Processable cyclic peptide nanotubes with tunable interiors

Article abstract of DOI:10.1021/ja2063082

A facile route to generate cyclic peptide nanotubes with tunable interiors is presented. By incorporating 3-amino-2-methylbenzoic acid in the d,l-alternating primary sequence of a cyclic peptide, a functional group can be presented in the interior of the

Full text of DOI:10.1021/ja2063082

COMMUNICATION
pubs.acs.org/JACS
Processable Cyclic Peptide Nanotubes with Tunable Interiors
Rami Hourani,† Chen Zhang,† Rob van der Weegen,‡ Luis Ruiz,§ Changyi Li,† Sinan Keten,§
Brett A. Helms,*,‡ and Ting Xu*,†
†Department of Materials Science and Engineering, University of California, Berkeley, California 94720-1760, United States
‡The Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States
§Department of Civil & Environmental Engineering and Mechanical Engineering, Northwestern University, Evanston,
Illinois 60208-3111, United States
S Supporting Information
b
important to modify the exterior of CPNs for dispersion and
ABSTRACT: A facile route to generate cyclic peptide
nanotubes with tunable interiors is presented. By incorpor-
ating 3-amino-2-methylbenzoic acid in the D,L-alternating
primary sequence of a cyclic peptide, a functional group can
be presented in the interior of the nanotubes without
compromising the formation of high aspect ratio nanotubes.
The new design of such a cyclic peptide also enables one to
modulate the nanotube growth process to be compatible
with the polymer processing window without compromis-
ing the formation of high aspect ratio nanotubes, thus
opening a viable approach toward molecularly defined
porous membranes.
directed nanotube growth in polymeric matrices. Also, the
number and placement of noncannonical amino acids can
have undesirable effects on the conformation of the ring and its
potential for self-assembly. In many cases, the CP ring contorts
owing to ring strain or otherwise rotates the backbone amide
bonds. As a result, the efficacy of inter-ring hydrogen bonding
is reduced and, in some cases, the formation of high aspect
ratio CPNs is compromised.
Prototypical CPs have a strong tendency to form nanotube
aggregates that are highly insoluble.15 The solubility and the
aspect ratio of CPNs can be tailored by attaching polymers to
the exterior of nanotubes.16 However, due to the obvious
difficulties in vertically aligning 1-D nanotubes with a high
aspect ratio in thin films, as is required for membrane fabri-
cation, there is a need to modulate the assembly of CPN
growth in a confined framework to macroscopically align
nanotubes. There may also be additional opportunities to
modulate the assembly through structural modifications to the
primary sequence.3
We present here a minimalist approach that overcomes barriers
to both generate peptide nanotubes with interior modification and
dynamically tune the assembly process within the processing
window of polymeric membranes. As an initial demonstration, a
methyl group is introduced to the interior, although other
functional groups might also be incorporated using orthogonal
peptide chemistries and will be explored separately.17 This was
achieved by substituting one of the L-Leu in the primary sequence
of a protypical CP sequence cyclo-[L-Lys-D-Ala-L-Leu-D-Ala]2
(8CP) using a single aromatic amino acid, 3-amino-2-methyl-
benzoic acid (γ-Mba-OH). The chemical structure of the Mba-
modified cyclic peptide, cyclo-(L-Lys-D-Ala-L-Leu-D-Ala-L-Lys-D-
Ala-γ-Mba-D-Ala) (Mba-8CP, Figure 1a), is shown alongside the
8CP (Figure 1b) for comparison.
3-(9-Fluorenylmethyloxycarbonyl)amino-2-methylbenzoic
acid (Fmoc-γ-Mba-OH) was prepared from 3-amino-2-methyl-
benzoic acid and N-(9-fluorenylmethoxycarbonyloxy) succini-
mide (Fmoc-OSu) in 80% yield. Both starting materials are
commercially available, and the resulting Fmoc-protected amino
acid can be made in large quantities and readily used in the SPPS.
