Reassociation of all-trans-3,4-dihydroanhydrorhodovibrin with lh1 subunits isolated from rhodospirillum rubrum: Selective binding of all-trans isomer from mixture of cis- and trans-Isomers

Article abstract of DOI:10.1246/bcsj.20120230

The reassociation of carotenoids with the photosynthetic light-harvesting 1 (LH1) pigmentprotein complex has proven to be a powerful technique to investigate the function of carotenoids. We have recently investigated the reassociation of the carotenoids, spirilloxanthin (number of conjugated C=C bonds (n) = 13), anhydrorhodovibrin (n = 12), rhodopin (n = 11), and spheroidene (n = 10)) with carotenoid-depleted LH1 subunits isolated from Rhodospirillum rubrum. In the case of the rhodopin complex, small anomalies were observed in the absorption maximum of the Qy band of bacteriochlorophyll a (BChl a Qy) and the singletsinglet energy-transter efficiency of 1carotenoid 1BChl a within the LH1carotenoid complex. The reason for these anomalies could probably be ascribed to the rhodopin hydroxy group. To investigate this further, reassociation experiments of the LH1 complex with hydroxy-protected rhodopin, all-trans-3,4-dihydroanhydrorhodovibrin (1), were conducted. We observed a normal absorption maximum (883 nm) of the BChl a Qy band in the LH11 complex. The energy-transter efficiency of 1carotenoid (11)1BChl a was found to be 53%, giving a good correlation with the other carotenoids. Furthermore, we found a preferential complexation of alltrans- 1 to carotenoid-depleted LH1 from a 1:1 mixture of all-trans-1 and cis-1 (cis-1 consists of a 1:1 mixture of 5-cis-5- trans-1 and 5-trans-5-cis-1).

Full text of DOI:10.1246/bcsj.20120230

©
2013 The Chemical Society of Japan
Bull. Chem. Soc. Jpn. Vol. 86, No. 1, 121­128 (2013)
121
Reassociation of All-trans-3,4-Dihydroanhydrorhodovibrin with LH1
Subunits Isolated from Rhodospirillum rubrum: Selective Binding
of All-trans Isomer from Mixture of cis- and trans-Isomers
Mai Yamamoto, Tomoko Horibe,2,4 Yoshiaki Nishisaka,2,4 Shuichi Suzuki,1,3 Masatoshi Kozaki,1
1
Ritsuko Fujii,3,4 Matsumi Doe, Mamoru Nango, Keiji Okada,*1,3,4 and Hideki Hashimoto*2,3,4
1
2,4
1Department of Chemistry, Graduate School of Science, Osaka City University,
Sugimoto, Sumiyoshi-ku, Osaka 558-8585
2Department of Physics, Graduate School of Science, Osaka City University,
Sugimoto, Sumiyoshi-ku, Osaka 558-8585
3The Osaka City University Advanced Research Institute for Natural Science and Technology (OCARINA),
Sugimoto, Sumiyoshi-ku, Osaka 558-8585
4JST/CREST, 4-1-8 Hon-cho, Kawaguchi, Saitama 332-0012
Received August 27, 2012; E-mail: okadak@sci.osaka-cu.ac.jp, hassy@sci.osaka-cu.ac.jp
The reassociation of carotenoids with the photosynthetic light-harvesting 1 (LH1) pigment­protein complex has
proven to be a powerful technique to investigate the function of carotenoids. We have recently investigated the
reassociation of the carotenoids, spirilloxanthin (number of conjugated C=C bonds (n) = 13), anhydrorhodovibrin (n =
12), rhodopin (n = 11), and spheroidene (n = 10)) with carotenoid-depleted LH1 subunits isolated from Rhodospirillum
rubrum. In the case of the rhodopin complex, small anomalies were observed in the absorption maximum of the Qy band
1
1
of bacteriochlorophyll a (BChl a Qy) and the singlet­singlet energy-transter efficiency of carotenoid* ¼ BChl a*
within the LH1­carotenoid complex. The reason for these anomalies could probably be ascribed to the rhodopin hydroxy
group. To investigate this further, reassociation experiments of the LH1 complex with hydroxy-protected rhodopin,
all-trans-3,4-dihydroanhydrorhodovibrin (1), were conducted. We observed a normal absorption maximum (883 nm) of
1
1
1
the BChl a Qy band in the LH1­1 complex. The energy-transter efficiency of carotenoid* ( 1*) ¼ BChl a* was found
to be 53%, giving a good correlation with the other carotenoids. Furthermore, we found a preferential complexation of all-
trans-1 to carotenoid-depleted LH1 from a 1:1 mixture of all-trans-1 and cis-1 (cis-1 consists of a 1:1 mixture of 5-cis-5¤-
trans-1 and 5-trans-5¤-cis-1).
