Preparation of a reversible redox-controlled cage-type molecule linked by disulfide bonds

Article abstract of DOI:10.1002/ejoc.200900044

A unique trithiol macromolecule, with an intrinsic conformational propensity for dimerization (cage) instead of oligomerization upon oxidation, was prepared straightforwardly through rational design. The quantitative conversion and the reversibility betwe

Full text of DOI:10.1002/ejoc.200900044

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200900044
Preparation of a Reversible Redox-Controlled Cage-Type Molecule Linked by
Disulfide Bonds
Yih-Chern Horng,*[a] Tzung-Lung Lin,[a] Cheng-Yi Tu,[a] Ting-Ju Sung,[a]
Chang-Chih Hsieh,[a] Ching-Han Hu,[a] Hon Man Lee,[a] and Ting-Shen Kuo[b]
Keywords: Dimerization / Self-assembly / Sulfur / Preorganization
A unique trithiol macromolecule, with an intrinsic conforma-
tional propensity for dimerization (cage) instead of oligomer-
ization upon oxidation, was prepared straightforwardly
through rational design. The quantitative conversion and the
X-ray structure of the synthesized cage-type molecule repre-
sents the first successful example of a redox-controlled re-
versible dimeric capsule linked through covalent disulfide
bonds.
reversibility between the cage and trithiols through redox re- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
actions were assayed by 1H NMR spectroscopic analysis. The
Germany, 2009)
addition, preorganization of the thiol groups at the right
geometry and orientation in the monomer is required for
the favorable kinetic formation of capsules. Thus, it is ex-
tremely difficult to obtain a thermodynamically and kinet-
ically stable synthetic capsule with tris(disulfide) bonds. In
addition, there is no structural information on disulfide-
bond-based synthetic capsules in the literature.
Herein we report the design, synthesis, and structural de-
termination of a novel reversible dimeric macromolecular
capsule linked by tri(disulfide) bonds.
Introduction
Synthetic macromolecular capsules have attracted much
attention because of their potential applications in bio-
logical systems, for example, in molecular recognition,[1] in
drug delivery,[2] and as miniature reactors.[3] Three main
methods are employed in the literature for the formation of
synthetic capsules: (i) Synthesis through permanent cova-
lent bonds under kinetic control.[4] (ii) Formation through
noncovalent interactions or metal–ligand coordination un-
der thermodynamic control.[5] (iii) Production through re-
versible covalent bonds under thermodynamic control.[6]
Although many molecular capsules have been made by ap-
plying the first and second methods, only a few examples
exist of synthetic capsules that have been formed by using
the third method.[7,8] Because the reversible thiol–disulfide
interchange reaction is an important factor for controlling
the protein folding process, and the strong covalent S–S
bonds are thought to be the key for stabilizing the struc-
tures of small proteins,[9] the utilization of dynamically re-
versible covalent disulfide bonds holds great potential for
synthesizing macromolecular capsules.
Results and Discussion
Inspired by the synthetic trithiol compounds 1[12] and
2[13] utilized to form a 3:1 subsite differentiated [4Fe–4S]
cluster for modeling the active site of aconitase (Scheme 1),
we planned to make a similar trithiol compound (i.e., 3) by
a more convenient method and convert it into a cage-like
macromolecule through oxidation under kinetic or thermo-
dynamic control. The fundamental topological features of
both 1 and 2 are (i) based on a hexasubstituted benzene
ring, (ii) a staggered conformation, which allows the func-
tional groups carrying the thiol donors to orient on the
same side of the aromatic base, and (iii) the thiol donors
turned inward toward a common area due to the steric de-
mands of the methyl substitutes of 4,6-dimethylphenyl
group and the benzo ring of the indolyl moiety. All three
features promised the formation of a dimeric molecular
capsule linked through three disulfide bonds. Most import-
antly, the third feature of the thiol donors turned inward
would prevent oligomerization due to steric hindrance.
However, the synthesis of 1 requires tremendous effort and
the preparation of both 1 and 2 uses mercury acetate, which
is potentially hazardous to human health and the environ-
However, attempts to synthesize macromolecular cap-
sules, especially dimeric tris(disulfide)s, have often resulted
in oligomeric products to some extent.[8,10] The restricted
CSSC dihedral angle of about 90°[11] thermodynamically li-
mits the formation of an S–S-bond-linked capsule. In
[a] Department of Chemistry, National Changhua University of
Education,
1 Jin-De Road, Changhua 50058, Taiwan
Fax: +886-4-721-1190
E-mail: ychorng@cc.ncue.edu.tw
[b] Instrumentation Center, Department of Chemistry, National
Taiwan Normal University,
Taipei 11677, Taiwan
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2009, 1511–1514
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1511

