in the green sulfur bacterium Chlorobaculum tepidum, called chlorosomes, consist of
self-aggregates of bacteriochlorophyll (BChl) c. In the biosynthesis of BChl c, BciC demethoxycarbonylase removes the
2
C13 -methoxycarbonyl group to facilitate the self-aggregation of BChl c. We previously reported the in vitro BciC-enzymatic
reactions and discussed the function of this enzyme in the biosynthesis of BChl c. This study aims to examine the substrate
specificity of BciC in detail using several semi-synthetic (bacterio)chlorophyll derivatives. The results indicate that the
substrate specificity of BciC is measurably affected by structural changes on the A/B rings including the bacteriochlorin
π-systems. Moreover, BciC showed its activity on a Zn-chelated chlorophyll derivative. On the contrary, BciC recognized
structural modifications on the D/E rings, including porphyrin pigments, which resulted in the significant decrease in the
enzymatic activity. The utilization of BciC provides mild conditions that may be useful for the in vitro preparation of various
chemically (un)stable chlorophyllous pigments.
Keywords Bacteriochlorophyll · Biosynthesis · Chlorosome · Demethoxycarbonylase · Green sulfur bacterium
Abbreviations
tR
Retention time
3-Vinyl
3
Ac
3-Acetyl
3V
BChl
Bacteriochlorophyll
BChlide Bacteriochlorophyllide
Cba.
Chl
Chlorobaculum
Introduction
Chlorophyll
Chlide
E.
Chlorophyllide
In nature, a number of chlorophylls (Chls) are found in dif-
ferent organisms and function as primary photosyntheti-
cally active pigments. The biosynthetic pathway of Chls
is branched from that of other physiologically important
compounds, such as hemes and bilins, by the insertion
of magnesium at the central position of protoporphyrin
IX (Walker and Willows 1997; Papenbrock et al. 2000).
Escherichia
ESI
Electrospray ionization
3-(1-Hydroxyethyl)
Liquid chromatography–mass spectrometry
Photodiode array
Pheophorbide
3
HE
LCMS
PDA
Pheide
RP
2
Reversed phase
The free C13 -carboxy group of Mg-protoporphyrin IX
is regioselectively esterified with S-adenosylmethionine
to its monomethyl ester by a methyltransferase (BchM or
ChlM) (Bollivar et al. 1994; Gibson and Hunter 1994). The
resulting Mg-protoporphyrin IX monomethyl ester then
converted to 3,8-divinyl-protochlorophyllide a through the
formation of an isocyclic ring (E ring), which is a signa-
ture moiety of Chl pigments that differentiates them from
other cyclic tetrapyrrole pigments (Ouchane et al. 2004;
Minamizaki et al. 2008; Chen et al. 2018). This step is
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s11120-018-0573-1) contains
supplementary material, which is available to authorized users.
*
Hitoshi Tamiaki
tamiaki@fc.ritsumei.ac.jp
1
2
Graduate School of Life Sciences, Ritsumeikan University,
Kusatsu, Shiga 525-8577, Japan
1
accompanied by the formation of a C13 -oxo group and a
Department of Medical Biochemistry, Kurume University
School of Medicine, Kurume, Fukuoka 830-0011, Japan
2
C13 -methoxycarbonyl group on the E ring (see Fig. 1, left).
Vol.:(0
1
123456789)
3

Photosynthesis Research
cell lysate of Escherichia (E.) coli expressing bciC gene,
the precipitates of the cells harvested and broken were sus-
pended in 50 mM Tris–HCl buffer (pH 7.8), and used for
enzymatic assays.
3
7
A
B
8
N
N
N
N
N
N
N
N
BciC
Mg
Mg
The Chl c substrate was extracted as a mixture of Chl c
1
D
C
and Chl c from commercially available diatoms Chaetoc-
2
17
E
13
eros gracilis using a mixture of pyridine and chloroform as
1
32
1
31
previously reported (Mizoguchi et al. 2009). A mixture of
O
O
O
OCH3
Chl c and Chl c , in approximate ratio of 1:1, was used as
1 2
O
O
OH
OH
pyroChlide a
the substrate in the BciC-enzymatic reaction.
Chlide a
Chl a was extracted from commercially available, cyano-
bacterium, Spirulina geitleri with a mixture of petroleum
ether and methanol (1:1, v/v) and purified according to pre-
vious report (Iriyama et al. 1974).
2
Fig. 1 C13 -demethoxycarbonylation of chlorophyllide a to pyrochlo-
rophyllide a by BciC
Chl b was extracted from commercially available spinach
powder with a mixture of petroleum ether and methanol (1:1,
v/v) (Oba et al. 1997). After centrifugation of the extract, the
resulting supernatant was partitioned between diethyl ether
and distilled water. The ethereal phase was washed with dis-
tilled water several times and concentrated. The residue was
purified by reversed phase (RP) HPLC to yield Chl b.
