Synthesis of the Chlorofusin Cyclic Peptide: Assignment of the Asparagine Stereochemistry

Article abstract of DOI:10.1021/ol036083g

(Equation presented) An efficient synthesis of two diastereomers of the chlorofusin cyclic peptide bearing either the L-Asn3/D-Asn-4 or D-Asn3/L-Asn4 stereochemistry is detailed. Four key subunits were prepared, sequentially coupled, and cyclized to provi

Full text of DOI:10.1021/ol036083g

ORGANIC
LETTERS
2003
Vol. 5, No. 26
5047-5050
Synthesis of the Chlorofusin Cyclic
Peptide: Assignment of the Asparagine
Stereochemistry
Pankaj Desai, Steven S. Pfeiffer, and Dale L. Boger*
Department of Chemistry and the Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 North Torrey Pines Road,
La Jolla, California 92037
boger@scripps.edu
Received October 24, 2003
ABSTRACT
An efficient synthesis of two diastereomers of the chlorofusin cyclic peptide bearing either the L-Asn3/D-Asn-4 or D-Asn3/L-Asn4 stereochemistry
is detailed. Four key subunits were prepared, sequentially coupled, and cyclized to provide the two diastereomeric macrocycles. The absolute
stereochemistry at the asparagine residues 3 and 4 was assigned as L and D, respectively, by correlating the NMR data of the two diastereomers
with that reported for the natural product.
Phosphoprotein p53 plays a crucial role in the prevention of
cancer.1-3 It controls cell proliferation and development of
genetic abnormalities by inducing G1 arrest or apoptosis
upon DNA damage.4,5 The protein MDM2 (HDM2) interacts
with the tumor suppressor p53 to form a stable complex6 in
which the DNA binding domain of p53 is concealed and
unable to function as a regulator of cell division.7-10 The
X-ray crystal structure of a p53-MDM2 complex revealed
that this interaction is mediated by three key hydrophobic
amino acid residues of p53 and a hydrophobic cleft in
MDM2.11 Molecules that bind this MDM2 hydrophobic cleft
can disrupt the p53-MDM2 interaction and restore the
inhibited function to p53.12-14
Chlorofusin is a fungal metabolite isolated as the major
component in the fermentation broth of Fusarium sp. 2202615
(updated as Microdochium caespitosum).16 It was found to
disrupt the interaction between p53 and MDM2 by directly
(1) Vogelstein, B.; Lane, D.; Levine, A. J. Nature 2000, 408, 307-310.
(2) Review: Vousden, K. H. Cell 2000, 103, 691-694.
(3) Vousden, K. H.; Vande Woude, G. F. Nat. Cell Biol. 2000, 2, E178-
E180.
(4) Review: Shen, Y.; White, E. AdV. Cancer Res. 2001, 82, 55-84.
(5) Review: Lakin, N. D.; Jackson, S. P. Oncogene 1999, 18, 7644-
7655.
(6) Oliner, J. D.; Pietenpol, J. A.; Thiagalingam, S.; Gyuris, J.; Kinzler,
K. W.; Vogelstein, B. Nature 1993, 362, 857-860.
(7) Oliner, J. D.; Kinzler, K. W., Meltzer, P. S., George, D. L.; Vogelstein,
B. Nature 1992, 358, 80-83.
(8) Freedman, D. A.; Levine, A. J. Cancer Res. 1999, 59, 1-7.
(9) Haupt, Y.; Maya, R.; Kazaz, A.; Oren, M. Nature 1997, 387, 296-
299.
(10) Kubbutat, M. H. G.; Jones, S. N.; Vousden, K. H. Nature 1997,
387, 299-303.
(11) Kussie, P. H.; Gorina, S.; Marechal, V.; Elenbaas, B.; Moreau, J.;
Levine, A. J.; Pavletich, N. P. Science 1996, 274, 948-953.
(12) Review: Zhang, R.; Wang, H. Curr. Pharm. Des. 2000, 6, 393-
416.
(13) Review: Picksley, S. M.; Dart, D. A.; Mansoor, M. S.; Loadman,
P. M. Exp. Opin. Ther. Pat. 2001, 11, 1825-1835.
(14) Review: Che`ne, P. Nature 2003, 3, 102-109.
10.1021/ol036083g CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/27/2003

