Total synthesis of enantiopure β-D-mannosyl phosphomycoketides from Mycobacterium tuberculosis

Article abstract of DOI:10.1021/ja060499i

The first stereoselective total synthesis of a β-d-mannosyl phosphomycoketide is reported. To introduce the stereogenic centers in the chain, three linear chiral building blocks were prepared using two different asymmetric catalytic conjugate addition protocols. Coupling of the various linear fragments was affected using a Julia-Kocienski sequence. This approach constitutes a general and convergent method for the construction of saturated oligoisoprenoid chains of any length and stereochemistry. In addition, an alternative approach for the formation of the difficult β-mannosyl phosphate linkage was shown to be successful. Biological evalutation of the all-S compound revealed that its antigenic potency for T cells is identical to that of the natural product. This result implies that the fine structure of the lipid part has a strong influence on biological activity and that the T cell response is discriminating between different stereoisomers. Copyright

Full text of DOI:10.1021/ja060499i

Published on Web 03/21/2006
Total Synthesis of Enantiopure â-D-Mannosyl Phosphomycoketides from
Mycobacterium tuberculosis
Ruben P. van Summeren,† D. Branch Moody,‡ Ben L. Feringa,*,† and Adriaan J. Minnaard*,†
Stratingh Institute, UniVersity of Groningen, Nijenborgh 4, 9747 AG, Groningen, The Netherlands, and
Brigham and Women’s Hospital, HarVard Medical School, Boston, Massachusetts 02115
Received January 22, 2006; E-mail: B.L.Feringa@rug.nl; A.J.Minnaard@rug.nl
â-Mannosyl phosphomycoketides (MPMs) 1 and 2 are potent
mycobacterial antigens for T cells that were isolated in 2000 from
Mycobacterium tuberculosis and Mycobacterium aVium, respec-
tively (Figure 1).1 Initial structural characterization did not include
assignment of the stereochemistry of the highly unusual saturated
oligoisoprenoid chain. However, the recent elucidation of the
biosynthetic pathway of the side chain shows that all methyl-
substituted stereogenic centers are introduced via the same mech-
anism through the iterative action of a polyketide synthase.2
In 2002, a synthesis of 2 was reported, confirming the overall
molecular structure.3 However, as a stereorandom alkyl chain was
used, the stereochemistry of the lipid part remained unsolved. Since
the side chain is the only structural feature that distinguishes foreign
antigens 1 and 2 from endogenous â-mannosyl phosphodolichols,
its fine structure must be the basis on which the immune system
discriminates. Further insight into the relationship between structure
and activity requires a general synthetic protocol to synthesize
MPMs with lipid chains of any desired length and stereochemistry.
Our efforts toward a stereoselective total synthesis of 1 were
inspired by the notion that it would be an ideal target to test our
recently reported conjugate addition (CA) protocols. As outlined
Figure 1. Structures of 1 and 2 and retrosynthetic analysis.
The construction of building block 3 started with a Wittig reaction
of the readily accessible aldehyde 9 with PPh3CHCOSEt, to afford
R,â-unsaturated thioester 10 (Scheme 1).7 CA of MeMgBr to trans-
10 in the presence of CuBr‚SMe2 and chiral ligand L* resulted in
the formation of (3R)-7 with excellent selectivity (93% ee).5 LiAlH4
reduction of the thioester proceeded readily to provide the corre-
sponding primary alcohol 11. Finally, 11 was converted into sulfone
(3R)-3 via Mitsunobu reaction with 1-tert-butyltetrazole-5-thiol,
followed by oxidation with mCPBA.8
in the retrosynthetic scheme (Figure 1), we envisaged that 1 could
be assembled by connection of four chiral building blocks (3-6).
