Detection of 132-carboxy-chlorin produced by the in vitro BciC enzymatic hydrolysis of zinc chlorophyllide

Article abstract of DOI:10.1016/j.bmcl.2021.127931

Green photosynthetic bacteria with an efficient light-harvesting system contain special chlorophyll molecules, called bacteriochlorophylls c, d, e, in their main antennae. In the biosynthetic pathway, a BciC enzyme is proposed to catalyze the hydrolysis of the C13²-methoxycarbonyl group of chlorophyllide a, but the resulting C13²-carboxy group has not been detected yet because it is spontaneously removed due to the instability of the β-keto-carboxylic acid. In this study, the in vitro BciC enzymatic reactions of zinc methyl (13¹R/S)-hydroxy-mesochlorophyllides a were examined and a carboxylic acid possessing the C13²S-OH was first observed as the hydrolyzed product of the C13²-COOCH₃.

Full text of DOI:10.1016/j.bmcl.2021.127931

Bioorg. Med. Chem. Lett. 40 (2021) 127931
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Detection of 132-carboxy-chlorin produced by the in vitro BciC enzymatic
hydrolysis of zinc chlorophyllide
,
*
Mitsuaki Hirosea, Jiro Haradab, Hitoshi Tamiakia
a Graduate School of Life Sciences, Ritsumeikan University, Kusatsu, Shiga 525-8577, Japan
b Medical Biochemistry, Kurume University School of Medicine, Kurume, Fukuoka 830-0011, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
Green photosynthetic bacteria with an efficient light-harvesting system contain special chlorophyll molecules,
called bacteriochlorophylls c, d, e, in their main antennae. In the biosynthetic pathway, a BciC enzyme is pro-
posed to catalyze the hydrolysis of the C132-methoxycarbonyl group of chlorophyllide a, but the resulting C132-
carboxy group has not been detected yet because it is spontaneously removed due to the instability of the β-keto-
carboxylic acid. In this study, the in vitro BciC enzymatic reactions of zinc methyl (131R/S)-hydroxy-meso-
chlorophyllides a were examined and a carboxylic acid possessing the C132S-OH was first observed as the hy-
drolyzed product of the C132-COOCH3.
Bacteriochlorophyll
Biosynthesis
Carboxylic acid
Chlorophyll
Stereochemistry
Chlorophylls (Chls) play a key role in the initial stage of chlor-
ophotosynthesis including sunlight absorption, excited energy transfer,
and charge separation.1–3 Green photosynthetic bacteria possess effi-
cient light-harvesting antennae systems (chlorosomes), which contain a
large amount of special Chl molecules, called bacteriochlorophylls
(BChls) c, d, and e.3–7 The chemical structures of BChl c/d/e molecules
partially differ from those of all the other (B)Chl molecules; for example,
BChls c/d/e self-aggregating inside chlorosomes without protein support
can be structurally discriminated from (B)Chl a forming complexes with
proteins in (an)oxygenic phototrophs by the following two functional
substituents at the peripheral positions of cyclic tetrapyrroles (see
Fig. 1). (1) BChls c/d/e have a 1-hydroxyethyl group at the C3-position,
compared to the C3-vinyl group of Chl a or the C3-acetyl group of BChl
a. The coordination of the C31-OH to the central Mg and the hydrogen-
bonding of the C31-OH with the C13-C=O contributed to form their self-
aggregation in the chlorosomes.7–10 (2) BChls c/d/e lack a methox-
ycarbonyl group at the C132-position which is present in (B)Chl a. The
removal of a methoxycarbonyl group at the C132-position induces to
reduce the steric hindrance around the 13-keto-carbonyl group. These
chemical modifications enable the more interaction between the C13-
investigated by genetic studies.13 Liu and Bryant first reported that a
bciC gene affected the removal of the C132-methoxycarbonyl group
during the biosynthesis of BChl c (Fig. 2).14 Later, we examined the in
vitro BciC enzymatic reactions and proposed a pathway in which the
BciC enzyme would catalyze the hydrolysis of the methoxycarbonyl
group in chlorophyllide (Chlide) a, followed by a spontaneous decar-
boxylation of the resulting carboxylic acid to give pyroChlide a.15 The
β-keto-carboxylic acid in cyclic tetrapyrroles produced by this enzymatic
reaction is extremely unstable to be decarboxylated immediately, so it
has not been observed yet. It should be noted that the linear tetrapyrrole
metabolites bearing the C132-carboxy group were visible during the
breakdown of the Chls.16,17
Herein, we report the in vitro BciC enzymatic reactions of synthetic
zinc Chl a derivatives possessing a C131-hydroxy group and the first
detection of a 132-carboxy-chlorin as the hydrolyzed product, which has
been predicted but was not observed previously.