Carotenoids are important components of the light-harvest-
C=C bonds (n) = 13), anhydrorhodovibrin (n = 12), rhodopin
ing 1 and 2 (LH1 and LH2) antenna pigment­protein com-
plexes. The successful crystal structure analysis of the LH2
complex from Rhodopseudomonas acidophila strain10050
has provided definitive information on the structure of this
antenna pigment­protein complex.1 However, the location
and the structure of carotenoids remain uncertain for the
(n = 11), and spheroidene (n = 10), isolated from different
5
bacteria (Figure 1). The absorption maximum of the Q band
y
of bacteriochlorophyll a (BChl a) existing as a dimer form
in the reassociated complexes is located around 881­884 nm
depending on the environment of the BChl a and the carotenoid
incorporated, i.e., when the complex binds spirilloxanthin,
anhydrorhodovibrin, or spheroidene, the absorption maximum
was observed at 883­884 nm, whereas when the complex binds
rhodopin, the maximun appeared at 881 nm (Figure 2). This
2
LH1 complex. Generally, carotenoids, which are generated in
these complexes by biosynthesis, exist as a mixture of different
3
species. Therefore, it has been difficult to extract the functions
of a single particular carotenoid from these complexes. The
reassociation of a pure carotenoid with the carotenoid-depleted
LH1 has proven to be a powerful technique to investigate the
slightly decreased wavelength shift of the BChl a Q band
y
in the LH1­rhodopin complex was considered to be the result
of the interaction between the polar hydroxy group of rhodopin
4
5
functions of carotenoids: In this way, the precise spectroscopic
and BChl in LH1. Furthermore, we evaluated the efficiencies
1
1
characterization and determination of the singlet energy-trans-
fer efficiencies, from the incorporated carotenoid to the chloro-
phyll moieties, can be obtained as a function of carotenoid
structure in the reassociated complex. We have recently investi-
gated the reassociation of the LH1 complex of the purple
photosynthetic bacterium Rhodospirillum (Rs.) rubrum with
the four carotenoids, spirilloxanthin (number of the conjugated
of singlet energy transfer from carotenoid* ¼ BChl a*
within the LH1­carotenoid complexes, by comparing their
absorption and fluorescence (907 nm)-excitation spectra. The
singlet energy-transter efficiency was found to increase with
a decrease in the number of C=C bonds n for the four
5
carotenoids. However, the plot slightly deviates for the LH1­
rhodopin complex (Figure 3 (right)). Thus, anomalies have
Published on the web January 10, 2013; doi:10.1246/bcsj.20120230

1
22
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
3,4-Dihydroanhydrorhodovibrin­LH1 Complex
been observed in both the absorption wavelength of the BChl
a Q band and the efficiency of the singlet energy transfer in the
y
reassociated LH1­rhodopin complex.
In order to clarify these anomalies, in this study, a hydroxy-
protected analogue of rhodopin was designed and synthe-
sized for reassociation experiments with the LH1 complex.
The carotenoid, 3,4-dihydroanhydrorhodovibrin (1: O-methyl-
rhodopin) was found to be a suitable carotenoid for this
purpose (Figure 1). Although carotenoid 1 has been found
in Rs. rubrum ST4 mutants, it is only a minor carotenoid and
6
its isolation from this natural source is inefficient. In such
cases, chemical synthesis is a powerful tool, and in this study,
we synthesized 1 and investigated the reassociation of the
LH1­1 complex.
Results and Discussion
Synthesis of All-trans-3,4-Dihydroanhydrorhodovibrin
(
1).
Scheme 1 illustrates the synthetic route of 3,4-di-
hydroanhydrorhodovibrin (1) from the known 7-methoxy-3,7-
7
,8
dimethyloct-2-enal (2). In the transformation from 2 to 4, four
7
,9
carbon atoms can be extended using phosphonate 3 through
a Horner­Wadsworth­Emmons reaction. The cyano group in
Figure 1. Chemical structures of some typical carotenoids.
Figure 2. Change in the absorption spectrum of the carotenoid-depleted LH1 antenna in Tris-HCl buffer containing LDAO during
the addition of 1 (left) and the purified all-trans-1­LH1 complex after DE52 chromatography, sucrose density gradient
centrifugation, and PD-10 column chromatography (right).
Figure 3. Absorption and emission spectra of the reassociated LH1­all-trans-1 complex (left), the comparison of the absorption and
the fluorescence excitation spectra (middle), and the correlation between the singlet­singlet energy-transter efficiency vs. the
number of conjugated C=C bonds (right).

M. Yamamoto et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
123
Scheme 1. Synthesis of 3,4-dihydroanhydrorhodovibrin (1).
4
was transformed to aldehyde 5 through a diisobutyl-
spectrum (red line in Figure 2 (left)) displayed a sharp band
at around 880 nm. The observed red shift of this band is
ascribed to the formation of the aggregated LH1 subunits
stabilized by 1.
aluminium hydride (DIBAH) reduction. Repeating the same
procedure provided the eight-carbon extended (from 2) alde-
hyde, 15-methoxy-3,7,11,15-tetramethylhexadeca-2,4,6,8,10-
pentaenal (7). The Ti(0)-mediated cross-coupling (McMurry
olefin synthesis using TiCl (43 mmol)­LiAlH (22 mmol) in
THF, rt, overnight) of 7 (20 mmol) with 3,7,11,15-tetramethyl-
hexadeca-2,4,6,8,10,14-hexaenal (8: 19 mmol)10 produced a
mixture of carotenoids. The mixture was crudely separated into
three portions by column chromatography; lycopen (9), 3,4-
dihydroanhydrorhodovibrin (1), and 3,4,3¤,4¤-tetrahydrospiril-
loxantine (10) moieties in a molar ratio of 11:40:13. Each of
these carotenoid portions consisted of a complex mixture of
trans- and cis-stereoisomers. Pure all-trans 1 was obtained from
the second fraction by repeating recrystallization from n-hexane
at ¹20 °C and it was used for the reassociation experiments. The
structure of all-trans-1 was confirmed by detailed NOE experi-
The crude reassociated complex was further purified using
ion-exchange chromatography (DE52) and sucrose density
3
4
5
gradient centrifugation. NaCl was removed by passing the
complex through a prepacked disposable PD-10 column to
give the pure reassociated LH1 complex, as judged from its
absorption spectrum, which displayed the dimeric BChl a Qy
band at 883 nm (Figure 2 (right)). The vibrational structure of
the S ¼ S transition (around 489 nm) of 1 is clearly observed,
0
2
indicating that 1 was systematized into the reassociated LH1
complex. The peaks assigned to the BChl a Q , Q , and Soret
y
x
absorptions were clearly observed at 883, 591, and 376 nm,
respectively.