Y.-C. Horng et al.
SHORT COMMUNICATION
Scheme 1. Trithiol molecules.
Scheme 2. Syntheses of trithiol 3, its methyl derivative 4, and macromolecular capsule 7. Reagents and conditions: (a) K2CO3, CH3CN,
72 h, 85 °C; (b) NaBH4, DMF, 10 min, 4 °C; (c) O2 or I2, CH2Cl2, r.t.; (d) NaOMe, MeI, CH3CN, 12 h, r.t.
ment, for the deprotection of the trithiol. Thus, an alterna-
tive approach toward the creation of a new trithiol with
features (i) to (iii), described above, was conducted
(Scheme 2).
Before performing the experiments, the most stable con-
formation of 3 was evaluated by density functional theory
computations. It is well known that the three substituents
in 2,4,6-tris(substituent)-1,3,5-triethylbenzene tend to ori-
ent on the opposite side of the benzene plane relative to the
ethyl groups.[14] The optimized conformation with all three
thiol donors turned inward is more stable than the outward
Figure 1. The optimized structure of 3 with all three thiols turned
inward (3a, left) and outward (3b, right). Hydrogen atoms bound
to carbon atoms are omitted for clarity.
one by 20.4 kcalmol–1, according to density functional as a precipitate. This product was then further reduced with
theory (B3LYP/6-31G) computation (Figure 1). Thus, the NaBH4 in a one-pot reaction to give designed trithiol mole-
computational results indicated a favorable preorganization cule 3. The identification of 3 was established by 1H NMR,
of 3 for forming a dimeric capsule.
Protection reagents, required for syntheses involving (EA).
13C NMR, and IR spectroscopy and elemental analysis
highly reactive thiolates, were unnecessary for the very
Although we were not able to obtain crystals of 3, the
straightforward synthesis of 3. When an acetonitrile solu- molecular structure of its methyl derivative (i.e., 4) was de-
tion containing commercially available 2,4,6-tris(bromo- termined by X-ray analysis (Figure 2), which confirmed the
methyl)-1,3,5-triethylbenzene (5) and 2,2Ј-aminophenyl di- right geometries and orientations of the ethyl groups and S
sulfide (6) was heated at reflux for 72 h in the presence of atoms, and that it was comparable to the structure of 3a
an excess amount of K2CO3, a major product was formed simulated by computation.
1512
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 1511–1514

