Self-aggregation of synthetic protobacteriochlorophyll-d derivatives

Article abstract of DOI:10.1562/2004-08-02-RA-254.1

3¹-Racemically pure zinc 3¹-hydroxy-13 ¹-oxo-porphyrins (zinc methyl 17,18-dehydro-bacteriopheophorbides-d) as well as their 3¹-demethyl form were prepared by modifying chlorophyll-a through oxidation by 2,3-dichloro-5,6-dicyano-benzo-quinone. From visible, circular dichroism and infrared spectral analyses, these synthetic pigments self-aggregated in 1%(vol/ vol) tetrahydrofuran and cyclohexane to give large oligomers by an intermolecular bonding of 13-C=O...H-O(3 ¹)... Zn(central) and π-π interaction of the porphyrin chromophores. The supramolecular structures are similar to those of the corresponding chlorins and a core part of extra-membranous light-harvesting antennas of photosynthetic green bacteria. The 17,18-dehydrogenation of a chlorin to porphyrin moiety did not disturb its self-aggregation, and the synthetic zinc porphyrins are good models for naturally occurring, self-aggregative bacteriochlerophylls.

Full text of DOI:10.1562/2004-08-02-RA-254.1

Self-aggregation of Synthetic Protobacteriochlorophyll- Derivatives
d
Author(s): Hitoshi Tamiaki, Hiroyuki Kitamoto, Takuya Watanabe, and Reiko Shibata
Source: Photochemistry and Photobiology, 81(1):170-176.
Published By: American Society for Photobiology
DOI: http://dx.doi.org/10.1562/2004-08-02-RA-254.1
URL: http://www.bioone.org/doi/full/10.1562/2004-08-02-RA-254.1
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and
environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published
by nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.

Photochemistry and Photobiology, 2005, 81: 170–176
Self-aggregation of Synthetic Protobacteriochlorophyll-d Derivatives{
Hitoshi Tamiaki*, Hiroyuki Kitamoto, Takuya Watanabe and Reiko Shibata
Department of Bioscience and Biotechnology, Faculty of Science and Engineering, Ritsumeikan University, Kusatsu, Japan
Received 31 July 2004; accepted 1 October 2004
tigations (2–4,7–11): a special intermolecular bonding of 13-
C5Oꢀ ꢀ ꢀH–O(31)ꢀ ꢀ ꢀMg(central) and p–p interaction of composite
ABSTRACT
31-Racemically pure zinc 31-hydroxy-131-oxo-porphyrins (zinc
chlorophyllous p-systems.
methyl 17,18-dehydro-bacteriopheophorbides-d) as well as
their 31-demethyl form were prepared by modifying chloro-
phyll-a through oxidation by 2,3-dichloro-5,6-dicyano-benzo-
quinone. From visible, circular dichroism and infrared spectral
analyses, these synthetic pigments self-aggregated in 1%(vol/
vol) tetrahydrofuran and cyclohexane to give large oligomers
by an intermolecular bonding of 13-C5Oꢀ ꢀ ꢀH–O(31)ꢀ ꢀ ꢀ
Zn(central) and p–p interaction of the porphyrin chro-
mophores. The supramolecular structures are similar to
those of the corresponding chlorins and a core part of extra-
membranous light-harvesting antennas of photosynthetic green
bacteria. The 17,18-dehydrogenation of a chlorin to porphyrin
moiety did not disturb its self-aggregation, and the synthetic
zinc porphyrins are good models for naturally occurring,
self-aggregative bacteriochlorophylls.
We earlier reported the self-aggregation of synthetic zinc BChl-
c/d/e derivatives including Zn-4/6 (Fig. 1) in nonpolar organic
solvents as chlorosomal models (12–16), indicating that linear
location of hydroxy, keto-carbonyl groups and a central co-
ordinative metal including Mg, Zn and Cd in a molecule is
necessary for such a self-aggregation. Natural self-aggregative
BChl-c/d/e have chlorin (5 17,18-dihydroporphyrin) p-systems,
and we have found that a bacteriochlorin (5 7,8,17,18-tetrahy-
droporphyrin) moiety was also useful for its self-aggregation (17).
In this study, we report on the synthesis of zinc 31-hydroxy-131-
oxo-porphyrins Zn-1/2 (Fig. 1) by modifying naturally occurring
Chl-a and their self-aggregation. A few reports are available on
self-aggregation of synthetic hydroxy- and carbonyl-functionated
porphyrins prepared from artificial meso-arylporphyrins (18,19)
and octaethylporphyrin (20), but this is the first example of
chlorosomal self-aggregation of porphyrins based on the natural
chlorophyllous pigments (21).
INTRODUCTION
In this study, we used the following tentative nomenclature.
Protochlorophylls refer to 17,18-dehydro-chlorophylls and are
currently widely used; we therefore used the prefix ‘‘proto’’ for
17,18-dehydrogenated compounds. Protobacteriochlorophyll-d
(5 17,18-dehydro-BChl-d) has a fully conjugated porphyrin p-
system. Protobacteriopheophorbide-d possessing 8-ethyl and 12-
methyl groups can be termed 31-hydroxy-phytoporphyrin on the
basis of IUPAC-IUB nomenclature (22).
