Total syntheses of phycocyanobilin derivatives bearing a modified A-ring toward the structure/function analysis of phytochrome

Article abstract of DOI:10.1246/cl.2000.1398

Phycocyanobilin (PCB) derivatives bearing a modified A-ring, including 3,3′-dihydrogenated PCB derivatives, were efficiently synthesized in free acid forms by applying our original methods for the preparation of A-, A/B-, and C/D-ring components toward th

Full text of DOI:10.1246/cl.2000.1398

1398
Chemistry Letters 2000
Total Syntheses of Phycocyanobilin Derivatives Bearing a Modified A-Ring
toward the Structure/Function Analysis of Phytochrome
Daisuke Sawamoto, Hiroshi Nakamura, Hideki Kinoshita, Shuhei Fujinami, and Katsuhiko Inomata*
Department of Chemical Science, Graduate School of Natural Science and Technology, Kanazawa University,
Kakuma, Kanazawa, Ishikawa 920-1192
(Received September 18, 2000; CL-000864)
Phycocyanobilin (PCB) derivatives bearing a modified A-
ring, including 3,3'-dihydrogenated PCB derivatives, were effi-
ciently synthesized in free acid forms by applying our original
methods for the preparation of A-, A/B-, and C/D-ring compo-
nents toward the structure/function analysis of phytochrome.
Phytochrome, the best-characterized light-sensing photo-
receptor in plants, mediates a variety of light-responsive develop-
mental processes in photoreversible manner. The chromophores
named phytochromobilin (PΦB) and its substitute phyco-
cyanobilin (PCB, 2 in Figure 1), that differs from PΦB only by
substitution of the vinyl group at C-18 with an ethyl group, are
linear tetrapyrrole derivatives and covalently bound to their
apoproteins at A-ring.1
Little is known about the structural requirements of the chro-
mophore for holophytochrome functions, because of the difficulty
of chemical synthesis of the chromophore. Therefore, we have
been studying on the syntheses of phycobilin derivatives,2 and
succeeded for the first time in synthesizing PCB (2)2b,c and PΦB2d
in free acid forms.
On the other hand, A-ring 12f for 3,3'-dihydrogenated PCB
derivative 6 was prepared starting from 2,2-dimethylsuccinic
anhydride (13) as shown in Scheme 2. However, the method was
not applicable to the synthesis of A-ring of PCB analog 7, as it
was difficult to discriminate between ethyl and methyl groups for
monothiocarbonylation with Lawesson’s reagent. Therefore,
A/B-ring component of 7 was constructed in an alternative man-
ner as will be described later.
In this paper, we wish to report the syntheses of PCB deriva-
tives bearing a modified A-ring, such as 2-norPCB (1), 2-
methylPCB (3), 2-homoPCB (4), and 3-homoPCB (5), for the
structure/function analysis (Figure 1). Further, 3,3'-dihydrogen-
ated PCB derivatives (6 and 7) were also synthesized to investi-
gate their non-covalent interaction with apoprotein toward the
development of affinity chromatography to purify apoprotein.
A-rings 12a–f thus prepared were coupled with ylide 17 as
B-ring to afford A/B-rings 18a–f (Scheme 3). It was essential to
add DBU to couple 12f with 17, probably due to the poor nucle-
ophilicity of the thiocarbonyl group of 12f as compared to those
of 12a–e by lack of conjugation with the exo-olefinic bond.
A-rings (12a–e) for PCB and/or its analogs (1–5) were pre-
pared according to our method starting from compound 8,
maleic or citraconic imide (9a,b) as shown in Scheme 1.2b
Copyright © 2000 The Chemical Society of Japan