We reported earlier that Chlide a as a natural substrate was C132-
demethoxycarbonylated in vitro by an action of the recombinant BciC
protein derived from the bciC gene of Cba. tepidum.18 Moreover, the in
vitro BciC enzymatic reactions of zinc complexes of Chlide a and its
methyl ester in the C17-propionate residue have been reported to pro-
ceed successfully.19 Since all the substrates have a C131-oxo moiety, the
β-keto-carboxylic acids produced by the BciC-catalyzed hydrolysis of the
C132-methoxycarbonyl group are so unstable under the enzymatic re-
action conditions in phosphate buffer (pH 7.0) at 45℃ that the
1
11
–
–
C
O and C3 -OH groups.
The green photosynthetic bacterium, Chlorobaculum (Cba.) tepidum
synthesizes BChl c as a self-aggregated pigment in the chlorosome. A
genome sequential analysis of its TLS strain has already been
completed,12 and the biosynthetic pathways of BChl c have been
* Corresponding author.
E-mail address: tamiaki@fc.ritsumei.ac.jp (H. Tamiaki).
https://doi.org/10.1016/j.bmcl.2021.127931
Received 28 December 2020; Received in revised form 23 February 2021; Accepted 25 February 2021
Available online 9 March 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

M. Hirose et al.
Bioorganic & Medicinal Chemistry Letters 40 (2021) 127931
respectively (Fig. S1). According to the correlation of 3JH-H with the ϕ
angle reported by Karplus,27 the major and minor products were
confirmed to be (131S)-Zn-1a and (131R)-Zn-1b, respectively. The
assignment is further supported by their NOESY data that the C131-H of
Zn-1a was not correlated with its C132-H, but apparent correlation be-
tween those of Zn-1b was observed (Figs. S8 and S9). Additionally, Zn-
1c was prepared by hydrogenation23 and zinc metalation28 of methyl
pheophorbide a.
The BciC enzymatic reaction of Zn-1c bearing the C3-ethyl and C131-
oxo groups was examined in an aqueous 50 mM phosphate buffer so-
lution (pH 7.0) containing 1% dimethyl sulfoxide at 45 ℃ in the dark.
After an incubation for 1 h, the band #1, corresponding to the Zn-1c
substrate (retention time, tR = 7.3 min) decreased, and a new band #2
(tR = 9.1 min) was detected (the red to blue profiles in Fig. 3). The on-
line visible absorption spectrum for the band #2 was almost the same as
that for the band #1. The on-line mass spectrum indicated that the
product gave a parent peak (m/z) at 613, which corresponds to the
calculated mass number of protonated zinc methyl mesopyr-
opheophorbide a (Zn-3c): m/z = 613 (MH+). Moreover, the retention
time for the band #2 was consistent with that of the authentic product
sample Zn-3c,29 and the band #2 was not detected in the control
experiment without BciC protein (the green profile in Fig. 3). The results
show that the BciC enzyme reacted with Zn-1c to give Zn-3c. Based on
our recent report,15 the BciC enzyme would catalyze the hydrolysis of
the C132-methoxycarbonyl group, and the resulting β-keto-carboxylic
acid Zn-2c should be (non-enzymatically) decarboxylated.
Fig. 1. Structures of (B)Chl a (left) and BChl c/d/e molecules (right); Chl a: R3
= CHCH2, C7=C8; BChl a: R3 = CH2CH3, C7H–C8H; BChl c: R7 = R20 = CH3;
BChl d: R7 = CH3, R20 = H; BChl e: R7 = CHO, R20 = CH3; R8 = CH2CH3,
CH2CH2CH3, CH2CH(CH3)2, or CH2C(CH3)3; R12 = CH3 or CH2CH3; the asterisk
(*) shows the 31-chirality.
decarboxylation immediately occurs to give the corresponding pyro-
Chlides lacking the C132-COOCH3. To enhance the stability of the hy-
drolyzed product, the in vitro BciC enzymatic reactions were examined
using synthetic substrates Zn-1a/b bearing the C131-hydroxy group
instead of the C131-oxo moiety in zinc methyl mesopheophorbide a (Zn-
1c) possessing the C3-ethyl group which is chemically more stable than
the C3-vinyl group in Chlides a.
