Solid-phase synthesis of C-terminal peptide aldehydes from amino acetals anchored to a backbone amide linker (BAL) handle

Article abstract of DOI:10.1016/S0040-4039(00)00950-3

Peptide aldehydes were synthesized, starting from amino acetals, by a Solid-phase backbone amide linker (BAL) strategy. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00950-3

Tetrahedron Letters 41 (2000) 6131±6135
Solid-phase synthesis of C-terminal peptide aldehydes
from amino acetals anchored to a backbone amide
linker (BAL) handle
Fanny Guillaumie,a Joseph C. Kappel,b Nicholas M. Kelly,a George Baranyb,*
and Knud J. Jensena,*
aDepartment of Organic Chemistry, Technical University of Denmark, 2800 Lyngby, Denmark
bDepartment of Chemistry, University of Minnesota, Minneapolis, MN 55455, USA
Received 23 March 2000; accepted 7 June 2000
Abstract
Peptide aldehydes were synthesized, starting from amino acetals, by a solid-phase backbone amide linker
(BAL) strategy. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: peptide aldehydes; solid-phase synthesis; BAL; amino acetals.
C-Terminal peptide aldehydes are potential serine, cysteine, and aspartic protease inhibitors
that have emerged as promising therapeutic agents for the treatment of, for example, viral
infections such as HIV and human rhinovirus 3C.1 4 While peptide aldehydes have been prepared
previously by solution synthesis, combinatorial syntheses of such compounds are more likely
facilitated by solid-phase strategies. Current methods include release from a Weinreb amide-
based handle with LiAlH4, from a semicarbazone handle with dilute acid, or from an ole®nic
linker by ozonolysis.5 7 Recently, we reported the synthesis of a peptide aldehyde with a
C-terminal glycinal residue, starting from 2,2-dimethoxyethylamine anchored to a backbone
amide linker (BAL) handle.8 Here we report on the extension of our rapid and ecient strategy to
allow for the general synthesis of complex peptide aldehydes starting from amino acid-derived
amino acetals.
Initial studies focused on peptide aldehydes with C-terminal alaninal (R=Me) or phenylalaninal
(R=Bn). Amino acetals were derived from Na-Z-amino acids by a four-step sequence (Scheme 1).
First, the Weinreb amides were prepared and then reduced with LiAlH4 at 0 to 25ꢀC to give the
corresponding aldehydes. Treatment with ethylene glycol in re¯uxing toluene, in the presence of a
catalytic amount of TsOH, established the 1,3-dioxolane moiety, following which removal of the
* Corresponding authors. E-mail: okki@pop.dtu.dk, barany@garnet.tc.umn.edu
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00950-3

