Synthesis of the Bycroft-Gowland structure of micrococcin P1

Article abstract of DOI:10.1021/ol991115e

(formula presented) We describe the chemical synthesis of the accepted structure of micrococcin P1, a member of the thiostrepton group of antibiotics, and we show that this architecture does not correspond to that of the natural product. Methods developed during the present study should greatly facilitate ongoing efforts centering on the determination of the actual structure of microccin P1, in addition to being applicable to the synthesis of more complex thiostrepton congeners.

Full text of DOI:10.1021/ol991115e

ORGANIC
LETTERS
1
999
Vol. 1, No. 11
843-1846
Synthesis of the Bycroft−Gowland
Structure of Micrococcin P1
1
Marco A. Ciufolini* and Yong-Chun Shen
Department of Chemistry, Rice UniVersity, 6100 Main Street,
Houston, Texas 77005-1892, and UniVersit e´ Claude Bernard Lyon 1 &
Ecole Sup e´ rieure de Chimie, Physique, Electronique de Lyon, 43,
Bd. du 11 NoVembre 1918, 69622 Villeurbanne Cedex, France
ciufi@cpe.fr
Received October 4, 1999
ABSTRACT
We describe the chemical synthesis of the accepted structure of micrococcin P1, a member of the thiostrepton group of antibiotics, and we
show that this architecture does not correspond to that of the natural product. Methods developed during the present study should greatly
facilitate ongoing efforts centering on the determination of the actual structure of microccin P1, in addition to being applicable to the synthesis
of more complex thiostrepton congeners.
1
Micrococcin P is a cytotoxic agent that expresses its activity
The structure of micrococcins P was originally assigned
by Walker, Lukacs, and collaborators as 1 (Figure 1).6
However, this architecture proved to be untenable and it was
2
by strongly inhibiting protein synthesis. This substance is
one of the structurally simpler members of the thiostrepton
group of antibiotics, and it is obtained as a mixture of two
3
7
revised to 2 by Bycroft and Gowland on the basis of
components termed micrococcins P1 (MP1) and P2 (MP2).
These differ in that a D-(R)-(-)-1-amino-2-propanol (D-
isoalaninol)4 segment present in in the side chain of MP1
extensive degradation studies. Bycroft and Gowland were
unable to assign the stereochemistry of the threonine-derived
thiazole bound to the pyridine. Nonetheless, other researchers
have often represented that molecular subunit as having an
L-threonine-like stereochemistry, as apparent, e.g., from
recent papers describing synthetic studies on micrococcins.8
Tacit acceptance of this assignment appears to have ensued.
,5
(
(
major component) becomes an aminoacetone unit in MP2
minor).
(
1) Isolation from Bacillus pumilus: Fuller, A. T. Nature 1955, 175,
7
22. Micrococcin P is very similar, possibly identical, to an antibiotic
isolated from a Micrococcus species and named “micrococcin” by: Su, T.
L. Brit. J. Exp. Path. 1948, 29, 473. No structural work on the Su
micrococcin appears to have been described in the literature.
(6) (a) Walker, J.; Olesker, A.; Valente, L.; Rabanal, R.; Lukacs, G. J.
Chem. Soc., Chem. Commun. 1977, 706. (b) Reference 6a, p 708, top of
the page.
(7) Bycroft, B. W.; Gowland, M. S. J. Chem. Soc., Chem. Commun. 1978,
256.
(8) (a) Nakamura, Y.; Shin, C.-G.; Umemura, K.; Yoshimura, J. Chem.
Lett. 1992, 1005. (b) Okumura, K.; Shigekuni, M.; Nakamura, Y.; Shin,
C.-G. Chem. Lett. 1996, 1025.
(
2) (a) Rosendahl, G.; Douthwaite, S. Nucl. Acids Res. 1994, 22, 357.
b) Cundliffe, E.; Thompson, C. J. Eur. J. Biochem. 1981, 118, 47.
3) Review: Pestka, S. In Antibiotics; Corcoran, J. W.; Hahn, F. E.,
Eds.: Springer-Verlag: New York, 1975; Vol. 3, p 480 ff.
(
(
(
4) Mijovic, M. P. V.; Walker, J. J. Chem. Soc. 1960, 909.
(5) The side chain amino alcohol was originally assigned as alaninol
(
ref 4) and later ascertained to be isoalaninol (ref 6). In either case, the
(9) (a) Murakami, T.; Holt, T. G.; Thompson, C. T. J. Bacteriol. 1989,
171, 1459. (b) Holmes, D. G.; Caso, J. L.; Thompson, C. T. EMBO J. 1993,
8, 3183. (c) Yun, B.-S.; Hidaka, T.; Furihata, K.; Seto, H. J. Antibiot. 1994,
47, 510.
leVo isomer is the one of D-(R)-configuration: Klyne, W. Buckingham, J.
Atlas of Stereochemistry: Absolute Configurations of Organic Molecules,
2
nd ed.; Oxford University Press: New York, 1978.
1
0.1021/ol991115e CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/21/1999

