Selective photochemical cleavage of an α-ketoamide in a highly functionalised macrolide ascomycin

Article abstract of DOI:10.1016/j.tetlet.2004.02.014

A novel photochemical amide cleavage reaction of (6S)-methoxyascomycin opening a pathway for the selective cleavage of the pipecolic acid, is described. The scope of this reaction with several analogues carrying suitable protecting groups is examined.

Full text of DOI:10.1016/j.tetlet.2004.02.014

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 2527–2530
Selective photochemical cleavage of an a-ketoamide in a highly
functionalised macrolide ascomycin
*
€
Murty A. R. C. Bulusu, Peter Waldstatten, Thomas Tricotett and Gerhard Schulz
Novartis Institute for BioMedical Research, Brunnerstrasse 59, A-1235 Vienna, Austria
Received 14 November 2003; revised 29 January 2004; accepted 4 February 2004
Abstract—A novel photochemical amide cleavage reaction of (6S)-methoxyascomycin opening a pathway for the selective cleavage
of the pipecolic acid, is described. The scope of this reaction with several analogues carrying suitable protecting groups is examined.
Ó 2004 Elsevier Ltd. All rights reserved.
The immunomodulatory macrolactam ascomycin (1,
Fig. 1) is a fermentation product from Streptomyces
hygroscopicus var. ascomyceticus, originally isolated due
to its antifungal activities.1 Pimecrolimus 2, (the 33-
epichloro derivative of 1, Elidel,â SDZ ASM 981) has
shown high therapeutic efficacy in patients with
inflammatory skin diseases and pimecrolimus cream 1%
is on the market for the topical treatment of atopic
dermatitis.2 Ascomycin is a highly functionalised mole-
cule containing several free hydroxy groups, an aldol
functionality, and a reactive tricarbonyl unit attached to
pipecolic acid. As demonstrated by studies with the
structurally closely related fermentation product FK506
(tacrolimus), the pipecolic acid moiety of the macro-
lactam is involved in the binding to its receptor macro-
philin.3 It would, therefore, be of interest to examine the
impact of modifications in this region of the molecule on
its biological activities. However, because of the lack of
functionalities in this part of the molecule, selective
modifications, in particular on the amino acid unit,
require complex synthetic strategies.4 We have earlier
described the phototransformation of 1 into (6S)-
methoxy-9-hydroxy ascomycin 3 (MeOH, k P 366 nm,
0–5 °C, 8 h, 75%) and its subsequent selective Cu(II)-
catalysed oxidation leading to (6S)-methoxyascomycin
4.5a;b Using these reactions we were able to synthesise
several analogues of 1 with modified amino acid units.5c
Recently we have found that photolysis of 4 featuring
C(6)–OMe under the same reaction conditions follows a
different pathway leading to selective cleavage of the
amide bond. This reaction opens a potential strategy for
the selective removal of the pipecolic acid and replace-
ment by other amino acids.
1
33
R
HO
O
O
O
O
21
O
O
OH
1
O
OH
6
N
N
O
2
O
3
R
O
The phototransformation of 1 into 3 involves abstrac-
tion of the equatorial C(6)–H by the excited C(9)@O
leading to a diradical, and its reorganisation to the
zwitterion Z1 (Fig. 2), which is trapped by attack of
MeOH on C(6).5;6 (6S)-Methoxyascomycin 4 still pos-
sesses an hydrogen atom on C(6) in the equatorial
position, and can, in principle, undergo a similar pho-
toreaction, which would initially lead to the intermedi-
ate Z4. Attack of MeOH on C(6) of Z4, as observed
with Z1, appeared to be unfavourable now because of
the resulting steric congestion due to the additional
OMe group. We were therefore interested to know the
fate of this intermediate, and to this end we looked at
the photochemistry of 4.
O
R
9
O
OH
O
OH
O
O
O
O
O
1: Ascomycin: R1 = (R)-OH
2: Pimecrolimus: R1 = (S)-Cl;
3: R2 = H, R3 = OH
4: (R2, R3) = O
Figure 1.