Mba-8CP and 8CP were synthesized on and cleaved from a
2-chlorotrityl chloride resin prior to head-to-tail cyclization at
rganic nanotubes have unique advantages over carbon
O
nanotubes and inorganic counterparts since their supramo-
lecular assembly, often governed by secondary interactions,
is fully reversible. They are useful building blocks to fabricate
membranes for applications such as carbon capture, water
desalination, and protective coatings.1À3 There have been
extensive efforts to design and synthesize nanotubes using
dendrimers,4 peptides,5 peptidomimetics,6 DNAs,7 foldamers,8
and J-type rosettes.9 However, in order to fabricate technologi-
cally relevant membranes using organic nanotubes, there are two
barriers: synthesizing nanotubes with a molecularly defined size,
shape, and interior chemistry to control selectivity;10 and mod-
ulating the nanotube assembly process to be compatible with
polymers processing toward membrane fabrication.3
Cyclic peptides (CPs), comprised of an even number of alter-
nating D- and L-α-amino acids, are particularly attractive since
hydrogen bonding between adjacent cyclic peptides leads to
stable 1-D hollow cyclic peptide nanotubes (CPNs) that exhibit
transport properties similar to those seen in transmembrane
proteins.11 To further control mass transport through CPNs,
attempts to modify their interiors have focused on incorporating
artificial amino acids containing cyclohexanes and aromatic rings,
or unsaturated amino acids at multiple positions in the canonical
CP sequence to project a specific chemical functionality toward
the pore interior.10,12,13 Most of these approaches involve amino
acid derivatives that are synthetically nontrivial, making it difficult
to fabricate membranes at large scales. Synthesizing CPs using the
solid phase peptide synthesis (SPPS) methodologies provides full
control of the number and location of modifications in the pep-
tide sequence.14 These aspects of the synthesis are particularly
Received: July 7, 2011
Published: September 06, 2011
r
2011 American Chemical Society
15296
dx.doi.org/10.1021/ja2063082 J. Am. Chem. Soc. 2011, 133, 15296–15299
|

Journal of the American Chemical Society
COMMUNICATION
Figure 3. 2D NOESY spectra in d6-DMSO of molecularly dissolved
Mba-8CP indicating the through space correlations between the aro-
matic methyl group’s protons in purple (a) and the aromatic protons
in red (b) with the proton resonances of the amides in orange. The key
cross-peaks are shown in red.
nanotubes showed that the mutation disproportionately elon-
gates the ring structure and disrupts the symmetry of its amide
bonds, and consequently the alignment of the carbonyl and
amide groups in the antiparallel assembly. As a result, the number
of hydrogen bonds between adjacent cyclic peptide rings is
reduced from an average of 7.3 hydrogen bonds for 8CP to 5.9
for Mba-8CP. The average inter-ring distance is ∼4.8 Å in both
cases (Figure S1).
Figure 1. Chemical structures of (a) Mba-8CP and (b) its conventional
analogue 8CP; Snapshots of equilibrium structures calculated from
molecular dynamics (MD) simulations, showing cross-sectional (c,d)
and lateral (e,f) views of Mba-8CP and 8CP nanotubes respectively. The
internal diameters of Mba-8CP and 8CP are ∼4.7 and ∼7.6 Å, respec-
tively, based on van der Waals radii.
Mba-8CP readily assembles into nanotubes in acetonitrile
(ACN). Inter-ring hydrogen bonding for Mba-8CP was clearly
observed at 3284 cmÀ1 using Fourier transform infrared (FTIR)
spectroscopy (Figure 2a). Characteristic peaks for the amide Ia,
amide Ib, and amide II regions at ν 1673, 1628, and 1538 cmÀ1
were consistent with those reported for 8CP.18 Transmission
electron microscopy (TEM) revealed bundles of nanotubes,
∼100À500 nm in length (Figure 2b). The width of the bundles
is between 20 and 40 nm, which corresponds to approximately
over 20 individual Mba-8CPNs in a single aggregate. The
dynamic light scattering (DLS) profile of Mba-8CP showed a
nominal nanotube length of ∼235 nm, slightly smaller than that
of ∼377 nm for 8CP (Figure S2).