Reversible Redox-Controlled Cage-Type Molecule Linked by Disulfide Bonds
ume of the cavity for encapsulation can be calculated to be
87.1 Å3 (diameter: 4.46 Å; height: 5.58 Å). With the volume
of the cavity and a gate size of 2.4 Å,[19] capsule 7 can easily
accommodate molecules such as dichloromethane (55 Å3)
and chloroform (72 Å3).[20]
The quantitative conversion and the reversibility between
the cage and trithiols are essential for its biological applica-
tions. Under equilibrium conditions, the conversion of 3
into 7 in C6D6 was followed in the presence of air by using
1H NMR spectroscopy (Figure 4). During the entire con-
version process, no precipitates were observed, which en-
sures that no other products were formed except the ones
observed from the NMR measurements. The complete oxi-
dation of 3 only afforded 7; no oligomeric products were
formed, and one or two intermediates appeared during the
conversion process. Our DFT computation revealed that
the reaction energy needed for two moles of 3 to form cap-
sule 7 plus three moles of H2 was 11.6 kcalmol–1, which is
much lower than what is needed for 2,4,6-tris(mercapto-
methyl)-1,3,5-triethylbenzene (8) to form a dimeric mole-
cule (36.13 kcalmol–1). Entropic effects will further favor
capsule formation. Experimentally, the oxidation of 8 under
kinetic conditions (I2, CHCl3) only yielded the oligomeric
product as a result of the highly strained CSSC dihedral
angle on dimer formation.[8a]
Figure 2. The molecular structure (left) and space-filled model
(right) of 4. The thermal ellipsoids are drawn at a 35% probability
level. Nitrogen and sulfur atoms are represented as shaded ellip-
soids. All of the hydrogen atoms are omitted for clarity.
Next, attempts were made to synthesize tris(disulfide)
molecular capsules by the oxidation of 3 under thermo-
dynamic control (O2) or kinetic control (I2, not under high-
dilution conditions) through self-assembly. Both attempts
gave product 7 with a high yield (about 90%). Surprisingly,
this product was identical to the one from the reaction be-
tween 5 and 6. Thus, the formation of 7 is thermodynami-
cally and kinetically favored over other larger oligomers.
1
The identity of 7 was confirmed by EA and H and 13C
NMR spectroscopy, whereas the highly symmetric struc-
1
ture, as suggested by the H and 13C NMR spectra, was
further supported by X-ray structure determination of 7,
whose crystals were obtained from the recrystallization in
CH2Cl2 solution (Figure 3).
Figure 3. The molecular structure of capsule 7; side view (left) and
top view (right). The thermal ellipsoids are drawn at a 35% prob-
ability level. Nitrogen and sulfur atoms are represented as shaded
ellipsoids. Severely disordered solvent molecules in the void spaces
inside and outside of 7 were not identified and refined. All of the
hydrogen atoms are omitted for clarity. Key bond parameters for
7: S–S 2.065(2) Å, C–S–S–C torsion angle 96.8(2)°.
Figure 4. 1H NMR spectra (400 MHz, 298 K, C6D6) of (a) purified
3, (b) oxidation of 3 in air for 2 weeks at room temperature, (c)
further oxidation of 3 from (b) in air for 1 d at 70 °C, and (d)
complete oxidation of 3.
The molecular structure of dimeric capsule 7 shows an
¯
R3c symmetry, which is rare in terms of the cage molecules
Interestingly, when the conversion was conducted in
reported to date.[15] All six phenyl rings orient perpendicu- DMF solution aerobically at room temperature, crystals of
lar to the benzene bases to form a staggered conformation DMF-encapsulated molecule 7·DMF were found in the pre-
(top view), with the ethyl groups pointing outward, and all cipitate. The identification of 7·DMF was established by 1H
six amino protons oriented toward the center. The CSSC and 13C NMR spectroscopy, EA, and X-ray structure deter-
dihedral angle and S–S distance of 7 are very similar to mination (Figure 5). To the best of our knowledge, this is
those of starting reagent 6 [89.5° and 2.061(4) Å, respec- the first time that the X-ray structure of a reversible molec-
tively],[16] and bis[2,6-bis(pivaloylamino)phenyl] disulfide ular capsule encapsulating a guest molecule using disulfide
[98.9° and 2.079(4) Å, respectively].[17] The cavity can be linkages has been determined. Although there is no obvious
viewed as a cylinder in the core of 7, with a diameter of hydrogen bonding between the guest and 7, a weak lone
8.06 Å[18] and a height of 8.98 Å. If we account for the pair (carbonyl)···π interaction between the DMF molecule
van der Waals radius of a sulfur atom (1.80 Å) and the and the aromatic base of 7 was observed in the crystal
thickness of the arene ring (1.70 Å), the van der Waals vol- structure.[21]
Eur. J. Org. Chem. 2009, 1511–1514
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1513