Keywords: Phototransformation; Photochemical amide cleavage; Zwit-
terion.
* Corresponding author. Tel.: +43-01-866-59-535; fax: +43-01-866-59-
785; e-mail: murty.bulusu@pharma.novartis.com
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.02.014

2528
M. A. R. C. Bulusu et al. / Tetrahedron Letters 45 (2004) 2527–2530
3.81 (s, 3H, CO2Me), 3.65 (s, 3H, MeO–C(6)), and the
MeOH
R
13C NMR (125 MHz, CDCl3) signals at d 174.2
(C(8)@O), 172.2 (C(1)@O), 164.7 (C(6)@N), 73.3 (C(9)),
52.7 (MeO–C(8)), 52.4 (MeO–C(6)). The NMR spectra
of the isomeric mixture 5 showed similar sets of signals
for all the isomeric components, thus indicating the
same basic structure resulting through amide cleavage
for all the isomers. The formation of 5 could be
explained on the basis of the postulated intermediate Z4
(Fig. 2). It appears that attack of MeOH on C(6), or the
intramolecular nucleophilic attack by C(9)–O on C(6),
are indeed, not the favoured pathways for Z4. Instead, it
undergoes attack by MeOH on the electrophilic amide
carbonyl, leading to the amide cleavage product 5.
Alternatively, the zwitterion Z4 could undergo cleavage
leading to a hydroxy ketene, which is then trapped by
MeOH.6
O
O
O
O
O
R
hν
N+
MeOH
O
6
N
H
O
O
9
Z1: R = H
Z4: R = OMe
1: R = H
4: R = OMe
Figure 2. Partial structures of 1 and 4 and the intermediates Z1 and Z4
in the photolysis.
Irradiation of 4 in MeOH, as described above, for 4.5 h,
followed by separation on preparative HPLC, afforded
unreacted 4 (7%) and the amide cleavage product 5 as a
mixture of C(9)- and C(10)-isomers in 58% yield
(Scheme 1).7 Through repeated HPLC a pure sample of
5 (27%) consisting of one single isomer could be
The phototransformation of 4 into 5 features the
cleavage of the strongest (amide) bond, with retention of
the tricarbonyl backbone, in the unprotected multi-
obtained. Characteristic of
(500 MHz, CDCl3) signals at d 4.20 (s, 1H, H–C(9)),
5
are the 1H NMR
33
RO
O
RO
O
O
O
21
O
O
O
1
hν
OR
OR
6
6
1
O
O
N
N
8
9
O
MeOH
0-5 oC
O
OH
OH
O
MeO2C
O
9
OH
O
14
O
O
O
O
4 (R = H)
6 (R = t-BDMS)
5 (R = H: 58%)
7 (R = t-BDMS: 33%)
1) AllOCOCl, DMAP,
CH3CN: 83%; 2) Pd(Ph3P)4,
THF, 50 ˚C: 89%
RO
O
RO
O
RO
O
O
O
O
O
O
O
O
O
OR
hν
OR
OR
O
O
N
N
N
O
+
O
H
MeOH
0-5 oC
O
O
O
OH
OH
MeO2C
MeO2C
O
O
O
O
O
O
O
TMSCl, NaI
CH3CN, RT
O
O
O
O
8
9 (16%)
10 (58%, two steps)
Dess Martin
CH2Cl2, RT
3 h: 44%
Dess Martin
CH2Cl2, RT
3 h: 60%
RO
O
RO
RO
O
O
O
O
O
2 eq LiOH
O
O
O
O
O
O
OR
OR
OR
THF/H2O: 3/1
RT, 4 h: 35%
O
N
H
N
N
O
O
H
O
O
O
MeO2C
HO2C
MeO2C
O
O
O
O
O
O
O
13
O
O
O
O
O
12
For 4-5: R = H; for 6-13: R = t-BDMS
Scheme 1. Photochemical amide cleavage reactions of ascomycin derivatives and their further transformations.
