was treated with HF(aq) for 2 h, neutralized with NaOH(aq),
and extracted into CH2Cl2/EtOAc. Finally, hydrogenolysis
cleaved the Cbz protective groups (42%; two steps). How-
ever, this product was spectroscopically distinctively different
from natural aeruginosin 298-A; apparently it represented a
diastereomer.23
Upon reinspection of the configuration of other members
of the aeruginosin family, we began to suspect that the
stereochemical assignment of the leucine residue might be
incorrect.24 Thus, we also prepared the D-leucine analogue
of aeruginosin 298-A using the same strategy (Scheme 6).
stereochemistry of (+)-aeruginosin 298-A. A direct com-
parison between synthetic and natural samples is unfortu-
nately not possible, since the natural product is no longer
available.
In conclusion, a concise synthesis of the novel bicyclic
amino acid 2 from L-tyrosine has been developed that can
readily be adapted to the preparation of analogues and
peptidomimetic scaffolds. The total synthesis of (+)-aerugi-
nosin 298-A demonstrates the potentially biomimetic incor-
poration of an oxidative cyclization product of L-tyrosine
into the highly functionalized backbone of this potent
thrombin inhibitor. Key steps of the total synthesis include
the efficient construction of other nonproteinogenic building
blocks in aeruginosin 298-A as well as an extensive
optimization of coupling strategies.
Scheme 6. Synthesis of D-Leu-aeruginosin 298-Aa
The preparation of LLeu- as well as DLeu-aeruginosin 298-A
compels one to a reassignment of the configuration of the
natural product. As a consequence, the binding mode of
aeruginosin 298-A to thrombin is as expected,27 with a
D-leucine occupying the D-S3 binding site of the enzyme.
This structural reassignment also explains the similarity28
between the binding modes of 298-A and 98-B29 (98-B has
a D-alloisoleucine residue at P3) to serine proteases.30
Acknowledgment. This work has been supported by the
National Institutes of Health (AI/GM-33506) and by a
graduate fellowship to J.M. from FCAR (Que´bec). We thank
Dr. H. Takahashi for preliminary experiments on the
thermodynamic equilibration of 11 to 12, Professor Mu-
1
rakami for providing H and 13C spectra of the natural
product, and Professor Tulinsky for helpful discussions
regarding the X-ray structural assignment.
a Reagents and conditions: (a) NH3Cl-D-Leu-OBn, DEPC,
iPr2NEt, CH2Cl2, 68%; (b) H2, Pd/C, EtOH, 95%; (c) Bu3SnH,
Pd(PPh3)4, AcOH, CH2Cl2; (d) DLeu-22, DEPBT, iPr2NEt, CH2Cl2,
59%; (e) HF (aq); H2, Pd/C, EtOH, 34%.
Supporting Information Available: Experimental pro-
cedures and spectroscopic data. This material is available
OL006759X
Indeed, the 1H and 13C NMR data of DLeu-1 matched exactly
those reported for aeruginosin 298-A.25,26
(24) The 1H NMR data for 298-A (L-Leu) closely parallel those of 298-
B-major, 89-A, 89-B, and desulfated, dimethyl acetal-derivatized 89-A and
89-B (all have D-Leu) but are very different from 298-B-minor (L-Leu);
see Supporting Information.
While at this stage we cannot yet explain the discrepancy
between our assignment and the X-ray structure of (+)-
aeruginosin 298-A, we are confident that the revised structure
Leu-1, which matches all available spectroscopic data of the
(25) Marfey analysis of the acid hydrolysate of both LLeu- and DLeu-1
confirmed that the leucine residues were indeed in the L- and D-form,
respectively (procedure: Adamson, J. G.; Hoang, T.; Crivici, A.; Lajoie,
G. A. Anal. Biochem. 1992, 202, 210-214; see Supporting Information).
D
natural product, is indeed representative of the actual
(18) Griffith, W. P.; Ley, S. V. Aldrichimica Acta 1990, 23, 13-19.
(19) Cbz-Choi(TBS)-OMe (L-3) was converted into Ac-Choi-OMe, which
(26) [R]D:
(0.36, H2O).
L
Leu-1 -18 (0.25, H2O), DLeu-1 +16 (0.17, H2O), lit. +22
1
by H and 13C NMR and [R]D is identical to that prepared by Bonjoch et
(27) Based on the well-established binding mode of D-Phe-Pro-Arg
chloromethyl ketone (Bode, W.; Turk, D.; Karshikov, A. Protein Sci. 1992,
1, 426-471).
(28) The P1-P4 sequences of both 298-A and 98-B occupy the same
binding sites: the D-allo-Ile and D-Hpla of 98-B hydrogen bond to Gly216
and Gly219, respectively; the Leu and D-Hpla of 298-A also bind to Gly216
and Gly219, respectively.
(29) Sandler, B.; Murakami, M.; Clardy, J. J. Am. Chem. Soc. 1998,
120, 595-596.
(30) After submission of this manuscript, another synthesis of aeruginosin
298-A was published that is in agreement with our stereochemical
conclusions: Valls, N.; Lopez-Canet, M.; Vallribera, M.; Bonjoch, J. J.
Am. Chem. Soc. ASAP, Nov. 1, 2000 Web publication date.
al. (Bonjoch, J.; Catena, J.; Isabal, E.; Lopez-Canet, M.; Valls, N.
Tetrahedron: Asymmetry 1996, 7, 1899-1902).
(20) Rodriguez, M.; Llinares, M.; Doulut, S.; Heitz, A.; Martinez, J.
Tetrahedron Lett. 1991, 7, 923-926. Fresh NaBH4 was required to suppress
lactam formation.
(21) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-
7287. Ireland, R. E.; Liu, L. J. Org. Chem. 1993, 58, 2899.
(22) DEPBT: 3-(diethoxyphosphoryloxy)-1,2,3-benzotriazin-4(3H)-one.
Li, H.; Jiang, X.; Ye, Y.; Fan, C.; Romoff, T.; Goodman, M. Org. Lett.
1999, 1, 91.
(23) An alternative synthetic strategy whereby the Hpla-Leu side chain
was attached first gave identical material.
4216
Org. Lett., Vol. 2, No. 26, 2000