Self-aggregation of porphyrins and chlorophylls (Chls) has attracted
much attention in the investigation of natural photosynthetic systems
as well as in the construction of artificial photoactive devices. Two
self-aggregates of porphyrinoids (cyclic tetrapyrroles) without any
assistance of peptide scaffolds are naturally observed: (I) self-
aggregates of Fe(III)-protoporphyrin-IX in malaria haemozoin (1)
and (II) self-aggregates of bacteriochlorophyll(BChl)s-c/d/e in
extramembranous light-harvesting antenna of photosynthetic green
bacteria (2–6). The latter systems are called ‘‘chlorosomes’’ because
of their green colored, submicrometer-sized apparatus (green þ
body5chloroþsome). In a chlorosome, a large quantity of BChl-c/
d/e molecules self-aggregate with no specific interaction of peptides
to form oligomers, which absorb sunlight efficiently and transfer the
harvesting light energy rapidly to an acceptor at the chlorosomal
surface. The supramolecular structures have not yet been identified,
but their local interaction was determined from various inves-
MATERIALS AND METHODS
Apparatus. Proton nuclear magnetic resonance (1H NMR) spectra were
measured with a Bruker AC-300 spectrometer (Rheinstetten, Germany);
chemical shifts (d) are expressed in parts per million relative to CHCl3 (7.26
ppm) as an internal reference. Fast atomic bombardment (FAB) mass spectra
were measured with a JEOL GCmate II spectrometer (Akishima, Japan).
Visible absorption and circular dichroism (CD) spectra were measured with
a Hitachi U-3500 spectrophotometer (Tokyo, Japan) and a Jasco J-720W
spectropolarimeter (Hachioji, Japan), respectively. Infrared spectra were
measured at room temperature with a Shimadzu FTIR-8600 spectropho-
tometer (Kyoto, Japan); a solution was measured in a KBr cell and a solid
film on a thin aluminum-coated glass plate by means of reflection absorption
spectroscopy through a Shimadzu AIM-800R microscope. High-perfor-
mance liquid chromatography (HPLC) was done with a Shimadzu LC-10AS
pump, an SPD-10AV visible detector and a C-R6A chromatopac. Flash
column chromatography (FCC) was performed with silica gel (Merck,
Kieselgel 60, 9385 Darmstadt, Germany).
{Posted on the website on 6 October 2004.
*To whom correspondence should be addressed: Department of Bioscience
and Biotechnology, Faculty of Science and Engineering, Ritsumeikan
University, Kusatsu, Shiga 525-8577, Japan. Fax: 81-77-561-2659;
e-mail: tamiaki@se.ritsumei.ac.jp
Abbreviations: BChl, bacteriochlorophyll; CD, circular dichroism; Chl,
chlorophyll; DDQ, 2,3-dichloro-5,6-dicyano-benzoquinone; FAB, fast
atomic bombardment; FCC, flash column chromatography; FWHM, full
width at half-maximum; HPLC, high-performance liquid chromatogra-
phy; 1H NMR, proton nuclear magnetic resonance; THF, tetrahydrofuran.
Chemicals. Tetrahydrofuran (THF) was distilled from CaH2 before use.
Solvents for visible and CD spectra were purchased from Nacalai Tesque
(spectroscopy grade, Kyoto, Japan). Methyl pyropheophorbide-a (3)
(12,13), zinc methyl pyropheophorbide-a (Zn-3) (23), methyl bacterio-
ꢀ 2005 American Society for Photobiology 0031-8655/05
170

Photochemistry and Photobiology, 2005, 81 171
Figure 1. Molecular structures of a naturally occurring bacteriochloro-
phyll-d (left), its synthetic chlorin-type models Zn-4/6 (center, 17CH–
18CH) and the porphyrintype analogues Zn-1/2 (right, 17C518C).
Methyl 31S-protobacteriopheophorbide-d (1S). Similar to synthesis of 1,
HPLC-separated Zn-4S was DDQ oxidized and treated with acid to give
1S; VIS (THF) kmax 5 639 (relative intensity, 0.01), 585 (0.05), 561 (0.08),
521 (0.05), 429 (0.54), 418 (1.00) nm; MS (FAB) found: m/z 565. Calcd for
C54H36N4O4: MHþ, 565.
Methyl 31-demethyl-protobacteriopheophorbide-d (2). Zinc metallation
of chlorin 5 (67 lmol, 36.9 mg) gave Zn-5 (see below), which was DDQ
oxidized and zinc demetallated (see above) to afford porphyrin 8 (23). The
resulting residue containing 8 without purification by FCC was dissolved in
dry dichloromethane (20 mL) and reduced by t-BuNH2BH3 (335 lmol,
29.1 mg). After stirring at room temperature for about 1 h, the reaction
mixture was quenched with aq. 4% NaHCO3 and water, dried over Na2SO4
and evaporated in vacuo. The resulting residue was purified with FCC (1%
MeOH–CH2Cl2) to give pure 2 (35 lmol, 19.2 mg) in 52% total yield from
5; 2: melting point 258–2608C (from CH2Cl2 and hexane); VIS (THF)
k
max 5 637 (relative intensity, 0.01), 584 (0.05), 562 (0.08), 521 (0.05),
430 (0.53), 417 (1.00) nm; see other spectral data of 2 in Ref. 23.
Zinc methyl protobacteriopheophorbide-d (Zn-1). To a dichloromethane
solution of 1 (31R/S 5 1/1), a methanol solution saturated with zinc acetate
dihydrate was added. After stirring at room temperature for 1 h, the reaction
mixture was quenched with aq. 4% NaHCO3. Produced white precipitates of
zinc carbonate were filtered off. The filtrate was extracted with dichloro-
methane, and the combined organic layers were washed with water twice,
dried over Na2SO4 and evaporated in vacuo to give pure Zn-1 (31R/S51/1)
in an almost quantitative yield; dark green solid; melting point . 3008C
(from CH2Cl2 and hexane); VIS (THF) kmax 5607 (relative intensity, 0.11),
560 (0.05), 427 (1.00), 377 (0.12) nm; VIS (acetone) kmax 5 607 (e, 2.5 3
104), 560 (1.73104), 426 (2.53105) nm; 1H NMR (CDCl3—3% pyridine-
d5) d 10.15, 9.92, 9.61 (each 1H, s, 5-, 10-, 20-H), 6.57 (1H, q, J57 Hz, 3-
CH), 5.65 (2H, s, 131-CH2), 3.95 (2H, q, J58 Hz, 8-CH2), 3.84, 3.72, 3.61,
3.47, 3.38 (each 3H, s, 2-, 7-, 12-, 18-CH3, 172-COOCH3), 3.92, 2.95 (each
2H, t, J58 Hz, 17-CH2CH2), 2.25 (3H, q, J57 Hz, 3-CH3), 1.82 (3H, t, J58
Hz, 81-CH3); MS (FAB) found: m/z 626. Calcd for C54H34N4O4Zn: Mþ, 626.