Under the same conditions described above, the BciC-catalyzed re-
action of zinc (131S)-hydroxy-chlorin Zn-1a (m/z = 673 for MH+) was
examined (Fig. 4). After an incubation for 1 h, the band #1, corre-
sponding to the Zn-1a substrate (tR = 14.8 min) decreased, and two new
bands #2 (tR = 2.9 min) and #3 (tR = 13.6 min) were apparently
detected (the red to blue profiles in Fig. 4). The on-line visible absorp-
tion spectrum for the band #3 exhibited a large Soret band at 408 nm
and a small absorption band in the red-light region (Fig. S2). The on-line
mass spectrum indicated that the product gave a parent peak at 671
(=673 – 2) (Fig. S3). Since the band #3 was also detected in the control
experiment without BciC protein (the green profile in Fig. 4), it was
assigned to Zn-4a, which was produced by the oxidation (17,18-dehy-
As shown in Scheme 1, methyl pheophorbide a,20–23 obtained from
the chemical modification of Chl a, was first reduced at the C13-keto-
carbonyl group using NaBH4,24 then zinc-metalated22 and epimerically
separated by HPLC, and finally hydrogenated at the C3-vinyl group25 to
give Zn-1a and Zn-1b (see their 1H NMR, Vis, and MS data in Supple-
mentary Data). The 1H NMR spectra of the 131-epimers in 3% pyr-
idine‑d5 and CDCl3 showed different coupling constants (3JH-H values)
between the 131-H and the 132-H (see the right, lower drawings in
Scheme 1). The major product provided a constant of 1 Hz, and the
minor one gave a higher value of 6 Hz. Energy-minimized molecular
models of trans-Zn-1a and cis-Zn-1b were constructed by MM+/MP3
calculations,26 and the former and latter dihedral angles (ϕ) between the
C131–H and C132–H bands were estimated to be 104.9◦ and 0.6◦,
drogenation) of chlorin Zn-1a to the porphyrin π-skeleton.
The on-line visible absorption spectrum for the band #2 (Soret band
at 406 nm and Qy band at 619 nm) was almost the same as that obtained
for Zn-1a (Soret band at 403 nm and Qy band at 619 nm) (Fig. S2). The
results indicate that the band #2 has a chlorin π-skeleton. The band #2
Fig. 2. Hydrolysis of the C132-methoxycarbonyl group in Chlide a catalyzed by a BciC enzyme, and successive decarboxylation during the biosynthesis of BChls c/
d/e.
2

M. Hirose et al.
Bioorganic & Medicinal Chemistry Letters 40 (2021) 127931
Scheme 1. Synthesis of zinc methyl mesopheophorbides a Zn-1a/b/c from methyl pheophorbide a: (i) NaBH4, THF, MeOH; (ii) Zn(OAc)2⋅2H2O, MeOH, CH2Cl2; (iii)
H2, Rh/Al2O3, Me2CO; (iv) H2, Pd/C, CH2Cl2; (v) Zn(OAc)2, Me2CO, CH2Cl2.
3

M. Hirose et al.
Bioorganic & Medicinal Chemistry Letters 40 (2021) 127931
#1
#2
6
8
10
12
Retention time / min
Fig. 3. BciC activity assay with Zn-1c. From top to bottom, the profiles show HPL chromatograms for Zn-1c before (red) and after incubation without BciC for 1 h
(green) or with BciC for 1 h (blue) and authentic samples Zn-3c (black): Cosmosil 5C18-AR-II, 3.0 ϕ × 150 mm, MeOH/H2O = 90/10, 0.5 mL minꢀ 1, 35 ◦C.
Fig. 4. BciC activity assay with Zn-1a. From top to bottom, the profiles show HPL chromatograms for Zn-1a before (red) and after incubation without BciC for 1 h
(green) or with BciC for 1 h (blue), as well as for the reaction mixture after the treatment of the band #2 with CH2N2 (black), and Zn-3a/b (gray): Cosmosil 5C18-AR-
II, 3.0 ϕ × 150 mm, MeOH/H2O = 80/20, 0.5 mL minꢀ 1, 35 ◦C.
4

M. Hirose et al.
Bioorganic & Medicinal Chemistry Letters 40 (2021) 127931
4 Holzwarth AR, Griebenow K, Schaffner K. Chlorosomes, photosynthetic antennae
with novel self-organized pigment structures. J Photochem Photobiol A: Chem. 1992;
65:61–71.
was eluted so rapidly in RP-HPLC that it should be more hydrophilic
than Zn-1a, so it can be speculated that it possesses a free carboxy
moiety. The on-line mass spectrum indicated that the band #2 gave a
parent peak (m/z) at 659, which corresponds to the calculated mass
number of the protonated species of a compound produced by hydro-
lyzing one of the methyl esters in Zn-1a. These observations indicated
that the band #2 to be 132-carboxy-chlorin Zn-2a.30 It was not possible
to isolate the product related to the band #2 because of its low stability.