Y.-C. Horng et al.
SHORT COMMUNICATION
Supporting Information (see footnote on the first page of this arti-
cle): Details of the syntheses of 3, 4, 7, and 7·DMF; conversion
studies between 3 and 7; crystallographic data for 4, 7, and 7·DMF;
computational studies of 3 and 7.
Acknowledgments
We are grateful to the National Science Council (Taiwan) for their
financial support of this work.
[1] S. M. Biros, J. Rebek Jr, Chem. Soc. Rev. 2007, 36, 93–104.
[2] K. R. West, S. Otto, Curr. Drug Discov. Technol. 2005, 2, 123–
160.
[3] a) T. S. Koblenz, J. Wassenaar, J. N. Reek, Chem. Soc. Rev.
2008, 37, 247–262; b) L. R. MacGillivray, G. S. Papaefstathiou,
T. Friscic, T. D. Hamilton, D. K. Bucar, Q. Chu, D. B. Varsh-
ney, I. G. Georgiev, Acc. Chem. Res. 2008, 41, 280–291.
[4] a) A. Jasat, J. C. Sherman, Chem. Rev. 1999, 99, 931–968; b) J.
Sherman, Chem. Commun. 2003, 1617–1623.
[5] a) J. Rebek Jr, Angew. Chem. Int. Ed. 2005, 44, 2068–2078; b)
W. Y. Sun, M. Yoshizawa, T. Kusukawa, M. Fujita, Curr. Opin.
Chem. Biol. 2002, 6, 757–764; c) S. Liu, B. C. Gibb, Chem.
Commun. 2008, 3709–3716.
Figure 5. The molecular structure (left) and space-filled model
(right) of 7·DMF. The thermal ellipsoids are drawn at a 35% prob-
ability level. The DMF molecule is disordered with D3h symmetry
and only one disorder component is shown. The atoms of the en-
capsulated DMF molecule are represented as shaded ellipsoids. All
of the hydrogen atoms are omitted for clarity. Key bond parameters
for 7·DMF: S–S 2.049(3) Å, C–S–S–C torsion angle 96.4(2)°,
O(1)···centroid (aromatic base) 3.26 Å, aromatic plan-centroid–
O(1) axis 76.7°.
[6] S. J. Rowan, S. J. Cantrill, G. R. Cousins, J. K. Sanders, J. F.
Stoddart, Angew. Chem. Int. Ed. 2002, 41, 898–952.
[7] a) S. Ro, S. J. Rowan, A. R. Pease, D. J. Cram, J. F. Stoddart,
Org. Lett. 2000, 2, 2411–2414; b) C. Naumann, S. Place, J. C.
Sherman, J. Am. Chem. Soc. 2002, 124, 16–17; c) A. L. Kieran,
S. I. Pascu, T. Jarrosson, J. K. M. Sanders, Chem. Commun.
2005, 1276–1278.
[8] a) S. W. Tam-Chang, J. S. Stehouwer, J. Hao, J. Org. Chem.
1999, 64, 334–335; b) K. R. West, K. D. Bake, S. Otto, Org.
Lett. 2005, 7, 2615–2618.
[9] J. D. Hartgerink, Curr. Opin. Chem. Biol. 2004, 8, 604–609.
[10] J. Houk, G. M. Whitesides, J. Am. Chem. Soc. 1987, 109, 6825–
6836.
[11] D. Jiao, M. Barfield, J. E. Combariza, V. J. Hruby, J. Am.
Chem. Soc. 1992, 114, 3639–3643.
[12] T. D. P. Stack, R. H. Holm, J. Am. Chem. Soc. 1988, 110, 2484–
2494.
[13] C. Walsdorff, W. Saak, S. Pohl, J. Chem. Soc., Dalton Trans.
1997, 1857–1862.
[14] G. Hennrich, E. V. Anslyn, Chem. Eur. J. 2002, 8, 2218–2224.
[15] Z. Xie, C. W. Tsang, F. Xue, T. C. W. Mak, Inorg. Chem. 1997,
36, 2246–2247.
[16] J. D. Lee, M. W. R. Bryant, Acta Crystallogr., Sect. B: Struct.
Crystallogr. Cryst. Chem. 1970, 26, 1729–1735.
[17] N. Ueyama, T.-a. Okamura, Y. Yamada, A. Nakamura, J. Org.
Chem. 1995, 60, 4893–4899.
Conclusions
We designed an extremely straightforward method to
synthesize in remarkably high yield a novel dimeric capsule
that can be assembled by tris(disulfide) bonds and deter-
mined its structure. We also demonstrated that the capsule
is easily opened up through the reduction of disulfide
bonds. The encapsulation of a neutral guest molecule in 7
during the oxidation process of 3 was observed. The infor-
mation contained in the molecular structure of 7 improves
our understanding of the rational design for a dimeric mac-
romolecular capsule containing S–S bonds and the revers-
ible opening and closing of 7 through its dynamic covalent
bonds. Furthermore, the results described here help to ad-
dress the problems of applications of disulfide-linked mo-
lecular capsules in biological and medicinal systems. The
next challenge will be to create a reversible water-soluble
molecular capsule that employs compact and rigid building
blocks similar to 7 for greater biological relevance.
[18] The diameter can be estimated as that of a sphere in contact
with all the midpoints of the three S–S bonds.
[19] The distance between two nearby S atoms (6.0 Å), not bound
together, minus the diameter of an S atom (3.6 Å).
[20] a) A. Bondi, J. Phys. Chem. 1964, 68, 441–451; b) Y. H. Zhao,
M. H. Abraham, A. M. Zissimos, J. Org. Chem. 2003, 68,
7368–7373.
Experimental Section
The experimental details can be found in the Supporting Infor-
mation.
CCDC-704226 (for 4), -704227 (for 7), and -707649 (for 7·DMF)
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[21] M. Egli, S. Sarkhel, Acc. Chem. Res. 2007, 40, 197–205.
Received: January 16, 2009
Published Online: February 17, 2009
1514
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 1511–1514