Detailed spectroscopic characterizations using 1H and
1
1HÀ H-COSY (Figure S3) NMR experiments were used to
confirm the structure of Mba-8CP. A through-space 2D NOESY
experiment was also carried out on a molecularly dissolved
(i.e., disassembled) sample in DMSO to determine the orienta-
tion of the substituent at the 2-position of the aromatic amino
acid in the modified CP (Figure S4). The methyl group of
γ-Mba-OH at δ 2.19 ppm showed distinctive through-space
interactions with each of the amide resonances in the CP
backbone between 7.80 and 9.20 ppm (Figure 3a). This cor-
relation was absent for the aromatic protons, except for the
two neighboring amide protons on either side of the aromatic
ring (Figure 3b). Similar studies were carried out in the ACN
solution but were unsuccessful. The 2D NOSEY results reflect
the solution conformation of the Mba-8CP, which most likely
represent the favorable conformation of the monomer in a
CPN. This agrees well with the MD simulation that the methyl
group was directed toward the inside of the CPN, while the
aromatic protons were directed outward. Thus, by simply
inserting this single aromatic amino acid into the primary
sequence of the cyclic peptide 8-mer, its associated functional
groups were presented in the interior of the nanotubes without
compromising the formation of high aspect ratio assemblies.
Figure 2. (a) FTIRspectrumofa thin film ofMba-8CPcast fromACN at
room temperature. (b) TEM image of Mba-8CP nanotubes that aggre-
gated into bundles, likely during the drying process. Scale bar: 200 nm.
high dilution in the presence of propane phosphonic acid
anhydride (T3P) in 85% yield. After deprotection of the Boc
groups on the lysine side chains, the crude materials were purified
using preparative HPLC in an overall yield of 30%.
Molecular dynamics (MD) simulations were carried out to
compare the structure and stability of the assemblies of Mba-8CP
and 8CP. The simulations were performed using the COMPASS
force field. Top views and side views of the equilibrated struc-
tures obtained from MD simulations are shown in Figure 1cÀf.
For Mba-8CP, the L-Leu to γ-Mba mutation points the methyl
group into the interior of the pore as designed. This prominent
hydrophobic group substantially reduces the internal diameter
which is estimated to be ∼4.7 ( 0.6 Å based on van der Waals
radii (Figure 1c), a 38% reduction in size from the ∼7.6 Å pore
size of the 8CP nanotube (Figure 1d). The aromatic rings of
γ-Mba-OH are configured along one side of the nanotube, with
slight tilts. The lateral views (Figure 1e,f) of the formed
15297
dx.doi.org/10.1021/ja2063082 |J. Am. Chem. Soc. 2011, 133, 15296–15299

Journal of the American Chemical Society
COMMUNICATION
toward a plateau at ∼40 μM. The dissociation constant Kd was
found to be 21 ( 5 μM. The assembly process of Mba-8CP was
determined to be completely thermoreversible within the experi-
mental conditions applied. Figure 4b relates the magnitude of the
Cotton effect at 210 nm for Mba-CP derived nanotubes in ACN
at different concentrations upon heating from 20 to 70 °C. The
effect of temperature on the CD signal profile was dependent on
the concentration of the Mba-8CP. As the temperature is
increased, the CD signals of all profiles approach a similar equi-
librium state closer to the molecularly dissolved state. Similar studies
with the regular 8CP could not be achieved due to its lower solubility
and much stronger hydrogen bonding among the rings, which
showed little thermal losses in the CD signal. The dependence of
the reversible assembly process on concentration and temperature
in the case of CPNs derived from Mba-8CP points to distinct
opportunities to process smooth polymer thin films from solution.
