Acetyl-terminated and template-assembled collagen-based polypeptides composed of Gly-Pro-Hyp sequences. 2. Synthesis and conformational analysis by circular dichroism, ultraviolet absorbance, and optical rotation

Article abstract of DOI:10.1021/ja961260c

Template-assembled collagen-based polypeptides KTA-[Gly-(Gly-Pro-Hyp)(n)-NH₂]₃ (n = 1, 3, 5, 6; KTA is cis,cis-1,3,5-trimethylcyclohexane-1,3,5-tricarboxylic acid, also known as the Kemp triacid) and acetyl-terminated single-chain co

Full text of DOI:10.1021/ja961260c

J. Am. Chem. Soc. 1996, 118, 10351-10358
10351
Acetyl-Terminated and Template-Assembled Collagen-Based
Polypeptides Composed of Gly-Pro-Hyp Sequences. 2.
Synthesis and Conformational Analysis by Circular Dichroism,
Ultraviolet Absorbance, and Optical Rotation
Yangbo Feng, Giuseppe Melacini, Joseph P. Taulane, and Murray Goodman*
Contribution from the Department of Chemistry & Biochemistry, UniVersity of California at
San Diego, La Jolla, California 92093-0343
ReceiVed April 17, 1996X
Abstract: Template-assembled collagen-based polypeptides KTA-[Gly-(Gly-Pro-Hyp)n-NH2]3 (n ) 1, 3, 5, 6; KTA
is cis,cis-1,3,5-trimethylcyclohexane-1,3,5-tricarboxylic acid, also known as the Kemp triacid) and acetyl-terminated
single-chain collagen-based analogs Ac-(Gly-Pro-Hyp)n-NH2 (n ) 1, 3, 5, 6, 9) were synthesized by solid phase
segment condensation methods. The triple-helical propensities of these collagen analogs were investigated using
circular dichroism, ultraviolet absorbance, optical rotation, and nuclear magnetic resonance measurements. The acetyl
analogs, Ac-(Gly-Pro-Hyp)n-NH2 (n ) 6, 9), assume a stable triple-helical conformation in H2O (0.2 mg/mL) at
room temperature. By contrast, Ac-(Gly-Pro-Hyp)5-NH2 adopts a triple-helical conformation in H2O only below 18
°C at a concentration of 0.2 mg/mL. For the template-assembled collagen analogs, results show that KTA-[Gly-
(Gly-Pro-Hyp)n-NH2]3 (n ) 5, 6) peptides form triple-helical structures which have melting temperatures above 70
°C in H2O. These melting temperatures are much higher than those of the corresponding acetyl analogs, demonstrating
the significant triple-helix-stabilizing effects of the KTA template. In addition, the KTA template facilitates triple-
helical structures by dramatically accelerating triple-helix formation.
Introduction
colleagues.22,23 The most interesting compounds among these
collagen analogs are the monodisperse polypeptides containing
Gly-Pro-Pro or Gly-Pro-Hyp sequences.14,15,20,21,24,25 Charac-
terization of these compounds has demonstrated the effectiveness
of the Hyp residue at position Y in stabilizing triple-helical
conformations. These Gly-Pro-Hyp sequence-based peptides
can form stable triple-helical structures at or above room
Collagen is the most common protein in connective tissues.
It possesses many unique properties that contribute directly to
its role as a structural material. The tertiary structure of collagen
is known as a triple helix. It consists of three extended left-
handed polyproline-II-like peptide chains1,2 which are arranged
in a parallel direction with a one-residue shift intertwined to
form a right-handed superhelix.3,4 The primary structure of the
peptide chains is composed mainly of trimer repeats. The first
residue of these trimers is always a glycine followed by two
variable residues: Gly-X-Y. The second (X) and third (Y)
positions can be any amino acid. Proline and hydroxyproline
are commonly found at these positions (X , Y), and they account
for about 20% of the total amino acid composition in collagen
sequences.5,6
(8) Brown, F. R., III; Di Corato, A.; Lorenzi, G. P.; Blout, E. R. J. Mol.
Biol. 1972, 63, 85-99.
(9) Brown, F. R., III; Hopfinger, A. J.; Blout, E. R. J. Mol. Biol. 1972,
63, 101-115.
(10) Ananthanarayanan, V. S.; Brahmachari, S. K. Biopolymers 1976,
15, 707-716.
(11) Bansal, M.; Ramachandran, G. N. Int. J. Pept. Protein Res. 1978,
11, 73-81.
(12) Cabrol, D.; Broch, H.; Vasilescu, D. Biopolymers 1981, 20, 269-
282.
(13) Doyle, B. B.; Traub, W.; Lorenzi, G. P.; Blout, E. R. Biochemistry
1971, 10, 3052-3060.
Mimicry of collagen structures can help elucidate the unique
triple-helical conformations and provide insights into the
preparation of novel collagen-like biomaterials. Many sequen-
tial polytripeptides composed of collagen-like unique sequences
Gly-X-Y have been synthesized to mimic collagen-like triple-
helical structures.7-21 Pioneering work on polytripeptides was
reported in the early 1960s by Katchalski-Katzir and his
(14) Inouye, K.; Sakakibara, S.; Prockop, D. J. Biochim. Biophys. Acta
1976, 420, 133-141.
(15) Inouye, K.; Kobayashi, Y.; Kyogoku, Y.; Kishida, Y.; Sakakibara,
S.; Prockop, D. J. Arch. Biochem. Biophys. 1982, 219, 198-203.
(16) Miller, M. H.; Nemethy, G.; Scheraga, H. A. Macromolecules 1980,
13, 910-913.
(17) Segal, D. M.; Traub, W. J. Mol. Biol. 1969, 43, 487-496.
(18) Rapaka, R. S.; Bhatnagar, R. S. Int. J. Pept. Protein Res. 1975, 7,
119-128.
X Abstract published in AdVance ACS Abstracts, October 15, 1996.
(1) Bornstein, P.; Traub, W. In The Proteins; Neurath, H., Hill, R. L.,
Eds.; Academic Press: New York, 1979; Vol. 2, pp 412-632.
(2) Ramachandran, G. N. In Aspects of Protein Structure; Ramachandran,
G. N., Ed.; Academic Press: New York, 1963, pp 39.
(3) Bella, J.; Eaton, M.; Brodsky, B.; Berman, H. M. Science 1994, 266,
75-81.
(19) Scatturin, A.; Tamburro, A. M.; Del Pra, A.; Bordignon, E. Int. J.
Pept. Protein Res. 1975, 7, 425-435.
(20) Sakakibara, S.; Kishida, Y.; Kikuchi, Y.; Sakai, R.; Kakiuchi, K.
Bull. Chem. Soc. Jpn. 1968, 41, 1273.
(21) Sakakibara, S.; Inouye, K.; Shuda, K.; Kishida, Y.; Kobayashi, Y.;
Prockop, D. J. Biochim. Biophys. Acta 1973, 303, 198-202.
(22) Engel, J.; Kurtz, J.; Berger, A.; Katchalski-Katzir, E. J. Mol. Biol.
1966, 17, 255-272.
(4) Venugopal, M. G.; Ramshaw, J. A. M.; Braswell, E.; Zhu, D.;
Brodsky, B. Biochemistry 1994, 33, 7948-7956.
(5) Traub, W.; Piez, K. AdV. Protein Chem. 1971, 25, 243-341.