Fragments 4 and 5 can both be constructed starting from the same
versatile saturated isoprenoid building block 8. Recently, we
disclosed an enantioselective protocol to access all diastereoisomers
of 8 using Cu-phosphoramidite-catalyzed iterative CA of Me2Zn
to cycloocta-2,7-dienone, followed by oxidative ring opening.4 To
prepare fragment 3, we planned to use the Cu-catalyzed 1,4-addition
of MeMgBr to linear R,â-unsaturated thioesters as the key step.5
The strategy, as outlined above, is both highly convergent and
general for the assembly of enantiopure oligoisoprenoid chains of
any desired length and configuration. However, as the stereogenic
centers are introduced at a relatively early stage, it is essential that
the valuable linear fragments 3-5 are cross-coupled in an efficient
manner. To this end, we examined several carbon-carbon bond-
forming reactions (vide infra). Eventually, a Julia-Kocienski
sequence proved to be the best method.6
Scheme 1. Catalytic Asymmetric Synthesis of Fragment (3R)-3
Coupling of the resulting alkyl chain to mannose via a â-man-
nosyl phosphate linkage is a challenge in itself, considering the
known difficulty of generating such linkages. Application of the
elegant â-mannosylation protocol developed by Crich and Dudkin
was in our case undesirable, as it would require 3 equiv of precious
side chain.3 We therefore decided to first install a â-phosphate
moiety on a suitably protected mannopyranose (6) and then to
couple to the aliphatic chain. Incidentally, this strategy also reduced
the overall sequence with four steps and made the R-phosphate
analogue accessible.
To prepare fragments 4 and 5 (Scheme 2), the hydroxyl moiety
of (3R,7S)-84 was protected as a silyl ether, after which the methyl
ester was reduced to afford the common precursor 12. Sulfone
(3R,7S)-4 was subsequently obtained from 12 analogously to the
synthesis of 3 from 11. For the assembly of 5, 12 was converted
into the corresponding tosylate and then subjected to a Cu-catalyzed
chain elongation reaction with n-pentylmagnesium bromide to give
13. Finally, treatment of 13 with TBAF furnished the primary
alcohol, which was oxidized with TPAP/NMO to give aldehyde
(2S,6S)-5.9
† University of Groningen.
‡ Harvard Medical School.
9
4546
J. AM. CHEM. SOC. 2006, 128, 4546-4547
10.1021/ja060499i CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2. Synthesis of Fragments 4 and 5 and Cross-Coupling
We were pleased to find that coupling of a 2-fold excess of readily
available 6 to (all-S)-17 (1.0 equiv) in the presence of 2,4,6-
triisopropylbenzenesulfonyl chloride (3.0 equiv) proceeded well to
give 20 in a gratifying 79% yield.14 Finally, deacetylation using
NaOMe gave the target molecule (all-S)-1 in quantitative yield.15
Biological evaluation of (all-S)-1 revealed that its antigenic
potency for T cells is identical to that of the natural product within
the margin of error, while stereorandom 2 is significantly less
potent.16 The latter result implies that the stereochemistry of the
lipid part has an unexpectedly strong influence on T cell response.
In summary, we disclose the first (general) procedure for the
total synthesis of enantiopure MPMs. To this end, a fully catalytic
method to synthesize saturated oligoisoprenoids was developed. In
addition, an alternative approach for the formation of the â-man-
nosyl phosphate linkage was shown to be successful. The synthetic
viability of this new protocol is demonstrated by the concise total
synthesis of (all-S)-1 with an overall yield of 6.7% and a longest
linear sequence of 18 steps.
Acknowledgment. We thank Tan-yun Cheng for conducting
the bioassays. T. D. Tiemersma-Wegman (GC and HPLC), A.
Kiewiet (MS), and W. Kruizinga (NMR) are acknowledged for
technical support. This work was supported by the Dutch Organiza-
tion for Scientific Research (NWO) and the NIH (AR 48632).
Supporting Information Available: Detailed experimental pro-
cedures and spectroscopic (1H and 13C NMR) and analytical data of
all reaction products. This material is available free of charge via the
Internet at http://pubs.acs.org.
Coupling of 4 with a small excess of 5 (1.15 equiv) in the
presence of LiHMDS (1.05 equiv) yielded diisoprenoid 14 (pre-
dominantly trans) in a rewarding 74%. The highest yields were
obtained when 4 and 5 were mixed prior to addition of the base.10
Similar procedures using Wittig or Horner-Wadsworth-Emmons
analogues of 4 consistently gave inferior results. Direct formation
of a C-C single bond by Cu-catalyzed cross-coupling of the
corresponding alkyl halides either under Grignard conditions or in
the presence of SmI2 was highly unreliable and invariably gave
low yields in our hands.11 To make 14 suitable for coupling to
fragment 3, the primary alcohol was deprotected and oxidized to
give aldehyde 15 (cf. conversion of 13 into 5).