Chl d was obtained from cyanobacterium Acaryochloris
marina cells using acetone and methanol (4:1, v/v) (Mizo-
guchi et al. 2006), similarly to the extraction of Chl b, and
purified by RP-HPLC.
The C13-keto carbonyl group is conserved in the molecular
structure of all Chls found as components of photosynthetic
apparatuses. On the other hand, some Chl pigments lacking
2
the C13 -methoxycarbonyl group are observed in specific
organisms, while most of Chls conserve this functional
group.
Bacteriochlorophylls (BChls) c/d/e/f are found in unique
light-harvesting antennas called chlorosomes of some pho-
tosynthetic bacteria, green sulfur bacteria, filamentous
anoxygenic phototrophs, and phototrophic acidobacteria
(
Olson 1998; Blankenship and Mastuura 2003; Bryant et al.
BChl a (see below) was oxidized to 3-acetyl-Chl a
(3Ac-Chl a) via selective dehydrogenation of C7H–C8H to
C7=C8 using 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(Teramura and Tamiaki 2018). The C3-acetyl group of 3Ac-
Chl a was reduced using tert-butylamine borane in tetrahy-
drofuran to yield 3-(1-hydroxyethyl)-Chl a (3HE-Chl a) as
2
007; Harada et al. 2018). The chlorosomal pigments self-
aggregate by intermolecular interactions to be packed in
2
chlorosomes. The absence of the C13 -methoxycarbonyl
group in their chemical structures results in less steric hin-
drance around the C13-keto group, thus enhancing aggre-
gation (Oba and Tamiaki 1999). Liu and Bryant reported
that the bciC gene in the green sulfur bacterium Chloro-
baculum (Cba.) tepidum, is related to the removal of the
1
a 1:1 mixture of the 3 R/S-epimers (Kunieda et al. 2004).
Chlorophyllase derived from Chenopodium album
(Tsuchiya et al. 1999) was prepared by expressing in E. coli
BL21(DE3), as previously described (Tsukatani et al. 2013).
It was used to remove the phytyl group at the C17-propionate
residue of Chl a, Chl b, Chl d, 3Ac-Chl a, and 3HE-Chl a.
Each purified Chl in acetone was incubated with the E. coli
lysate containing chlorophyllase in aqueous Tris–HCl buffer
(pH 7.8) at room temperature to yield Chlide a, Chlide b,
Chlide d, 3Ac-Chlide a, or 3HE-Chlide a (Teramura and
Tamiaki 2018).
2
C13 -methoxycarbonyl group in the biosynthesis of BChl
c (Liu and Bryant 2011). Later, we reported the in vitro
reaction using BciC enzyme (Fig. 1), which functions as a
demethoxycarbonylase. In that work, we concluded that the
2
C13 -demethoxycarbonylation of chlorophyllide a (Chlide
a) branches the biosynthetic pathway of BChl c from that
of the other Chls (Teramura et al. 2016a). Here, we report a
more detailed analysis on the substrate specificity of BciC
demethoxycarbonylase derived from Cba. tepidum.
Chlide a was converted into zinc pheophorbide a (Zn-
Pheide a) by demetalation with H SO followed by a meta-
2
4
lation reaction with zinc acetate dihydrate in acetone (Tera-
Materials and methods
mura and Tamiaki 2018).
BChl a was extracted from the cultured cells of the pur-
ple bacterium Rhodobacter sphaeroides using acetone and
methanol (4:1, v/v) (Tamiaki et al. 1998). BChl a was puri-
fied by normal phase HPLC.
Preparation of BciC enzyme and substrates
The BciC enzyme derived from Cba. tepidum was prepared
according to previously reported methods (Teramura et al.
3-Vinyl-BChl a (3V-BChl a) was obtained from the cul-
tured ΔbchF mutant cells of R. sphaeroides (Teramura et al.
2
016a). Because BciC was not able to be isolated from the
1
3

Photosynthesis Research
2
016a) in a manner similar to the extraction of BChl a, and
(3.0 ϕ × 75 mm), MeOH/50 mM ammonium acetate (pH
−
1
was purified by RP-HPLC. The fraction containing 3V-BChl
a was dialyzed with 50 mM phosphate buffer (pH 7.8) for
the following enzymatic reaction.
5.2)=75/25 at 0.4 mL min ; for 3HE-Chlide a, Cosmosil
2.5C -MS-II (3.0 ϕ×75 mm), MeOH/50 mM ammonium
1
8
−
1
acetate (pH 5.2) = 65/35 at 0.6 mL min ; for 3Ac-Chlide
BChl a and 3V-BChl a in acetone were treated with a
chlorophyllase solution at 50 °C to yield BChlide a and
a, Cosmosil 2.5C -MS-II (3.0 ϕ×75 mm), MeOH/50 mM
1
8
−
1
ammonium acetate (pH 5.2)= 85/15 at 0.6 mL min ; for
3
V-BChlide a according to the aforementioned hydrolysis
Zn-Pheide a, Cosmosil 2.5C -MS-II (3.0 ϕ × 75 mm),
1
8
of Chls to Chlides (Teramura and Tamiaki 2018).