11

M. A. R. C. Bulusu et al. / Tetrahedron Letters 45 (2004) 2527–2530
2529
functional molecule ascomycin under simple reaction
conditions and hence is very attractive. To date there are
no methods available for selective cleavage of the amide
bond in ascomycin even after introducing suitable pro-
tecting groups, especially because of the sensitive tri-
carbonyl moiety. Previously modifications in the amino
acid region have involved cleavage of the C(1)–C(9) unit
followed by reconstruction of the tricarbonyl and other
units.4;8 With a view to exploring the scope and potential
of this photo amide cleavage reaction for developing a
shorter semi-synthetic approach, we looked at the pho-
toreaction of the protected ascomycins 6 and 8. Com-
pound 6 was prepared from 4 through silylation (4,
3 equiv t-BDMSCl, 3.5 equiv imidazole, DMF, 75 °C,
70%). The allyl protected derivative 8 was prepared
from 6 through allyloxycarbonylation of C(10)–OH (6,
4.4 equiv allyloxycarbonyl chloride, 10 equiv DMAP,
CH3CN, rt, 4 h, 83%), followed by its decarboxylation9
(0.02 equiv Pd(Ph3P)4, THF, 50 °C, 5 min, 89%) in a
good overall yield.
It should be noted here that all the derivatives studied so
far, except 8, bear a free OH on C(10) of ascomycin.5 In
solution these hemiketal structures, featuring a dicar-
bonyl chromophore, exist in equilibrium with the corres-
ponding tricarbonyl structures. Hence, one cannot be
sure which chromophore is actually involved in the
photoreaction. The successful photoreaction of 8, fea-
turing a locked dicarbonyl chromophore, clearly dem-
onstrates that these reactions can, indeed, also proceed
efficiently through the dicarbonyl chromophore even
at the longer wavelengths (>366 nm) employed in the
reaction, which are necessary for avoiding photoreac-
tion at other sites.5 This is a crucial finding as this allows
for protection of the sensitive tricarbonyl moiety in the
form of the dicarbonyl chromophore with retention
of the photoreactivity.
The a-hydroxyesters 9 and 10 could be independently
ꢀ
oxidised (Dess Martin, pyridine, 4 A MS, CH2Cl2, rt,
3 h) leading to the ketoesters 12 (44%) and 11 (60%),
respectively. Saponification (THF/H2O; 3:1, 2 equiv of
LiOH, rt, 4 h) of the ester 12 afforded the required acid
13 (35%) with concomitant hydrolysis of the imine-ether
occurring either during the reaction or during acidic
workup.
Irradiation of 6 and 8 in MeOH, as described before,
resulted in total consumption of the starting materials
giving the amide cleavage products 7 and 9, respectively,
as mixtures of isomers. Single isomers of 7 and 9 could
be isolated in 33% and 16% yields, respectively, after
chromatography on SiO2 pretreated with 5% NaHCO3
and which were susequently characterised. The 1H
NMR spectrum of crude 9 showed the presence of
varying amounts of its hydrolysis product 10. In fact,
the crude photolysate mixture could be transformed into
10 (58%) by treatment10 with TMSCl/NaI/CH3CN at rt
for 28 h, and purification by chromatography, thus
giving an experimental protocol leading to the isolation
of stable single products in good yields.