(6) Ramachandran, G. N.; Ramakrishnan, C. In Biochemistry of Collagen;
Ramachandran, G. N., Reddi, A. H., Eds.; Plenum Press: New York, 1976;
pp 45-84.
(7) Brown, F. R., III; Carver, J. P.; Blout, E. R. J. Mol. Biol. 1969, 39,
307-313.
(23) Steinberg, I. Z.; Harrington, W. F.; Berger, A.; Sela, M.; Katchalski-
Katzir, E. J. Am. Chem. Soc. 1960, 82, 5263-5279.
(24) Sakakibara, S.; Kishida, Y.; Okuyama, K.; Tanaka, N.; Ashida, T.;
Kakudo, M. J. Mol. Biol. 1972, 65, 371-373.
(25) Kobayashi, Y.; Sakai, R.; Kakiuchi, K.; Isemura, T. Biopolymers
1970, 9, 415-425.
S0002-7863(96)01260-7 CCC: $12.00 © 1996 American Chemical Society

10352 J. Am. Chem. Soc., Vol. 118, No. 43, 1996
Feng et al.
temperature when the peptide chains contain 10 or more Gly-
Pro-Hyp repeats.14,15,21,25 All these reported compounds, (Gly-
Pro-Hyp)n, possess free amines on the N-termini and free acids
on the C-termini.
Template-assembled synthetic protein (TASP) methods have
been used in the design and synthesis of many protein
mimics.26-32 In this method, template molecules are used to
fix peptide loops and to induce R-helical (or helical bundles)
and â-turn structures of polypeptides. For example, Schneider
and Kelly have reported the use of templates to form â-struc-
tures.32 Muller et al. disclosed anthrancene-type tricyclic
structures that can bridge two antiparallel peptide â-strands or
induce â-turns.28 Muller et al. also discovered that Kemp triacid
condensed with glycine or alanine can act as a template for
inducing R-helicity of an attached polypeptide.28 Ghadiri et al.
have prepared a 3-R-helix bundle by means of a ruthenium metal
bipyridyl complex.30
Figure 1. Structure of the Kemp triacid. According to NMR studies,42
the most stable conformation for this tricarboxylic acid compound is a
chair conformer with the carboxyl groups in the axial positions.
Scheme 1
Similar template-assembling methods can also be used to
mimic collagen-like triple-helical structures. In the collagen
triple helix, three parallel polypeptide chains are supercoiled
around a common axis and each of them is shifted by one
residue along the axis as compared to the adjacent chains. A
template with three functional groups that can be connected to
the three peptide chains, if designed properly, will help the three
chains to adopt this special array and register. Heidemann et
al. have used lysine dimer and 1,2,3-propanetricarboxylic acid
to prepare covalently bridged synthetic polypeptides which were
found to assemble into a triple helix.33-37 Fields et al.38-40 and
Tanaka et al.41 employed two consecutively connected lysine
residues with three functional groups to link the three peptide
chains at the C-termini or the N-termini. These templates are
able to stabilize triple-helical structures by changing the entropy
effects.
In our approach, a more conformationally constrained organic
molecule, cis,cis-1,3,5-trimethylcyclohexane-1,3,5-tricarboxylic
acid (the Kemp triacid, KTA, Figure 1),42 was selected to induce
the collagen-like triple-helical structure for very short peptide
chains.43 The KTA contains three axial carboxylic acid func-
tionalities, each of which can be coupled to a strand of the triple
helix. In addition, its three axial functional groups are parallel
to each other, which can facilitate the interactions of the three
peptide chains to form the desired triple-helical conformation.
Rather than coupling KTA directly to the N-termini of the
peptides, a glycine spacer was attached to each of the carboxyl
groups. The spacer reduces the steric hindrance of the three
peptide chains reacting with the KTA functional groups and
provides the flexibility to allow the three peptide chains to adopt
the proper one-residue register shift necessary for a collagen-
like triple-helical conformation. The template-assembled col-
lagen-based polypeptides KTA-[Gly-(Gly-Pro-Hyp)n-NH2]3 (n
) 1, 3, 5, 6), were synthesized using KTA as the template and
glycine as the spacer. Acetyl compounds, Ac-(Gly-Pro-Hyp)n-
NH2 (n ) 1, 3, 5, 6, 9), were also prepared for comparison
studies to demonstrate the effects of the KTA template.
In this paper, the synthesis of these template-assembled
collagen-based polypeptides is reported along with the synthesis
of the N-terminal-acetylated single-chain sequential polypep-
tides. The biophysical characterization of both the acetyl-
terminated and template-assembled collagen-based analogs is
presented and is based on circular dichroism (CD), ultraviolet
(UV) absorbance, optical rotation, and nuclear magnetic reso-
nance (NMR) measurements. The stabilizing effect of the KTA
template with glycine residue spacers on triple-helical confor-
mation is shown. Solvent and concentration effects on triple
helicity are also discussed.
(26) Tuchscherer, G.; Domer, B.; Sila, U.; Kamber, B.; Mutter, M.
Tetrahedron 1993, 49, 3559-3575.
(27) Vuilleumier, S.; Mutter, M. Biopolymers 1993, 33, 389-400.
(28) Muller, K.; Obrecht, D.; Knierzinger, A.; Stankovic, C.; Spiegler,
C.; Bannwarth, W.; Trzeciak, A.; Englert, G.; Labhardt, A. M.; Schonholzer,
P. In PerspectiVe in Medicinal Chemistry; Testa, B., et al., Eds.; Verlag:
Basel, 1993; pp 513-531.
(29) Ikeura, Y.; Honda, Y.; Kurihara, K.; Kunitake, T. Chem. Lett. 1990,
169-172.
(30) Ghadiri, M. R.; Case, M. A. Angew. Chem., Int. Ed. Engl. 1993,
32, 1594-1597.
(31) Whitesell, J. K.; Chang, H. K. Science 1993, 261, 73-76.
(32) Schneider, J. P.; Kelly, J. W. J. Am. Chem. Soc. 1995, 117, 2533-
2546.
(33) Greiche, Y.; Heidemann, E. Biopolymers 1979, 18, 2359-2361.
(34) Roth, W.; Heidemann, E. Biopolymers 1980, 19, 1909-1917.
(35) Thakur, S.; Vadolas, D.; Germann, H. P.; Heidemann, E. Biopoly-
mers 1986, 25, 1081-1086.
(36) Dolz, R.; Heidemann, E. Biopolymers 1986, 25, 1069-1080.
(37) Germann, H. P.; Heidemann, E. Biopolymers 1988, 27, 157-163.
(38) Fields, G. B. J. Theor. Biol. 1991, 153, 585-602.
(39) Fields, C. G.; Lovdahl, C. M.; Miles, A. J.; Matthias Hagen, V. L.;
Fields, G. B. Biopolymers 1993, 33, 1695-1707.
(40) Fields, C. G.; Mickelson, D. J.; Drake, S. L.; McCarthy, J. B.; Fields,
G. B. J. Biol. Chem. 1993, 268, 14153-14160.
(41) Tanaka, T.; Wada, Y.; Nakamura, H.; Doi, T.; Imanishi, T.; Kodama,
T. FEBS Lett. 1993, 334, 272-276.
(42) Kemp, D. S.; Petrakis, K. S. J. Org. Chem. 1981, 46, 5140-5143.
(43) Goodman, M.; Feng, Y.; Melacini, G.; Taulane, J. P. J. Am. Chem.
Soc. 1996, 118, 5156-5157. This is paper 1 of our series on collagen-
based structures.