Just after separating the band #2, the compound was treated with
ethereal diazomethane and analyzed by HPLC (the black profile in
Fig. 4). In the chromatogram, methyl ester Zn-1a (#1) and its dehy-
drogenated porphyrin Zn-4a (#3) were visible. Therefore, the band #2
was confirmed to correspond to Zn-2a bearing a carboxy group at the
C132-position. The further decarboxylated product Zn-3a31 (#5) was
not detected in the chromatogram due to the absence of C131-oxo
moiety in the β-hydroxycarboxylic acid Zn-2a.32
5 Olson JM. Chlorophyll organization and function in green photosynthetic bacteria.
Photochem Photobiol. 1998;67:61–75.
6 Orf GS, Blankenship RE. Chlorosome antenna complexes from green photosynthetic
bacteria. Photosynth Res. 2013;116:315–331.
7 Matsubara M, Tamiaki H. Supramolecular chlorophyll aggregates inspired from
specific light-harvesting antenna “chlorosome”: Static nanostructure, dynamic
construction process, and versatile application. J Photochem Photobiol C: Photochem
Rev. 2020;45:1000385.
8 Tamiaki H. Supramolecular structure in extramembraneous antennae of green
photosynthetic bacteria. Coord Chem Rev. 1996;148:183–197.
9 Ganapathy S, Oostergetel GT, Wawrzyniak PK, et al. Alternating syn-anti
bacteriochlorophylls form concentric helical nanotubes in chlorosomes. Proc Nat
Acad Sci USA. 2009;106:8525–8530.
10 Luo SC, Khin Y, Huang SJ, et al. Probing the spatial organization of
bacteriochlorophyll c by solid-state nuclear magnetic resonance. Biochemistry. 2014;
53:5515–5525.
11 Oba T, Tamiaki H. Why do chlorosomal chlorophylls lack the C132-methoxycarbonyl
moiety? An in vitro model study. Photosynth Res. 1999;61:23–31.
12 Eisen JA, Nelson KE, Paulsen IT, et al. The complete genome sequence of Chlorobium
tepidum TLS, a photosynthetic, anaerobic, green-sulfur bacterium. Proc Natl Acad Sci
USA. 2002;99:9509–9514.
The BciC-catalyzed reaction of (131R)-epimeric Zn-1b was examined
in a similar way (Fig. S4). After reaction for 1 h, no Zn-2b was detected
as the C132-carboxylic acid product. A hydroxy group at the C131-po-
sition and a methoxycarbonyl group at the C132-position are situated in
the cis configuration for Zn-1b, so an intramolecular hydrogen bond
between the former and latter groups, 131-O-H… O=C-132 or 131-O-
H… O-C-131, would unstabilize the complex of Zn-1b with BciC to pro-
duce no desired enzymatic reaction.
13 Bryant DA, Hunter CN, Warren MJ. Biosynthesis of the modified tetrapyrrolesꢀ the
pigments of life. J Biol Chem. 2020;295:6888–6925.
14 Liu Z, Bryant DA. Identification of a gene essential for the first committed step in the
biosynthesis of bacteriochlorophyll c. J Biol Chem. 2011;286:22393–22402.
15 Hirose M, Harada J, Tamiaki H. In vitro hydrolysis of zinc chlorophyllide a
homologues by a BciC enzyme. Biochemistry. 2020;59:4622–4626.
16 Christ B, Schelbert S, Aubry S, et al. MES16, a member of the methylesterase protein
family, specifically demethylates fluorescent chlorophyll catabolites during
chlorophyll breakdown in arabidopsis. Plant Physiol. 2012;158:628–641.
The in vitro C132-hydrolysis of the C132-methoxycarbonyl group in
zinc Chl a derivatives Zn-1a/b bearing the C131-hydroxy group were
examined using the BciC enzyme derived from one of the green sulfur
bacteria species, Cba. tepidum. A desired product bearing the C132-car-
boxy group, Zn-2a, was first observed in the BciC enzymatic reaction of
(132S)-Zn-1a. This result supports the fact that the BciC enzyme cata-
lyzes the hydrolysis of the (132R)-methoxycarbonyl group in Chlide a.
Additionally, the present semi-synthetic Chl derivatives might be useful
for screening the candidates of sensitizing reagents applied to photo-
dynamic therapy.33
¨
17 Krautler B. Breakdown of chlorophyll in higher plants–Phyllobilions as abundant, yet
hardly visible signs of ripening, senescence, and cell death. Angew Chem Int Ed. 2016;
55:4882–4907.