To ensure the reversible assembly process of Mba-8CP does
not compromise the formation of high aspect ratio CPNs, we
monitored the nanotube growth using polyethylene glycol
(PEG)-covered CPNs of Mba-8CP. HO-PEG-NHCOCH2CH2-
COOH (Mw = 3000 g/mol) was conjugated to Mba-8CP by
coupling to the amino groups of lysine residues of Mba-8CP as
described previously.3 The nanotube growth process of PEG-
conjugated Mba-8CP in the solid state was studied in situ using
FTIR spectroscopy by monitoring the NÀH stretching vibration
at 3292 cmÀ1 upon heating from 30 to 180 °C. The nanotubes
reformed upon cooling with only a slight hysteresis (Figure 5a).
Upon spin-casting onto a silicon substrate with a native silicon
oxide layer, the PEG-conjugatedMba-8CPNs thatare 50À150 nm
in length can be easily visualized (Figure 5b). The width of the
formed nanotube is ∼5.5 nm (the height is ∼0.5 nm), which
corresponds to individual PEG-covered Mba-8CPNs.16 This con-
firms the advantage of attaching polymers to CPNs in reducing
their tendency to aggregate, hence rendering them more proces-
sable. The assemblyof the casted CPNscan bereadilydisrupted by
thermal or solvent annealing; typical treatments to prepare poly-
meric membranes (Figure 5c). The high aspect ratio nano-
tubes reform upon thermal annealing followed by slow cooling
(Figure 5d).
Figure 4. (a) CD spectra of Mba-8CP in ACN at different concentra-
tions upon heating from 20 to 70 °C. The inset illustrates the
dependence of molar ellipticity at 210 nm as a function of concentration.
(b) The CD signal of Mba-8CP in ACN at 210 nm as a function of
temperature for different concentrations. The heating rate is 5 °C/min
with an equilibration time of 5 min.
Since the assembly of cyclic peptides is a highly directional 1-D
growth process and the overall properties of nanotubes are
influenced by the molecular scale building blocks, we see a viable
path toward organic nanotubes with molecularly defined inter-
iors to mimic transmembrane proteins for enhanced selectivity in
molecular recognition, transport, and separation processes. The
added advantage of this new design is the ability to manipulate
the nanotube formation to be compatible with the processing
window of polymeric membranes so that subunits or short
nanotubes, rather than high aspect ratio nanotubes, can be
incorporated into a polymer matrix and, subsequently, grow
nanotubes in situ to fabricate functional membranes.
Figure 5. (a) Normalized absorptions of PEG-conjugated Mba-8CP for
the heating and cooling cycles (30À180 °C) as a function of tempera-
ture using the intensity of the peak maximum at υ = 3292 cmÀ1 at 30 °C.
AFM images of a spin-casted THF solution of PEG-conjugated Mba-
8CP nanotubes (b), followed by solvent-annealing and quenching (c),
followed by thermal annealing at 80 °C for 1 h and slow cooling (d)
(Scale bar: 50 nm). Reversible growth of high aspect ratio PEG-covered
CPNs of Mba-8CP can be clearly seen.
Theoretical calculations were also carried out to investigate the
incorporation of two or more unnatural amino acids in the
modified cyclic peptides. However, introduction of multiple
modifications lead to either blocked pores or puckered cyclic
peptide structures largely incapable of assembling into CPNs, as
illustrated in Figure S5.
The mutation in the Mba-8CP sequence also suppressed to
some extent the nanotube growth. The solution phase assembly
of Mba-8CP was studied as a function of concentration and
temperature using circular dichroism (CD) spectroscopy. CPNs
from Mba-8CP in ACN showed a distinctive negative Cotton
effect at λ = 210 nm (Figure 4a). Increasing the concentration of
Mba-8CP increased the intensity of the CD signal at 210 nm,
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental and computational
b
details and characterization of materials synthesized. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
bahelms@lbl.gov; tingxu@berkeley.edu
15298
dx.doi.org/10.1021/ja2063082 |J. Am. Chem. Soc. 2011, 133, 15296–15299

Journal of the American Chemical Society
COMMUNICATION
’ ACKNOWLEDGMENT
(15) (a) Khazanovich, N.; Granja, J. R.; McRee, D. E.; Milligan,
R. A.; Ghadiri, M. R. J. Am. Chem. Soc. 1994, 116, 6011–6012. (b)
Ghadiri, M. R.; Kobayashi, K.; Granja, J. R; Chadha, R. K.; McRee, D. E.