The observed absorption maximum of the BChl a Q band
y
1
ments for all vinyl protons (see H NMR (600 MHz) spectrum in
(883 nm) in the LH1­1 complex appears at longer wavelength
than that of the LH1­rhodopin complex (881 nm), and is close
to the maximum observed for the reassociated LH1 complexes
containing the other three carotenoids (883­884 nm), as well as
Supporting Information). The experimental procedures for the
syntheses of 3 and 8 are described in Supporting Information.
Reassociation of All-trans-1 with Carotenoid-Depleted
LH1: Synthesis of the LH1­All-trans-1 Complex. In the
reassociation experiments, pure all-trans-1 was used (hereafter,
5
that of the natural LH1 complex from Rs. rubrum S1. Thus, it
is clear that the protection of hydroxy group of rhodopin gives
1
denotes all-trans-1 unless otherwise noted). Figure 2 (left)
shows the absorption spectral changes during the addition of
to the carotenoid-depleted LH1 antenna in a Tris-HCl buffer
rise to the usual absorption of the BChl a Q band. This finding
y
supports our hypothesis that the shorter wavelength observed
for the LH1­rhodopin complex can be ascribed to the inter-
action between the BChl a chromophore and the polar hydroxy
1
containing lauryldimethylamine-oxide (LDAO, surfactant). At
t = 0 h (blue line in Figure 2 (left)), the absorption at 780
5
group of rhodopin.
nm can be assigned to the Q absorption band of the BChl
Energy Transfer from the Excited Singlet State of 1 to
BChl a within the LH1­1 Complex. Figure 3 (left) shows
the absorption and emission spectra of the purified LH1­1
complex. Upon excitation at the main peak of carotenoid 1 at
y
a monomer in the LH1 subunit complex, while the absorptions
at 820 and 869 nm can be assigned to the BChl a dimer and its
aggregates in the complex, respectively.4 During the addition
of a solution of 1 in acetone, the 780 nm band decreased
in intensity and the 869 nm band gradually shifted to longer
wavelengths while increasing in intensity. The final absorption
,5
489 nm, emission from BChl a Q was observed at 902 nm.
y
The fluorescence excitation spectrum (Figure 3 (middle), red
line) was measured by monitoring the fluorescence intensity at

1
24
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
3,4-Dihydroanhydrorhodovibrin­LH1 Complex
Scheme 2. Stereoisomers of 1.
9
02 nm. The excitation spectrum was compared to the absorp-
shows an HPLC trace of the re-extracted carotenoid fraction
from the purified reassociated LH1­1 complex. In both traces, a
tion spectrum at 375 nm at the same intensity (Soret band of
BChl a). The overall efficiency of the singlet­singlet energy
transfer from 1 to the BChl a in the reassociated LH1 complex
was evaluated by comparing the area of the carotenoid absorp-
tion band (420­570 nm) with that of the excitation spectrum
and was determined to be 53%. We have previously measured
singlet energy-transfer efficiencies for LH1 complexes with
spirilloxanthin, anhydrorhodovibrin, rhodopin, and spheroid-
ene, and plotted these efficiencies against n, where a small
deviation was observed for the plot of rhodopin (n = 11).
Replacing the plot of rhodopin (n = 11) with that of 3,4-di-
hydroanhydrorhodovibrin (1, n = 11) resulted in a straight line
with a good correlation coefficient (99.9% in Figure 3 (right)).
Preferential Complexation of All-trans-1 to LH1 from a
cis­trans Mixture of 3,4-Dihydroanhydrorhodovibrin (1).
Several stereoisomers of 3,4-dihydroanhydrorhodovibrin (1)
constituted the second fraction of carotenoids separated by
column chromatography (Scheme 2). We could obtain all-
trans-1 as a pure solid by recrystallization, while the cis-1
isomers were obtained by HPLC separation as an inseparable
small unidentified signal was observed at R = 8.8 min, which
t
could be a different cis-isomer possibly induced by room light.
Interestingly, in the analysis of the re-extracted carotenoids
from the reassociated complex, the all-trans-1 isomer was
found to constitute > 80% of the mixture, while for the cis-1
isomers only a small shoulder was observed (Figure 4 (right)).
This experiment clearly indicates that the carotenoid-depleted
LH1 preferentially reassociates with the all-trans-1 as opposed
to the cis-1 isomers, suggesting that the binding space for
carotenoids in the associated LH1 complex is tight enough to
differentiate the 5- and 5¤-stereoisomers of 3,4-dihydroanhy-
drorhodovibrin (1).
Conclusion
To study the roles of carotenoids in light-harvesting antenna
pigment­protein complexes, we have investigated the reasso-
ciation of the carotenoids, spirilloxanthin (n = 13), anhydro-
rhodovibrin (n = 12), rhodopin (n = 11), and spheroidene (n =
10) with the carotenoid-depleted LH1 subunit obtained from the
5
mixture of 5-cis-5¤-trans-1 and 5-trans-5¤-cis-1 (5-cis-5¤-trans-
purple photosynthetic bacterium Rs. rubrum. Compared to the
1
1
/5-trans-5¤-cis-1 µ 1:1 by H NMR). The structural assign-
other reassociation complexes, the LH1­rhodopin reassociated
complex exhibited small anomalies in the absorption maximum
ment of these isomers is shown in Supporting Information.