Synthesis
Synthesis of the N-Terminal Acetyl Compounds. The
peptide chains were assembled by the solid phase segment
condensation method that was first introduced by Sakakibara
et al.20 The resin utilized was 4-methylbenzhydrylamine
(MBHA), and Boc(tert-butyloxycarbonyl) chemistry was em-
ployed in the solid phase synthesis (Scheme 1). At each
segment condensation, 1.2-1.5 equiv of building block Boc-
Gly-Pro-Hyp(OBz)-OH was used with the coupling reagents
diisopropylcarbodiimide (DIC) and N-hydroxybenzotriazole

Collagen-Based Polypeptides. 2
J. Am. Chem. Soc., Vol. 118, No. 43, 1996 10353
Scheme 2a
Scheme 3
a (a) Gly-OBz, EDC/HOBt, DMF, room temperature, 4 h, 98%; (b)
H2/Pd, MeOH, 97%.
(HOBt). Trifluoroacetic acid (TFA) was used to remove the
Boc protecting group after each coupling step, and the resin
was subsequently neutralized with triethylamine (TEA). The
Kaiser ninhydrin test was used to monitor the coupling reaction.
After a specific chain length was reached, the N-terminal amines
were acetylated by treatment with acetic anhydride. The
products were removed from the resin using HF cleavage
methods. Since MBHA resin was used, the final products were
all amides on the C-termini. Reverse phase HPLC was utilized
to separate the products and determine the purity. The structures
were all verified using mass spectrometry and 1H-NMR
spectroscopy.
Generally, in solid phase peptide synthesis more than 2 equiv
of the free acid component is used to drive each coupling
reaction to completion. Here only 1.2-1.5 equiv of the building
block tripeptide Boc-Gly-Pro-Hyp(OBz)-OH was used at each
coupling step because of the synthetic effort required for the
preparation of this tripeptide. Nevertheless, this coupling
strategy was found to be successful up to the ninth coupling
stage.
Synthesis of the Template KTA-(Gly-OH)3. Peptide-
coupling methods in solution were used to prepare the template
structure KTA-(Gly-OH)3, and the synthesis is described in
Scheme 2. The KTA-(Gly-OBz)3 was prepared by coupling
the amino acid ester Gly-OBz to the carboxyl groups of KTA
using 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide (EDC)
and HOBt as the coupling reagents. The benzyl groups were
then removed by hydrogenation to yield the template with three
attached glycine residues bearing free carboxyl groups, KTA-
(Gly-OH)3.
Synthesis of the Template-Assembled Compounds. To
prepare template-assembled collagen-like polypeptides, the
peptide chains were built up on the resin and the template was
directly coupled to the N-termini of these chains before removal
from the resin. The synthetic route is described in Scheme 3,
and the coupling reagents were DIC and HOBt. In order to
obtain a high yield of the desired three-chain product and to
minimize side products, KTA-(Gly-OH)3 was used as the
limiting reagent. The couplings were monitored by the Kaiser
ninhydrin test to completion, and the HF cleavage method was
used to remove the products from the resin. Reverse phase
HPLC was utilized to separate the template-assembled com-
pounds and determine their purity. Mass spectrometry and 1H-
NMR spectroscopy were used to characterize the peptide
structures.
features have been used as a basis to provide information about
the presence of a polyproline-II-like or collagen-like triple-
helical structure in solution for natural and synthetic pep-
tides.4,7,15,49 In order to investigate the triple-helical profiles
of our synthetic collagen-based peptides, their CD spectra in
both H2O and ethylene glycol/H2O (EG/H2O, v/v, 2:1) solvents
were determined. Ethylene glycol is known to stabilize helical
structures and therefore can be very useful to amplify and detect
very weak triple-helical propensities.7,8,27,50 Typical CD spectra
are shown in Figure 2. The spectroscopic parameters for the
polypeptides are listed in Table 1. The CD parameters listed
in Table 1 for collagen7,8 and (Gly-Pro-Hyp)10-OH4 are based
on previously reported data.
The 30-residue peptide (Gly-Pro-Hyp)10-OH has been re-
ported to be triple helical in H2O,3,4,15,21,51,52 while Ac-Gly-Pro-
Hyp-NH2 and KTA-[Gly-Gly-Pro-Hyp-NH2]3 lack the structural
requirements to form a triple-helical structure.53 However, these
three compounds along with all the other compounds listed in
Table 1 show similar CD spectral shapes and peak positions
for the minimum, crossover, and maximum, which are similar
to those of a typical polyproline-II-like conformation.7,45,46,48,54
These CD spectral shapes and peak positions do not show
whether the polyproline-II-like peptide chains are associated to
triple-helical structures. Therefore these features, although
necessary, can not be used conclusively to establish the presence
of a collagen-like triple-helical conformation.
A parameter related to the CD peak intensities was found to
be useful in establishing triple-helical conformations in solution.
This parameter, Rpn, denotes the ratio of positive peak intensity
over negative peak intensity. The Rpn values for the synthetic
collagen-based peptides are listed in Table 1. The Rpn values
for natural collagen and (Gly-Pro-Hyp)10-OH are also included
in Table 1 and are estimated from previously reported CD
spectra.4,7,8 From Table 1, it can be seen that a sharp increase
in Rpn values occurs at specific chain lengths. For the acetyl
compounds, Ac-(Gly-Pro-Hyp)n-NH2, the sharp increase is seen
when n ) 6 in H2O and n ) 5 in EG/H2O (v/v, 2:1). For the
template-assembled analogs, KTA-[Gly-(Gly-Pro-Hyp)n-NH2]3,
there is a sharp increase in Rpn values at n ) 3 in both H2O
and EG/H2O (v/v, 2:1). The Rpn value at the point of sharp
increase is solvent dependent and found to be 0.12 in H2O and
0.15 in EG/H2O (v/v, 2:1). From Table 1, it can also be seen
that the Rpn values for our collagen-like polypeptides are
essentially the same as the Rpn values exhibited by collagen
Biophysical Studies
(48) Sreerama, N.; Woody, R. W. Biochemistry 1994, 33, 10022-10025.
(49) Toumadje, A.; Johnson, W. C., Jr. J. Am. Chem. Soc. 1995, 117,
7023-7024.
(50) Choma, C. T.; Lear, J. D.; Nelson, M. J.; Dutton, P. L.; Roberson,
D. E.; DeGrado, W. F. J. Am. Chem. Soc. 1994, 116, 856-865.
(51) Brodsky, B.; Li, M.; Long, C. G.; Apigo, J.; Baum, J. Biopolymers
1992, 32, 447-451.
(52) Li, M.; Fan, P.; Brodsky, B.; Baum, J. Biochemistry 1993, 32, 7377-
7387.
(53) Melacini, G.; Feng, Y.; Goodman, M. J. Am. Chem. Soc. 1996, 118,
10359-10364.
(54) Ramachandran, G. N. In Treatise of Collagen, 1st ed.; Ramachan-
dran, G. N., Ed.; Academic Press: London, 1967; Vol. 1, pp 103-184.
CD Spectra. Collagen-like triple-helical and polyproline-
II-like structures exhibit CD spectra in solutions which are
characterized by a large negative peak around 200 nm and a
small positive peak around 215-227 nm.4,7,8,44-49 These
(44) Brodsky-Doyle, B.; Leonard, K. R.; Reid, K. B. M. Biochem. J.
1976, 159, 279-286.
