Rotaxanes of cyclic peptides

Article abstract of DOI:10.1021/ja057206q

The synthesis of rotaxanes derived from the synthetic peptide macrocycles cyclo(l-ProGly)₄ and cyclo(l-ProGly)₅ and diammonium threads is described. [2]Rotaxanes are formed in good yields (56-63%), despite the disruption of internal

Full text of DOI:10.1021/ja057206q

Published on Web 01/19/2006
Rotaxanes of Cyclic Peptides
Vincent Aucagne, David A. Leigh,* Julia S. Lock, and Andrew R. Thomson
UniVersity of Edinburgh, School of Chemistry, The King’s Buildings, West Mains Road, Edinburgh EH9 3JJ, UK
Received October 22, 2005; E-mail: David.Leigh@ed.ac.uk.
a
Scheme 1. Synthesis of Cyclic Peptide [2]Rotaxanes 3-6(SbF6)2
Until recently the existence of mechanically interlocked peptide
backbones in nature was far from clear-cut. However, in the past
few years, first knots,1 then catenanes2 and, finally, rotaxanes3 have
been unambiguously characterized in naturally occurring peptides
and proteins.4 Although the implications of kinetically stable entan-
glements on peptide and protein structure, properties, function, and
folding remain almost completely unknown, already studies on the
few interlocked peptides available have shown them to possess a
wealth of intriguing properties, including high resistance to pepti-
dases,5a unique modes of antimicrobial and antiviral action,5b-d
impressive membrane transport characteristics,5e and stability to
thermal and chemical denaturing.5f However, few artificial or simp-
ler analogues have been investigated to date since no methodology
yet exists for synthetically interlocking peptide fragments.6,7 En
route to this goal, here we describe the preparation of rotaxanes from
a class of well-known cyclic peptides and some nonpeptidic threads.
Efficient synthetic routes to catenanes and rotaxanes generally
rely on strong-binding mutual recognition elements on each
component to direct a threading or macrocyclization reaction.8 The
problem with applying such a strategy to peptides is that the amide
groups of each free component will normally be largely self-satisfied
in terms of hydrogen bonding through folding, and little driving
force will exist for interlocking. Indeed, several attempts3b,c to
synthesize microcin J25, a 21-residue oligopeptide with a rotaxane
architecture, by conventional peptide synthesis strategies have failed.
However, many natural and unnatural cyclic peptides are known
to bind cations efficiently. Crown ether9 and cucurbituril10 organic
cation binding provides some of the most effective rotaxane
template syntheses known, and we wondered whether a similar
interaction could also be used to promote rotaxane formation with
peptidic macrocycles. As prototypical systems we investigated
cyclic octa- (1) and deca (2)-peptides derived from the L-ProGly
repeat unit.11 High stability constants (Ka ≈ 103-105 M-1) have
been reported for 1:1 metal and 1:1 and 1:2 protonated amino acid
ester complexes of these cyclopeptides in acetonitrile.11c Accord-
ingly, we constructed a series of cationic diol threads and examined
their efficacy in rotaxane-forming “stoppering” reactions with a
bulky acid chloride and macrocycles 1 and 2 (Scheme 1).12
Cyclo(L-ProGly)4 (1) and cyclo(L-ProGly)5 (2) were prepared
via a solid-phase backbone amide-linking approach (see Supporting
Information (SI)). As they are water soluble, a simple aqueous
extraction removes either macrocycle from a rotaxane-forming
reaction mixture, and they can easily be recycled. In contrast, the
chromatographic separation of the [2]rotaxanes from the corre-
sponding threads is rather difficult, and it therefore proved