When a 1:1 mixture of all-trans 1 and the cis-isomeric
mixture was employed for the reassociation with the carote-
noid-depleted LH1, the reassociated complex after purifica-
tion exhibited an absorption maximum at 883 nm, identical to
the reassociated LH1­all-trans-1 (Figure 2 (right)). Then, the
incorporated carotenoid component in the reassociated LH1
complex was extracted from the complex (for details see
Experimental Section) and analyzed by HPLC.
of the BChl a Q band and the efficiency of singlet­singlet
y
1
1
energy transfer from carotenoid* ¼ (BChl a)*. We antici-
pated that these small anomalies could result from the inter-
action between the polar OH group of rhodopin and BChl a.
5
In order to obtain further insight into these observations,
we investigated the reassociation of the LH1 complex with
3,4-dihydroanhydrorhodovibrin (1, n = 11), which is the OH-
protected (OMe) variant of rhodopin. Because of the low
availability of 1 from natural sources, we chemically synthe-
sized carotenoid 1, which could be obtained as the pure all-
trans-1 by recrystallization. The reassociation complex LH1­1
exhibited an absorption maximum at 883 nm and indicated no
Figure 4 (left) shows an HPLC profile of a cis­trans mixture
of 1 before the reassociation experiment, showing peaks
stemming from all-trans-1 (retention time (R ) = 9.9 min) and
t
its cis-isomers (R = 9.6 min, not separated). Figure 4 (right)
t

M. Yamamoto et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
125
Figure 4. HPLC profiles of a cis­trans mixture of 1 (all-trans-1 and 5-cis-1 isomers) before (left) and after (right) reassociation with
carotenoid-depleted LH1.
specific interaction between the OMe of 1 and BChl a, thus
supporting our previous consideration. We also determined
the efficiency of singlet­singlet energy transfer for LH1­1
Growth of the Cells of Purple Photosynthetic Bacteria
and Preparation of Chromatophores. Cells of Rs. rubrum
wild-type strain S1, Rhodopseudomonas (Rps.) palustris were
grown anaerobically in the light in C-succinate medium at
29 °C. These cells were harvested by centrifugation (12000
rpm for 5 min at 4 °C), resuspended in 20 mM Tris-HCl
buffer (pH 8.0), and stored in a freezer (¹30 °C) until use. The
preparation of chromatophores was carried out essentially as
5
(
(
53% for n = 11). The plots provided an excellent straight line
r = 99.9%).
As a result of the chemical synthesis, several stereoisomers
of 1 were also obtained. When the reassociation experiment
was carried out using a 1:1 mixture of all-trans-1 and the cis-1
isomers (5-cis-5¤-trans-1/5-trans-5¤-cis-1 µ 1:1), the carote-
noid-depleted LH1 subunits selectively bound to all-trans-1
4
d,11
described previously:
The cells were homogenized in 50
mM PBS buffer (pH 7.0) and a small amount of DNase and
(
>80%). In natural systems, carotenoids exist as a mixture
MgCl were added. The cells were broken by two passages
2
of different species in LH1 and LH2 through biosynthetic
processes. For this reason, these complexes would have a
relatively loose space for recognition of all-trans-carotenoids.
However, high selectivity was observed for all-trans-carote-
noids/cis-carotenoids, indicating a moderately tight space to
differentiate all-trans- and cis-carotenoids.
through a French press (140 MPa). Unbroken cells and cell-
wall materials were removed by centrifugation (12000 rpm for
90 min at 4 °C) and then, the chromatophores were pelleted by
ultracentrifugation (50000 rpm, for 90 min at 4 °C).
Isolation of Carotenoid-Depleted LH1 Subunits from Rs.
rubrum S1. The native LH1 complex was isolated and puri-
fied from the chromatophores of Rs. rubrum S1 as described
Experimental
4d,11b
previously:
The freeze-dried pellets (800 mg) were sus-
General. 1H NMR spectra were recorded on a JEOL JNM-
LA400 and a Brucker Avance 600. H­H COSY and H­H
NOESY spectra were recorded on a Brucker Avance 600. FAB-
MS spectra were recorded on a JEOL JMS-AX-700 spectrom-
eter. Infrared spectra were measured using KRS-5 plates and
KBr plates on a SHIMADZU FTIR-8700 spectrometer.
Absorption and fluorescence spectra were measured on a
pended in benzene (200 mL). The suspension was well stirred
for 20 min and centrifuged (10000 rpm for 10 min at 10 °C).
These procedures were repeated (three times) until the dried
chromatophores became green (carotenoid-depleted LH1, 395
mg). Thus obtained carotenoid-depleted LH1 complex, con-
taining LH1-¡ and LH1-¢ polypeptides and Bchl a (as mono-
mers, dimers, and aggregates), was suspended in 40 mM Tris-
HCL buffer (32 mL, pH 8.0) containing 10 mM sodium ascor-
bate and 0.3% LDAO (lauryldimethylamine-oxide). The whole
suspension was well stirred for 45 min in a refrigerator (at 4 °C)
and centrifuged (7650 rpm for 40 min at 4 °C). The supernatant
solution was kept in a freezer (at ¹80 °C) before use.
