Synthesis and CD spectra of fluoro- and hydroxy-substituted β-peptides

Article abstract of DOI:10.1002/hlca.200390150

β-Amino acids 1-3 with OH and F substituents in the α-position have been prepared (Scheme) from the natural (S)-α-amino acids alanine, valine, and leucine, and incorporated into β-hexa- and β-heptapeptides 4-12. The peptide syntheses were performed according to a conventional solution strategy (Boc/Bn protection) with fragment coupling. The new β-peptides with (series a) and without (series b) terminal protection were isolated in HPLC-pure form and characterized by NMR spectroscopy and MALDI mass spectrometry. The chemical properties as well as the patterns of the CD spectra (Figs. 3-5) depend upon constitution (OH, F, F₂ substitution) and configuration (l or u) of the amino acid residues, upon the total number of OH and F substituents in the peptide chain, and upon the solvent used (H₂O, MeOH, CF₃CH₂OH, (CF₃)₂CHOH). No reliable clues regarding the structures can be obtained from these CD spectra. Only a full NMR analysis will be able to answer the questions: a) with which known secondary structures (Figs. 1 and 2) of β-peptides are the OH and F derivatives compatible? b) Are new secondary structures enforced by the polar and/or H-bonding backbone substituents? Furthermore, the β-peptides described here will enable us to study changes in chemical, enzymatic, and metabolic stability, and in physiological properties caused by the heteroatoms.

Full text of DOI:10.1002/hlca.200390150

1862
Helvetica Chimica Acta Vol. 86 (2003)
Synthesis and CD Spectra of Fluoro- and Hydroxy-Substituted b-Peptides
Preliminary Communication
by FrancÀois Gessier1), Christian Noti2), Magnus Rueping3), and Dieter Seebach*
Laboratorium f¸r Organische Chemie der Eidgenˆssischen Technischen Hochschule, ETH-Z¸rich-
Hˆnggerberg, Wolfgang-Pauli-Strasse 10, CH-8093 Z¸rich
b-Amino acids 1 3 with OHand F substituents in the a-position have been prepared (Scheme) from the
natural (S)-a-amino acids alanine, valine, and leucine, and incorporated into b-hexa- and b-heptapeptides 4 12.
The peptide syntheses were performed according to a conventional solution strategy (Boc/Bn protection) with
fragment coupling. The new b-peptides with (series a) and without (series b) terminal protection were isolated
in HPLC-pure form and characterized by NMR spectroscopy and MALDI mass spectrometry. The chemical
properties as well as the patterns of the CD spectra (Figs. 3 5) depend upon constitution (OH, F, F2
substitution) and configuration (l or u) of the amino acid residues, upon the total number of OHand F
substituents in the peptide chain, and upon the solvent used (H2O, MeOH, CF3CH2OH, (CF3)2CHOH). No
reliable clues regarding the structures can be obtained from these CD spectra. Only a full NMR analysis will be
able to answer the questions: a) with which known secondary structures (Figs. 1 and 2) of b-peptides are the OH
and F derivatives compatible? b) Are new secondary structures enforced by the polar and/or H-bonding
backbone substituents? Furthermore, the b-peptides described here will enable us to study changes in chemical,
enzymatic, and metabolic stability, and in physiological properties caused by the heteroatoms.
Structural evidence from NMR, X-ray, molecular-modelling, and CD investigations
of b-peptides containing the side chains of proteinogenic amino acids has provided
rules for the design of the secondary structures 314 helix and strands [1], while the
possible substitution patterns and configurations of amino acids in the b-peptidic
hairpin turn and 12/10 helix have not been probed as well [2]. As outlined in Fig. 1,
steric hindrance should prevent non-H-atoms from occupying axial positions in the 314
helix and in-plane position in a sheet (see black spheres in Fig. 1,a and b), which has
been concluded mainly from investigations with Me substituents (A, B, and C in Fig. 2).
The Van der Waals radius of a Me group is of course much larger than that of a H-atom,
and we wondered whether an F- or O-atom, or an OHgroup might fit the forbidden
position next to the CˆO group (cf. the A values in Fig. 1,c, and D, E, F, and G in
Fig. 2). As compared to the H-atom, these atoms have, of course, not only a somewhat
different size but also an electronic effect (Pauling electronegativities H2.2, C 2.6, O
3.4, F 4.0) and the ability of forming H-bonds4). Thus, another issue to be addressed by
the present study was to see what influence the substituents F and OHmight exert on
1
)
Postdoctoral research 2002/2003, financed by ETH-Z¸rich, Schweizerischer Nationalfonds (Project
No. 2000-058831.99), and Novartis Pharma AG, Basel.
2
3
4
)
)
)
Part of the Master Thesis (Diplomarbeit) of C. N., ETH-Z¸rich, 2002/2003.
Part of the Dissertation of M. R. (No. 14677), ETH-Z¸rich, 2002.
For a discussion of the H-bond-acceptor properties of organic fluoride, see [3].