MeOH/50 mM ammonium acetate (pH 5.2) = 80/20 at
−
1
0
.6 mL min ; for BChlide a, Cosmosil 2.5C -MS-II
18
Activity assays of BciC in vitro
(3.0 ϕ × 75 mm), MeOH/50 mM ammonium acetate (pH
−
1
5
.2)=80/20 at 0.6 mL min ; for 3V-BChlide a, Cosmosil
The enzymatic assay of BciC derived from Cba. tepidum
was performed in 1 mL of 25 mM phosphate buffer (pH 7.0)
containing 100 mM NaCl, 5% dimethyl sulfoxide, 10 µM
chlorophyllous pigments as substrates, and 13 µg (or 650 µg
for a 50-fold amount) of proteins containing BciC. After
incubation for 10 min at 45 °C in the dark, 1.5 mL of acetone
was added and the mixture was centrifuged to precipitate
the proteins. Acetone was evaporated from the supernatant
under a stream of nitrogen gas, the resulting solution was
filtered with a COSMONICE® filter (Nacalai Tesque, Japan)
for HPLC analyses.
2.5C -MS-II (3.0 ϕ×75 mm), MeOH/50 mM ammonium
18
−
1
acetate (pH 5.2)=85/15 at 0.6 mL min .
2
Assignments of C13 ‑demethoxycarbonylated
products
The BciC-catalyzed products, pyro(B)Chls, were character-
ized by their visible absorption and mass spectra in LCMS.
The slow elution of the products in RP-HPLC supported the
2
removal of a methoxycarbonyl group at the C13 -position,
which increases their hydrophobicity (Teramura and Tami-
aki 2018).
3
V-BChlide a was not isolated due to its instability, such
2
as demetalation; thus, the hydrolysis reaction by chlorophyl-
lase and BciC treatment were carried out sequentially. A
buffer mixture containing 3V-BChl a was incubated with
chlorophyllase for 10 min at room temperature, then BciC
was added, and the mixture was incubated and analyzed
similarly to the assays of the other pigments.
13 -Demethoxycarbonyl-Chlide b (pyroChlide b): VIS
(MeOH/aq. 50 mM AcONH =75/25) λ =651 (relative
4
max
intensity, 0.34), 603 (0.12), 469 nm (1.00); MS (ESI) found:
+
m/z 571.5, calcd for C H N O Mg: MH , 571.2.
3
3
31
4
4
2
13 -Demethoxycarbonyl-Chlide d (pyroChlide d): VIS
(MeOH/aq. 50 mM AcONH =75/25) λ =702 (relative
4
max
intensity, 1.00), 456 (0.76), 401 nm (0.69); MS (ESI) found:
+
Reaction analysis by HPLC
m/z 559.5, calcd for C H N O Mg: MH , 559.2.
3
2
31
4
4
2
3
-Acetyl-13 -demethoxycarbonyl-Chlide a (3Ac-pyro-
The enzymatic reactions were analyzed by HPLC using a
Shimadzu Prominence liquid chromatography system con-
sisting of a CBM-20A system controller, an SPD-M20A
photodiode array detector (PDA), LC-20AD pumps, a DGU-
Chlide a): VIS (MeOH/aq. 50 mM AcONH = 85/15)
4
λ
= 688 (relative intensity, 1.00), 444 (0.97),
max
397 nm (0.91); MS (ESI) found: m/z 573.3, calcd for
+
C H N O Mg: MH , 573.2.
3
3
33
4
4
2
2
0A3 degasser, and a CTO-20AC column oven (Japan). For
13 -Demethoxycarbonyl-3-(1-hydroxyethyl)Chlide a (3HE-
all the HPLC analysis, the column oven was set to 35 °C,
and the PDA detector was set to 300–800 nm. Liquid chro-
matography–mass spectrometry (LCMS) was performed
using a Shimadzu LCMS-2010EV system consisting of an
electrospray ionization (ESI) quadrupole mass spectrometer.
All the solvents used for HPLC were of HPLC grade and
purchased from Nacalai Tesque or Wako Pure Chemical Ind.
RP-HPLC analyses of the enzymatic reactions were
carried out under the following conditions: for Chl c /c ,
pyroChlide a): VIS (MeOH/aq. 50 mM AcONH =65/35)
4
λ
=659 (relative intensity, 1.00), 608 (0.22), 429 (0.89),
max
412 nm (0.88); MS (ESI) found: m/z 575.5, calcd for
+
C H N O Mg: MH , 575.3.