Zinc methyl 31R-protobacteriopheophorbide-d (Zn-1R). Similar to the
synthesis of Zn-1, 1R was zinc metallated to give Zn-1R; VIS (THF)
Scheme 1. Synthesis of methyl protobacteriopheophorbide-d (1) and its 31-
epimers 1R/1S; (i) HBr–AcOH, H2O, CH2N2–Et2O; (ii) Zn(OAc)2ꢁ 2H2O/
MeOH–CH2Cl2; (iii) DDQ–CH3COCH3, aq. HCl–CH2Cl2; (iv) reverse-
phase HPLC separation.
pheophorbide-d (4) (13), zinc methyl bacteriopheophorbide-d (Zn-4) (13),
zinc methyl 31R-bacteriopheophorbide-d (Zn-4R) (13), zinc methyl 31S-
bacteriopheophorbide-d (Zn-4S) (13), methyl pyropheophorbide-d (5) (12),
methyl 31-demethyl-bacteriopheophorbide-d (6) (12), zinc methyl 31-
demethyl-bacteriopheophorbide-d (Zn-6) (12), methyl protopyropheophor-
bide-a (7) (23) were prepared according to the reported procedures. Other
cyclic tetrapyrroles were synthesized in the following.
Methyl protobacteriopheophorbide-d (1). According to reported proce-
dures (23), Zn-4 was transformed to 1 as follows. To a distilled acetone
solution (40 mL) of Zn-4 (31R/S 5 1/1, 67 lmol, 42.1 mg) was slowly
added a distilled acetone solution (20 mL) of 2,3-dichloro-5,6-dicyano-
benzoquinone (DDQ; 100 lmol, 27.7 mg), and the reaction mixture was
stirred at room temperature. After the disappearance of the 650 nm peak
characteristic of the chlorin moiety, the reaction mixture was quenched with
aq. 4% KHSO4 and extracted with dichloromethane. The combined organic
phases were washed with aq. 4% NaHCO3 and water. After treatment of the
organic layer with concentrated HCl for about 1 min with stirring at room
temperature, the reaction mixture was quenched with aq. 4% NaHCO3 and
water, dried over Na2SO4 and evaporated in vacuo. The resulting residue
was purified with FCC (1.5% MeOH–CH2Cl2) to give pure 1 (31R/S 5 1/1,
45 lmol, 25.2 mg) in 67% yield; see spectral data of 1 in Ref. 23.
Alternatively, the 3-vinyl group of 7 was hydrated by treatment of 30%
HBr in acetic acid at 608C for 3 h (to 3-CHBrMe), water [to 3-CH(OH)Me]
and ethereal diazomethane (the resulting 172-COOH to 172-COOMe)
(see (i) in Scheme 1 and Refs. 13,24) to give a 31-racemic mixture (1:1) of
1 in 46% yield.
k
max 5 607 (relative intensity, 0.11), 560 (0.05), 427 (1.00), 377 (0.12) nm;
MS (FAB) found: m/z 626. Calcd for C54H34N4O4Zn: Mþ, 626.
Zinc methyl 31S-protobacteriopheophorbide-d (Zn-1S). Similar to the
synthesis of Zn-1, 1S was zinc metallated to give Zn-1S; VIS (THF) kmax 5
607 (relative intensity, 0.11), 560 (0.05), 427 (1.00), 377 (0.12) nm; MS
(FAB) found: m/z 626. Calcd for C54H34N4O4Zn: Mþ, 626.
Zinc methyl 31-demethyl-protobacteriopheophorbide-d (Zn-2). Similar
to the synthesis of Zn-1, 2 was zinc metallated to give Zn-2; VIS (THF)
kmax 5 608 (relative intensity, 0.10), 590 (0.04, sh), 560 (0.05), 545
(0.04, sh), 524 (0.03), 428 (1.00), 406 (0.31, sh) nm; see other spectral data
of Zn-2 in Ref. 25.
RESULTS AND DISCUSSION
Methyl 31R-protobacteriopheophorbide-d (1R). Similar to synthesis of 1,
HPLC-separated Zn-4R was DDQ oxidized and treated with acid to give
1R; VIS (THF) kmax 5639 (relative intensity, 0.01), 585 (0.05), 561 (0.08),
521 (0.05), 429 (0.54), 418 (1.00) nm; MS (FAB) found: m/z 565. Calcd for
C54H36N4O4: MHþ, 565.