Helvetica Chimica Acta Vol. 86 (2003)
1863
the backbone conformation of a b-peptide when placed in allowed positions (green
spheres in Fig. 1), an effect that cannot possibly be studied in a-peptides5)6). Finally,
the F- and OH-substituted b-peptides will offer an opportunity to determine what
effects the enhanced electrophilicity of the amide CˆO groups has on their biological
properties (such as antimicrobial and haemolytic activity, enzymatic and metabolic
stability, or biodegradability7).
The required enantiomerically pure N-Boc-protected 3-amino-2-fluoro and 3-
amino-2-hydroxy carboxylic acids with the side chains of alanine, valine, and leucine
were prepared as shown in the Scheme. The N,N-dibenzyl-protected aldehydes
prepared from the amino acids by standard procedures [8] were subjected to a trans-
cyanohydrination process (cyanohydrine of acetone/KCN/cat. Bu4NI [9]) to give
mixtures (1:2 to 1:1) of epimeric cyanohydrins. This nonselective reaction was
practical in the present investigation, since we needed to have both, the l- and the u-
diastereoisomers of the 3-amino-2-hydroxyalkanoates, and since it did not make any
difference which epimer was used for the conversion to the difluoro derivatives.
Methanolysis of the epimeric dibenzylamino-cyanohydrins, followed by chromato-
graphic separation of the hydroxy methyl esters, debenzylation, N-Boc protection, and
saponification, led to the configurationally pure acids 1a 1c. On the other hand, the
diastereoisomeric N,N-dibenzyl-protected hydroxy methyl esters were converted to the
2-fluoromethyl esters by retentive OH/F substitution with DAST [10][11], saponified,
debenzylated, and N-Boc-protected to give the acids 2a 2c of either (R,S)- or (S,S)-
configuration. Oxidation of the epimeric mixtures of 2-hydroxy methyl esters to 2-keto
esters, which do not racemize if chromatographic purification is avoided, treatment
with DAST [10], saponification, debenzylation, and N-Boc protection provided the 2,2-
difluoro carboxylic acids 3a 3c (Scheme). For the C-terminal amino acid residues of
the b-peptides to be prepared (see 4 13) by solution coupling of N-Boc-protected
hydroxy and fluoro amino acids, the benzyl esters of 1c, 2c, and 3c were required; they
were prepared by treatment of the corresponding Cs salts with BnBr [12].
With the suitable hydroxy and fluoro amino acids at hand, we synthesized the b-tri-,
b-hexa-, and b-heptapeptides 4 13 with various numbers of hydroxy or fluoro amino
acids incorporated. All syntheses were carried out in solution and according to the
same principle: starting from the benzyl ester of the C-terminal Leu-derived hydroxy or
fluoro amino acid, a tripeptide was built up8), which was then employed for a
dimerizing fragment coupling to an N- and C-protected hexapeptide 6a, 10a, 11a, and
12a. When a central OH-, F-, or F2-substituted b-amino acid was to be incorporated, the
b-tripeptide with the side chains of Val, Ala, and Leu was first coupled with the
corresponding heterosubstituted b-amino acid, and, to the resulting tetrapeptide,
another b3hVal-b3hAla-b3hLeu fragment was attached to give the b-heptapeptides 4a,
5a, 7a, 8a, and 9a. Full deprotection (Bn and N-Boc cleavage) of b-peptides 4a 9a,
5
)
)
For a general discussion of OHsubstitution in b- and g-peptides, see [4].
6
We know already that a b-peptide consisting of unlike-3-amino-2-hydroxy carboxylic acid moieties (cf. 11)
folds to a novel 28 helix (probably due to H-bond formation of OH with the neighboring CˆO O-atom
[5]), rather than adopting a sheet structure [6].
For investigations of these properties of −normal× b-peptides, see [7], and refs. cit. therein.
Boc-b3hVal-b3hAla-b3hLeu-OBn has been used by us many times before [1].
7
8
)
)