3
3 35 4 4
2
Zinc 13 -demethoxycarbonyl-Chlide a (Zn-pyroChlide
a): VIS (MeOH/aq. 50 mM AcONH = 80/20) λ = 661
4
max
(relative intensity, 0.84), 615 (0.20), 430 nm (1.00); MS
+
(ESI) found: m/z 596.3, calcd for C H N O Mg: M , 596.2
33
32
4
3
(Teramura and Tamiaki 2018).
1
2
2
Inertsil ODS-P (4.6 ϕ × 150 mm, GL Science Inc.), and
13 -Demethoxycarbonyl-BChlide a (pyroBChlide a): VIS
MeOH/50 mM ammonium acetate (pH 5.2) = 85/15
(MeOH/aq. 50 mM AcONH =80/20) λ =777 (relative
4
max
−
1
at 1.0 mL min ; for Chl a, Cosmosil 2.5C -MS-II
intensity, 1.00), 613 (0.24), 363 nm (0.93); MS (ESI) found:
1
8
−
1
+
(
3.0 ϕ×75 mm, Nacalai Tesque), MeOH at 0.6 mL min ;
m/z 575.5, calcd for C H N O Mg: MH , 575.3.
3
3
35
4
4
2
for Chlide b, Cosmosil 2.5C -MS-II (3.0 ϕ × 75 mm),
13 -Demethoxycarbonyl-3-vinyl-BChlide a (3V-pyroB-
1
8
MeOH/50 mM ammonium acetate (pH 5.2) = 75/25 at
Chlide a): VIS (MeOH/aq. 50 mM AcONH = 85/15)
4
−
1
0
.4 mL min ; for Chlide d, Cosmosil 2.5C -MS-II
λ
= 728 (relative intensity, 0.67), 574 (0.24), 352 nm
1
8
max
1
3

Photosynthesis Research
(
1.00); MS (ESI) found: m/z 559.6, calcd for C H N O Mg:
of porphyrin (PChlide a and DV-PChlide a, see Table 1)
or bacteriochlorin pigments (3V-BChlide a) under the
examined conditions. In this study, Chl c was used for
the BciC assay as a different type of porphyrin substrate
that possesses an acrylate residue at the C17-position,
instead of the C17-propionate residue (Table 1). We used
a mixture of Chl c and Chl c as substrates but none of
33
35
4
3
+
MH , 559.3.
Results and discussion
Effect of the π‑conjugation degree
1
2
the desired products were detected in the HPLC chroma-
togram of the reaction mixture (Fig. S1). Furthermore,
using a 50-fold amount of BciC in comparison with the
previous report, no BciC-catalyzed reaction proceeded.
No porphyrin pigments could be demethoxycarbonylated
We previously reported that BciC recognized the
π-conjugation degree of its cyclic tetrapyrrole substrate
(
Teramura et al. 2016a). BciC showed its activity on chlo-
rin pigments (Chlide a and DV-Chlide a, see Table 1),
2
while BciC did not catalyze the demethoxycarbonylation
at the C13 -position by the treatment with BciC.
Table 1 Enzymatic activity of BciC on various Chlide derivatives
R3
R7
7
R8
8
N
N
M
N
N
1
8
1
7
1
32
O
O
OCH3
O
O
R17
Substrate
Core π-systema
M
R3
R7
R8
R17
BciC activity
−
Chl c /c2
P
P
P
Mg
Mg
Mg
Mg
Mg
Mg
Mg
Mg
Mg
Mg
Mg
Mg
Mg
Mg
H2
CHCH2
CHCH2
CHCH2
CHCH2
CHCH2
CHCH2
CHCH2
CH(OH)CH3
COCH3
COCH3
CHCH2
CHCH2
CHO
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CHO
CHO
CH3
CH3
CH3
CH3
CH3
CH3
CH CH /CHCH
CH CH
CHCH2
CH CH
CH CH
CH CH
CHCH2
CH CH
CH CH
CH CH
CH CH
CH CH
CH CH
CH CH
CH CH
CH CH
Hb
H
H
phytyl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
1
2
3
3
2
d
PChlide a
DV-PChlide a
Chl a
Chlide a
Chlide a’
−
−
2
d
C
C
C
C
C
C
C
C
C
C
C
C
C
B
B
−
2
3
3
3
d
++
2
c
d
−
2
d
DV-Chlide a
++
3HE-Chlide a
3Ac-Chlide a
3Ac-Chlide a’
++
++
−
++
−
2
3
3
3
3
3
3
3
3
3
3
3
2
c
2
Chlide b
Chlide b’
Chlide d
Chlide d’
Pheide a
Zn-Pheide a
BChlide a
2
c
2
++
−
2
c
CHO
2
d
CHCH2
CHCH2
COCH3
CHCH2
−
2
Zn
Mg
Mg
++
+
+
2
CH CH
CH CH
2
3
V-BChlide a
2
a
P: porphyrin with C7=C8 and C17=C18; C: chlorin with C7=C8 and C17(S)H–C18(S)H; B: bacteriochlorin with C7(R)H–C8(R)H and
C17(S)H–C18(S)H
b
1
7-Acrylate residue; the other substrates possess the 17-propionate residue
c
2
(
13 S)-Epimer
d
Teramura et al. (2016a)
1
3

Photosynthesis Research
Fig. 2 BciC activity assay with 3V-BChlide a. HPLC profiles show
before (upper) and after the 50-fold enzymatic reaction (lower). Reac-
tion and HPLC conditions are shown in “Materials and methods” sec-
tion
1
1
Fig. 3 BciC activity assay with a (3 R)- and (3 S)-epimeric mixture
of 3HE-Chlide a. HPLC profiles show before (upper) and after the
enzymatic reaction (lower). Reaction and HPLC conditions are shown
in “Materials and methods” section
Next, we examined the reaction of 3V-BChlide a with
et al. 2016a). This showed that BciC activity might be
relatively unaffected by changes of peripheral substituents.