convenient to use an excess of the cyclic peptide to maximize the
ratio of rotaxane to thread. Although only traces of interlocked
products were formed with monocationic ammonium or pyridinium
threads, we were delighted to find that [2]rotaxanes 3-6(SbF6)2
were formed in 56-63% yield from ethane-1,2-diammonium and
butane-1,4-diammonium templates.12
a Reagents and conditions: i. (4-ClC6H4)3CCH2COCl, CHCl3:CH3CN
3:7, RT, 7 days. ii. sat. NaHCO3, then 1 M HCl. For clarity, atom labels
are only shown for one repeat unit of the cyclopeptide in [2]rotaxanes
3-6(SbF6)2
and 8Cl2 and macrocycle 1) provide some insights into the structure
of the rotaxanes and the nature of the template interaction. Free
cyclo(L-ProGly)4 exists as a mixture of rotamers in acetonitrile,11c
∼30% in the all-trans conformer form (shown in light blue in Figure
1c) in which four γ-turn Gly-to-Gly H-bonds give rise to a relatively
planar structure with the glycine carbonyls pointing toward the
center and the proline carbonyls directed away. However, a single
set of signals is observed for each constitutionally distinct proton
of the cyclic peptide in both 3Cl2 (Figure 1b) and 5Cl2 (Figure
1d). Cyclo(L-ProGly)4 is known11c to adopt a geometry in 1:2 host-
guest ammonium cation complexes which is also an all-trans
rotamer, but is cylindrical rather than flat, with the glycine carbonyls
rotated toward one face and the proline carbonyls to the other,
destroying the internal H-bond network. The H and 13C (see ref
1
11c) NMR are consistent with a similar structure existing in
rotaxanes 3Cl2 and 5Cl2. Rapid spinning of the thread in the cavity
(the macrocycle resonances remain symmetrical even at 240 K in
CD3CN) enables each of the peptide carbonyls to interact with the
two ammonium groups through an alternating network of hydrogen
bonds and electrostatic CdOδ-‚‚‚N+ interactions.
Confirmation of the change in conformation of cyclo(L-ProGly)4
upon rotaxane formation is seen in the changes in the resonances
of the glycine methylene groups. The geminal GlyHR protons in
1
The H NMR spectra (Figure 1) of the cyclo(L-ProGly)4-based
[2]rotaxanes 3Cl2 and 5Cl2 and their free components (threads 7Cl2
9
1784
J. AM. CHEM. SOC. 2006, 128, 1784-1785
10.1021/ja057206q CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
H-bonding with the longer template occurs to each ammonium
group from both directions (obviously not simultaneously). In other
words, the cyclopeptide is loosely held in 5Cl2 and able to access
the full length of the thread.
The ability of cation-amide interactions to disrupt internal
amide-amide H-bonding networks augurs well for their use to over-
come peptide folding and provide a thermodynamic driving force
for the formation of kinetically stable peptide and protein entangle-
ments.
Supporting Information Available: Experimental details for the
synthesis of the macrocycles, rotaxanes, threads, and precursors.
References
(1) (a) Mao, B. J. Am. Chem. Soc. 1989, 111, 6132-6136. (b) Liang, C.;
Mislow, K. J. Am. Chem. Soc. 1994, 116, 11189-11190. (c) Liang, C.;
Mislow, K. J. Am. Chem. Soc. 1995, 117, 4201-4213. (d) Taylor, W. R.
Nature 2000, 406, 916-919. (e) Rosengren, K. J.; Daly, N. L.; Plan, M.
R.; Waine, C.; Craik, D. J. J. Biol. Chem. 2003, 278, 8606-8616. (f)
Taylor, W. R.; Lin, K. Nature 2003, 421, 25.
(2) (a) Duda, R. L. Cell 1998, 94, 55-60. (b) Wikoff, W. R.; Liljas, L.;
Tsuruta, H.; Hendrix, R. W.; Johnson, J. E. Science 2000, 289, 2129-
2133.
(3) (a) Brayro, M. J.; Mukhopadhyay, J.; Swapna, G. V. T.; Huang, J. Y.;
Ma L.-C.; Sineva, E.; Dawson, P. E.; Montelione, G. T.; Ebright, R. H.