binding to the N-terminal domain of MDM2 (IC50 ) 4.6
µM, KD ) 4.7 µM).15,17 Thus, chlorofusin represents an
exciting lead for antineoplastic intervention that acts by a
rare disruption of a protein-protein interaction.18
On the basis of spectroscopic studies, the chlorofusin
structure was proposed to be composed of a densely
functionalized chromophore linked through the terminal
amine of ornithine to a 27-membered cyclic peptide com-
posed of nine amino acid residues (Figure 1).15 Two of the
Figure 2. Chlorofusin cyclic peptide.
the Scho¨llkopf reagent (S)-119 with 1-iodooctane (n-BuLi,
THF, -78 °C, 94%, >96% de), Scheme 1. Hydrolysis of 2
Scheme 1
Figure 1. Proposed structure of chlorofusin.
cyclic peptide amino acids possess a nonstandard or modified
side chain, and four possess the D-configuration. Although
the spectroscopic studies of chlorofusin permitted assignment
of the unusual chromophore structure and relative stereo-
chemistry, the absolute stereochemistry at C-4, C-8, and C-9
could not be assigned. Similarly, the two cyclic peptide
asparagine residues Asn3 and Asn4 were established to have
opposite stereochemistries (L and D), although the respective
assignments were not possible.
(0.5 M aq HCl, 25 °C, 24 h, quant.) and Cbz amine protection
(3 equiv CbzCl, 3 equiv of Na2CO3, EtOAc-H2O 1:1, 0-25
°C, 18 h, 95%) to assist chromatographic purification and
permit prolonged storage provided 4 (>96% ee),20 and
subsequent Cbz deprotection (H2, 10% Pd/C, EtOH, 25 °C,
1 h, quant.) afforded D-Ada-OMe (3).
Herein, we report the synthesis of the two possible Asn
diastereomers (L,D and D,L) of the chlorofusin cyclic peptide
and the assignment of the natural product stereochemistry
as L-Asn3 and D-Asn4. Four key subunits were assembled,
sequentially coupled, and cyclized to provide the 27-
membered cyclic peptide core (Figure 2). The coupling and
macrocyclization sites were carefully chosen to minimize the
use of protecting groups and maximize the convergency of
the synthesis. Deliberate late-stage incorporation of the
subunit bearing the two asparagine residues allowed conve-
nient access to both diastereomers required to assign the
absolute stereochemistry.
The preparation of the key subunits and assemblage of
the heptapeptide 16 common to both diastereomers is detailed
in Scheme 2 and the preparation of the two required
asparagine dipeptides is summarized in Scheme 3. Dipeptide
5 was obtained by coupling Boc-D-Leu-OH and L-Thr-OBn
(EDCI, HOAt, DMF, 0-25 °C, 18 h, 94%), and subsequent
debenzylation (H2, 10% Pd/C, EtOH, 25 °C, 1 h) provided
the corresponding free acid 6. In a similar manner, dipeptide
7 was obtained from Boc-D-Leu-OH and D-Ada-OMe (3,
EDCI, HOAt, DMF, 0-25 °C, 18 h, 95%), which was further
subjected to Boc deprotection (4 M HCl, dioxane, 0 °C, 1
h) to provide 8. Coupling of 6 and 8 (EDCI, HOAt, NaHCO3,
DMF, 0-25 °C, 18 h, 85%) afforded tetrapeptide 9 that was
subsequently hydrolyzed (LiOH, THF-H2O 1:1, 0 °C, 18
h) to provide acid 10. Dipeptide 11 was prepared by coupling
Boc-L-Thr-OH and L-Ala-OBn (EDCI, HOAt, DMF, 0-25
°C, 18 h, 86%) and was followed by Boc deprotection (4 M
The D-2-aminodecanoic acid (D-Ada) residue bearing the
only nonstandard side chain was prepared by alkylation of
(15) Duncan, S. J.; Gru¨schow, S.; Williams, D. H.; McNicholas, C.;
Purewal, R.; Hajek, M.; Gerlitz, M.; Martin, S.; Wrigley, S. K.; Moore, M.
J. Am. Chem. Soc. 2001, 123, 554-560.
(16) Duncan, S. J.; Williams, D. H.; Ainsworth, M.; Martin, S.; Ford,
R.; Wrigley, S. K. Tetrahedron Lett. 2002, 43, 1075-1078.
(17) Duncan, S. J.; Cooper, M. A.; Williams, D. H. Chem. Commun.
2003, 316-317.
(19) Groth, U.; Scho¨llkopf, U. Synthesis 1983, 1, 37-38.
(20) Established by chiral HPLC analysis: CHIRALCEL OD 0.46 ×
25 cm column, 12% EtOAc-hexanes, 1 mL/min; retention times ) 10.2
min (S), 11.3 min (R); [R]D -5.0 (c 1.0, CHCl3).
(18) Boger, D. L.; Desharnais, J.; Capps, K. Angew. Chem., Int. Ed. 2003,
42, 4138-4176.
5048
Org. Lett., Vol. 5, No. 26, 2003