(45) Rabanal, F.; Ludevid, M. D.; Pons, M.; Giralt, E. Biopolymers 1993,
33, 1019-1028.
(46) Madison, V.; Schellman, J. Biopolymers 1970, 9, 511-567.
(47) Siligardi, G.; Drake, A. F. Biopolymers (Pept. Sci.) 1995, 37, 1-12.

10354 J. Am. Chem. Soc., Vol. 118, No. 43, 1996
Feng et al.
Figure 2. CD spectra of Ac-(Gly-Pro-Hyp)n-NH2 (n ) 1, 3, 5, 6, 9) (A) and KTA-[Gly-(Gly-Pro-Hyp)n-NH2]3 (n ) 1, 3, 5, 6) (B) in H2O (0.2
mg/mL) at 20 °C.
Table 1. Circular Dichroism Data for the Synthesized Collagen-Based Polypeptides Composed of Gly-Pro-Hyp Sequencesa
H2O
EG/H2O (v/v, 2:1)
compounds
min (nm)
cross (nm)
max (nm)
Rpnb
min (nm)
cross (nm)
max (nm)
Rpnb
Ac-Gly-Pro-Hyp-NH2
Ac-(Gly-Pro-Hyp)3-NH2
Ac-(Gly-Pro-Hyp)5-NH2
Ac-(Gly-Pro-Hyp)6-NH2
Ac-(Gly-Pro-Hyp)9-NH2
198 (-2.1 × 104)
200 (-2.3 × 104)
200 (-2.6 × 104)
198 (-2.9 × 104)
198 (-3.1 × 104)
214
218
218
218
218
220 (1.4 × 103) 0.07 198 (-1.9 × 104)
223 (1.8 × 103) 0.08 199 (-2.4 × 104)
224 (1.7 × 103) 0.07 199 (-2.9 × 104)
224 (3.4 × 103) 0.12 198 (-3.4 × 104)
224 (4.4 × 103) 0.14 197 (-4.0 × 104)
214
217
217
217
217
221 (1.9 × 103) 0.10
223 (2.2 × 103) 0.09
223 (4.4 × 103) 0.15
223 (5.3 × 103) 0.16
223 (6.7 × 103) 0.17
KTA-[Gly-Gly-Pro-Hyp-NH2]3
199 (-1.7 × 104)
217
217
218
218
222 (0.8 × 103) 0.05 199 (-1.7 × 104)
223 (3.7 × 103) 0.14 200 (-3.1 × 104)
224 (4.0 × 103) 0.14 200 (-2.7 × 104)
224 (4.1 × 103) 0.12 199 (-2.2 × 104)
217
217
218
218
223 (1.0 × 103) 0.06
224 (5.0 × 103) 0.16
224 (4.4 × 103) 0.16
224 (3.7 × 103) 0.17
KTA-[Gly-(Gly-Pro-Hyp)3-NH2]3 200 (-2.6 × 104)
KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3 198 (-2.9 × 104)
KTA-[Gly-(Gly-Pro-Hyp)6-NH2]3 198 (-3.4 × 104)
collagenc
198 (-5.4 × 104)
213
218
220 (7.1 × 103) 0.13 197 (-5.7 × 104)
213
220 (9.5 × 103) 0.17
(Gly-Pro-Hyp)10-OHc
198 (-3.4 × 104)
225 (4.3 × 103) 0.13
a CD spectra were obtained at 20 °C using a peptide concentration of 0.2 mg/mL. The peak intensities are included in parentheses. b Rpn
represents the ratio of positive peak intensity over negative peak intensity (absolute values). c Estimated from published CD spectra: refs 8 and 9
for collagen in H2O, ref 7 for collagen in EG/H2O (v/v, 2:1), and ref 4 for (Gly-Pro-Hyp)10-OH in H2O.
and (Gly-Pro-Hyp)10-OH. On the basis of these Rpn values,
the presence of triple-helical conformations is established for
the following synthetic polypeptides, at 20 °C, at a concentration
of 0.2 mg/mL: Ac-(Gly-Pro-Hyp)n-NH2 (n ) 6, 9) and KTA-
[Gly-(Gly-Pro-Hyp)n-NH2]3 (n ) 3, 5, 6) in both H2O and EG/
H2O (v/v, 2:1) and Ac-(Gly-Pro-Hyp)5-NH2 in EG/H2O (v/v,
2:1).
temperatures of 37 and 36 °C (Table 2) were obtained from
UV absorbance and optical rotation measurements, res-
pectively.
Based on the results of melting studies (Table 2), it can be
concluded that KTA-[Gly-(Gly-Pro-Hyp)n-NH2]3 (n ) 3, 5, 6)
and Ac-(Gly-Pro-Hyp)n-NH2 (n ) 6, 9) in both H2O and EG/
H2O (v/v, 2:1) as well as Ac-(Gly-Pro-Hyp)5-NH2 in EG/H2O
(v/v, 2:1) form triple-helical conformations at 20 °C, at a
concentration of 0.2 mg/mL. However, Ac-(Gly-Pro-Hyp)5-
NH2 is triple helical in H2O only below 18 °C, at a concentration
of 0.2 mg/mL. It is important to point out that the results
obtained from these melting studies are consistent with those
obtained from Rpn values with regard to the presence of triple-
helical conformations in solution.
Acetyl-Terminated Collagen Analogs. Tables 1 and 2 show
that Ac-Gly-Pro-Hyp-NH2 and Ac-(Gly-Pro-Hyp)3-NH2 cannot
assume triple-helical structures in either H2O or EG/H2O (v/v,
2:1) solvents. This is indicated by their low Rpn values (Table
1) and the lack of observable melting transitions (Table 2). On
the other hand, Ac-(Gly-Pro-Hyp)n-NH2 (n ) 6, 9) peptides form
very stable triple-helical structures with melting temperatures
of 36 °C (n ) 6 in H2O), 52 °C (n ) 6 in EG/H2O, v/v, 2:1),
and 67 °C (n ) 9 in H2O). At a concentration of 0.2 mg/mL,
Ac-(Gly-Pro-Hyp)5-NH2 forms a stable triple-helical conforma-
tion in EG/H2O (v/v, 2:1) with a melting temperature of 32 °C.
However in H2O, a triple-helical structure is observed only
below a temperature of 18 °C (Table 2). These results suggest
that this compound represents a transition between the triple-
helical and the non-triple-helical conformations, in H2O. Ad-
ditions of alcohol solvents such as EtOH and ethylene glycol
Melting Studies. In addition to CD spectroscopy, melting
curve measurements were also used to determine the triple
helicity of the synthetic collagen-based polypeptides. Measure-
ments of the optical rotation (R)15,20,21 and the CD molar
ellipticity (θ)4,39,51,52 as a function of temperature have been
used to monitor the thermal denaturation of collagen-like triple-
helical conformations. Similarly, UV absorbance has also been
used for this purpose.55 For natural collagen, denaturation is
always accompanied by a hyperchromic effect in UV absorb-
ance.56 A similar effect is observed for our synthetic acetyl-
terminated and template-assembled peptides.
Both optical rotation and UV absorbance measurements were
used to construct the melting curves for our synthetic collagen-
based peptides. The melting temperatures obtained in both H2O
and EG/H2O (v/v, 2:1) are listed in Table 2. The melting curves
obtained by optical rotation measurements in H2O are presented
in Figure 3. The results obtained from both UV and optical
rotation methods are consistent with each other. For example,
for Ac-(Gly-Pro-Hyp)6-NH2 in H2O (0.2 mg/mL), melting
(55) von Hippel, P. H. In Treatise on Collagen, 1st ed.; Ramachandran,
G. N., Ed.; Academic Press: London, 1967; Vol. 1, pp 253-338.
(56) Wood, G. C. Biochem. Biophys. Res. Commun. 1963, 13, 95.