J. Am. Chem. Soc. 2003, 125, 12382-12383. (b) Rosengren, K. J.; Clark,
R. J.; Daly, N. L.; Goransson, U.; Jones, A.; Craik, D. J. J. Am. Chem.
Soc. 2003, 125, 12464-12474. (c) Wilson, K.-A.; Kalkum, M.; Ottesen,
J.; Yuzenkova, J.; Chait, B. T.; Landick, R.; Muir, T.; Severinov, K.;
Darst, S. A. J. Am. Chem. Soc. 2003, 125, 12475-12483.
(4) One reason for these architectures remaining unknown for so long is that
they are often difficult to spot in heavily folded protein structures.1d,f Also
Edman degradation does not reveal interlocked character; thus, such
compounds can easily be misassigned.
(5) (a) Blond, A.; Pe´duzzi, J.; Goulard, C.; Chiuchiolo, M. J.; Barthe´le´my,
M.; Prigent, Y.; Salomo´n, R. A.; Far´ıas, R. N.; Moreno, F.; Rebuffat, S.
Eur. J. Biochem. 1999, 259, 747-756. (b) Delgado, M. A.; Rintoul, M.
R.; Far´ıas, R. N.; Salomo´n, R. A.; J. Bacteriol. 2001, 183, 4543-4550.
(c) Yuzenkova, J.; Delgado, M.; Nechaev, S.; Savalia, D.; Epshtein, V.;
Artsimovitch, I.; Mooney, R. A.; Landick, R.; Farias, R. N.; Salomon,
R.; Severinov, K. J. Biol. Chem. 2002, 277, 50867-50875. (d) Mukho-
padhyay, J.; Sineva, E.; Knight, J.; Levy, R. M.; Elbright, R. H. Mol.
Cell 2004, 14, 739-751. (e) Bellomio, A.; Vincent, P. A.; deArcuri, B.
F.; Salomo´n, R. A.; Morero, R. D.; Far´ıas, R. N. Biochem. Biophys. Res.
Commun. 2004, 325, 1454-1458. (f) Blond, A.; Cheminant, M.;
Destoumieux-Garzon, D.; Segalas-Milazzo, I.; Pe´duzzi, J.; Goulard, C.;
Rebuffat, S. Eur. J. Biochem. 2002, 269, 6212-6222.
(6) For the synthesis of an unnatural protein catenane based on the entwined
dimeric domain of the tumor suppressor protein p53, see: Yan, L. Z.;
Dawson, P. E.; Angew. Chem., Int. Ed. 2001, 40, 3625-3627.
(7) For rotaxanes derived from short (nonfolding) peptide threads see: (a)
Leigh, D. A.; Murphy, A.; Smart, J. P.; Slawin, A. M. Z. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 728-732. (b) Wurpel, G. W. H.; Brouwer, A.
M.; van Stokkum, I. H. M.; Farran A.; Leigh, D. A. J. Am. Chem. Soc.
2001, 123, 11327-11328. (c) Asakawa, M.; Brancato, G.; Fanti, M.;
Leigh, D. A.; Shimizu, T.; Slawin, A. M. Z.; Wong, J. K. Y.; Zerbetto,
F.; Zhang, S. J. Am. Chem. Soc. 2002, 124, 2939-2950. (d) Brancato,
G.; Coutrot, F.; Leigh, D. A.; Murphy, A.; Wong, J. K. Y.; Zerbetto, F.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4967-4971. (e) Da Ros, T.; Guldi,
D. M.; Farran Morales, A.; Leigh, D. A.; Prato, M.; Turco, R. Org. Lett.
2003, 5, 689-691. (f) Hannam, J. S.; Kidd, T. J.; Leigh, D. A.; Wilson,
A. J. Org. Lett. 2003, 5, 1907-1910. Bottari, G.; Leigh, D. A.; Pe´rez, E.
M. J. Am. Chem. Soc. 2003, 125, 13360-13361. (g) Hannam, J. S.; Lacy,
S. M.; Leigh, D. A.; Saiz, C. G.; Slawin, A. M. Z.; Stitchell, S. G. Angew.
Chem., Int. Ed. 2004, 43, 3260-3264. (h) Leigh, D. A. et al. Angew.