Julia-Kocienski coupling of 3 (1.2 equiv) to 15 using LiHMDS
(1.1 equiv) as the base yielded 16 (predominantely trans) in a
satisfying 80% yield. Exposure to hydrogen over Pd/C simulta-
neously deprotected the primary alcohol and hydrogenated the
double bonds to give saturated isoprenoid (all-S)-17.12
References
(1) Moody, D. B.; Ulrichs, T.; Mu¨hlecker W.; Young, D. C.; Gurcha, S. S.;
Grant, E.; Rosat, J.-P.; Brenner, M. B.; Costello, C. E.; Besra, G. S.;
Porcelli, S. A. Nature 2000, 404, 884-888.
(2) Matsunaga, I.; Bhatt, A.; Young, D. C.; Cheng, T.-Y.; Eyles, S. J.; Besra,
G. S.; Briken, V.; Porcelli, S. A.; Costello, C. E.; Jacobs, W. R., Jr.;
Moody, D. B. J. Exp. Med. 2004, 200, 1559-1569. On the basis of the
biosynthetic pathway of mycoketides and the prediction by the authors
that the absolute stereochemistry of the alkyl chain is most likely all-S,
we chose to prepare the molecule with that configuration.
(3) Crich, D.; Dudkin, V. J. Am. Chem. Soc. 2002, 124, 2263-2266.
(4) Van Summeren, R. P.; Reijmer, S. J. W.; Feringa, B. L.; Minnaard, A. J.
Chem. Commun. 2005, 11, 1387-1389.
(5) Des Mazery, R.; Pullez, M.; Lo´pez, F.; Harutyunyan, S. R.; Minnaard A.
J.; Feringa, B. L. J. Am. Chem. Soc. 2005, 127, 9966-9967.
(6) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley, A. Synlett 1998,
26-28.
(7) Keck, G. E.; Boden, E. P.; Mabury, S. A. J. Org. Chem. 1985, 50, 709-
710.
Assembly of fragment 6 (Scheme 3) started with the reaction of
hemiacetal 18 with limiting diphenyl chlorophosphate in the
presence of DMAP to give diphenyl â-mannosyl phosphate 19.13
(8) Kocienski, P. J.; Bell, A.; Blakemore, P. R. Synlett 2000, 365-366.
(9) Griffith, W. P.; Ley, S. V.; Whitcombe, G. P.; White, A. D. J. Chem.
Soc., Chem. Commun. 1987, 1625-1627.
(10) Under these conditions, no epimerization at the R position of the aldehyde
was observed.
Scheme 3. Synthesis of Fragment 6 and Coupling to (all-S)-17
(11) Berkowitz, W. F.; Wu, Y. Tetrahedron Lett. 1997, 38, 8141-8144.
(12) GC analysis (see Supporting Information) showed a single isomer of 17,
although it is likely that a trace of the diastereomer originating from
incorporation of (3S)-3 is still present.
(13) Sabesan, S.; Neira, S. Carbohydr. Res. 1992, 223, 169-185.
(14) Warren, C. D.; Liu, I. Y.; Herscovics, A.; Jeanloz, R. W. J. Biol. Chem.
1975, 250, 8069-8078.
(15) The anomeric configuration of (all-S)-1 was unequivocally determined
by NOE correlation between the anomeric hydrogen and H’s 3 and 5 of
the sugar moiety and by a cross-ring fragmentation (m/z 587.5) and a
dehydration (m/z 689.5) in the low-energy ESI CID spectrum. Additional
support was given by comparison of the chemical shift of the anomeric
proton of 1 with that of the anomeric proton of the independently
1
synthesized (all-S)-R-1 analogue as well as by comparison of the JCH
coupling of both anomers (R ) 169 Hz, â ) 159 Hz).
(16) Bioassays were conducted at the Brigham and Women’s Hospital, Harvard
Medical School. Details will be reported in due course.
Hydrogenation over Adams’ catalyst, followed by quenching with
pyridine, afforded pyridinium â-mannopyranosyl phosphate salt 6.
JA060499I
9
J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006 4547