Figure 1. Proposed structures of micrococcins P1 and P2 and retrosynthetic dissection thereof.
Several members of the thiostrepton family induce expres-
sion of genes of unknown function in Streptomyces species,
an observation that has stimulated much interest in the
molecular biology arena.10 This behavior has not yet been
reported for 2, but because all thiostrepton substances display
9
similar biological properties, it seems likely that MP1/MP2
may also be gene inducers. Despite these exciting observa-
tions, little synthetic work has been recorded in the thio-
1
1,12
strepton area.
Recently, Shin has announced the synthesis
1
3
of two substances described as micrococcin P and micro-
1
4
Scheme 1a
coccin P1. However, these synthetic compounds are
epimers of the Walker-Lukacs structure 1a and of the
Bycroft-Gowland structure 2a, respectively, in that L-(S)-
Scheme 2a
a
(
a) (R)-isomer, DCC, CH
Cl , then aqueous HCl, 93%; (d) catalytic DMAP, Et
Cl , 99%; (e) aqueous THF, 99%; (f) EtOOCCl, Et N, CH Cl
, 94%; (g) PhH, reflux; (h) KHCO , DME, then
TFAA, pyridine, 0 °C, 72% f-h; (i) CH Cl , 99%.
2
Cl
2
; (b) pyridine, 96% a-b; (c) TFA,
CH
CH
2
2
3
N,
a
2
2
3
2
2
,
(a) (a) CH
N, CH
aqueous HCl, THF; (f) CHCl
2
Cl
2
, 91%; (b) CH
, (e) first TFA, CH
, 88% d-f.
2
Cl
2
, 99%; (c) DCC, CH
2 2
Cl , 97%;
then aqueous NH
3
3
(d) Et
3
2
Cl
2
2
Cl (-BOC), then 0.2 N
2
2
2
3
1844
Org. Lett., Vol. 1, No. 11, 1999