The two asparagine dipeptides 17 (from Fmoc-D-Asn(Trt)-
OH and L-Asn(Trt)-OBn) and 19 (from Fmoc-L-Asn(Trt)-
OH and D-Asn(Trt)-OBn) were prepared simultaneously
(Scheme 3). Fmoc deprotection immediately prior to use
under standard conditions (0 °C, 10% piperidine in CH2Cl2)
provided 18 and 20, respectively.
Scheme 2
As depicted in Scheme 4, heptapeptide 16 could either be
coupled with dipeptide 18 to afford the D,L-linear peptide
Scheme 4
HCl, dioxane, 0 °C, 1 h) to provide 12. Subsequent coupling
of 12 with Boc-L-Orn(SES)-OH21 (EDCI, HOAt, NaHCO3,
DMF, 0-25 °C, 18 h, 89%) afforded tripeptide 14 after Boc
deprotection (4 M HCl, dioxane, 0 °C, 1 h) of compound
13. The tetrapeptide 10 was coupled with tripeptide 14
(EDCI, HOAt, NaHCO3, DMF, 0-25 °C, 18 h, 92%) to
provide the heptapeptide 15, which was further debenzylated
(H2, 10% Pd/C, EtOH, 25 °C, 1 h) to afford free acid 16.
Scheme 3
21 in 71% yield or coupled with dipeptide 20 to afford the
corresponding L,D-diastereomer 24 in 72% yield. Each
diastereomer was further subjected to a sequential depro-
tection of the Boc group (bromocatecholborane BCB, CH2-
Cl2, 0 °C, 1 h) and benzyl group (H2, 10% Pd/C, EtOH, 25
°C, 1 h). Subsequent macrocyclization under high dilution
conditions (EDCI, HOAt, 0.01 M DMF, 0-25 °C, 24 h)
afforded the diastereomeric macrocycles 27 and 29 in 60%
yield in each case. TFA-mediated deprotection of the trityl
groups (TFA-H2O 20:1, 25 °C, 5 h) cleanly provided the
corresponding cyclic peptides 28 and 30.
(21) Jiang, W.; Wanner, J.; Lee, R. J.; Bounaud, P.; Boger, D. L. J. Am.
Chem. Soc. 2002, 124, 5288-5290. Jiang, W.; Wanner, J.; Lee, R. J.;
Bounaud, P.; Boger, D. L. J. Am. Chem. Soc. 2003, 125, 1877-1887.
Org. Lett., Vol. 5, No. 26, 2003
5049