Collagen-Based Polypeptides. 2
J. Am. Chem. Soc., Vol. 118, No. 43, 1996 10355
Table 2. Melting Results of Collagen-Based Polypeptides Composed of Gly-Pro-Hyp Sequences from UV and Optical Rotation
Measurements
compounds
H2O
EG/H2O (v/v, 2:1)
Ac-Gly-Pro-Hyp-NH2
Ac-(Gly-Pro-Hyp)3-NH2
Ac-(Gly-Pro-Hyp)5-NH2
Ac-(Gly-Pro-Hyp)6-NH2
Ac-(Gly-Pro-Hyp)9-NH2
no transition observed
no transition observed
18 °Cb
no transition observed
no transition observed
32 °Cb
37 °C,a 36 °Cb
67 °Cb
52 °Ca
KTA-[Gly-Gly-Pro-Hyp-NH2]3
KTA-[Gly-(Gly-Pro-Hyp)3-NH2]3
KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3
KTA-[Gly-(Gly-Pro-Hyp)6-NH2]3
no transition observed
30 °Cb
no transition observed
50 °Cb
72 °C,a 70 °Cb
82 °C,a 81 °Cb
ca. 95 °Ca,c
>95 °Ca,c
Ac-(Gly-Pro-Hyp)5-NH2
(Gly-Pro-Hyp)5-OHd
(Gly-Pro-Hyp)10-OHd
23 °C,b in 50% aqueous EtOH
5 °C, in 50% aqueous EtOH
58 °C, in 50% aqueous EtOH
a Results obtained from UV measurements. The concentrations are 0.04 mg/mL. b Results obtained from optical rotation measurements. The
wavelength was set at 365 nmHg. Peptide concentrations are 0.2 mg/mL. c The denaturation commences at about 90 °C. Complete melting curves
could not be obtained because the measurements could not be carried out above 100 °C. d Reference 21.
Figure 3. Temperature dependence of specific rotations at 365 nmHg for Ac-(Gly-Pro-Hyp)n-NH2 (n ) 5, 6, 9) (A) and KTA-[Gly-(Gly-Pro-
Hyp)n-NH2]3 (n ) 3, 5, 6) (B) in H2O (0.2 mg/mL).
increase the melting temperatures, indicating an enhancement
of triple helicity (Table 2). In addition, in more concentrated
H2O solutions, higher melting temperatures were obtained
presumably because of concentration effects (see discussion
below). However, Ac-(Gly-Pro-Hyp)3-NH2 does not show any
evidence of triple helicity even at a concentration of ca. 2 mg/
mL, at 5 °C.53 These results indicate that at least five Gly-
Pro-Hyp repeats are required to form a collagen-like triple-
helical structure in H2O, in the absence of a template.
Template-Assembled Collagen Analogs. The effect of the
KTA template on triple-helical structures can be demonstrated
by comparing the properties of KTA-[Gly-(Gly-Pro-Hyp)n-
NH2]3 with those of the corresponding acetyl compounds, Ac-
(Gly-Pro-Hyp)n-NH2 (n ) 3, 5, 6). For the n ) 5 analogs, only
the template-assembled compound shows a typical triple-helical
CD spectrum in H2O, at 20 °C (Table 1). The melting
temperatures of KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3 are 70 °C in
H2O and ca. 95 °C in EG/H2O (v/v, 2:1). In contrast, much
lower melting temperatures are seen for Ac-(Gly-Pro-Hyp)5-
NH2: 18 °C in H2O and 32 °C in EG/H2O (v/v, 2:1). Similar
results were also obtained for the n ) 6 analogs. As shown in
Table 2, a melting temperature of 81 °C was observed for KTA-
[Gly-(Gly-Pro-Hyp)6-NH2]3 in H2O, and a melting temperature
of 36 °C was obtained for Ac-(Gly-Pro-Hyp)6-NH2 in H2O (0.2
mg/mL). The increase in melting temperatures for the template-
assembled analogs shows the significant stabilizing effects of
the KTA-based template on triple-helical conformations. Sa-
kakibara et al. observed a melting temperature of 58 °C for
(Gly-Pro-Hyp)10-OH in 50% EtOH/H2O,21 which is 53 °C higher
than that obtained for (Gly-Pro-Hyp)5-OH (5 °C in 50% EtOH/
H2O.21 Therefore, the effect of the KTA template, with glycine
residue spacers, on the thermal stability of the triple helices
formed with five Gly-Pro-Hyp repeat analogs is similar to the
addition of five more trimer repeats to the peptide chain. The
triple-helix-inducing effect of the KTA template is even more
dramatic in the n ) 3 compounds. KTA-[Gly-(Gly-Pro-Hyp)3-
NH2]3 is triple helical in H2O at room temperature. In contrast,
no signs of triple helicity were observed for Ac-(Gly-Pro-Hyp)3-
NH2 at lower temperatures in H2O and EG/H2O (v/v, 2:1)
solvents (Table 2) as well as at higher concentrations (ca. 2
mg/mL).53
In addition to the triple-helix-stabilizing effects discussed
above, the KTA template also facilitates triple-helical structures
by accelerating triple-helix formation. Triple-helix denaturation-
reformation experiments were carried out in H2O, at concentra-
tions of 0.2 mg/mL. The samples were thermally denatured,
and the triple-helix reformation was monitored by optical
rotation measurements at a temperature lower than the melting
temperature of each sample. It was found that several hours
were necessary for the triple helices of Ac-(Gly-Pro-Hyp)n-NH2
(n ) 5, 6) to reform. For the longer chain acetyl analog Ac-
(Gly-Pro-Hyp)9-NH2, the triple helix reformed in 1 h. In
contrast the template-assembled analogs, KTA-[Gly-(Gly-Pro-
Hyp)n-NH2]3 (n ) 3, 5, 6), reformed their triple helices
instantaneously (<1 min). These results substantiate the
dramatic kinetic effects of the KTA-based template on triple-
helix formation.
Concentration Effects on Triple Helicity. For single-chain
acetyl-terminated collagen analogs, triple-helix formation in-
volves the interaction of three molecules. Therefore, the

10356 J. Am. Chem. Soc., Vol. 118, No. 43, 1996
Feng et al.
Figure 4. (A) Temperature dependence of specific rotations at 365 nmHg for Ac-(Gly-Pro-Hyp)5-NH2 in H2O at different concentrations and (B)
relationship between the melting temperature and the concentration for Ac-(Gly-Pro-Hyp)5-NH2 in H2O. The critical triple-helical concentration
(CTHC) is ca. 1 mg/mL.
concentration is expected to have a significant effect on triple-
helical propensity, especially when the peptide chain is short.