Based on the above, we prepared chlorophyllous pigments
bearing different substituents as substrates for the BciC
assay. The reaction of Chl a with BciC did not afford the
a 50-fold amount of BciC. Figure 2 shows a new peak at
3
.5 min along with the decrease of the 2.7-min band which
corresponds to 3V-BChlide a in the reaction mixture. The
new peak was assigned to the demethoxycarbonylated prod-
uct, 3V-pyroBChlide a, based on its visible absorption and
mass spectra (see “Materials and methods”). It is noteworthy
that the relatively small amount of BciC, which is enough
for demethoxycarbonylation of Chlide a, was not able to
produce 3V-pyroBChlide a in a given time. The structural
difference between Chlide a and BChlide a is only whether
the C7=C8 double bond is reduced to C7–C8 single bond,
i.e., 3V-BChlide a has a bacteriochlorin while Chlide a has a
chlorin as their structural skeletons (Table 1). This indicates
that BciC recognizes a bacteriochlorin pigment (3V-BChlide
a) as substrate, but the reaction occurred more slowly than
that of the corresponding chlorin π-conjugate (Chlide a).
2
C13 -demethoxycarbonylated product (see Fig. S2). This
result would be ascribable to the bulkiness and/or the hydro-
phobicity of the esterifying moiety (the phytyl group) of
the C17-propionate residue in Chl a structure. BciC might
not be able to accommodate this long and bulky group in
its substrate-binding site, or key amino acids to remove the
2
C13 -methoxycarbonyl group may not access to the sub-
strate. The other possible reason is that Chl a has a lower
solubility in the buffer under the present conditions due to
its hydrophobicity, and it might make it difficult for BciC
to get access to the substrate. Next, three C3-modified
Chlides a, i.e., 3HE-Chlide a [3-CH(OH)Me], 3Ac-Chlide a
(
3-COMe), Chlide d (3-CHO), were examined as substrates.
1
Effect of the peripheral substituents
As shown in Fig. 3, a 3 -epimeric mixture of 3HE-Chlide a
was completely consumed by the enzymatic action of BciC
(45 °C, 10 min), and two new peaks appeared at 15.2 and
18.6 min. They were assigned to the R- and S-epimers of
3HE-pyroChlide a, respectively. BciC successfully removed
As previously reported, BciC catalyzes demethoxycar-
bonylation of DV-Chlide a possessing the 8-vinyl group,
and of Chlide a possessing the 8-ethyl group (Teramura
1
3

Photosynthesis Research
Fig. 4 BciC activity assay with 3Ac-Chlide a. HPLC profiles show
before (upper) and after the enzymatic reaction (lower). Reaction and
HPLC conditions are shown in “Materials and methods” section
Fig. 5 BciC activity assay with Chlide d. HPLC profiles show before
(upper) and after the enzymatic reaction (lower). Reaction and HPLC
conditions are shown in “Materials and methods” section
2
the C13 -methoxycarbonyl group in the hydrated derivative
of Chlide a. It is worth noting that no epimerization at the
t = 9.0 min (Fig. 6). As with Chl a′ (see above), no
R
1
C3 -position occurred during the enzymatic reaction. Fig-
demethoxycarbonylation of Chlide b′ (t = 10.5 min)
R
ure 4 shows that 3Ac-Chlide a at 2.1 min was enzymatically
converted into 3Ac-pyroChlide a at 3.4 min, similar to 3HE-
Chlide a to 3HE-pyroChlide a conversion (vide supra). The
BciC-catalyzed reaction proceeded with the 3-acetylchlo-
rin, the oxidized (ketone) form of 3HE-Chlide a (second-
ary alcohol). As shown in Fig. 5, Chlide d at 5.1 min was
also demethoxycarbonylated by BciC and pyroChlide d was
detected at 8.5 min. This demonstrates that BciC is active
was visible. The BciC activity was thus observed despite
chemical modifications at the C7-position of substrate:
C7-methyl and formyl groups.