JASCO V-530 (experimental conditions: 4 °C under N ) and a
2
JASCO FP-6600 (experimental conditions: 0 °C under N2),
respectively. HPLC was performed using a Shimadzu LC-20
equipped with SPD-20AV (column: LiChrosorb Si-60, 5 ¯m,
2
4
.6 © 300 mm , eluent: 2.0 vol % ethyl acetate in hexane,
wavelength: 350­520 nm). For the chromatographic separation
of the synthesized materials, column chromatography was
used (silica gel 60, 100­200 mesh). All commercially available
compounds were reagent grade and used without further puri-
fication. Toluene and THF were dried and distilled over
sodium. All procedures using carotenoids and photosynthetic
bacteria were carried out in dim red light.
Reassociation of LH1­All-trans-1 Complex. The above
carotenoid-depleted LH1 solution (ca. 1.5 mL) was diluted with
40 mM Tris-HCl buffer (45 mL). After addition of 900 ¯L of
5% [v/v] LDAO, the solution was further diluted with 40 mM
Tris-HCl buffer (105 mL). Into this solution, a freshly prepared
acetone solution of 1 (all-trans-1, O.D. (474 nm) = ca. 1.0) was
gradually added dropwise under nitrogen at 0 °C. The progress

1
26
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
3,4-Dihydroanhydrorhodovibrin­LH1 Complex
of the reaction was monitored by measuring the absorption
spectra (Figure 3 (left), recorded at 4 °C) during the addition
of all-trans-1. During the addition, the absorption maximum of
(m, 1H), 6.22­6.14 (two doublet like), 5.95 (d-like, J =
11.2 Hz, 1H), 5.15 (s, 1H), 3.17 (s, 3H), 2.28­2.08 (m, 4H),
2.19 (s, 3H), 1.85 (s, 3H), 1.53­1.40 (m, 2H), 1.14 (s, 6H); IR
¹1
BChl a Q band shifted to a longer wavelength. When the
(KSR-5/cm ): 2972, 2941, 2826, 2206, 1634, 1599, 1574,
y
+
+
+
absorption maximum reached 880 nm (Figure 2 (left), after
961, 739; MS (FAB ): m/z = 248 [M + H] ; HRMS (EI ):
found m/z 247.1936; calcd for C H NO m/z 247.1936. This
6
h), the solution was kept overnight in a refrigerator to
1
6
25
4d,11b
stabilize the reassociation complex.
material was used for the next step without further purification.
11-Methoxy-3,7,11-trimethyldodeca-2,4,6-trienal (5). A
Purification of LH1­All-trans-1 Complex.
The crude
reassociated LH1­1 complex was purified according to the
previous procedures5 using ion-exchange chromatography
20 mL of toluene solution containing 4 (4.9 g, 20 mmol) in a
300-mL three-necked flask was cooled to ¹5 °C in an ice­salt
bath. DIBAH (1.0 M in a hexane solution, 22 mL, 22 mmol)
was slowly added to the toluene solution using a separatory
funnel, while maintaining the temperature below ¹5 °C. After
stirring for an additional 1 h, methanol (12 mL) was slowly
added to the reaction mixture at a temperature below ¹3 °C,
(DEAE­cellulose DE52 column, Whatman) and sucrose density
gradient centrifugation (0.4 M/0.8 M/1.2 M). After the centri-
fugation, the target LH1­1 complex was accumulated in a 0.8 M
sucrose layer. Then the sucrose solution was loaded onto a
DE52 column prequilibrated with 20 mM Tris-HCl buffer
(
pH 8.0) containing 0.03% LDAO. Free (uncomplexed) all-
trans-1 was eluted with 20 mM Tris-HCl buffer containing
0 mM NaCl plus 0.04% Triton X-100. The LH1­1 complex
followed by a saturated aqueous NH Cl solution (ca. 20 mL).
4
The ice­salt bath was removed and the reaction mixture was
stirred for 30 min at room temperature, after which the reaction
mixture was filtered through a celite pad. The celite pad
was washed with diethyl ether. The combined filtrate was
washed with water and a saturated aqueous NaCl solution. The
resulting organic layer was dried over Na SO , then filtered and
5
was eluted with 20 mM Tris-HCl buffer containing 250 mM
NaCl plus 0.03% LDAO. The NaCl was removed by passing
through a PD-10 desalting column. Finally, the concentration
of LDAO was adjusted to 0.01% by dilution with cold 50 mM
Tris-HCl buffer (pH 8.0) to stabilize the LH1­all-trans-1
complex in an LDAO micellar solution.
2
4
concentrated under reduced pressure. The residue was purified
by silica gel column chromatography using hexane­diethyl
ether (1:1 v/v) as an eluent, to give 11-methoxy-3,7,11-
trimethyldodeca-2,4,6-trienal (5, 3.7 g, 76%) as an orange
Re-extraction of Carotenoids from the Reassociated
LH1­1 Complex.
The reassociated LH1­1 complex was
1
loaded onto a DE52 column prequilibrated with water. The
LDAO detergent involved in the reassociated LH1­1 complex
was easily removed by eluting with water. The water remaining
oil. 5: C H O ; MW 250.38; H NMR (400 MHz, CDCl ) as
1
6
26
2
3
an isomeric mixture (ca. 1:1): ¤ 10.10 (10.101, 10.097) (two d,
J = 8.1 Hz, 1H), 7.03­6.93 (two dd like, J = 15 Hz, 11.4 Hz,
1H), 6.30­6.20 (two d, J = 15 Hz, 1H), 6.02 (d, J = 11 Hz,
1H), 5.95 (d, J = 8 Hz, 1H), 3.18 (3.178, 3.170) (two s, 3H),
2.30 (2.305, 2.293) (two s, 3H), 2.30­2.10 (m, 4H), 1.87 (broad
in the DE52 was also removed by pressurizing with N . Then,
2
eluting with a mixture of acetone and methanol (7:2 v/v)
provided a fractions of carotenoids. The carotenoid fraction
was concentrated under reduced pressure and investigated
by HPLC.