Ac-(Gly-Pro-Hyp)5-NH2, in H2O, is a good candidate to study
concentration effects since it has a low melting temperature (18
°C at 0.2 mg/mL) and is the non-template-assembled collagen
analog with the shortest chains capable of forming a triple-
helical conformation near room temperature.
is consistent with a maximal triple helicity considering the end
effect and the one-residue register shift in a collagen-like triple
helix.53
The CTHC values for Ac-(Gly-Pro-Hyp)n-NH2 (n ) 6, 9)
occur at a concentration of 0.02 mg/mL in H2O or less. As
demonstrated in Figure 3, the transition magnitude for Ac-(Gly-
Pro-Hyp)6-NH2 is almost the same as that obtained for KTA-
[Gly-(Gly-Pro-Hyp)6-NH2]3. On the basis of the above discus-
sion for Ac-(Gly-Pro-Hyp)5-NH2 and KTA-[Gly-(Gly-Pro-
Hyp)5-NH2]3, at a concentration of 0.2 mg/mL, the triple helicity
of Ac-(Gly-Pro-Hyp)6-NH2 is similar to that of KTA-[Gly-(Gly-
Pro-Hyp)6-NH2]3, which is maximal according to our NMR
studies.53 The specific rotations for Ac-(Gly-Pro-Hyp)n-NH2
(n ) 6, 9) do not vary from 0.2 to 0.02 mg/mL in H2O (20 °C),
which indicates that the polypeptides Ac-(Gly-Pro-Hyp)n-NH2
(n ) 6, 9) are completely triple helical even at a concentration
of 0.02 mg/mL in H2O.
NMR Spectra. The presence of a triple-helical conformation
in solution can also be established by NMR spectroscopy. The
formation of a triple-helical structure results in unique interchain
NOEs and the appearance of a new set of NMR resonances
which can not be observed for the fully unfolded analogs.51,52
The 1-D spectra of KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3 is pre-
sented here to demonstrate the use of NMR to study triple-
helical conformations. A comprehensive conformational analy-
sis by NMR for our synthetic collagen analogs is presented in
the following paper.53 Among the resonances of the triple-
helical set, the high field δ-H of the proline residue at 3.2 ppm
is well resolved and not overlapped with any resonances of the
denatured conformations. The resonance at 3.2 ppm (Figure
5) can therefore be used unambiguously to identify the triple-
helical structure.53 For KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3 (D2O,
19 °C, 2 mg/mL), the peak at 3.2 ppm is strong, but after heating
to 75 °C it almost disappears, indicating denaturation of the
triple-helical structure which is fully consistent with the melting
transition measurements (Table 2).
As shown in Figure 4A, the melting temperature of Ac-(Gly-
Pro-Hyp)5-NH2, in H2O, increased from 17 to 31 °C when the
concentration was varied from 0.1 to 6.1 mg/mL. The relation-
ship between melting temperature and concentration is provided
in Figure 4B. Two linear relationships are observed. In the
lower concentration range, the melting temperature increases
significantly with increasing concentration, with a slope of 8.2
°C/mg. However, when the concentration is higher than ca. 1
mg/mL, the melting temperature increases with concentrations
with a slope of only ca. 1 °C/mg. The intersection of these
two slopes defines a critical triple-helical concentration (CTHC).
For Ac-(Gly-Pro-Hyp)5-NH2, in H2O, the CTHC is ca. 1 mg/
mL. On the other hand, our control experiments show that the
concentration dependence of the melting temperature of KTA-
[Gly-(Gly-Pro-Hyp)5-NH2]3 is small at high dilution; the melting
temperature increases ca. 1 °C from 0.2 to 1.5 mg/mL. This
dependence is similar to that of Ac-(Gly-Pro-Hyp)5-NH2 above
its CTHC.
In addition to stabilizing triple helices, higher concentrations
for the single-chain peptides also increase the percentage of
triple-helical conformations in solution. This is evident for Ac-
(Gly-Pro-Hyp)5-NH2, in H2O, where the magnitude of melting
transition range increases with increasing concentration (Figure
4A). The magnitude of melting transition is defined as the
difference in specific rotations immediately before and im-
mediately after the transition range. The larger the transition
magnitude, the greater the percentage of triple-helical conforma-
tions. The percentage of triple-helical structures in H2O for
Ac-(Gly-Pro-Hyp)5-NH2 is close to a maximum when the
concentration is greater than its CTHC. This is demonstrated
by the similar transition magnitudes between Ac-(Gly-Pro-
Hyp)5-NH2, at concentrations higher than its CTHC (Figure 4A),
and KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3, in H2O (Figure 3B). Our
NMR results show that the average number of Pro residues in
each chain in a triple-helical environment for both KTA-[Gly-
(Gly-Pro-Hyp)5-NH2]3 and Ac-(Gly-Pro-Hyp)5-NH2 in H2O (ca.
2 mg/mL) is 4 (out of 5 total Pro residues on each chain), which
Conclusions
The chemistry presented in this paper describes a unique
synthetic approach to collagen-based polypeptides. The as-
sembling of peptide chains and the coupling of the template to
these chains are both carried out on a resin. It is necessary to
insert glycine residues as spacers between the template and the
peptide chains considering both the steric effects involved in
the synthesis and the conformational requirements for triple-
helical conformations.

Collagen-Based Polypeptides. 2
J. Am. Chem. Soc., Vol. 118, No. 43, 1996 10357
Scientific. Triethylamine, Pd/C, benzyl chloride, and p-nitrophenol
were purchased from Aldrich. HOBt, TFA (HPLC-grade), and HCl/
dioxane (4 N) were purchased from Chem-Impex International.
General Procedures. All reactions in solution were monitored by
thin-layer chromatography (TLC) carried out on precoated silica gel
60F-54 plates (Merck) using various solvent systems. Compounds were
visualized by UV illustration or ninhydrin or bromocresol spray
reagents. Silica gel 60 (Merck, 230-400 mesh ASTM) was used for
column chromatography. Two HPLC instruments were used to analyze
and purify the products. The first was a Waters (510 pump, 484
detector) system. The second was the Waters MILLENNIUM 2010
system (715 Ultra WISP sample processor, 996 photodiode array
detector, two 510 pumps) with a NEC PowerMate 486/33I computer.
Solvents used in HPLC included (A) H2O with or without 0.1% TFA
and (B) CH3CN with or without 0.1% TFA. The flow rate was 10
mL/min for the preparatory column (Vydac, C-18, 25 × 2.2 cm), 4
mL/min for the semipreparatory column (Vydac, C-18, 25 × 1.0 cm),
and 1.0 mL/min for the analytical column (Vydac, C-18, 25 × 0.46
cm).
To allow for proper equilibration of triple-helix formation, all
solutions for biophysical studies were stored in a refrigerator (ca. 4
°C) for at least 24 h prior to each experiment and for another 2 h at the
specified temperature before acquiring data. CD measurements were
carried out on a modified Cary 61 spectropolarimeter which was
modified by replacing the original Pockel cell with a 50 kHz
photoelastic modulator (Hinds International FS-5/PEM-80). The
original Cary linear polarizer was replaced with a MgF2 linear polarizer
supplied by AVIV, Inc. An EGG Princeton Applied Research model
128A lock in amplifier was used to integrate the phase-detected output
of the original end-on PMT and preamp. System automation, multiple
scan signal averaging, and base line subtraction were accomplished
with an AT286 PC interfaced directly to both the Cary 61 and the 128A
amplifier. The system software and custom hardware interfaces were
designed by Allen MicroComputer Services Inc. and the UC San Diego
Department of Chemistry & Biochemistry Computer Facility. The CD
spectra were obtained using a 0.02 or 0.05 cm path length cell by signal
averaging 10 scans from 190 to 300 nm at a scan speed of 1.0 nm/s.