Angew. Chem., Int. Ed. Engl. 1995, 34, 93–95. (c) Kobayashi, K.; Granja,
J. R.; Ghadiri, M. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 95–98. (d)
Clark, T. D.; Ghadiri, M. R. J. Am. Chem. Soc. 1995, 117, 12364–12365.
(e) Ghadiri, M. R.; Granja, J. R.; Milligan, R. A.; McRee, D. E.;
Khazanovich, N. Nature 1993, 366, 324–327.
(16) (a) ten Cate, M. G. J.; Severin, N.; Borner, H. G. Macromolecules
2006, 39, 7831–7838. (b) Couet, J.; Biesalski, M. Small 2008,
4, 1008–1016. (c) Couet, J.; Jeyaprakash, J. D.; Samuel, S.; Kopyshev,
A.; Santer, S.; Biesalski, M. Angew. Chem., Int. Ed. 2005, 44, 3297–3301.
(17) (a) Hermanson, G. T. Bioconjugate Techniques; Academic Press,
Elsevier Inc.: London, U.K., 2008. (b) Chan, C. W.; White, P. D. Fmoc
solid phase peptide synthesis: A practical approach; Oxford University
Press: New York, USA, 2000.
We acknowledge support from the DOE-EFRC for Gas
Separations Relevant to Clean Energy Technologies under
Award Number DE-SC0001015 (R.H., C.Z., B.H., T.X.); Army
Research Office under Contract No. W91NF-09-1-0374
(R.vdW., C.L., T.X.). R.H. thanks FQRNT (Quꢀebec, Canada)
for a postdoctoral fellowship. S.K. and L.R. acknowledge a super-
computing grant from Northwestern University Quest HPC
system and support from CEE & ME departments. Portions of
this work (e.g., synthesis of γ-Mba-OH, CPs, and spectroscopic
characterization) were performed as a User project at the Molec-
ular Foundry, funded by the DOE under Contract No. DE-AC02-
05CH11231. We thank H. Dong for the TEM measurements
and J. Pelton for help and discussions about the NMR studies.
(18) (a) Krimm, S.; Bandekar, J. In Advances in Protein Chemistry;
Anfinsen, C. B., Edsall, J. T., Richards, F. M., Eds.; Academic Press:
Orlando, FL, 1986; pp 181À364. (b) Haris, P. I.; Chapman, D.
Biopolymers (Peptide Sci.) 1995, 37, 251–263. (c) Clark, T. D.; Buehler,
L. K.; Ghadiri, M. R. J. Am. Chem. Soc. 1998, 120, 651–656.
’ REFERENCES
(1) (a) Gin, D. L.; Noble, R. D. Science 2011, 332, 674–676.
(b) Kaucher, M. S.; Peterca, M.; Dulcey, A. E.; Kim, A. J.; Vinogradov,
S. A.; Hammer, D. A.; Heiney, P. A.; Percec, V. J. Am. Chem. Soc. 2007,
129, 11698–11699.
(2) Shannon, M. A.; Bohn, P. W.; Elimelech, M.; Georgiadis, J. G.;
Marinas, B. J.; Mayes, A. M. Nature 2008, 452, 301–310.
(3) Xu, T. Z., N.; Ren, F.; Hourani, R.; Lee, M. T.; Shu, J. Y.; Mao, S.;
Helms, B. A. ACS Nano 2011, 5, 1376–1384.
(4) Percec, V.; Dulcey, A. E.; Balagurusamy, V. S. K.; Miura, Y.;
Smidrkal, J.; Peterca, M.; Nummelin, S.; Edlund, U.; Hudson, S. D.;
Heiney, P. A.; Hu, D. A.; Magonov, S. N.; Vinogradov, S. A. Nature 2004,
430, 764–768.