6132
Z-protecting group with 1,4-cyclohexadiene in the presence of Pearlman's catalyst gave the
desired amino acetals.9
i
.
Scheme 1. Synthesis of amino 1,3-dioxolanes. Reagents and conditions: (a) HN(Me)OMe HCl (1.5 equiv.), Pr2NEt
(2.6 equiv.), HBTU (1.1 equiv.), HOBt (1.1 equiv.), DMF, 25ꢀC, 2 h; R=Me, 90%; R=Bn, 95%; (b) LiAlH4 (3.0
equiv.), THF, 0ꢀC to 25ꢀC, 0.5 h; R=Me, 70%; R=Bn, 90%; (c) (CH2OH)2 (10 equiv.), TsOH (0.1 equiv.), PhMe,
re¯ux, 6 h; R=Me, 92%; R=Bn, 64%; (d) Pd(OH)2 (1 equiv.), 1,4-cyclohexadiene (10 equiv.), EtOAc, 4 h; R=Me,
71%; R=Bn, 99%
Alternatively, Na-Fmoc-amino acids were converted to the corresponding Weinreb amides
which, upon treatment with LiAlH4 at ^78ꢀC gave the Na-Fmoc-protected amino aldehydes
(Scheme 2).10,11 DIBAL-H also proved ecient for the reduction. However, both reductive pro-
cedures su€ered from partial cleavage of the Fmoc moiety under the reaction conditions, and the
initial crude products were of low purity. In a superior approach, the Na-Fmoc-amino acids were
treated12 with iBuOCOCl to form the corresponding mixed anhydrides, which were subsequently
reduced with NaBH4 to give the Na-Fmoc-amino alcohols. Swern oxidation cleanly converted the
alcohols to the aldehydes, i.e. Na-Fmoc-Phe-H in 65% and Na-Fmoc-Ala-H in 67% yield. Pro-
tection of the aldehyde moieties was then accomplished by treatment with trimethyl orthoformate
in methanol, in the presence of a catalytic amount of TsOH at 25ꢀC, to form the corresponding
acetals under mild conditions (Scheme 2). Selective removal of the Fmoc group was dicult, with
some methods giving little or no product.13 The relatively best method was treatment of the
acetals with 4 N aq. NaOH±MeOH±dioxane (1:9:30) for 10 min, which gave the expected amino
acetals. Preparative HPLC immediately after Fmoc-removal from the Phe-derivative yielded 21%
of phenylalaninal dimethyl acetal;14 other puri®cation procedures were inferior.
Scheme 2. Synthesis of Na-Fmoc-amino aldehydes and amino dimethyl acetals. Reagents and conditions: (a)
ꢀ
.
HN(Me)OMe HCl (1.1 equiv.), Et3N (2.1 equiv.), HBTU (1.0 equiv.), CH2Cl2, 25 C, 18 h; R=Me, 91%; R=Bn, 98%;
(b) DIBAL-H (1.5 equiv.), CH2Cl2, ^78ꢀC, 1 h; R=Me, 30%; R=Bn, 49%; (c) NMM (1.0 equiv.), BuOCOCl (1.0
i
equiv.), DME, ^15ꢀC, 1 min; then NaBH4 (1.5 equiv.), H2O, ^15ꢀC, 30 sec; R=Me, 91%; R=Bn, 94%; (d) (COCl)2
(1.5 equiv.), DMSO (3.0 equiv.), NEt3 (6.0 equiv.), CH2Cl2, ^60ꢀC, 1 h; R=Me, 67%; R=Bn, 65%; (e) (MeO)3CH
(2.0 equiv.), TsOH (0.1 equiv.), MeOH, 25ꢀC, 2 h; R=Me, 79%; R=Bn, 97%; (f) R=Me, DBU (4.0 equiv.), DMF,
25ꢀC, 10 min (in situ procedure); R=Bn, 4 N aq NaOH±MeOH±dioxane (1:9:30), 10 min, 25ꢀC, 21%

6133
The methodologies described in the preceding paragraph gave amino 1,3-dioxolanes (Scheme
1) or amino dimethyl acetals (Scheme 2). These were coupled to PALdehyde-PS or PEG-PS resins
by NaBH3CN-promoted reductive aminations in DMF±HOAc (99:1) (Scheme 3). However,
given the low yields and associated diculties in the preparation of amino dimethyl acetals from
their Fmoc precursors (Scheme 2), a simpler alternative in situ strategy was developed. Thus, the
Fmoc-protected alaninal acetal was dissolved in DMF±DBU (9:1); after 10 min, HOAc was
added, and the solution was transferred to PALdehyde resin together with NaBH3CN to
complete the reductive amination in 1 h (Scheme 3).
Scheme 3. Solid-phase synthesis of peptide aldehydes
Peptide synthesis proceeded with acylation of the secondary BAL-linked amine using either
Fmoc-amino acid as preformed symmetrical anhydrides in CH2Cl2 or the Fmoc-amino acid
activated in situ by HATU/DIEA in CH2Cl2±DMF (9:1). Past this point, chain elongation fol-
lowed standard Fmoc procedures. Treatment of completed peptidyl-resins with TFA:±H2O (19:1)
released the ®nal products (Scheme 3). Concomitant cleavage of the acetal moiety to free the
C-terminal aldehyde functionality was con®rmed by MS. Although both the preformed and the in
situ-generated amino acetals gave the expected peptide aldehydes, the in situ protocol gave
simpler and more economical access to peptide aldehydes. Sequences synthesized by this strategy
include: Na-Fmoc-Ala-Ala-Pro-Ala-H, Na-Fmoc-Asp-Phe-Val-Ala-H, Na-Fmoc-Ala-Ala-Pro-
Phe-H, and Na-Fmoc-Glu-Val-Lys-Phe-H from dimethyl acetals and Na-Fmoc-Asp-Phe-Val-Ala-
H and Na-Fmoc-Glu-Val-Val-Phe-H from 1,3-dioxolanes.15 Na-Fmoc-Asp-Phe-Val-Ala-H was
also obtained when the completed dimethyl acetal peptidyl-resin was treated with TFA±thio-
anisole±ethanedithiol (EDT)±anisole (90:5:3:2; Reagent R). Interestingly, thioacetal adduct(s) of
Na-Fmoc-Ala-Ala-Pro-Phe-H with EDT were obtained upon treatment of the completed
peptidyl-resin with TFA±EDT (19:1).16
To address the question of racemization17 of the amino aldehyde moiety, the route to
Na-Fmoc-Glu-Val-Val-Phe-H, starting from Z-Phe-OH, was repeated starting from Na-Z-d-Phe-
OH to provide a reference to the corresponding diastereomeric Fmoc-protected peptide aldehyde.
Starting with the same protected peptide-resins, ®nal Fmoc deprotection steps were carried out
on resin, followed by TFA cleavage, to provide the free peptides H-Glu-Val-Val-Phe-H and