Chem., Int. Ed. 2005, 44, 3062-3067.
Figure 1. 1H NMR spectra (400 MHz, CD3CN, 298 K) of (a) ethane-1,2-
diammonium thread 7Cl2; (b) [2]rotaxane 3Cl2; (c) cyclo(L-ProGly)4 (1);
(d) [2]rotaxane 5Cl2; (e) butane-1,2-diammonium thread 8Cl2. The labeling
corresponds to that shown in Scheme 1. The resonances of the all-trans
rotamer of cyclo(L-ProGly)4 are shown in light blue, and those from minor
rotamers, only present in part (c), in dark blue.
the H-bonded all-trans rotamer of free 1 are in similar proximities
to the shielding region of the adjacent proline carbonyl groups and
appear at similar chemical shifts (Figure 1c). However, in both 3Cl2
and 5Cl2 H-bonding of the glycine carbonyls to the ammonium
groups rotates the NHCO groups, causing the positions of the
geminal methylene protons relative to the adjacent proline carbonyls
to differ such that the four GlyHR1 proton resonances are shifted
downfield by 0.35 ppm, whereas the four GlyHR2 proton resonances
are shifted upfield by 0.30 ppm. Despite the loss of the internal
H-bonding network, the amide resonances still experience a slight
net upfield shift in the rotaxanes as a result of the inductive effect
from the strong interaction of glycine carbonyls with the ammonium
groups.
In the ethane-1,2-diammonium rotaxane, 3Cl2, it is clear that
one ammonium group (Hi′, Figure 1b) H-bonds principally to the
glycine carbonyls and one (Hi1 and Hi2, Figure 1b) to the proline
carbonyls. The greater shielding of the Hi′ protons (indicative of
stronger H-bonding) is, again, consistent with the previous studies12c
on cyclo(L-ProGly)4-ammonium ion host-guest complexes. The
fact that the -CH2N+- protons internal to the template (Hj) are
shielded in the rotaxane compared to the thread, while those external
to the template (Hh) are deshielded, indicates that each ammonium
group is largely H-bonded from just one direction, with the
macrocycle located over the central ethane group.
(8) Amabilino, D. B.; Stoddart, J. F. Chem. ReV. 1995, 95, 2725-2828.
(9) See, for example: (a) Kolchinski, A. G.; Busch, D. H.; Alcock, N. W. J.
Chem. Soc., Chem. Commun. 1995, 1289-1291. (b) Ashton, P. R.; Glink,
P. T.; Stoddart, J. F.; Tasker, P. A.; White, A. J. P.; Williams, D. J. Chem.
Eur. J. 1996, 2, 729-736. (c) Gong, C.; Gibson, H. W. Angew. Chem.,
Int. Ed. 1998, 37, 310-314. (d) Loeb, S. J.; Wisner, J. A. Chem. Commun.
1998, 2757-2758.
(10) (a) Mock, W. L.; Irra, T. A.; Wepsiec, J. P.; Adhia, M. J. Org. Chem.
1989, 54, 5302-5308. (b) Kim, K. Chem. Soc. ReV. 2002, 31, 96-107.
(11) (a) Deber, C. M.; Blout, E. R. Isr. J. Chem. 1974, 12, 15-29. (b) Deber,
C. M.; Blout, E. R. J. Am. Chem. Soc. 1974, 96, 7566-7568. (c) Madison,
V.; Deber, C. M.; Blout, E. R. J. Am. Chem. Soc. 1977, 99, 4788-4798.
(12) To maintain the ammonium template the “stoppering” reaction must be
carried out in the absence of base. HCl is generated during the reaction,
autocatalyzing ester formation. Since there is the potential for a mixture
of counterions in the initially formed rotaxane salts, anion exchange was
performed to ensure uniformity.
While the shifts in the -H2N+- signals in rotaxane 5Cl2 are
less informative, the internal and external -CH2N+- groups are
both split into shielded and deshielded resonances, indicating that
JA057206Q
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 1785