The effect of C3- and C7-substituents in Chlides on
the BciC activity was investigated at the initial stage of
enzymatic reactions (Fig. S3). The production rates of the
demethoxycarbonylation of Chlides modified on the A
ring, i.e., 3Ac-Chlide a, 3HE-Chlide a, and Chlide d, by
BciC suppressed to the one-third of Chlide a, while that
of the Chlide modified on the B ring, Chlide b, decreased
to the two-thirds of Chlide a. The latter is identical with
the previous observation that the BciC activity on DV-
Chlide a, one of the B-ring modified Chlides, was the
two-thirds on Chlide a (Teramura et al. 2016a). It is wor-
thy to note that the enzymatic demethoxycarbonylation
of the C3-acetyl-type of bacteriochlorin (BChlide a) was
partially observed with a 50-fold amount of BciC (Fig. 7):
1
on Chlide d, the C3 -demethylated form of 3Ac-Chlide a.
The starting materials of 3Ac-Chlide a and Chlide d
2
contained small amounts of their (13 S)-stereoisomers,
3
Ac-Chlide a′ and Chlide d′. After the enzymatic reactions
described above (Figs. 4, 5), the epimers remained at reten-
tion times (t )=3.9 (3Ac-Chlide a′) and 9.9 min (Chlide d′).
R
2
Thus, BciC catalyzed only the (13 R)-epimers, 3Ac-Chlide
a and Chlide d, but not their epimers (prime forms). This
stereospecificity is consistent with the previous observation
for Chlide a/a′ (Teramura et al. 2016a).
t = 1.1 (BChlide a) → 1.3 min (pyroBChlide a). Based
R
Chlide b possesses a formyl group at the C7-posi-
tion and its BciC reaction yielded the expected
on the above results, structural changes at the C3- and
2
C7-positions of 17 -carboxychlorins have less effects on
2
C13 -methoxycarbonylated product (pyroChlide b) at
the substrate recognition of BciC, but trans-hydrogenation
1
3

Photosynthesis Research
Fig. 6 BciC activity assay with Chlide b. HPLC profiles show before
Fig. 7 BciC activity assay with BChlide a. HPLC profiles show
before (upper) and after the 50-fold enzymatic reaction (lower). Reac-
tion and HPLC conditions are shown in “Materials and methods” sec-
tion
(
upper) and after the enzymatic reaction (lower). Reaction and HPLC
conditions are shown in “Materials and methods” section
at the C7=C8 bond of chlorin to bacteriochlorin remarka-
bly slowed down the enzymatic reactions, and the stereoin-
et al. 1996), tolerate replacement of a magnesium by a zinc.
Because BciC potentially catalyzes demethoxycarbonylation
of such zinc complexes, a variety of Chl derivatives will be
synthesized using BciC.
2
2
2
version at the C13 -position of 13 R to 13 S (prime form)
suppressed the BciC-catalyzed reactions in a given time.
Effect of the central metal
It was previously demonstrated that BciC did not show its
activity on Pheide a (Teramura et al. 2016a), indicating that
BciC required a magnesium at the center of substrates. In
this study, we prepared a zinc-chelated Pheide a, Zn-Pheide
a. This substrate was successfully demethoxycarbonylated
Conclusion
2
BciC-catalyzed C13 -demethoxycarbonylation was toler-
ant for the chemical modification of substituents on the
A and B rings of the analyzed substrates, and, to a lower
extent, hydrogenation of the B ring (Table 1). However,
structural changes on the D and E rings, including esteri-
fication of the C17-propionate residue and epimerization
by BciC, as shown in Fig. 8: t = 6.9 (Zn-Pheide
R
a)→11.6 min (Zn-pyroPhlide a). All natural BChls func-
tioning in chlorosomes have a magnesium ion at the cen-
tral position. This experiment demonstrates that the central
magnesium of BciC substrates may be replaced with a zinc
ion. Zn-chelated derivatives are often used for model pig-
ments of naturally occurring (B)Chls owing to their in vitro
high chemical stability in comparison with acid-labile Mg-
chelated Chls. It is also known that many enzymes work-
ing on chlorophyllous pigments in nature (Griffiths 1980;
Helfrich and Rüdiger 1992; Kiesel et al. 2015; Teramura
et al. 2016b, 2018), as well as Chl-binding proteins (Wakao
2
of C13 -position, completely abolished the BciC activity.