Synthesis of zinc methyl protobacteriopheophorbides-d
Zn-1/2
Methyl bacteriopheophorbide-d (4, see Scheme 1) possessing ethyl
and methyl groups at the 8- and 12-positions, respectively, was

172 Hitoshi Tamiaki et al.
Scheme 2. Synthesis of zinc methyl protobacteriopheophorbides-d, Zn-1
(31-epimeric 1:1 mixture), Zn-1R (31R-epimer) and Zn-1R (31R-epimer),
and its 31-demethyl analog Zn-2; (ii) Zn(OAc)2ꢁ 2H2O/MeOH–CH2Cl2.
prepared by hydration of the 3-vinyl group in methyl pyropheo-
phorbide-a (3), which was easily available by modifying
Chl-a (13). After standard zinc metallation, zinc complex of 4,
Zn-4, was given as a 31-epimeric mixture (1:1) (13). According to
the reported oxidation by DDQ in acetone (23), Zn-4 (chlorin 5
trans-17,18-dihydroporphyrin) was transformed to the correspond-
ing porphyrin and successive demetallation gave methyl proto-
bacteriopheophorbide-d (1) as a pure form after purification by
FCC. It is noted that direct oxidation of 4 to 1 could not occur by
the action of DDQ because 4 was less electrochemically oxidized
than Zn-4 (23). The resulting 1 was a 1:1 racemic mixture at the
31-position. The racemate 1 was analytically separated by chiral
HPLC (26), but the preparative separation of 1 to 1R and 1S was
very difficult in this stage. On the other hand, a 31-epimeric
mixture of Zn-4 was separated by a single run of reverse-phase
HPLC to afford Zn-4R and Zn-4S in a mg order (13). Each
separated epimer Zn-4R/S was 17,18-dehydrogenated by DDQ and
successively demetallated by an acid to give the corresponding
metal-free porphyrin 1R/1S, whose stereochemistry was deter-
mined by chiral HPLC. During the procedures, no change in the 31-
stereochemistry was observed, and thus, racemically pure 1R and
1S were obtained in a preparative scale. Finally, zinc metallation of
1(R/S) gave desired zinc methyl protobacteriopheophorbide-d [Zn-
1(R/S), see Scheme 2] with retention of the 31-stereochemistry.
Such zinc porphyrins were once produced after 17,18-dehydroge-
nation of the zinc chlorins described above, but little could be
purified because of their lower solubility in usual organic solvents.
Therefore, after demetallation of the oxidized products, pure free
base porphyrins were isolated by purification of FCC. Their
remetallation gave pure Zn-1(R/S) by recrystallization alone
because zinc metallation proceeded smoothly and cleanly.
Scheme 3. Synthesis of methyl 31-demethyl-protobacteriopheophorbide-d
(2); (ii) Zn(OAc)2ꢁ 2H2O/MeOH–CH2Cl2; (iii) DDQ–CH3COCH3, aq. HCl/
CH2Cl2; (v) OsO4/THF–NaIO4/aq. AcOH; (vi) t-BuNH2BH3/CH2Cl2.
group in 2 through the 3-formyl group in 8 (23). The total yield
from 3 to 2 via Zn-3 was 39% (23), which was comparable to the
former (43%), but the former was an easier route to 2 than the latter
because all the present porphyrins were less soluble than the
corresponding chlorins, and a large quantity of solvent was
necessary for alternation of the peripheral substituents on the
porphyrin p-system. It is noteworthy that 1 was alternatively
afforded by hydration of the 3-vinyl group in porphyrin 7, and the
yield (46%) was less than that of hydration of chlorin 3 to 4 (82%).
Standard zinc metallation of 2 gave pure Zn-2 quantitatively
(Scheme 2). All 31-hydroxy-131-oxo-porphyrins (Zn-)1(R/S)/2
were identified by their 1H NMR, visible and/or mass spectra,
indicating that they did not contain any chlorin moiety character-
ized by the intense Qy-absorption band at around 650–660 nm in
an organic solution.
According to the similar access to (Zn-)1 from 3-(1-hydroxy-
ethyl)chlorin 4, 3-hydroxymethyl-chlorin 6 (531-demethyl form of
4, see Scheme 3), which was prepared by oxidation of the 3-vinyl
group in 3 and successive reduction of the formyl group in the
resulting methyl pyropheophorbide-d (5) (12), was applied to
synthesis of the corresponding porphyrin 2. After zinc metallation,
DDQ oxidation of Zn-6 gave a complex mixture as the product,
and the desired corresponding porphyrin could not be detected
in the mixture. The 3-hydroxymethyl group was so reactive that
undesired overoxidation occurred at the less-sterically hindered
benzylic position in Zn-6 (primary alcohol) than in Zn-4 (second-
ary alcohol). Therefore, Zn-5 possessing a formyl group at the 3-
position was similarly 17,18-dehydrogenated and demetallated to
give 3-formyl-porphyrin 8. The 3-formyl group was selectively
reduced by t-BuNH2BH3 to give desired 2 (23) in a total yield of
43% from 3 to 2 via Zn-5. 3-Vinyl-chlorin Zn-3 was also oxidized
to give the porphyrin 7 after demetallation, and the transformation
of the 3-vinyl group in 7 was transformed to the 3-hydroxymethyl
Self-aggregation of zinc methyl protobacteriopheophorbide-d
Zn-1(R/S)
A racemic mixture of Zn-1 (31R/S 5 1/1) was dissolved in THF
(0.2 mM) to give an intense absorption band at 427 nm as its Soret
peak (broken line in Fig. 2). A relatively weak (one-ninth high)
peak was observed at 607 nm, as its Qy-absorption band possessing
a full width at half-maximum (FWHM) of 350 cmꢂ1. These
features indicated that Zn-1 was monomeric in the THF solution.
In 1%(vol/vol) THF and cyclohexane (0.2 mM), both the bands
were shifted to longer wavelengths and broadened in comparison
with the monomeric ones, as shown by the solid line in Fig. 2.

Photochemistry and Photobiology, 2005, 81 173
Figure 2. Visible spectra of 2 3 10ꢂ4 M Zn-1 in THF (– – –) and 1%(vol/
vol) THF and cyclohexane (—) in a 1 mm cell.