6134
H-Glu-Val-Val-d-Phe-H. These free peptides were resolved cleanly by HPLC; in the l-Phe-H
peptide, 12% of the opposite diastereomer (d) was present, while in the d-Phe-H peptide, 26% of
l was present. Further racemization17 of the l-Phe-H peptide was not observed upon standing
overnight in TFA.
In conclusion, we have developed a general strategy for the synthesis of C-terminal peptide
aldehydes that relies on anchoring of amino acid-derived acetals through a BAL handle to a solid
support. Final peptide aldehydes were released with TFA±H2O (19:1), with concomitant cleavage
of the acetal moiety. Extension of this approach to trifunctional Na-Fmoc-amino acids with
appropriate TFA-labile side-chain protecting groups would give precursors to peptides with
C-terminal trifunctional aldehydes.
Acknowledgements
We are grateful to Annette Nordestgaard from Phytera Symbion for LC/MS analyses, and
Professor Fernando Albericio for valuable discussions. We thank the Leo Foundation (K.J.), the
Lundbeck Foundation (K.J.), and the NIH (GM 42722 to GB) for ®nancial support. J.K. was
supported by an NIH Biotechnology Training Grant (GM 08347).
References
1. McConnell, R. M.; Barnes, G. E.; Hoyng, C. F.; Gunn, J. M. J. Med. Chem. 1990, 33, 86±93.
2. Sarubbi, E.; Seneci, P. F.; Angelastro, M. R.; Peet, N. P.; Denaro, M.; Islam, K. FEBS Lett. 1993, 319, 253±256.
3. Semple, J. E.; Minami, N. K.; Tamura, S. Y.; Brunck, T. K.; Nutt, R. F.; Ripka, W. C. Bioorg. Med. Chem. Lett.
1997, 7, 2421±2426.
4. Shepherd, T. A.; Cox, G. A.; McKinney, E.; Tang, J.; Wakulchik, M.; Zimmerman, R. E.; Villarreal, E. C. Bioorg.
Med. Chem. Lett. 1996, 6, 2893±2896.
5. Fehrentz, J. A.; Paris, M.; Heitz, A.; Velek, J.; Winternitz, F.; Martinez, J. J. Org. Chem. 1997, 62, 6792±6796.
6. Murphy, A. M.; Dagnino Jr., R.; Vallar, P. L.; Trippe, A. J.; Sherman, S. L.; Lumpkin, R. H.; Tamura, S. Y.;
Webb, T. R. J. Am. Chem. Soc. 1992, 114, 3156±3157.
7. Pothion, C.; Paris, M.; Heitz, A.; Rocheblave, L.; Rouch, F.; Fehrentz, J. A.; Martinez, J. Tetrahedron Lett. 1997,
38, 7749±7752.
8. Jensen, K. J.; Alsina, J.; Songster, M. F.; Vagner, J.; Albericio, F.; Barany, G. J. Am. Chem. Soc. 1998, 120,
5441±5452.
9. The described conditions gave a good yield and clean product. More standard catalytic hydrogenolysis conditions
gave lower yields, and showed unreacted starting material and/or the formation of by-products. For example, H2
in the presence of Pd/C gave the desired product in only 22% yield after puri®cation. H2 in the presence of
Pd(OH)2 gave erratic results, with the range of product formed being 26±62%.
10. Fehrentz, J. A.; Castro, B. Synthesis 1983, 676±678.
11. Wen, J. J.; Crews, C. M. Tetrahedron: Asymmetry 1998, 9, 1855±1858.
12. Rodriguez, M.; Llinares, M.; Doulut, S.; Heitz, A.; Martinez, J. Tetrahedron Lett. 1991, 32, 923±926.
13. Other conditions tried include (with the associated problems): piperidine:DMF (1:4), dicult to remove
piperidine; 5 equiv. aq. NaOH in THF for 3 h, incomplete reaction; 5 equiv. aq. NaOH in 1,4-dioxane for 3 h,
solubility problems, dicult to reproduce, loss during work up; ammonium formate in the presence of Pd/C in
MeOH±dioxane (1:1) under argon, no reaction after 1 day; DBU in EtOAc at 70ꢀC for 3 h, amino acetal could not
be isolated.
14. Optical rotation for the phenylalaninal dimethyl acetal was [ꢀ]D ^32.6 (c 1.5, MeOH, at 21ꢀC); lit. [ꢀ]D ^27.2 (c 1.6,
MeOH), Gacek, M.; Undheim, K. Tetrahedron 1974, 30, 4233±4237.
15. The purities (HPLC, 265/220 nm) and m/z of the Fmoc-protected crude peptide aldehydes were: (from dimethyl
acetals) Glu-Val-Lys-Phe-H 50%, 728.6 [M+H]+; Asp-Phe-Val-Ala-H, 54%, 657.4 [M+H]+; Ala-Ala-Pro-Ala-H,