The UV melting curves were constructed using a Cary-1E UV
spectrometer. Data were collected at 223 nm with a heating rate of
0.2 °C/min. Optical rotations were measured with a Perkin-Elmer 241
polarimeter equipped with a model 900 isotemp refrigerator circulator
(Fisher Scientific). Data were collected at 365 nmHg. At each
temperature point, the samples were allowed to equilibrate until the
optical rotation was time independent. Triple-helix denaturation-
reformation experiments were carried out by optical rotation measure-
ments. The samples were thermally denatured in an oven at temper-
atures significantly higher than the melting temperatures, for a period
of 2 h. The triple-helix reformation was followed by monitoring the
optical rotations at a temperature lower than the melting temperature,
at different time intervals. The transfer of samples from the oven to
the polarimeter was done in less than 1 min.
Figure 5. NMR spectra of KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3 in D2O
at 19 and 75 °C. The sample concentration was 2 mg/mL. Gly (U)
and Pro (U) represent the signals of Gly and Pro which are not in the
triple-helical environment (unassembled); Gly (A) and Pro (A) denote
the signals of Gly and Pro which are in the triple-helical array
(assembled). The intensities of Gly (A) and Pro (A) are proportional
to the percentage of triple-helical conformations in solution.
A CD parameter, the Rpn, was introduced as an empirical
criterion to establish the presence of triple-helical conformations
in solution. It was used to detect the formation of a triple-
helical conformation on increasing the length of the peptide
chain. For collagen analogs composed of Gly-Pro-Hyp se-
quences, triple-helix formation is always accompanied by a sharp
increase in Rpn values. A critical triple-helical concentration
(CTHC) of ca. 1 mg/mL in H2O was observed for the single-
chain acetyl-terminated analog Ac-(Gly-Pro-Hyp)5-NH2. Above
the CTHC, its triple-helical percentage in solution is close to
that of the corresponding template-assembled analog KTA-[Gly-
(Gly-Pro-Hyp)5-NH2]3, which is completely triple helical in H2O
according to NMR spectroscopy. The single-chain structures
Ac-(Gly-Pro-Hyp)n-NH2 (n ) 6, 9) are completely triple helical
even at a concentration of 0.02 mg/mL, in H2O.
The integrated biophysical studies utilizing CD, UV, and
optical rotation measurements demonstrate that the KTA
template, with glycine residue spacers, can induce and facilitate
triple-helix formation and also substantially stabilize triple-
helical structures. Specifically, the KTA template reduces the
minimal chain length necessary to form a triple-helical structure
to only three Gly-Pro-Hyp repeats. In addition, the KTA
template significantly increases the melting temperatures of
triple-helical structures. Furthermore, triple-helix formation for
the template assembled collagen analogs is instantaneous (<1
min), while some time is required for triple-helix formation of
the single-chain acetyl analogs.
NMR spectroscopy was used to verify the structures of intermediates
and final products and to establish the conformations of KTA-[Gly-
(Gly-Pro-Hyp)5-NH2]3. For some compounds, 2-D NMR spectra were
measured to determine peak assignments. These spectra were obtained
on either a Bruker AMX 500 MHz spectrometer or a 360 MHz
spectrometer assembled in house with a Techmag pulse programmer
and digitizer and an Oxford Instruments superconducting magnet. For
conformational analysis, the NMR sample was prepared by dissolving
KTA-[Gly-(Gly-Pro-Hyp)5-NH2]3 in D2O to give a concentration of 2
mg/mL. The sample was then stored in a refrigerator (4 °C) for at
least 24 h and equilibrated for another 2 h at the specified temperature
before data acquisition. NMR spectra were recorded with a spectral
width of 7507.507 Hz and a time domain of 16K. For each spectrum,
80 scans were acquired with a relaxation delay of 1.3 s. An exponential
multiplication was used as the window function for the processing.
Mass spectra were obtained at UV Riverside and the Scripps Research
Institute. Fast atom bombardment (FAB), electrospray ionization (ESI),
or matrix-assisted laser desorption ionization (MALDI) methods were
used to verify product structures.
Experimental Section
Materials. All chiral amino acids used were of the L-configuration.
Protected amino acids, EDC, and DCC were purchased from Bachem.
Deuterated solvent (D2O) was purchased from Cambridge Isotope Labs.
Reagent-grade and HPLC-grade solvents (DCM, DMF, H2O, CH3CN,
CHCl3, MeOH, EtOAc, hexane, and THF) were purchased from Fisher
General Solid Phase Synthesis. A Boc chemistry approach was
used in the general solid phase synthesis. An MBHA resin‚HCl
(Bachem) (200-400 mesh, 1% DVB, 0.45 mmol/g substitution) was

10358 J. Am. Chem. Soc., Vol. 118, No. 43, 1996
Feng et al.
swollen in DCM for 2 h and washed with a solution of 10% TEA in
DCM followed by washing with DCM. A solution of 25% DMF in
DCM was used as the solvent (DMF was used to help dissolve the
HOBt). DIC (2 equiv) and HOBt (2 equiv) were used as the coupling
reagents along with 1.2-1.5 equiv of Boc-Gly-Pro-Hyp(OBz)-OH.
After the completion of each coupling step, the solution was removed
by filtration and the resin (with peptides attached) was washed with
MeOH (1×) and then DCM (3×). The N-terminus was deprotected
(removing Boc protection group) using a solution of 30% TFA in DCM
(0.5-1.0 mL of anisol was added) for 0.5 h. After the deprotection,
the solution was removed by filtration and the resin was washed with
MeOH (1×) and then DCM (2×). The resin was neutralized by
washing with a solution of 10% TEA in DCM (2×) and DCM (2×).
The peptide chain was then built up by consecutive coupling reactions
to achieve the desired length. To prepare the acetyl compound, the
N-terminal amine was acetylated using acetic anhydride in a solution
of 5% TEA in DCM. To obtain the template-assembled compounds,
the templates were attached to the free N-termini of the peptide chains
using DIC and HOBt as the coupling reagents. The resin with acetyl-
terminated or template-assembled peptides was then washed several
times with DCM and dried in vacuum overnight. The peptides were
cleaved from the resin using standard HF cleavage methods with an
apparatus manufactured by Immuno-Dynamics Inc. The HF cleavage
reactions were allowed to proceed for 0.5-1 h at -5-0 °C in the
presence of anisol (1-3 mL depending on the amount of the resin).
The resulting mixture (resin and product) was washed (3×) with either
anhydrous Et2O or a mixture of Et2O and hexane on a sintered glass
filter. The product was then separated from the resin by extracting
with H2O.
KTA-(Gly-OBz)3. A solution of the Kemp triacid (KTA) (1.0 g,
3.9 mmol), tosyl salt of Gly-OBz (6.5 g, 19 mmol), and HOBt (2.6 g,
19 mmol) in DMF (50 mL) was cooled to 0 °C. TEA (5 mL, 36 mmol)
was then added slowly. After the solution was stirred for 5 min, EDC
(3.7 g, 19 mmol) was added and the stirring was continued for 4 h, at
room temperature. The DMF was distilled under reduced pressure,
and the remaining mixture was poured into EtOAc (300 mL). The
EtOAc solution was washed with H2O (2 × 20 mL), saturated NaHCO3
(2 × 20 mL), brine (20 mL), saturated NaHSO4 (2 × 20 mL), and
brine again until the brine layer was neutral. The organic layer was
dried with Na2SO4, and the solvent was removed under reduced pressure
to obtain the crude product. Column chromatography using silica gel
(EtOAc/hexane, 3:2) was carried out to obtain KTA-(Gly-OBz)3 as a
white powder (2.65 g, 97.8%): Rf ) 0.55 (EtOAc/hexane, 3:1).