Typically, the Qy band moved to 649 nm as its peak and had
810 cmꢂ1 as its FWHM and a redshift of 1070 cmꢂ1 and about a
two-fold broadening occurred by decreasing polarity of the solvent
(THF–cyclohexane 5100/0 fi 1/99). The resulting Soret band was
too broad to give an apparent peak but gave a shoulder at around
475 nm, whose absorbance intensity was as strong as that of the
Qy peak. The maximum at ca 430 nm was ascribed to residual
monomer. Addition of THF to the self-aggregated solution gradu-
ally increased the monomeric bands and finally changed to the
monomeric solution. Considering that similar visible changes were
observed in self-aggregation of Chls (12,13), zinc porphyrin Zn-1
self-aggregated in the nonpolar organic solvent.
Figure 3. IR spectra of Zn-1 in THF (A) and 1%(vol/vol) THF and
cyclohexane (B) in a 0.1 and 0.5 mm KBr cell, respectively.
these spectra were slightly different from those of self-aggregates
of Zn-1 described above. The Qy peaks of Zn-1R/S were situated
at 652 nm and shifted to a 1140 cmꢂ1 longer wavelength than the
monomeric (607 nm). Their FWHMs were about 1000 cmꢂ1 and
larger than the value of Zn-1 (810 cmꢂ1). The Soret peaks clearly
appeared at 478 nm in Zn-1R/S. These visible absorption changes
indicate that Zn-1R/S are similar self-aggregates in the nonpolar
solvent with Zn-1. Slight differences show that Zn-1 self-
aggregated in the random order of the enantiomers, not in the se-
parated manner similar to a 1:1 mixture of Zn-1R and Zn-1S
self-aggregates.
In a THF solution of Zn-1, several vibrational bands were
observed at 3410, 1742, 1699 and 1607 cmꢂ1 (Fig. 3A), which were
assigned to O–H, ester C5O, keto C5O and porphyrin skeletal
C–C/C–N stretching, respectively, when compared with the
reported data of chlorins (12,27–29). The values of these O–H
and C5O vibrations showed that the functional groups were free
from any specific interaction and also that Zn-1 was monomeric in
THF. In 1%(vol/vol) THF and cyclohexane, self-aggregates of Zn-
1 gave 1742, 1666 and 1607cmꢂ1 vibrational peaks (Fig. 3B). A
large shift in the keto-carbonyl stretching peak (1699 ꢂ 1666 5 33
cmꢂ1) by self-aggregation of porphyrin Zn-1 is comparable with
shifts by the formation of C5Oꢀ ꢀ ꢀH–Oꢀ ꢀ ꢀZn in chlorin Zn-4 (1691
ꢂ 1655 5 36 cmꢂ1). Therefore, Zn-1 possessing a porphyrin p-
system self-aggregated to form similar supramolecular structures
with in vivo and in vitro self-aggregates of chlorosomal Chls (7–11)
and their models of synthetic chlorins including Zn-4
(13,15,27,30,31). Oligomeric Zn-1 was built by special bondings,
13-C5Oꢀ ꢀ ꢀ
In a THF solution (0.2 mM), neither of the enantiomers Zn-1R/S
gave apparent CD peaks at 300–800 nm. This is reasonable because
the 31-chiral functional group neither conjugates with the porphyrin
H–O(31)ꢀ ꢀ ꢀZn, as well as p–p interaction of composite porphyrin
moieties.
It is noteworthy that an 8 cmꢂ1 higher wavenumber shift in
monomeric 13-C5O stretching vibrational peaks was observed by
17,18-dehydrogenation as in Zn-4 fi Zn-1. This shift can be
explained by the following. Change of the p-conjugation from
chlorin to porphyrin enhances the steric repulsion between the 17-
propionate and 132-H2, and the 13-C5O in Zn-1 was more out of
the tetrapyrrole p-plane than that in Zn-4. Therefore, the 13-C5O
in Zn-1 was less conjugated with p-system of cyclic tetrapyrrole in
a molecule and gave a higher wavenumber in the stretching
vibration than that in Zn-4.
31-Racemically pure Zn-1R/S gave the same visible absorption
bands as the 1:1 mixture Zn-1 in THF and were monomeric species
In 1%(vol/vol) THF and cyclohexane (0.2 mM), Zn-1R afforded
the same visible spectrum as Zn-1S, as expected (Fig. 4A), but
Figure 4. Visible spectrum (A) of 2 3 10ꢂ4 M Zn-1R and CD spectra (B)
of 2 3 10ꢂ4 M Zn-1R (—) and Zn-1S (– – –) in 1%(vol/vol) THF and
cyclohexane in a 1 mm cell.

174 Hitoshi Tamiaki et al.
Figure 5. Visible spectra of 1 3 10ꢂ5 M Zn-2 in THF (– – –) and 1%(vol/
vol) THF and cyclohexane (—) in a 10 mm cell and of the solid film (ꢀ ꢀ ꢀ)
on a quartz glass.
p-system nor affects it from the steric aspect. In 1%(vol/vol) THF
and cyclohexane (0.2 mM), Zn-1R afforded S-shaped CD peaks
(þ/ꢂ) at both the redshifted Qy and Soret regions (solid line in
Fig. 4B). This observation can be ascribed to the formation of
ordered supramolecules by self-aggregation of chiral Zn-1R. Zn-
1R self-aggregated in the nonpolar organic solvent to mainly
form clockwise-twisted and helical supramolecular structures
based on the exciton chirality method (32). In contrast, self-
aggregates of Zn-1S gave reverse S-shapes (ꢂ/þ) in the CD
spectrum, which was a complete mirror image of that observed
in Zn-1R self-aggregates (Fig. 4B). As expected, Zn-1S self-
aggregated to form mirror-imaged (anticlockwise-twisted and
helical) supramolecules of Zn-1R self-aggregates. (Very recently,
Balaban et al. have reported similar observation of CD spectra in
self-aggregates of chiral porphyrins [33].) Generation of the CD
activity by self-aggregation of the 31-chiral compounds is
supported by the results that racemic mixture Zn-1 gave no
CD peaks in the self-aggregates and by the reports that both Zn-
4R/S self-aggregates gave giant CD peaks (13,27).