Synthesis of Materials. The reaction scheme (Scheme S1)
and procedures for the synthesis of 2, 3, and 8 are described
in Supporting Information.
¹1
s, 3H), 1.53­1.41 (m, 2H), 1.15 (s, 6H); IR (KSR-5/cm ):
2827, 2772, 2725, 1715, 1634, 1593, 1568, 961, 731; MS
+
+
+
(FAB ): m/z = 250 [M ]; HRMS (EI ): found m/z 250.1941;
calcd for C H O m/z 250.1933. This material was used for
1
6
26
2
the next step without further purification.
1
1-Methoxy-3,7,11-trimethyldodeca-2,4,6-trienenitrile
4). See Scheme 1. A 50 mL of THF solution of phosphonate
(5.86 g, 27 mmol) was placed in a 500-mL four-necked flask
15-Methoxy-3,7,11,15-tetramethylhexadeca-2,4,6,8,10-
pentaenenitrile (6). A 130 mL of THF solution of 3 (15.64 g,
71.8 mmol) was placed in a 1-L four-necked flask fitted with a
thermometer and a separatory funnel. The solution was cooled
to ¹10 °C in an ice­salt bath. n-Butyllithium (1.6 M in hexane,
45 mL, 72 mmol) was slowly added to the THF solution using
a separatory funnel, while maintaining the temperature below
¹7 °C. After stirring for an additional 30 min, 5 (15.97 g, 63.8
mmol) was slowly added to the reaction mixture, while the
temperature was maintained below ¹7 °C. After stirring for a
further 1 h, water (ca. 100 mL) and a saturated aqueous solution
(
3
fitted with a thermometer and a separatory funnel. The solution
was cooled to ¹12 °C in an ice­salt bath. n-Butyllithium (1.6 M
in hexane, 17 mL, 27 mmol) was slowly added to the THF
solution using a separatory funnel, while maintaining the tem-
perature below ¹10 °C. After stirring for an additional 30
min, 7-methoxy-3,7-dimethyloct-2-enal (2, 4.5 g, 24 mmol)
was slowly added to the reaction mixture, while the temper-
ature was maintained below ¹10 °C. After stirring a further
1
h, water (ca. 50 mL) and an aqueous solution (ca 50 mL) of
of NH Cl (ca. 100 mL) were added to the reaction mixture, and
4
saturated NH Cl were added to the reaction mixture, and the
the mixture was extracted with diethyl ether (300 mL © 2). The
organic layer was washed with water and a saturated aqueous
NaCl solution. The resulting organic layer was dried over
Na SO , filtered, and concentrated under reduced pressure. The
4
mixture was extracted with diethyl ether (200 mL © 2). The
organic layer was washed with water, and a saturated aqueous
solution of NaCl. The resulting organic layer was dried over
Na SO , filtered, and concentrated under reduced pressure.
2
4
residue was purified by silica gel column chromatography
using hexane­diethyl ether (7:3 v/v) as an eluent, to give
15-methoxy-3,7,11,15-tetramethylhexadeca-2,4,6,8,10-penta-
enenitrile (6, 19.1 g, 96%) as an orange oil. 6: C H NO; MW
2
4
The residue was purified by silica gel column chromatography
using hexane­diethyl ether (7:3 v/v) as an eluent, to give 11-
methoxy-3,7,11-trimethyldodeca-2,4,6-trienenitrile (4, 4.9 g,
2
1
31
1
1
8
2%) as an orange oil. 4: C H NO; MW 247.38; H NMR
313.48; H NMR (400 MHz, CDCl ) as an isomeric mixture
16
25
3
(
400 MHz, CDCl ) as an isomeric mixture (3:2): ¤ 6.88­6.70
(3:2): ¤ 6.93 (dd, J = 14.7 Hz, 11.0 Hz, 1H), 6.62 (dd, J =
3

M. Yamamoto et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
127
1
1
(
4.7 Hz, 11.0 Hz, 1H), 6.23 (d, J = 14.7 Hz, 1H), 6.14 (d, J =
1.0 Hz, 1H), 5.96 (d, J = 11.0 Hz, 1H), 5.17 (s, 1H), 3.17
broad s, 3H), 2.27­2.02 (m, 4H), 2.22 (s, 3H), 2.10 (t, J =
a saturated aqueous solution of NaCl. The organic layer was
dried over Na SO , filtered, and concentrated under reduced
2
4
pressure. The residue was separated by silica gel column
chromatography into three major fractions: Fr 1 as a mixture
6
3
2
8
.8 Hz, 2H), 2.00 (2.0045, 1.992) (two s, 3H), 1.83 (broad s,
¹
1
10b
H), 1.52­1.41 (m, 4H), 1.14 (s, 6H); IR (KRS-5/cm ): 2970,
of stereoisomers of lycopen (9, 0.37 g, 11%) as a red solid
941, 2872, 2826, 2205, 1632, 1601, 1568, 1254, 1207, 962,
60, 791, 731; MS (FAB ): m/z = 314 [M + H] ; HRMS
using hexane­benzene (10:1 v/v) as an eluent, Fr 2 as a
mixture of stereoisomers of 3,4-dihydroanhydrorhodovibrin (1)
(1.44 g, 40%) as a red solid using hexane­benzene (1:1 v/v) as
an eluent, and Fr 3 as a mixture of stereoisomers of 3,4,3¤,4¤-
+
+
+
(FAB ): found m/z 313.2416; calcd for C H NO m/z
21 31
3
13.24056. This material was used for the next step without
1
2
further purification.