Scheme 3a
a
Reagents and conditions: (a) aqueous THF; (b) DMAP, Et
Cl , Et N; (g) CHCl , 68% d-g; (h) 2-Me-2-butene, NaH PO
N, 0.02 M in DMF, 0 °C, 36 h, 65% j-l.
3
N, CH
2
Cl
2
; (c) BOP-Cl, Et
3
N, CH
3
CN, 84% a-c; (d) THF; (e) EtOAc; (f)
CN, 76% h-i; (j) aqueous THF; (k)
CH
2
2
3
3
2
4
, THF, H
2
O; (i) BOP-Cl, Et
3
N, CH
3
gas, dioxane; (l) Et
3
(
+)-isoalaninol, and not the D-(R)-(-)-isomer, is present as
Herein, we report the total synthesis of the Bycroft-
Gowland structure of micrococcin P1, 2a. The target
molecule may be considered as the result of fusion of a
tripeptide-like sector, 3, with subunit 4. A number of key
a subunit of the side chains. In addition, the work appears
to rest on the assumption that micrococcin P is 1a and
micrococcin P1 is 2a. This would be inaccurate in light of
refs 1-7. Until this confusion is resolved, and until synthetic
materials are compared with natural 2a, Shin’s important
work probably should not be regarded as embodying the total
synthesis of any known natural products.
15
observations directed us to manufacture segments 6, 8, and
10 as shown in Scheme 1. Precursor 6 to the short side chain
1
6
was made from the known threonine derivative 5 and
commercial (R)-(-)-isoalaninol. The previously described
1
2
7
, a building block for 14, was subjected to simultaneous
1
7
(10) E.g.: (a) McConkey, G. A.; Gogers, M. J.; McCutchan, T. F. J.
hydrolysis of oxazolone and ester units, leading to acid 8.
Amine 10 was obtained from L-valine by thiazole formation
Biol. Chem. 1997, 272, 2046. (b) Clough, B.; Strath, M.; Preiser, P.; Denny,
P.; Wilson, I. FEBS Lett. 1997, 406, 123. (c) Chiu, M. L.; Folcher, M.;
Griffin, P.; Holt, T.; Klatt, T.; Thompson, C. J. Biochemistry 1996, 35,
1
8
under Meyers conditions. The merger of 8 and 10 was
effected with DCC, and the resulting 11 was N-deblocked
and condensed with 5. Mesylation of the emerging 13,
2
332. (d) Lu, M.; Draper, D. E. Nucleic Acids Res. 1995, 23, 3426. (e)
Rosendahl, G.; Douthwaite, S. J. Mol. Biol. 1993, 234, 1013. (f) Egebjerg,
J.; Douthwaite, S. R.; Liljas, A.; Garrett, R. A. J. Mol. Biol. 1990, 213,
2
75 and references therein.
11) For pioneering work in this area, see: Kelly, T. R.; Jagoe, C. T.;
Gu, Z. Tetrahedron Lett. 1991, 32, 4263 as well as refs 8 and 13.
(
(16) E.g.: (a) Favara, D.; Omodei-Sale, A.; Consonni, P.; Depaoli, A.
Tetrahedron Lett. 1982, 23, 3105. (b) Garner, P. Tetrahedron Lett. 1984,
25, 5855.
(
(
12) Ciufolini, M. A.; Shen, Y.-C. J. Org. Chem. 1997, 62, 3804.
13) Shin, C.-G.; Okamura, K.; Shigekuni, M.; Nakamura, Y. Chem. Lett.
(17) (a) Flynn, D. L.; Zelle, R. E.; Grieco, P. A. J. Org. Chem. 1983,
48, 2424. (b) Ishizuka, T.; Kunieda, T. Tetrahedron Lett. 1987, 28, 4185.
(18) Aguilar, E.; Meyers, A. I. Tetrahedron Lett. 1994, 35, 2473.
Application of this protocol was essential to suppress racemization of 10.
(19) Attenburrow, J.; Cameron, A. F. B.; Chapman, J. H.; Evans, R.
M.; Hems, B. A.; Jansen, A. B. A.; Walker, T. J. Chem. Soc. 1952, 1094.
(20) Cf. Kraus, G. A.; Taschner, M. J. J. Org. Chem. 1980, 45, 1175.
(21) It is noteworthy that the elimination succeeded only when the
terminal threonine was fully deblocked (oxazoline formation occurred with
the blocked compound) and that the mesylate intermediate showed little
propensity to undergo intramolecular SN2 reaction with the free amino group.
Presumably, the rigidly defined s-trans conformation of the threonine amide
greatly disfavored approach of the NH2 group to the mesylate backside.
1
998, 139.
(14) Okamura, K.; Ito, A.; Yoshioka, D.; Shin, C.-G. Heterocycles 1998,
4
8, 1319.
(15) A route to an intermediate differing from 3 only at the level of
protecting groups has been described (ref 8a), but the present objectives
proved to be better served by an alternative avenue to that fragment. The
merger of 3 and 4 was best conducted by contributing the pyridine-thiazole
cluster as compound 14 (ref 13), by having the short side chain already
present on the pyridine-thiazole array during macrocyclization; by forming
the bonds between 3 and 4 in the order a first and then b, by introducing
dehydroamino acid units as threonines to be dehydrated at a well-defined
stage; and by performing the macrocyclization via an acyl azide.
Org. Lett., Vol. 1, No. 11, 1999
1845