Figure 6. IR spectra of Zn-2 in THF in a 0.1 mm KBr cell (A) and in the
solid film on an aluminum-coated glass (B).
than the Qy band. In 1%(vol/vol) THF and cyclohexane (0.01
mM), Zn-2 self-aggregated to give similar oligomeric peaks at
464 and 644 nm with Zn-1 self-aggregates (solid line in Fig. 5).
The redshifted and broadened Soret band was twice as intense as
the Qy band. The self-aggregated Qy band was redshifted in 920
cmꢂ1 (608 fi 644 nm) and about three times wider in the
FWHM (320 fi 920 cmꢂ1), compared with the monomeric ones.
At 20 times lower concentration (0.01 mM), most of Zn-2 self-
aggregated to form large oligomers, whereas Zn-1 remained
monomeric. Zn-2 (3-CH2OH) was more self-aggregative than
Zn-1 [3-CH(OH)Me] because of the sterically less-hindered 31-
hydroxy group, which is one of its most important interactive
substituents (12,31). Therefore, Zn-1 made oligomers more
easily than Zn-2 to form oligomeric precipitates just after
preparation of the 0.2 mM solution. The resulting solid film gave
exclusively oligomeric species possessing visible peaks at 463
and 649 nm and no residual monomeric peaks (dotted line in
Fig. 5). It is noted that both self-aggregates of porphyrins Zn-1/2
gave almost the same peak positions at around 650 nm (redshifts
5 ca 1000 cmꢂ1), whereas oligomers of more self-aggregative
chlorins Zn-6 (3-CH2OH) (12) had a more redshifted Qy band
(647 fi 741 nm, D 5 1960 cmꢂ1) compared with oligomeric Zn-
4 [3-CH(OH)Me] (13) possessing a Qy peak at 705 nm (‹ 645
nm, D 5 1300 cmꢂ1) (see Table 1).
Self-aggregation of zinc methyl
31-demethyl-protobacteriopheophorbide-d Zn-2
In THF (0.01 mM), Zn-2 (3-hydroxymethyl) was a monomeric
species possessing 428 and 608 nm peaks as Soret and Qy
maxima, respectively (broken line in Fig. 5), as well as Zn-1 [3-
(1-hydroxyethyl)]. The Soret band had 10 times greater intensity
Table 1. Visible absorption spectral data of zinc methyl (proto)bacterio-
pheophorbides-d Zn-1(R/S)/2/4/6 in monomeric and oligomeric solutions:
absorption maximum kmax (nm), FWHM (cmꢂ1) and redshift D (cmꢂ1) of
Monomeric Zn-2 in THF gave 3470, 1742, 1699 and 1605 cmꢂ1
vibrational bands (Fig. 6A), and oligomeric Zn-2 in the solid film
gave 3210, 1736, 1661 and 1609 cmꢂ1 vibrational bands (Fig. 6B).
Lower wavenumber shifts in 31-O–H and 13-C5O stretching
vibrational bands (3470 ꢂ 3210 5 260 cmꢂ1 and 1699 ꢂ 1661 5
38 cmꢂ1, respectively) by self-aggregation of Zn-2 were com-
parable with values of 240/36 cmꢂ1 in the corresponding chlorin
Zn-6 (12), indicating that oligomeric Zn-2 had supramolecular
structures similar to oligomeric Zn-1/6.
k
max by self-aggregation [5f1/kmax(monomer)ꢂ1/kmax(aggregates)g3107]
kmax [FWHM]
1%(vol/vol)
THF–cyclohexane
THF
427
607 [350]
427
607 [350]
428 [970]
608 [320]
645 [320]
647 [320]
D
Zn-1 (31R/S ¼ 1/1)
Zn-1R/Zn-1S
Zn-2
475 (sh)
649 [810]
478
652 [»1000]
464
644 [920]
705 [670]*
741 [750]
»2400
1070
2500
1140
1810
920
CONCLUSIONS
Zn-4 (31R/S ¼ 1/1)
1300
1960
Zinc methyl protobacteriopheophorbides-d (Zn-1/2) possessing a
porphyrin p-system were prepared by modifying naturally occur-
ring Chl-a (via 17,18-dehydrogenation by DDQ) as models of
Zn-6
*In 1%(vol/vol) CH2Cl2–cyclohexane.

Photochemistry and Photobiology, 2005, 81 175
intermolecular interaction within CH2Cl2-treated (31R)-type bacterio-
chlorophyll c solid aggregate. J. Phys. Chem. B 108, 2726–2734.
12. Tamiaki, H., M. Amakawa, Y. Shimono, R. Tanikaga, A. R. Holzwarth
and K. Schaffner (1996) Synthetic zinc and magnesium chlorin aggre-
gates as models for supramolecular antenna complexes in chlorosomes of
green photosynthetic bacteria. Photochem. Photobiol. 63, 92–99.
13. Tamiaki, H., S. Takeuchi, S. Tsudzuki, T. Miyatake and R. Tanikaga
(1998) Self-aggregation of synthetic zinc chlorins with a chiral 1-
hydroxyethyl group as a model for in vivo epimeric bacteriochloro-
phyll-c and d aggregates. Tetrahedron 54, 6699–6718.
14. Yagai, S., T. Miyatake and H. Tamiaki (2002) Regio- and
stereoisomeric control of the aggregation of zinc-chlorins possessing
inverted interactive hydroxyl and carbonyl groups. J. Org. Chem.