18 Teramura M, Harada J, Mizoguchi T, Yamamoto K, Tamiaki H. In vitro assays of BciC
showing C132-demethoxycarbonylase activity requisite for biosynthesis of
chlorosomal chlorophyll pigments. Plant Cell Physiol. 2016;57:1048–1052.
19 Hirose M, Teramura M, Harada J, Tamiaki H. BciC-catalyzed C132-
demethoxycarbonylation of metal pheophorbide a alkyl esters. ChemBioChem. 2020;
21:1473–1480.
20 Smith KM, Goff DA, Simpson DJ. Meso substitution of chlorophyll derivatives: Direct
route for transformation of bacteriopheophorbides d into bacteriopheophorbides c.
J Am Chem Soc. 1985;107:4946–4954.
21 Ma L, Dolphin D. Nucleophilic reaction of 1,8-diazabicyclo[5.4.0]undec-7-ene and
1,5-diazabicyclo[4.3.0]non-5-ene with methyl pheophorbide a. Unexpected
products. Tetrahedron. 1996;52:849–860.
Declaration of Competing Interest
22 Tamiaki H, Takeuchi S, Tsudzuki S, Miyatake T, Tanikaga R. Self-aggregation of
synthetic zinc chlorins with a chiral 1-hydroxyethyl group as a model for in vivo
epimeric bacteriochlorophyll-c and d aggregates. Tetrahedron. 1998;54:6699–6718.
23 Uliana MP, Pires L, Pratavieira S, et al. Photobiological characteristics of chlorophyll
a derivatives as microbial PDT agents. Photochem Photobiol Sci. 2014;13:1137–1145.
24 Ma L, Dolphin D. A novel and convenient conversion of chlorins to phytoporphyrins
during the modification of chlorophyll derivatives. Tetrahedron Lett. 1995;36:
7791–7794.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
25 Tamiaki H, Takekoshi D, Mizoguchi T. Reduction of vinyl groups in naturally
occurring chlorophylls-a. Bioorg Med Chem. 2011;19:52–57.
This work was partially supported by Japan Society for the Promo-
tion of Science (JSPS) KAKENHI Grant Number JP17H006436 in Sci-
entific Research on Innovative Areas “Innovation for Light-Energy
Conversion (I4LEC)”.
26 Kureishi Y, Tamiaki H. Synthesis and self-aggregation of zinc 20-halogenochlorins as
a model for bacteriochlorophylls c/d. J Porphyr Phthalocya. 1998;2:159–169.
27 Karplus M. Contact electron–spin coupling of nuclear magnetic moments. J Chem
Phys. 1959;30:11–15.
28 Hirose M, Teramura M, Harada J, Ogasawara S, Tamiaki H. In vitro C132-
dealkoxycarbonylations of zinc chlorophyll a derivatives including C132-substitutes
by a BciC enzyme. Bioorg Chem. 2020;102, 104111.
Appendix A. Supplementary data
29 Tamiaki H, Yagai S, Miyatake T. Synthetic zinc tetrapyrroles complexing with
pyridine as a single axial ligand. Bioorg Med Chem. 1998;6:2171–2178.
30 The BciC enzyme could not hydrolyze the esterifying group in the C17-propionate
residue,19 so 172-carboxy-chlorin, that is, zinc (131S)-hydroxy-pheophorbide a, was
ruled out as the mono-hydrolyzed product #2.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmcl.2021.127931.
31 Methyl pyropheophorbide a was reduced using NaBH4,32 hydroganated,20 and zinc-
References
metallated22 to give Zn-3a/b (major/minor).
1 Croce R, van Amerongen H. Light harvesting in oxygenic photosynthesis: Structural
biology meets spectroscopy. Science. 2020;369:eaay2058.
32 Tamiaki H, Monobe R, Koizumi S, Miyatake T, Kinoshita Y. Stereoselective
reduction, methylation, and phenylation of the 13-carbonyl group in chlorophyll
derivatives. Tetrahedron Asymm. 2013;24:677–682.
2 Chen JH, Wu H, Xu C, et al. Architecture of the photosynthetic complex from a green
sulfur bacterium. Science. 2020;370:eabb6350. https://doi.org/10.1126/science.
abb6350.
33 Hamblin MR. Photodynamic therapy for cancer: What’s past is prologue. Photochem
Photobiol. 2020;96:506–516.
3 Thiel V, Tank M, Bryant DA. Diversity of chlorophototrophic bacteria revealed in the
omics era. Annu Rev Plant Biol. 2018;69:21–49.
5