deprotection of the threonine subunit, and DBU treatment
gave 3 (Scheme 2).21
Indeed, Walker and Lukacs proposed the arrangement of 8,
10, and L-threonine as depicted in 1 on the basis of a
presumed structural similarity between micrococcin P, thio-
strepton, and nosiheptide, but alternative permutations of the
three subunits could not be ruled out. The Bycroft correc-
tion focused primarily on the structure of the side chain, but
it assumed that the topography of the macrocycle was correct.
Any error in the Walker structure would thus have been
propagated to the Bycroft-Gowland one.
Work directed toward the resolution of structural issues
is currently underway in our laboratory. These efforts are
hampered by the extreme rarity of micrococcin P1, but
furtunately, the chemical methodology developed in the
course of the present study should greatly assist us in these
endeavors. It also seems plausible that principles and
techniques demonstrated herein should be applicable to the
synthesis of more complex thiostrepton substances and of
chemically modified variants thereof. Hopefully, this will
stimulate additional interest in these remarkable natural
products.
Base hydrolysis of 14, introduction of an N-BOC group
on the oxazolone, and coupling of free acid 16 with 6 resulted
in installation of the short side chain. The diol obtained by
O-desilylation of 17 underwent chemoselective oxidation of
the benzylic-type carbinol to an aldehyde with Attenburrow
6
b
1
9
MnO
2
. Mesylation of the sole expressed alcohol gave 18,
DBU elimination of which provided 19. Oxidation of the
2
0
aldehyde to the free acid was effected with NaClO
the complete segment 4 thus obtained was condensed with
under the influence of BOP-Cl. The emerging 20 was
2
,
and
3
subjected to deblocking and macrocyclization with DPPA
to afford the target structure in 65% chromatographed yield
(Scheme 3). The longest linear sequence in our synthesis of
2a thus requires 23 steps from glycolonitrile, and the overall
yield along this pathway is 5.7%.
Spectra of our synthetic material were very similar, but
not identical, to those obtained from natural micrococcin P.
The accepted structure of the natural product must thus be
incorrect. The location of individual heterocycles in the
Acknowledgment. We thank the NIH (CA-55268), the
NSF (CHE 95-26183), the R. A. Welch Foundation (C-1007),
the A. P. Sloan Foundation, the MENRT, the CNRS, and
the R e´ gion-Rh oˆ ne-Alpes for support of our research program.
We also thank Professor Eric Cundliffe for a sample of
natural micrococcin P.
22
pyridine-thiazole cluster and the absolute configuration
2
3
4-6
of fragment 10, of the isoalaninol,
and of the L-
24
threonine components of micrococcin are secure. However,
attentive scrutiny of the micrococcin literature unravels
uncertainties regarding not only the stereochemistry of the
threonine-derived thiazole in 14/4 but also the order in which
8
, 10, and L-threonine are arranged within the macrocycle.
Supporting Information Available: Experimental pro-
cedures and spectral data for selected compounds. This
information is available free of charge via the Internet at
http://pubs.acs.org.
(
22) Ascertained by X-ray crystallography: James, M. N. G.; Watson,
K. J. J. Chem. Soc. C 1966, 1361.
23) Dean, B. M.; Mijovic, M. P. V.; Walker, J. J. Chem. Soc. 1961,
394.
24) Brookes, P.; Fuller, A. T.; Walker, J. J. Chem. Soc. 1957, 689.
(
3
(
OL991115E
1846
Org. Lett., Vol. 1, No. 11, 1999