1864
Helvetica Chimica Acta Vol. 86 (2003)
Fig. 1. Model of the (M)-314 helix of a b-peptide (a) and of the b-peptidic parallel sheet structure (b) and A values
for preference of equatorial vs. axial position on cyclohexane (c)
followed by reversed-phase HPLC purification, provided the N- and C-terminally
unprotected b-peptides 4b 9b.

Helvetica Chimica Acta Vol. 86 (2003)
1865
Scheme. Preparation of the N-Boc-Protected 3-Amino-2-fluoro, 3-Amino-2,2-difluoro, and 3-Amino-2-hydroxy
Carboxylic Acids 1 3 from the Proteogenic Amino Acids Alanine, Valine, and Leucine. The OH/F substitution
with DAST is carried out with the N,N-dibenzyl-protected, diastereomerically pure methyl 3-amino-2-hydroxy-
alkanoates and takes place with retention of configuration (as established by X-ray analysis of one of the
products and spectroscopic assigment by analogy). Boc ˆ t-(Butoxy)carbonyl, DASTˆ Et2NSF3.
a) HCl (gas)/MeOH. b) Chromatographic separation of epimers (SiO2; hexane/AcOEt). c) H2/Pd-C, MeOH.
d) Boc2O, Et3N, MeOH, 208. e) LiOH¥ H 2O, EtOH. f) DAST, CH2Cl2, 08. g) Swern oxidation: DMSO, SOCl2,
CH2Cl2, À 788.
All hexa- and heptapeptides, of which the formulae are shown herein, have been
isolated in high purity (> 97%) and characterized by CD, 13C- and 1H-NMR
spectroscopy, and by MALDI mass spectrometry. The starting materials (amino acid
derivatives) and the intermediate di-, tri-, and tetrapeptides have been fully
characterized by NMR, IR, MS, [a]D, m.p., and elemental analysis9).
Some CD spectra10) in 0.2 mm MeOHof the unprotected b3-heptapeptides, with
central F- and OH-substituted amino acid moieties, are shown in Fig. 3. They are most
puzzling, to say the least, because all show the 314-helix-typical pattern with a negative
Cotton effect between 215 and 220 nm, zero crossing between 210 and 205 nm, and a
positive Cotton effect near 200 nm no matter whether the F or OHgroup would
occupy a lateral (4b and 7b) or axial position (5b, 8b, and 9b) on such a 314 helix. As
shown in Fig. 4, the pattern of the CD spectra of the u-mono-hydroxy and of the u-
9
)
)
Full details will be presented in forthcoming full papers.
All CD spectra shown herein are non-normalized.
10