KTA-(Gly-OH)3. Nitrogen gas was passed through a solution of
KTA-(Gly-OBz)3 (2.65 g, 3.8 mmol) and MeOH (200 mL) for 10 min
in order to remove the dissolved air. Then, 10% Pd/C (0.2 g) was
added, and hydrogen was led into the solution. The hydrogenation
was continued until the consumption of hydrogen stopped. The catalyst
was removed by filtration, and MeOH was distilled under reduced
pressure to give the product as a white powder (1.55 g, 97%): Rf )
0.05 (CHCl3/MeOH/acetic acid, 85:15:3); analytical HPLC profile
showed a single homogeneous peak with tR ) 14 min (0.1% TFA in
the elution solvents, 10-20% B, 30 min); FAB-MS (M + H) calcd
for C18H27N3O9 430, obsd 430; 1H-NMR (360 MHz, DMSO-d6, 20 °C)
δ 7.93 (s, 3H, NH), 3.64 (s, 6H, Gly-R), 2.54 (d, 3H, CH2-equatorial),
1.24-1.18 (m, 12H, 9H for CH3, 3H for CH2-axial).
NH2 as a white powder (45 mg, 46%): analytical HPLC gave a
homogeneous single-peak chromatogram, tR ) 5.7 min (3-30% B in
25 min); FAB-MS accurate mass calcd for C14H22N4O5 327.1668, obsd
327.1678; 1H-NMR (500 MHz, D2O, using H2O for obtaining NH
signals, 27 °C, assignment by DQF-COSY method) δ 8.14 (s, 1H, Gly-
NH), 7.75 (s, 1H, C-terminal NH2), 7.01 (s, 1H, C-terminal NH2), 4.95
(dd, 0.15H, Pro-R), 4.74 (m, covered in H2O signal, Pro-R), 4.62 (m,
1H, Hyp-γ), 4.53 (t, 1H, Hyp-R), 4.15 (d, 1H, Gly-R-l, 4.00 (d, 1H,
Gly-R-h), 3.88 (d, 1H, Hyp-δ-l), 3.81 (dd, 1H, Hyp-δ-h), 3.60 (m, 2H,
Pro-δ), 2.34 (m, 2.1H, Hyp-â-l, Pro-â-l), 2.06 (m, 5.6 H, Hyp-â-h,
Pro-γ, acetyl CH3), 1.93 (m, 1.3H, Pro-â-l).
Ac-(Gly-Pro-Hyp)n-NH2 (n ) 3, 5, 6, 9). The syntheses of these
compounds are similar to that of Ac-Gly-Pro-Hyp-NH2. The detailed
procedures and the physicochemical results are provided in the
Supporting Information.
KTA-[Gly-Gly-Pro-Hyp-NH2]3. Boc-Gly-Pro-Hyp(OBz)-MBHA
(0.3 mmol based on resin substitution level) was prepared following
the procedures described in the General Solid Phase Synthesis section.
The Boc group was removed using a solution of 30% TFA in DCM
(15 mL) (1.0 mL of anisol was added as scavenger). The resin was
washed using DCM (2 × 10 mL), MeOH (2 × 10 mL), and DCM (2
× 10 mL) followed by a solution of 10% TEA in DCM (2 × 15 mL)
and DCM (2 × 15 mL) to give Gly-Pro-Hyp(OBz)-MBHA. KTA-
(Gly-OH)3 (35 mg, 0.08 mmol) and HOBt (50 mg) were added to the
reaction vessel, and a solution of 25% DMF in DCM (15 mL) was
used as the solvent (DMF was added to help dissolve the HOBt). Then,
DIC (1.0 M) in DCM (4 mL) was added. The Kaiser test showed the
absence of free amines after 3 days. The resin was washed with DCM
and MeOH and dried in vacuum overnight. The HF cleavage methods
were carried out to remove the peptide from the resin. The resulting
mixture of resin and product was washed with a mixture of hexane
and Et2O (3×). The product was separated from the resin by extracting
with H2O. Preparatory RP-HPLC was carried out to give the final
product as a white powder (55 mg, 56% yield based on the template
used): analytical HPLC profile showed a single homogeneous peak
with tR ) 13.2 min (no TFA, 7-50% B, 25 min); ESI-MS (M + Na)
1
calcd for C54H81N15O18 1251, obsd 1251; H-NMR (500 MHz, D2O,
27 °C, using H2O at 5 °C to obtain NH signals, DQF-COSY) δ 8.84
(m, 3H, Gly-NH), 8.57 (m, 3H, spacer Gly-NH), 7.89 (m, 3H, terminal
NH2), 7.33 (m, 3H, terminal NH2), 4.61 (m, 3H, Hyp-γ), 4.49 (t, 3H,
Hyp-R), 4.06 (q, 6H, Gly-R), 3.86-3.70 (m, 12H, spacer Gly-R, Hyp-
δ), 3.67-3.54 (m, 6H, Pro-δ), 2.65 (d, 3H, methylene-equatorial),
2.37-2.26 (m, 6H, Pro-â, Hyp-â), 2.10-2.00 (m, 9H, Pro-γ, Hyp-â),
1.97-1.90 (m, 3H, Pro-â), 1.33 (d, 3H, methylene-axial), 1.26 (s, 9H,
methyl).
KTA-[Gly-(Gly-Pro-Hyp)n-NH2]3 (n ) 3, 5, 6). The syntheses of
these compounds are similar to that of KTA-[Gly-Gly-Pro-Hyp-NH2]3.
The detailed procedures and the physicochemical results are provided
in the Supporting Information.
Acknowledgment. This project is funded by grants from
the National Science Foundation (NSFDMR-9201133) and Ciba-
Vision Inc. We thank the Scripps Research Institute and the
University of California at Riverside for carrying out MS
spectrometric analyses and Ms. Yonghong Lai for synthesizing
Ac-(Gly-Pro-Hyp)9-NH2. We also wish to thank Dr. Elizabeth
Jefferson for her critical comments and discussion.
Ac-Gly-Pro-Hyp-NH2. Boc-Gly-Pro-Hyp(OBz)-MBHA (0.3 mmol
based on the resin substitution) was prepared following the procedures
described in the General Solid Phase Synthesis section. The Boc group
was removed using a solution of 30% TFA in DCM (15 mL), and
anisol (1 mL) was added as scavenger. The resin was washed with
DCM (2 × 10 mL), MeOH (2 × 10 mL), and DCM (2 × 10 mL)
followed by 10% TEA in DCM (2 × 15 mL) and DCM (2 × 15 mL).
The N-termini were acetylated by treatment with a solution of acetic
anhydride (0.5 mL) in DCM (15 mL) with 5% TEA to give Ac-Gly-
Pro-Hyp(OBz)-MBHA. After carrying out the HF cleavage, the crude
material (72 mg) was obtained after lyophilizing. Preparatory RP-
HPLC was carried out to obtain the final product Ac-Gly-Pro-Hyp-
Supporting Information Available: Synthesis procedures,
physicochemical data of all collagen-based structures, and 1-D
1H-NMR spectra, mass spectra, and analytical RP-HPLC profiles
for some key intermediates and the acetyl-terminated and
template-assembled collagen-based polypeptides (39 pages). See
any current masthead page for ordering and Internet access
instructions.
JA961260C