The above results have established the viability of this
approach for a semi-synthetic strategy leading to ana-
logues of 1 in which pipecolic acid is replaced by other
amino acids. However, attempts to cleave the secondary
ester in 13 resulted in formation of several products
arising through ß-elimination of the 24-silyloxy moiety
and subsequent reactions. To circumvent this problem
from the beginning, we studied the photochemistry of
the suitably protected derivative 148 (Scheme 2), which
Si
O
Si
O
O
O
O
O
Si
O
Si
O
Si
Si
hν
O
O
O
O
O
O
N
N
O
MeOH, 8 h
0-5 oC
69%
O
OH
O
O
OH
O
OH
O
O
O
O
O
14
15
Cu(OAc)2, cat pyridine
CH2Cl2, RT, 2 d: 92%
Si
O
Si
O
O
O
O
O
Si
O
Si
Si
Si
O
hν
O
O
O
O
O
O
O
N
N
O
MeOH, 6 h
0-5 oC
44%
MeO2C
OH
OH
O
O
O
OH
O
O
O
O
O
17
16
Scheme 2. Photoreactions of protected ascomycin derivatives.

2530
M. A. R. C. Bulusu et al. / Tetrahedron Letters 45 (2004) 2527–2530
Med. Surg. 2001, 20, 233–241; (d) Hebert, A.; Warken,
K. A.; Cherill, R. Semin. Cutaneous Med. Surg. 2001, 20,
260–267; (e) Wellington, K.; Jarvis, B. Drugs 2002, 62,
was prepared from 1 through monosilylation of 33-OH,
EvansÕ reduction of C(22)@O (1.1 equiv Me4Nþ
(OAc)3HBꢀ, CH3CN/AcOH (100:35), )5 to 3 °C, 12 h,
70%), followed by silylation (3.0 equiv (CliPr2Si)2O,
3.0 equiv imidazole, DMF, rt, 3 d, 70%).
€
817–840; (f) Grassberger, M.; Meingassner, J.; Stutz, A.;
Rappersberger, K.; Wolff, K. Ascomycins. In Psoriasis;
Roenigk, H. H., Maibach, H. I., Eds.; 3rd ed.; Marcel
Dekker: New York, Basel, Hong Kong, 1998; pp 769–779,
Chapter 66; (g) Rappersberger, K.; Komar, M.; Ebelin,
M. E.; Scott, G.; Burtin, P.; Greig, G.; Kehren, J.;
Chibout, S. D.; Holter, W.; Richter, L.; Oberbauer, R.;
€
Cordier, A.; Stutz, A.; Wolff, K. J. Invest. Dermatol. 2002,
119, 876–888; (h) Kapp, A.; Papp, K.; Bingham, A.;
€
Folster-Holst, R.; Ortonne, J. P.; Potter, P.; Gulliver, W.;
Paul, C.; Molloy, S.; Barbier, N.; Hultsch, T.; de Prost, Y.
Irradiation of 14 in MeOH for 8 h, followed by chro-
matography, afforded 15 (69%), which after oxidation
with Cu(II) acetate gave the 6-methoxy derivative 16
(92%), which is ready for the photochemical amide
cleavage. Indeed, irradiation of 16 in MeOH for 6 h
followed by chromatography led to the amide cleavage
product 17 (44%). This establishes a crucial reaction in
the context of the semi-synthetic strategy.
J. Allergy Clin. Immunol. 2002, 110, 277–284; (i) Meurer,
€
€
M.; Folster-Holst, R.; Wozel, G.; Weidinger, G.; Junger,
M.; Brautigam, M. Dermatology 2002, 205, 271–277; (j)
€
In summary, we have shown that (6S)-methoxyasco-
mycin and several of its protected derivatives undergo a
novel amide cleavage reaction upon irradiation in useful
yields. The selective cleavage of an amide bond in an
unprotected multifunctional molecule such as ascomycin
is a remarkable finding. The applicability of this reaction
for developing a semi-synthetic strategy leading to
analogues of 1 modified in the amino acid region will
be further explored.
Wahn, U.; Bos, J. D.; Goodfield, M.; Caputo, R.; Papp,
K.; Manjra, A.; Dobozy, A.; Paul, C.; Molloy, S.;
Graeber, M.; Cherill, R.; DeProst, Y. Pediatrics 2002,
110, e2; (k) Ho, V.; Gupta, A.; Kaufmann, R.; Todd, G.;
€
Vanachlocha, F.; Takaoka, R.; Folster-Holst, R.; Potter,
P.; Marshall, K.; Thurston, M.; Bush, C.; Cherill, R.