6135
69%, 535.5 [M+H]+; (from 1,3-dioxolanes) Asp-Phe-Val-Ala-H, 82%, 657.3 [M+H]+; Glu-Val-Val-Phe-H, 78%;
699.2 [M+H]+; Glu-Val-Val-d-Phe-H, 84%, 699.4 [M+H]+; for unprotected peptide aldehydes H-Glu-Val-Val-Phe-
H, 82%, 477.1 [M+H]+; H-Glu-Val-Val-d-Phe-H, 58%, 477.1 [M+H]+.
16. Thioacetal adduct(s) were the major product (44%), assigned to the hemimercaptal and/or dithioacetal structures,
and only small amounts of the desired peptide aldehyde (21%) formed under these cleavage conditions (LC/MS
calcd peptide aldehyde 610.28; found m/z 3.56 min: 743.2 [M+EDT+K]+, 765.3 [M+EDT+K+Na]+, and 687.2
[M+EDT±H2O+H]+; 3.86 min, 646.2 [M+2H2O]+). Compare this result to that from treatment with the EDT-
containing cocktail Reagent R, where no thioacetal formed upon cleavage. We speculate that formation of the
peptide aldehyde only upon cleavage with Reagent R could be due to residual water in the anisole or thioanisole
used.
17. The ®nal C-terminal peptide aldehydes, as well as the starting amino aldehydes, have been reported to racemize
under conditions of silica gel column chromatography and HPLC (Fehrentz, J. A.; Paris, M.; Heitz, A.; Velek, J.;
Liu C.-F.; Winternitz, F.; Martinez, J. Tetrahedron Lett. 1995, 36, 7871±7874; Ede, N. J.; Eagle, S. N.; Wickham,
G.; Bray, A. M.; Warne, B.; Shoemaker, K.; Rosenberg, S. J. Pept. Sci. 2000, 6, 11±18). Experiments are currently
under way to determine at which stage of the overall process our compounds racemize, and to develop protocols
to maximize the chiral integrity of the ®nal aldehyde products.