These observations can be explained by the structural factors
of substrates: the A/B rings are located in the “northern”
part of cyclic tetrapyrroles in (B)Chls, far from the reactive
2
C13 -position, while the D/E rings are in the “southern”
2
part, near the C13 -site. Since BciC was not only active
on Chlide a but also on 3HE-Chlide a, it is possible that
this demethoxycarbonylation occurs after the hydration of
the C3-vinyl group by BchF/BchV in the biosynthesis of
1
3

Photosynthesis Research
References
Blankenship RE, Mastuura K (2003) Antenna complexes from green
photosynthetic bacteria. In: Green BR, Parson WW (eds) Light-
harvesting antennas in photosynthesis. Kluwer Academic Publish-
ers, Dordrecht, pp 195–217
Bollivar DW, Jiang ZY, Bauer CE, Beale SI (1994) Heterologous
expression of the bchM gene product from Rhodobacter cap-
sulatus and demonstration that it encodes S-adenosyl-l-me-
thionine: Mg-protoporphyrin IX methyltransferase. J Bacteriol
1
76:5290–5296
Bryant DA, Costas AMG, Maresca JA, Chew AGM, Klatt CG, Bateson
MM, Tallon LJ, Hostetler J, Nelson WC, Heidelberg JF, Ward
DM (2007) Candidatus Chloracidobacterium thermophilum: an
aerobic phototrophic acidobacterium. Science 317:523–526
Chen GE, Canniffe DP, Barnett SFH, Hollingshead S, Brindley AA,
Vasilev C, Bryant DA, Hunter N (2018) Complete enzyme set for
chlorophyll biosynthesis in Escherichia coli. Sci Adv 4:eaaq1407
Gibson LC, Hunter CN (1994) The bacteriochlorophyll biosynthesis
gene, bchM, of Rhodobacter sphaeroides encodes S-adenosyl-
l-methionine: Mg protoporphyrin IX methyltransferase. FEBS
Lett 352:127–130
Griffiths WT (1980) Substrate-specificity studies on protochlorophyl-
lide reductase in barley (Hordeum vulgare) etioplast membranes.
Biochem J 186:267–278
Harada J, Teramura M, Mizoguchi T, Tsukatani Y, Yamamoto K, Tami-
aki H (2015) Stereochemical conversion of the 3-vinyl group to
1
-hydroxyethyl group in bacteriochlorophyll c by the hydratases
BchF and BchV: adaptation of green sulfur bacteria to limited-
light environments. Mol Microbiol 98:1184–1198
Harada J, Shibata Y, Teramura M, Mizoguchi T, Kinoshita Y, Yama-
moto K, Tamiaki H (2018) In vivo excited energy transfer of
bacteriochlorophyll c, d, e, or f to bacteriochlorophyll a in the
wild-type and mutant cells of the green sulfur bacterium Chlo-
robaculum limnaeum. ChemPhotoChem 2:190–195
Fig. 8 BciC activity assay with Zn-Pheide a. HPLC profiles show
before (upper) and after the enzymatic reaction (lower). Reaction and
HPLC conditions are shown in “Materials and methods” section
Helfrich M, Rüdiger W (1992) Various metallopheophorbides as sub-
strates for chlorophyll synthetase. Z Naturforsc C 47:231–238
Iriyama K, Ogura N, Takamiya A (1974) A simple method for extrac-
tion and partial purification of chlorophyll from plant material,
using dioxane. J Biochem 76:901–904
chlorosomal BChls. This hypothesis is consistent with the
substrate specificities of BchF/BchV that the hydratases can
hydrate the C3-vinyl group of substrates with or without
2
C13 -methoxycarbonyl group (Harada et al. 2015). This
Kiesel S, Wätzlich D, Lange C, Reijerse E, Bröcker MJ, Rüdiger
W, Lubitz W, Scheer H, Moser J, Jahn D (2015) Iron-sulfur
cluster-dependent catalysis of chlorophyllide a oxidoreductase
from Roseobacter denitrificans. J Biol Chem 290:1141–1154
Kunieda M, Mizoguchi T, Tamiaki H (2004) Diastereoselective self-
aggregation of synthetic 3-(1-hydroxyethyl)-bacteriopyrochlo-
rophyll-a as a novel photosynthetic antenna model absorbing
near the infrared regions. Photochem Photobiol 79:55–61
Liu Z, Bryant DA (2011) Identification of gene essential for the first
committed step in the biosynthesis of bacteriochlorophyll c. J
Biol Chem 286:22393–22402
allows the presence of multiple pathways for the chloroso-
mal BChl biosynthesis from the viewpoint of the substrate
specificity of enzymes. Although conventional methods to
2
remove the C13 -methoxycarbonyl group have been per-
formed by high-temperature treatments under basic condi-
tions (e.g., reflux in collidine for several hours) (Tamiaki
et al. 2012), the enzymatic use of BciC provides a new
method to prepare various Chl derivatives under mild con-
ditions (45 °C, 10 min).
Minamizaki K, Mizoguchi T, Goto T, Tamiaki H, Fujita Y (2008)
Identification of two homologous genes, chlA and chlA , that
I
II
Acknowledgements This work was partially supported by JSPS
are differentially involved in isocyclic ring formation of chlo-
rophyll a in the cyanobacterium Synechocystis sp. PCC 6803. J
Biol Chem 283:2684–2692
KAKENHI Grant Number JP17H06436 in the Scientific Research
on Innovative Areas “Innovation for Light-Energy Conversion
4
(
I LEC)” (to HT) as well as the Sasakawa Scientific Research Grant
Mizoguchi T, Shoji A, Kunieda M, Miyashita H, Tsuchiya T,
Mimuro M, Tamiaki H (2006) Stereochemical determination
of chlorophyll-d molecule from Acaryochloris marina and its
modification to a self-aggregative chlorophyll as a model of
green photosynthetic bacterial antennae. Photochem Photobiol
Sci 5:291–299
from the Japan Science Society and JSPS KAKENHI Grant Number
JP17J08860 in the JSPS Research Fellow program (to MT).