67, 49–58.
15. Tamiaki, H., M. Omoda, Y. Saga and H. Morishita (2003) Synthesis of
homologously pure bacteriochlorophyll-e and f analogues from BChls-
c/d via transformation of the 7-methyl to formyl group and self-
aggregation of synthetic zinc methyl bacteriopheophorbides-c/d/e/f in
non-polar organic solvent. Tetrahedron 59, 4337–4350.
chlorosomal BChl-d possessing a chlorin p-system. Synthetic zinc
porphyrins Zn-1/2 self-aggregated in 1%(vol/vol) THF and cyclo-
hexane to give large oligomers with a redshifted and broadened
Qy band at around 650 nm compared with the monomeric. The
supramolecular structures of their oligomers were similar to those
of the corresponding chlorins Zn-4/6 as well as of chlorosomal
self-aggregates, but the Qy peaks were situated at a higher energy
position than those of chlorin self-aggregates (see Table 1). Con-
sidering that zinc bacteriochlorins afforded similar self-aggregates
(17) and 7-, 8- and 17-substitution did not disturb the formation of
the self-aggregates (15,34–38), no alternation on the Qx axis in-
cluding the present formation of the 17,18-double bond affected
self-aggregation of chlorosome-type pigments. Therefore, our
previous finding (12,14) was supported: the 31-hydroxy, 13-keto-
carbonyl groups and coordinative central metal on the Qy axis are
important for chlorosomal self-aggregation.
16. de Boer, I., J. Matysik, M. Amakawa, S. Yagai, H. Tamiaki, A. R.
Holzwarth and H. J. M. de Groot (2003) MAS NMR structure of
a microcrystalline Cd-bacteriochlorophyll d analog. J. Am. Chem. Soc.
125, 13374–13375.
17. Kunieda, M., T. Mizoguchi and H. Tamiaki (2004) Diastereoselective
self-aggregation of synthetic 3-(1-hydroxyethyl)-bacteriopyrochloro-
phyll-a as a novel photosynthetic antenna model absorbing near the
infrared region. Photochem. Photobiol. 79, 55–61.
18. Balaban, T. S., A. Eichho¨fer and J.-M. Lehn (2000) Self-assembly by
hydrogen bonding and p-p interactions in the crystal of a porphyrin—
attempts to mimic bacteriochlorophyll c. Eur. J. Org. Chem.
4047–4057.
19. Balaban, T. S., A. D. Bhise, M. Fischer, M. Linke-Schaetzel,
C. Roussel and N. Vanthuyne (2003) Controlling chirality and optical
properties of artificial antenna systems with self-assembling porphyrins.
Angew. Chem. Int. Ed. 42, 2140–2144.
20. Tamiaki, H., S. Kimura and T. Kimura (2003) Self-aggregation of
synthetic 21-hydroxy-121/131-oxo-porphyrins. Tetrahedron 59,
7423–7435.
21. Tamiaki, H., T. Kubota and R. Tanikaga (1996) Aggregation of
synthetic zinc complexes of cyclotetrapyrroles. Chem. Lett. 639–640.
22. Hynninen, P. H. (1991) Chemistry of chlorophylls: modifications. In
Chlorophylls (Edited by H. Scheer), pp. 145–209. CRC Press, Boca
Raton, FL.
23. Tamiaki, H., T. Watanabe and T. Miyatake (1999) Facile synthesis
of 131-oxo-porphyrins possessing 3-vinyl or 3-formyl group, proto-
chlorophyll-a/d derivatives by 17,18-dehydrogenation of chlorins.
J. Porphyrins Phthalocyanines 3, 45–52.
24. Smith, K. M., G. M. F. Bisset and M. Bushell (1980) Partial syntheses
of optically pure methyl bacteriopheophorbides c and d from methyl
pheophorbide a. J. Org. Chem. 45, 2218–2224.
25. Tamiaki, H., T. Watanabe and M. Kunieda (2005) Self-aggregation of
synthetic zinc 17,18-cis-bacteriochlorophyll-d derivative. In Photosyn-
thesis: Fundamental Aspects to Global Perspective (Edited by A. der
Est and D. Bruce) Allen Press, Lawrence, KS.
26. Tamiaki, H., M. Kouraba, K. Takeda, S. Kondo and R. Tanikaga (1998)
Asymmetric synthesis of methyl bacteriopheophorbide-d and analogues
by stereoselective reduction of the 3-acetyl to the 3-(1-hydroxyethyl)
group. Tetrahedron: Asymmetry 9, 2101–2111.
27. Balaban, T. S., H. Tamiaki, A. R. Holzwarth and K. Schaffner (1997)
Self-assembly of methyl zinc (31R)- and (31S)-bacteriopheophorbides d.
J. Phys. Chem. B 101, 3424–3431.
28. Tamiaki, H., M. Amakawa, A. R. Holzwarth and K. Schaffner (2002)
Aggregation of synthetic metallochlorins in hexane. A model of
chlorosomal bacteriochlorophyll self-assemblies in green bacteria.
Photosynth. Res. 71, 59–67.
Acknowledgements—We thank Dr. Shin-ichi Sasaki, Mr. Hidetada
Morishita and Mr. Shoichiro Takeuchi of Ritsumeikan University for
their experimental assistance. This work was partially supported by
Grants-in-Aid for Scientific Research (15033271) on Priority Areas (417)
from the Ministry of Education, Culture, Sports, Science and Technology
(MEXT) of the Japanese Government and for Scientific Research (B)
(15350107) from the Japan Society for the Promotion of Science (JSPS).