Chemistry Letters 2000
1399
A/B-rings 18a–f were formylated with methyl orthofor-
mate in TFA accompanying decarboxylation to afford 19a–f,
which were coupled with C/D-ring 21 to give triallyl ester
derivatives 1'–6' of PCB and its analogs (Scheme 4).
Interestingly, two methyl groups at C-2 of 3' were not
identical on 1H NMR spectrum. Thus, the structure of the ana-
log 3" (R6 = Me) was determined by X-ray crystallography as
shown in Figure 2, which revealed that A-ring cannot conjugate
with B~D-rings by steric demand by an ester group at C-5.3
Now, we could have various kinds of PCB derivatives bear-
ing a modified A-ring in free acid forms in hand. Investigations on
the in vitro assembly of these PCB derivatives with recombinant
apophytochrome are in progress.
The present work was financially supported in part by Grant-
in-Aid for Scientific Research from the Ministry of Education,
Science, Sports, and Culture.
Finally, triallyl ester groups of 1'–6’' were deprotected all at
once by morpholine with Pd-catalyst, followed by decarboxyl-
ation at C-5 by treating with TFA to afford free acid forms 1–64
with all-Z, all-syn conformations (confirmed by NOESY). In the
case of 6', decarboxylation proceeded only when heated in TFA.
It looks to be due to the lack of stabilization of the cation at C-4,
produced by initial protonation to C-5, by the neighboring exo-
olefin. Under the same conditions, compound 7' (R5 = CO2Allyl,
R6 = Allyl; prepared in another way) was decomposed. Such situ-
ation also prompted us to develop an alternative method, which
does not require decarboxylation at C-5, for the construction of
A/B-ring of 7.
A precursor of A/B-ring 24 was obtained by the Wittig-type
coupling reaction of 5-tosylpyrrolinone 22 with a formylpyrrole
23 in high yield as previously reported.2a Compound 24 was then
reduced with aluminum amalgam to 25, followed by isomeriza-
tion in TFA accompanying decarboxylation to afford the unstable
A/B-ring 26.
C/D-ring 20 was formylated contrary to the cases of the con-
struction of 1'–6', in which A/B-rings were formylated, and the
resulting C/D-ring 27 was coupled with 26 to afford the tetrapyr-
role 7" in good yield. Allyl ester groups of 7" were deprotected
in a similar manner described for triallyl esters 1'–6' to yield the
racemic 74 as all-Z, all-syn conformations. It was confirmed that
methyl and ethyl groups of the A-ring are in trans configuration
from the coupling constant (5.86 Hz) between methine protons at
C-2 and C-3 of the final product 7.
References and Notes
1
M. Furuya and P.-S. Song, “Assembly and Properties of
Holophytochrome,” in “Photomorphogenesis in Plants,” 2nd ed., ed.
by R. E. Kendrick, G. H. M. Kronenberg, Kluwer Academic
Publishers, Dordrecht, The Netherlands (1994), Chap. 4.3, pp.
105–140; M. Stanek and K. Grubmayr, Chem. Eur J., 4, 1653 and
1660 (1998). See also the references cited therein.
2
a) K. Kohori, M. Hashimoto, H. Kinoshita, and K. Inomata, Bull.
Chem. Soc. Jpn., 67, 3088 (1994). b) T. Kakiuchi, H. Kato, K. P.
Jayasundera, T. Higashi, K. Watabe, D. Sawamoto, H. Kinoshita,
and K. Inomata, Chem. Lett., 1998, 1001. c) K. P. Jayasundera, H.
Kinoshita, and K. Inomata, Chem. Lett., 1998, 1227. d) T. Kakiuchi,
H. Kinoshita, and K. Inomata, Synlett, 1999, 901. e) A. Ohta, D.
Sawamoto, K. P. Jayasundera, H. Kinoshita, and K. Inomata, Chem.
Lett., 2000, 492. See also the references cited therein.
3
4
Crystal data for C40H48N4O8 at rt. Dark violet prism, fw = 712.84,
triclinic, space group P1– with Z = 4, a = 16.595(3), b = 17.552(3), c
= 13.445(7) Å, α = 91.20(2), β = 94.52(3), γ = 86.565(9)°, V =
3896(2) Å3, Dc = 1.215 g cm–3, R = 0.080 and Rw = 0.129 for 7934
data.
Spectral data of the final products 1-7 are given for UV/vis (MeOH)
λmax and HRMS (FAB) (M++1) in the following. 1: 364 (41700),
622 (13700) nm; Found: m/z 573.2708. Calcd for C32H37N4O6:
573.2715. 2: see ref. 2b. 3: 364 (58300), 615 (20600) nm; Found:
m/z 601.3023. Calcd for C34H41N4O6: 601.3028. 4: 364 (53400),
595 (17700) nm; Found: m/z 601.3036. Calcd for C34H41N4O6:
601.3028. 5: 365 (49700), 597 (16100) nm; Found: m/z 601.3018.
Calcd for C34H41N4O6: 601.3028. 6: 345 (30500), 584 (13700) nm;
Found: m/z 603.3166. Calcd for C34H43N4O6: 603.3185. 7: 345
(33000), 580 (14000) nm; Found: m/z 589.3017. Calcd for
C33H41N4O6: 589.3028.