1866
Helvetica Chimica Acta Vol. 86 (2003)
Fig. 2. General structures of b2,3- and b2,2,3-peptides A C [1] and of the heterosubstituted b-peptides D
G
described herein
Fig. 3. Comparison of the CD spectra (0.2 mm in MeOH) of unprotected b-heptapeptides carrying a central a-
heterosubstituted b-amino acid. All the new b-heptapeptides exhibit the CD pattern associated to an (M)-314-
helical structure [1].
mono-fluoro derivatives 5b and 8b does not change with the solvent (MeOH,
CF3CH2OH, (CF3)2CHOH), except when we switch to aqueous solution, which causes
a dramatic effect. Also, the protected and the unprotected11) b-peptides of this type
may exhibit totally different CD spectra as exemplified by the mirror-image-type
11
)
Cf. the stabilization of the 314 helix upon terminal deprotection [1b][13].

Helvetica Chimica Acta Vol. 86 (2003)
1867

1868
Helvetica Chimica Acta Vol. 86 (2003)
Fig. 4. Influence of the solvent on CD spectra (0.2 mm in MeOH, CF3CH2OH(TFE), (CF 3)2CHOH
i
(HFPrOH), and H2O) of the u-hydroxy-b-heptapeptide 5b (a) and of the u-fluoro-b-heptapeptide 8b (b).

Helvetica Chimica Acta Vol. 86 (2003)
1869
Fig. 5. CD Spectra of b-peptides 4 and 6 containing l-2-hydroxy-3-amino acid moieties. a) CD Spectra in MeOH
of the terminally protected and unprotected l-hydroxy-b-heptapeptides, 4a and 4b, respectively. b) Overlay of
the CD spectra of the protected hexapeptide 6a (in different solvents) and of its unprotected form 6b (in
MeOH).
curves obtained for the l-mono-hydroxy derivatives 4a and 4b in MeOH( Fig. 5,a). The
all-l-trihydroxy-hexapeptide, on the other hand, shows CD spectra of the same general
pattern, whether it is protected or unprotected, or whether MeOHor the fluorinated
solvents are used (Fig. 5,b).
Thus, on the basis of the CD spectra, it is impossible to arrive at any safe structural
assigment. If we would use analogy with the typical CD spectra of b-peptides, which,
according to NMR analysis, form a 314 helix [1][13], a 12/10 helix [1][14], or a hairpin
turn [6], we would, for instance, have to assign (M)-314-helical structure to 4b, 5b, 7b,
8b, and 9b, (P)-314-helical structure to 4a in MeOHand to 6a in (CF3)2CHOH, and
hairpin or 12/10-helical structure to 5b in H2O! Thus, CD spectroscopy of b-peptides,
once again, turns out to be useless for structure determination [15]. At this point, it can
provide only a preliminary hint as to which compounds might be most interesting to
investigate by NMR spectroscopy first! We have performed or are in the process of
performing extensive, state-of-the-art, high-field NMR measurements of most of the
novel b-peptides mentioned herein. The interpretation of these measurements will
require time12), and the results will be published in due course. There will be, no doubt,
many surprises.
REFERENCES
[1] a) D. Seebach, M. Overhand, F. N. M. K¸hnle, B. Martinoni, L. Oberer, U. Hommel, H. Widmer, Helv.
Chim. Acta 1996, 79, 913; b) D. Seebach, P. E. Ciceri, M. Overhand, B. Jaun, D. Rigo, L. Oberer, U.
Hommel, R. Amstutz, H. Widmer, Helv. Chim. Acta 1996, 79, 2043; c) D. Seebach, S. Abele, K. Gademann,
It usually takes ca. 3.5 days to perform the measurements (1D Presat, DQFCOSY, TOCSY, HSQC 13C/1H,
HMBC 13C/1H, ROESY with different mixing times in ms), and it may take several weeks or even months
for the evaluation of the data.
12
)