REFERENCES
1. Senge, M. O. and S. Hatscher (2000) The malaria pigment
haemozoin—a focal point of action for antimalarial drugs. Chem-
BioChem. 1, 247–249.
2. Blankenship, R. E., J. M. Olson and M. Miller (1995) Antenna
complexes from green photosynthetic bacteria. In Anoxygenic Photo-
synthetic Bacteria (Edited by R. E. Blankenship, M. T. Madigan and
C. E. Bauer), pp. 399–435. Kluwer Academic Publishers, Dordrecht,
The Netherlands.
3. Olson, J. M. (1998) Chlorophyll organization and function in green
photosynthetic bacteria. Photochem. Photobiol. 67, 61–75.
4. Blankenship, R. E. and K. Matsuura (2003) Antenna complexes from
green photosynthetic bacteria. In Light-Harvesting Antennas in
Photosynthesis (Edited by B. R. Green and W. W. Parson), pp. 195–
217. Kluwer Academic Publishers, Dordrecht, The Netherlands.
5. Frigaard, N.-U., A. Gomez, M. Chew, H. Li, J. A. Maresca and D. A.
Bryant (2003) Chlorobium tepidum: insight into the structure,
physiology, and metabolism of a green sulfur bacterium derived from
the complete genome sequence. Photosynth. Res. 78, 93–117.
6. Saga, Y. and H. Tamiaki (2004) Fluorescence spectroscopy of single
photosynthetic light-harvesting supramolecular systems. Cell Biochem.
Biophys. 40, 149–166.
7. Tamiaki, H. (1996) Supramolecular structure in extramembraneous
antennae of green photosynthetic bacteria. Coord. Chem. Rev. 148,
183–197.
8. Saga, Y., K. Matsuura and H. Tamiaki (2001) Spectroscopic studies on
self-aggregation of bacteriochlorophyll-e in non-polar organic solvents:
effects of stereoisomeric configuration at the 31-position and alkyl
substituents at the 81-position. Photochem. Photobiol. 74, 72–80.
9. van Rossum, B.-J., D. B. Steensgaard, F. M. Mulder, G. J. Boender,
K. Schaffner, A. R. Holzwarth and H. J. M. de Groot (2001) A refined
model of the chlorosomal antennae of the green bacterium Chlorobium
tepidum form proton chemical shift constraints obtained with high-field
2-D and 3-D MAS NMR dipolar correlation spectroscopy. Biochemistry
40, 1587–1595.
29. Yagai, S. and H. Tamiaki (2001) Synthesis and self-aggregation of zinc
chlorophylls possessing an x-hydroxyalkyl group: effect of distance
between interactive hydroxy group and chlorin moiety on aggregation.
J. Chem. Soc. Perkin Trans. 1, 3135–3144.
30. Tamiaki, H., H. Kitamoto, A. Nishikawa, T. Hibino and R. Shibata
(2004) Determination of 31-stereochemistry in synthetic bacteriochlo-
rophyll-d homologs and self-aggregation of their zinc complexes.
Bioorg. Med. Chem. 12, 1657–1666.
10. Mizoguchi, T., Y. Saga and H. Tamiaki (2002) Isolation and structure
determination of a complete set of bacteriochlorophyll-d homologs and
epimers from a green sulfur bacterium Chlorobium vibrioforme and
their aggregation properties in hydrophobic solvents. Photochem.
Photobiol. Sci. 1, 780–787.
11. Umetsu, M., J. G. Hollander, J. Matysik, Z.-Y. Wang, T. Adschiri,
T. Nozawa and H. J. M. de Groot (2004) Magic-angle spinning nuclear
magnetic resonance under ultrahigh field reveals two forms of

176 Hitoshi Tamiaki et al.
31. Yagai, S., T. Miyatake, Y. Shimono and H. Tamiaki (2001)
Supramolecular structure of self-assembled synthetic zinc-131-oxo-
chlorins possessing a primary, secondary or tertiary alcoholic 31-
hydroxyl group: visible spectroscopic and molecular modeling studies.
Photochem. Photobiol. 73, 153–163.
32. Pescitelli, G., S. Gabriel, Y. Wang, J. Fleischhauer, R. W. Woody and
N. Berova (2003) Theoretical analysis of the porphyrin—porphyrin
exciton interaction in circular dichroism spectra of dimeric tetraar-
ylporphyrins. J. Am. Chem. Soc. 125, 7613–7628.
33. Balaban, T. S., M. Linke-Schaetzel, A. D. Bhise, N. Vanthuyne and
C. Roussel (2004) Green self-assembling porphyrins and chlorins as
mimics of the natural bacteriochlorophylls c, d, and e. Eur. J. Org.
Chem. 3919–3930.
34. Tamiaki, H., S. Miyata, Y. Kureishi and R. Tanikaga (1996)
Aggregation of synthetic zinc chlorins with several esterified alkyl
chains as models of bacteriochlorophyll-c homologs. Tetrahedron 52,
12421–12432.
35. Tamiaki, H., M. Kubo and T. Oba (2000) Synthesis and self-assembly
of zinc methyl bacteriopheophorbide-f and its homolog. Tetrahedron
56, 6245–6257.
36. Oba, T. and H. Tamiaki (2001) Effect of C7-substituent on self-
assembly of chlorosomal chlorophylls. Supramol. Chem. 12, 369–378.
37. Sasaki, S., M. Omoda and H. Tamiaki (2004) Effects of C8-substituents
on spectroscopic and self-aggregation properties of synthetic bacterio-
chlorophyll-d analogues. J. Photochem. Photobiol. A: Chem. 162,
307–315.
38. Sasaki, S. and H. Tamiaki (2004) Self-assembly of synthetic
bacteriochlorophyll-f analogues possessing C8-formyl group. Bull.
Chem. Soc. Jpn. 77, 797–800.