Compliance with ethical standards
Mizoguchi T, Nagai C, Kunieda M, Kimura Y, Okamura A, Tamiaki H
Conflict of interest The authors declare that they have no conflict of
(
2009) Stereochemical determination of the unique acrylate moi-
interest.
ety at the 17-position in chlorophylls-c from a diatom Chaetoceros
1
3

Photosynthesis Research
calcitrans and its effect upon electronic absorption properties. Org
Biomol Chem 7:2120–2126
Teramura M, Harada J, Mizoguchi T, Yamamoto K, Tamiaki H (2016a)
2
In vitro assays of BciC showing C13 -demethoxycarbonylase
Oba T, Tamiaki H (1999) Why do chlorosomal chlorophylls lack the
activity requisite for biosynthesis of chlorosomal chlorophyll
pigments. Plant Cell Physiol 57:1048–1052
2
C13 -methoxycarbonyl moiety? An in vitro model study. Photo-
synth Res 61:23–31
Teramura M, Harada J, Tamiaki H (2016b) In vitro stereospecific
hydration activities of the 3-vinyl group of chlorophyll deriva-
tives by BchF and BchV enzymes involved in bacteriochlorophyll
c biosynthesis of green sulfur bacteria. Photosynth Res 130:33–45
Teramura M, Harada J, Tamiaki H (2018) In vitro enzymatic assays of
photosynthetic bacterial 3-vinyl hydratases for bacteriochlorophyll
biosyntheses. Photosynth Res 135:319–328
Oba T, Masada Y, Tamiaki H (1997) Convenient preparation of pheo-
phytin b from plant extract through the C7-reduced intermediate.
Bull Chem Soc Jpn 70:1905–1909
Olson JM (1998) Chlorophyll organization and function in green pho-
tosynthetic bacteria. Photochem Photobiol 67:61–75
Ouchane S, Steunou AS, Picaud M, Astier C (2004) Aerobic and
anaerobic Mg-protoporphyrin monomethyl ester cyclases in pur-
ple bacteria. A strategy adopted to bypass the repressive oxygen
control system. J Biol Chem 279:6385–6639
Tsuchiya T, Ohta H, Okawa K, Iwamatsu A, Shimada H, Masuda T,
Takamiya K (1999) Cloning of chlorophyllase, the key enzyme
in chlorophyll degradation: finding of a lipase motif and the
induction by methyl jasmonate. Proc Natl Acad Sci USA
96:15362–15367
Papenbrock J, Mock HP, Tanaka R, Kruse E, Grimm B (2000) Role of
magnesium chelatase activity in the early steps of the tetrapyrrole
biosynthetic pathway. Plant Physiol 122:1161–1170
Tsukatani Y, Yamamoto H, Harada J, Yoshitomi T, Nomata J, Kasahara
M, Mizoguchi T, Fujita Y, Tamiaki H (2013) An unexpectedly
branched biosynthetic pathway for bacteriochlorophyll b capa-
ble of absorbing near-infrared light. Sci Rep 3:1217. https://doi.
org/10.1038/srep01217
Tamiaki H, Kouraba M, Takeda K, Kondo S, Tanikaga R (1998) Asym-
metric synthesis of methyl bacteriopheophorbide-d and analogues
by stereoselective reduction of the 3-acetyl to the 3-(1-hydroxy-
ethyl) group. Tetrahedron Asymmetry 9:2101–2111
Tamiaki H, Machida S, Mizutani K (2012) Modification of 3-substitu-
ents in (bacterio)chlorophyll derivatives to prepare 3-ethylated,
methylated, and unsubstituted (nickel) pyropheophorbides and
their optical properties. J Org Chem 77:4751–4758
Wakao N, Yokoi N, Isoyama N, Hiraishi A, Shimada K, Kobayashi M,
Kise H, Iwaki M, Itoh S, Takaichi S (1996) Discovery of natu-
ral photosynthesis using Zn-containing bacteriochlorophyll in
an aerobic bacterium Acidiphilium rubrum. Plant Cell Physiol
37:889–893
Teramura M, Tamiaki H (2018) Semi-synthesis and HPLC analysis
of (bacterio)chlorophyllides possessing a propionic acid residue
at the C17-position. J Porphyrins Phthalocyanines 22:423–436
Walker JC, Willows DR (1997) Mechanism and regulation of Mg-
chelatase. Biochem J 327:321–333
1
3