The data obtained from COSY, HMBC, and HMQC NMR
experiments indicated that the proton and carbon chemical
shifts for peptide 30 closely matched those reported for the
natural product as compared to those for peptide 28 (Figure
3).22 Characteristic of this similarity in proton chemical shifts,
the Ada8 side chain for chlorofusin.15 A key long-range NOE
between both â-methylene hydrogens of Asn3 and the two
6
methines of Thr6 including the R-H (C2 -H) was also
observed. It was suggested that residues Leu5, Thr1, and
Ada8 may fold and cluster over the top of the macrocyclic
6
ring while Asn3 folds under and lies proximal to Thr6 (C2 -
H). A large geminal coupling constant of the Asn3 methylene
(J ) 15.2 Hz) indicative of a restricted conformation and
the slow exchange of the Asn3 carboxamide was suggested
to be potentially derived from intramolecular H-bonding to
the Thr6 hydroxyl accounting for the clearly defined Asn3/
Thr6 side chain NOEs. To fold in this manner, it was
suggested that Asn3 likely has the L-configuration and Asn4
has the D-configuration.15
The long-range NOEs observed in the NOESY and
ROESY NMR experiments for 28 and 30 were not as
extensively defined in our work as those in the Williams
study but were sufficient to indicate that peptide 30 satisfies
the above key NOEs while 28 does not.22 Notably, 30 not
only exhibited the diagnostic Leu5 side chain NOEs with
the Thr1 and Ada8 side chains indicative of their potential
clustering on the top of the cyclic peptide but also exhibited
the NOEs between the Asn3 methylene and the Thr6 R-H
and â-H indicating that they lie proximal to one another
potentially beneath the cyclic peptide. In contrast, 28
exhibited a different series of NOEs. Instead of Thr1, it was
Thr6 that exhibited an extensive series of NOEs with Leu5.
Thus, the Leu5 methyl groups exhibited NOEs with the Thr6
methyl and R-H as well as the Ada8 side chain methylenes,
and the Leu5 γ-methine and NH exhibited NOEs with the
Thr6 R-H. Both the Asn3 (observed with chlorofusin) and
Asn4 (not observed with chlorofusin) methylenes exhibited
NOEs with the Thr6 R-H, but the additional NOEs of Asn3
to the Thr6 side chain seen with chlorofusin and 30 were
not observed. Interestingly, neither isomer (28 or 30)
exhibited the chlorofusin Asn4 methylene NOE with the Thr1
R-H. The Asn3 methylene of 30 exhibited the large geminal
coupling constant characteristic of the natural product (J )
15.8 vs 15.2 Hz for chlorofusin). This large geminal coupling
constant was not observed with the methylene of Asn3 of
28, whereas that of Asn4 of 28 did exhibit a large geminal
coupling (J ) 16.3 Hz). These observations permit the
assignment of absolute configuration at Asn3 in chlorofusin
as L and that of Asn4 as D.23
Figure 3. Difference in H1 (top) and C13 (bottom) NMR chemical
shifts (DMSO-d6, 500 MHz).
the nine R-CH chemical shifts of 30 exhibited e 0.02 δ
differences from those reported for chlorofusin (av ∆δ )
0.016), whereas those of 28 exhibited larger differences of
0.1-0.6 δ (av ∆δ ) 0.34). Analogous comparisons of the
13C shifts of the R-carbons provide similar distinctions (28,
∆δ ) 0.5-3.0, av ∆δ ) 1.91 vs 30, ∆δ ) 0.0-0.8, av ∆δ
) 0.47). In addition, Williams et al. reported diagnostic long-
range NOEs between the methyl groups of Leu5 and the
methyl group of Thr1 as well as the methylene groups of
Extensions of these studies to the total synthesis of
chlorofusin and structural analogues are in progress and will
be reported in due course.23
Acknowledgment. We gratefully acknowledge the fi-
nancial support of the National Institutes of Health (CA
41101) and the Skaggs Institute for Chemical Biology. We
are especially grateful to Professor J. Chen (H. Lee Moffitt
Cancer Center) for conducting the MDM2/p53 binding
inhibition assays and Dr. M. Searcey for discussions leading
to our joint disclosures.
(22) HMBC experiments permitted the unambiguous assignment (con-
nectivity) of the Thr1 vs Thr6, Asn3 vs Asn4, and Leu5 vs Leu7 residues
in 28 (Ala2 Me T Ala2 NH T Thr1 R-C.; Leu5 Me T Leu5 â-CH2
T
Leu5 NH T Asn4 R-C). In contrast with 30, only Leu7 could be
unambiguously established (Leu7 â-CH2 T Leu7 CO T Ada8 NH: weak)
1
and the remaining residue assignments were made on the basis of H and
13C NMR chemical shift comparisons reported for chlorofusin.
(23) Both 28 and 30 and a series of Orn 9 derivatives (-NHSES f NH2,
NHCbz. NHFmoc, NHBoc, N-phthalimide, and NH-naphthalene acetamide)
were assayed for inhibition of MDM2/p53 binding in an ELISA assay
enlisting immobilized p53 and full length MDM2. None were active (IC50
> 250 µM), indicating that they are g100 times less effective than
chlorofusin.
Supporting Information Available: Full experimental
details for the preparation of 2-30. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL036083G
5050
Org. Lett., Vol. 5, No. 26, 2003