1870
Helvetica Chimica Acta Vol. 86 (2003)
G. Guichard, T. Hintermann, B. Jaun, J. L. Matthews, J. V. Schreiber, L. Oberer, U. Hommel and H.
Widmer, Helv. Chim. Acta 1998, 81, 932; d) D. Seebach, T. Sifferlen, P. A. Mathieu, A. M. H‰ne, C. M.
Krell, D. J. Bierbaum, S. Abele, Helv. Chim. Acta 2000, 83, 2849; e) D. Seebach, T. Sifferlen, D. J.
Bierbaum, M. Rueping, B. Jaun, B. Schweizer, J. Schaefer, A. K. Mehta, R. D. O×Connor, B. H. Meier, M.
Ernst, A. Gl‰ttli, Helv. Chim. Acta 2002, 85, 2877; and refs. cit. therein.
[2] R. Baron, D. Bakowies, W. F. van Gunsteren, X. Daura, Helv. Chim. Acta 2002, 85, 3872; and refs. cit.
therein.
[3] J. D. Dunitz, R. Taylor, Chem. Eur. J. 1997, 3, 89.
[4] M. Brenner, D. Seebach, Helv. Chim. Acta 2001, 84, 1181.
[5] A. H‰ne, K. Gademann, B. Jaun, D. Seebach, Angew. Chem. 2003, 115, 1573; Angew. Chem., Int. Ed. 2003,
42, 1534.
[6] D. Seebach, S. Abele, K. Gademann, B. Jaun, Angew. Chem. 1999, 111, 1700; Angew. Chem., Int. Ed. 1999,
38, 1595.
[7] a) J. Frackenpohl, P. I. Arvidsson, J. V. Schreiber, D. Seebach, ChemBioChem 2001, 2, 445; b) P. I.
Arvidsson, J. Frackenpohl, N. S. Ryder, B. Liechty, F. Petersen, H. Zimmermann, G. P. Camenisch, R.
Woessner, D. Seebach, ChemBioChem 2001, 2, 771; c) J. V. Schreiber, J. Frackenpohl, F. Moser, T.
Fleischmann, H.-P. Kohler, D. Seebach, ChemBioChem 2002, 3, 424; d) H. Wiegand, B. Wirz, A.
Schweitzer, G. P. Camenisch, M. I. Rodriguez Perez, G. Gross, R. Woessner, R. Voges, P. I. Arvidsson, J.
Frackenpohl, D. Seebach, Biopharm. Drug Dispos. 2002, 23, 251.
¬
¬
[8] a) J. M. Andres, R. Barrio, M. A. Martinez, R. Pedrosa, A. Perez-Encabo, J. Org. Chem. 1996, 61, 4210;
b) M. T. Reetz, M. W. Drewes, R. Schwickardi, Org. Synth. 2000, 76, 110.
[9] M. A. Schwindt, D. T. Belmont, M. Carlson, L. C. Franklin, V. S. Hendrikson, G. L. Karrick, R. W. Poe,
D. M. Sobieray, J. Van De Vusse, J. Org. Chem. 1996, 61, 9564.
[10] R. P. Singh, J. M. Shreeve, Synthesis 2002, 17, 2561.
[11] a) L. Somekh, A. Shanzer, J. Am. Chem. Soc. 1982, 104, 5836; b) D. Gani, P. B. Hitchcock, D. W. Young, J.
Chem, Soc. Perkin Trans. 1 1985, 1363.
[12] S.-S. Wang, B. F. Gisin, D. P. Winter, R. Makofske, I. D. Kulesha, C. Tzougraki, J. Meienhofer, J. Org. Chem.
1977, 42, 1286.
[13] D. Seebach, J. V. Schreiber, S. Abele, X. Daura, W. F. van Gunsteren, Helv. Chim. Acta 2000, 83, 34.
[14] D. Seebach, K. Gademann, J. V. Schreiber, J. L. Matthews, T. Hintermann, B. Jaun, L. Oberer, U. Hommel,
H. Widmer, Helv. Chim. Acta 1997, 80, 2033.
[15] A. Gl‰ttli, X. Daura, D. Seebach, W. F. van Gunsteren, J. Am. Chem. Soc. 2002, 124, 12972.
Received April 11, 2003