(5) Kholkin, A.; Amdursky, N.; Bdikin, I.; Gazit, E.; Rosenman, G.
ACS Nano 2010, 4, 610–614.
(6) Hartgerink, J. D.; Granja, J. R.; Milligan, R. A.; Ghadiri, M. R.
J. Am. Chem. Soc. 1996, 118, 43–50.
(7) Davis, J. T.; Spada, G. P. Chem. Soc. Rev. 2007, 36, 296–313.
(8) Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596–
1608.
(9) Sakai, N.; Kamikawa, Y.; Nishii, M.; Matsuoka, T.; Kato, T.;
Matile, S. J. Am. Chem. Soc. 2006, 128, 2218–2219.
(10) (a) Amorin, M.; Castedo, L.; Granja, J. R. J. Am. Chem. Soc.
2003, 125, 2844–2845. (b) Reiriz, C.; Brea, R. J.; Arranz, R.; Carrascosa,
J. L.; Garibotti, A.; Manning, B.; Valpuesta, J. M.; Eritja, R.; Castedo, L.;
Granja, J. R. J. Am. Chem. Soc. 2009, 131, 11335–11337. (c) Horne,
W. S.; Stout, C. D.; Ghadiri, M. R. J. Am. Chem. Soc. 2003,
125, 9372–9376. (d) Reiriz, C.; Amorin, M.; García-Fandi~no, R.;
Castedo, L.; Granja, J. R. Org. Biomol. Chem. 2009, 7, 4358–4361.
(11) (a) Ghadiri, M. R.; Granja, J. R.; Buehler, L. K. Nature 1994,
369, 301–304. (b) Brea, R. J.; Reiriz, C.; Granja, J. R. Chem. Soc. Rev.
2010, 39, 1448–1456.
(12) (a) Kubik, S.; Goddard, R. J. Org. Chem. 1999, 64, 9475–9486.
(b) Kubik, S. J. Am. Chem. Soc. 1999, 121, 5846–5855. (c) Ishida, H.;
Qi, Z.; Sokabe, M.; Donowaki, K.; Inoue, Y. J. Org. Chem. 2001, 66,
2978–2989. (d) Kubik, S.; Goddard, R. Chem. Commun. 2000, 633–634.
(e) Kubik, S. Chem. Soc. Rev. 2009, 38, 585–605. (f) Gauthier, D.;
Baillargeon, P.; Drouin, M.; Dory, Y. L. Angew. Chem., Int. Ed. 2001,
40, 4635–4638. (g) Leclair, S.; Baillargeon, P.; Skouta, R.; Gauthier, D.;
Zhao, Y.; Dory, Y. L. Angew. Chem., Int. Ed. 2004, 43, 349–353.
(13) (a) Amorin, M.; Castedo, L.; Granja, J. R. Chem.—Eur. J. 2005,
11, 6543–6551. (b) Amorín, M.; Villaverde, V.; Castedo, L.; Granja, J. R.
J. Drug Delivery Sci. Technol. 2005, 15, 87–93. (c) Brea, R. J.; Castedo, L.;
Granja, J. R. Chem. Commun. 2007, 3267–3269. (d) Brea, R. J.; Amorín,
M.; Castedo, L.; Granja, J. R. Angew. Chem., Int. Ed. 2005, 44, 5710–5713.
(e) Amorín, M.; Brea, R. J.; Castedo, L.; Granja, J. R. Org. Lett. 2005,
7, 4681–4684. (f) Amorín, M.; Castedo, L.; Granja, J. R. Chem.—Eur. J.
2008, 14, 2100–2111.
(14) Atherton, E.; Sheppard, R. C. Solid Phase Peptide Synthesis: A
Practical Approach; Oxford University Press; USA, 1989.
15299
dx.doi.org/10.1021/ja2063082 |J. Am. Chem. Soc. 2011, 133, 15296–15299