5-Methoxy-3,7,11,15-tetramethylhexadeca-2,4,6,8,10-
pentaenal (7). A 130 mL of toluene solution containing 6
18.1 g, 58 mmol) in a 500-mL three-necked flask was cooled to
10 °C in an ice­salt bath. DIBAH (1.0 M in hexane, 65 mL,
5 mmol) was slowly added to the toluene solution through a
tetrahydrospirilloxantine (10, 0.49 g, 13%) as a red solid
using hexane­benzene (1:10 v/v) as an eluent. The target
second fraction (Fr 2) was recrystallized three times from
hexane in a freezer (at ¹20 °C) to give pure all-trans-1 (See
Supporting Information): C H O; MW 568.91; mp 127 °C;
1
(
¹
4
1
60
1
6
H NMR (600 MHz, C6D6): ¤ 6.771 (dd, J = 15.0 Hz, 12.0 Hz,
separatory funnel at a temperature below ¹10 °C. After stirring
for an additional 1 h, methanol (30 mL) was slowly added to
the reaction mixture while maintaining the temperature below
1H), 6.767 (dd, J = 15.0 Hz, 12.0 Hz, 1H), 6.70 (m, 3H), 6.67
(dd, J = 15.0 Hz, 9.5 Hz, 1H), 6.48 (d, J = 15 Hz, 2H), 6.45 (d,
J = 15.1 Hz, 1H), 6.43 (d, J = 15.0 Hz, 1H), 6.36 (d, J =
12.4 Hz, 1H), 6.34 (m, 3H), 6.17 (d, J = 9.5 Hz, 1H), 6.16 (d,
J = 9.5 Hz, 1H), 5.23 (t, J = 6.0 Hz, 1H), 3.04 (s, 3H), 2.19 (q,
J = 6.0 Hz, 2H), 2.16 (t, J = 6.0 Hz, 2H), 2.08 (t, J = 7.6 Hz,
2H), 1.94 (s, 3H), 1.93 (s, 3H), 1.88 (s, 6H), 1.76 (s, 3H), 1.75
(s, 3H), 1.67 (s, 3H), 1.58 (m, 2H), 1.57 (s, 3H), 1.42 (m, 2H),
¹
5 °C. A saturated aqueous NH Cl solution (ca. 100 mL) was
4
added to the reaction mixture, after which the ice­salt bath was
removed and the reaction mixture was stirred for a further
3
0 min at room temperature. The reaction mixture was filtered
through a celite pad, which was washed with diethyl ether.
The combined filtrate was washed with water and a saturated
aqueous NaCl solution. The resulting organic layer was dried
over Na SO , filtered and concentrated under reduced pressure.
1
3
1.06 (s, 6H); C NMR (100 MHz, C D ): ¤ 139.3, 139.0,
6
6
138.0, 138.0, 136.8, 136.8, 136.4, 136.4, 136.1, 136.1, 133.4,
133.4, 133.4, 132.4, 131.6, 130.7, 130.7, 126.8, 126.7, 125.7,
125.7, 125.3, 125.3, 124.4, 74.0, 48.9, 40.9, 40.6, 40.0, 27.1,
25.9, 25.0, 22.4, 17.7, 16.9, 16.8, 12.9, 12.9, 12.8, 12.8; IR
2
4
The residue was purified by silica gel column chromatography
using hexane­diethyl ether (1:1 v/v) as an eluent, to give 15-
methoxy-3,7,11,15-tetramethylhexadeca-2,4,6,8,10-pentaenal
¹
1
(KBr/cm ): 3381, 2972, 2934, 2880, 2831, 2363, 2343, 1719,
+
(
7, 6.9 g, 38%) as an orange oil. 7: C H O ; MW 316.48;
1636, 1456, 1367, 1076, 974, 667; MS (FAB ): m/z = 568
21
32
2
1
+
+
H NMR (600 MHz, CDCl3) as an isomeric mixture (5:3): main
component: ¤ 10.1 (d, J = 8.4 Hz, 1H), 7.12 (dd, J = 15.0 Hz,
1.5 Hz, 1H), 6.64 (dd, J = 15.0 Hz, 11.5 Hz, 1H), 6.38 (d, J =
[M ]; HRMS (FAB ): found m/z 568.4633; calcd for C41H60O
m/z 568.46442.
1
15.1 Hz, 1H), 6.26 (d, J = 15.1 Hz, 1H), 6.22 (d, J = 11.5 Hz,
1H), 5.98 (two d, J = 8.8 Hz, 2H), 3.18 (s, 3H), 2.33 (s, 3H),
2.11 (t-like, J = 7.2 Hz, 2H), 2.03 (s, 3H), 1.84 (s, 3H), 1.48
HH and KO thank Specially Distinguished Research Grant
from Osaka City University for financial support. HH thanks
HFSP for financial support.