J. Pediatr. 2003, 142, 155–162.
3. VanDuyne, G. D.; Standaert, R. F.; Karplus, P. A.;
Schreiber, S. L.; Clardy, J. Science 1991, 252, 839–842.
4. (a) Askin, D.; Reamer, R. A.; Joe, D.; Volante, R. P.;
Shinkai, I. J. Org. Chem. 1990, 55, 5448–5450; (b) Askin,
D.; Joe, D.; Reamer, R. A.; Volante, R. P.; Shinkai, I.
J. Org. Chem. 1990, 55, 5451–5454.
5. (a) Bulusu, A. R. C. M.; Haidl, E. H.; Schulz, G.;
Waldstaetten, P.; Grassberger, M. Indian J. Chem. 1999,
38B, 1159–1164; (b) Bulusu, A. EP 0626385A; Sandoz
Acknowledgements
T.T. thanks the Leonardo da Vinci Program for a fel-
lowship.
€
Ltd, 1994 (c) Bulusu, A. R. C. M.; Waldstatten, P.; Schulz,
G.; Grassberger, M. Tetrahedron Lett. 2004, 45, 2523.
6. (a) Chesta, C. A.; Whitten, D. G. J. Am. Chem. Soc. 1992,
114, 2188–2197; (b) Aoyama, H.; Hasegawa, T.; Watabe,
M.; Shiraishi, H.; Omote, Y. J. Org. Chem. 1978, 43, 419–
422; (c) Aoyama, H.; Sakamoto, M.; Kuwabara, K.;
Yoshida, K.; Omote, Y. J. Am. Chem. Soc. 1983, 105,
1958–1964; (d) Aoyama, H.; Hasegawa, T.; Omote, Y.
J. Am. Chem. Soc. 1979, 101, 5343–5347.
7. All compounds were fully characterised using 1H NMR
(250, 400 or 500 MHz), 13C NMR (100 or 125 MHz) and
mass spectral data and elemental analysis. The 1H and
13C-signal assignments for representative compounds were
made through C–H correlation spectra and 1H-decoupling
experiments.
References and notes
1. (a) Arai, T.; Koyoma, Y.; Suenaga, T.; Honda, H.
J. Antibiot. Ser. A 1962, 15, 231–232; (b) Hatanaka, H.;
Iwami, M.; Kino, T.; Goto, T.; Okuhara, M. J. Antibiot.
1988, 41, 1586–1591; (c) Hatanaka, H.; Kino, T.; Miyata,
S.; Inamura, N.; Kuroda, A.; Goto, T.; Tanaka, H.;
Okuhara, M. J. Antibiot. 1988, 41, 1592–1601; (d) Mor-
isaki, M.; Arai, T. J. Antibiot. 1992, 45, 126–128.
2. (a) Meingassner, J.; Grassberger, M.; Fahrngruber, H.;
€
Moore, H. D.; Schuurman, H.; Stutz, A. Br. J. Dermatol.
1997, 137, 568–576; (b) Grassberger, M.; Baumruker, T.;
Enz, A.; Hiestand, P.; Hultsch, T.; Kalthoff, F.; Schuler,
W.; Schulz, M.; Werner, F.-J.; Winiski, A.; Wolff, B.;
8. Horvath, A. Ph.D. Thesis, University of Vienna, 1993.
9. Guibe, F.; Saint MÕLeux, Y. Tetrahedron Lett. 1981, 22,
3591–3594.
€
Zenke, G. Br. J. Dermatol. 1999, 141, 264–273; (c) Stutz,
A.; Grassberger, M.; Meingassner, J. G. Semin. Cutaneous
10. Olah, G. A.; Narang, S. C.; Gupta, B. G. B.; Malhotra, R.
J. Org. Chem. 1979, 44, 1247–1251.