(
(
(
m, 2H), 1.45 (m, 2H), 1.15 (s, 6H), minor component: ¤ 10.1
d, J = 8.2 Hz, 1H), 7.28 (dd, J = 15.1 Hz, 11.5 Hz, 1H), 6.78
d, J = 15.1 Hz, 1H), 6.64 (dd, J = 15.1 Hz, 11.5 Hz, 1H),
Supporting Information
Synthetic scheme (Scheme S1) and procedures for com-
1
6
.31 (d, J = 15.1 Hz, 1H), 6.08 (d, J = 11.5 Hz, 1H), 6.06 (d,
pounds 2 and 8 and H NMR spectra and assignments for all-
J = 11.5 Hz, 1H), 5.98 (d, J = 8.2 Hz, 1H), 3.18 (s, 3H), 2.35
trans-1 and the mixture of cis-1 (5-cis-5¤-trans-1/5-trans-5¤-
cis-1 µ 1:1). This material is available free of charge on the
Web at: http://www.csj.jp/journals/bcsj/.
(
s, 3H), 2.13 (t-like, J = 7.2 Hz, 2H), 2.02 (s, 3H), 1.85 (s, 3H),
¹1
1
2
7
.51 (m, 2H), 1.45 (m, 2H), 1.16 (s, 6H); IR (CCl /cm ):
4
959, 2928, 2858, 1717, 1684, 1651, 1458, 1362, 791, 772,
+
+
+
52, 741, 667; MS (FAB ): m/z = 316 [M ]; HRMS (FAB ):
References
found m/z 316.2397; calcd for C H NO m/z 316.2402. This
21
31
1
G. McDermott, S. M. Prince, A. A. Freer, A. M.
material was used for the next step without further purification.
,4-Dihydroanhydrorhodovibrin (1). A 120 mL of THF
Hawthornthwaite-Lawless, M. Z. Papiz, R. J. Cogdell, N. W.
Isaacs, Nature 1995, 374, 517.
3
solution of titanium(III) trichloride (6.70 g, 43 mmol) was
placed in a 500-mL three-necked flask fitted with a separatory
funnel and a reflux condenser. Under argon, lithium aluminum
hydride (0.84 g, 22 mmol) was added to the solution. The mix-
ture was stirred at room temperature for 2 h. An 80 mL of
THF solution containing 7 (6.22 g, 20 mmol) and 8 (5.51 g, 19
mmol) was slowly added to the mixture using a separatory
funnel. The mixture was stirred overnight at room temperature,
cooled to 0 °C in an ice-bath, and treated with a 2 M aqueous
solution of HCl (60 mL). The crude mixture was extracted with
diethyl ether and the organic layer was washed with water and
2
A. W. Roszak, T. D. Howard, J. Southall, A. T. Gardiner,
C. J. Law, N. W. Isaacs, R. J. Cogdell, Science 2003, 302, 1969.
P. Qian, K. Saiki, T. Mizoguchi, K. Hara, T. Sashima, R.
Fujii, Y. Koyama, Photochem. Photobiol. 2001, 74, 444.
a) H. Rademaker, A. J. Hoff, R. Van Grondelle, L. N. M.
3
4
Duysens, Biochim. Biophys. Acta, Bioenerg. 1980, 592, 240.
b) R. J. Cogdell, P. O. Andersson, T. Gillbro, J. Photochem.
Photobiol., B 1992, 15, 105. c) J. Akahane, F. S. Rondonuwu, L.
Fiedor, Y. Watanabe, Y. Koyama, Chem. Phys. Lett. 2004, 393,
184. d) K. Nakagawa, S. Suzuki, R. Fujii, A. T. Gardiner, R. J.
Cogdell, M. Nango, H. Hashimoto, Photosynth. Res. 2008, 95,

1
28
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
3,4-Dihydroanhydrorhodovibrin­LH1 Complex
3
39.
5
Acta 1985, 68, 1519. b) A. Valla, B. Valla, R. Le Guillou, D.
Cartier, L. Dufossé, R. Labia, Helv. Chim. Acta 2007, 90, 512.
c) We thank Dr. Takeshi Yamada (Present address: Kitasato
University, Kitasato Institute for Life Sciences & Graduate School
of Infection Control Sciences) for discussion on the synthesis of
8 using a new phosphorous reagent of Weinreb type amide: T.
Yamada, K. Okada, T. Shinada, Y. Ohfune, H. Hashimoto, Synlett
2012, 23, 2980.
11 a) R. J. Cogdell, A. R. Crofts, Biochim. Biophys. Acta,
Bioenerg. 1978, 502, 409. b) K. Nakagawa, S. Suzuki, R. Fujii,
A. T. Gardiner, R. J. Cogdell, M. Nango, H. Hashimoto,
Photosynth. Res. 2008, 95, 345.
K. Nakagawa, S. Suzuki, R. Fujii, A. T. Gardiner, R. J.
Cogdell, M. Nango, H. Hashimoto, J. Phys. Chem. B 2008, 112,
9
467.
6
M. Komori, R. Ghosh, S. Takaichi, Y. Hu, T. Mizoguchi, Y.
Koyama, M. Kuki, Biochemistry 1998, 37, 8987.
T. Marui, T. Yamada, T. Kobayashi, H. Hashimoto,
Carotenoid Sci. 2001, 4, 76.
a) H. Pauling, D. A. Andrews, N. C. Hindley, Helv. Chim.
Acta 1976, 59, 1233. b) P. Chabardes, Tetrahedron Lett. 1988, 29,
7
8
6
253.
9
A. F. L. Creemers, J. Lugtenburg, J. Am. Chem. Soc. 2002,
1
24, 6324.
a) A. Zumbrunn, P. Uebelhart, C. H. Eugster, Helv. Chim.
12 A. J. Aasen, S. L. Jensen, Acta Chem. Scand. 1967, 21, 371.
1
0