Cis-trans proline isomerization effects on collagen triple-helix stability are limited

Article abstract of DOI:10.1021/ja904177k

We investigated the effect of restricting cis-trans proline isomerization on collagen triplehelix stability. The Pro residues at the Xaa and Yaa positions of an (Xaa-Yaa-Gly) triplet were replaced by a Pro-trans-Pro alkene isostere in the host-guest pepti

Full text of DOI:10.1021/ja904177k

Published on Web 09/02/2009
Cis-Trans Proline Isomerization Effects on Collagen
Triple-Helix Stability Are Limited
Nan Dai and Felicia A. Etzkorn*
Department of Chemistry, Virginia Tech, Blacksburg, Virginia 24061
Received May 29, 2009; E-mail: fetzkorn@vt.edu
Abstract: We investigated the effect of restricting cis-trans proline isomerization on collagen triple-
helix stability. The Pro residues at the Xaa and Yaa positions of an (Xaa-Yaa-Gly) triplet were replaced
by a Pro-trans-Pro alkene isostere in the host-guest peptide, H-(Pro-Pro-Gly)10-OH. The resulting
alkene isostere peptide had a Tm value 53.6 °C lower than that of the control peptide. The Pro-trans-
Pro alkene isostere peptide had a Tm value 3.9 °C higher than that of the previously reported Pro-
trans-Gly alkene isostere peptide that did not involve cis-trans Pro isomerization (Jenkins, C. L.;
Vasbinder, M. M.; Miller, S. J.; Raines, R. T. Org. Lett. 2005, 7, 2619-22). Thus, single cis-trans
proline amide isomerization alone has limited contribution to the overall stability of the collagen triple
helix. Since collagen has a high content of imino acid residues, the cumulative effects of cis-trans
isomerization may be quite significant. The peptide containing the Pro-trans-Pro isostere was significantly
less stable than the previously reported Gly-trans-Pro alkene isostere peptide that retained the backbone
interchain hydrogen bond (Dai, N.; Wang, X. J.; Etzkorn, F. A. J. Am. Chem. Soc. 2008, 130,
5396-5397), which confirms that direct interchain backbone hydrogen bonding is a major force for
stabilizing the collagen triple helix.
Collagen is the most important structural protein in verte-
brates, comprising more than 25% of all vertebrate protein.1 It
consists of three parallel left-handed polyproline type II (PPII)
chains2 wound around a common axis.3-6 The primary structure
of collagen can be represented as tripeptide repeating units (Xaa-
Yaa-Gly)n, in which 10% of Xaa residues are proline (Pro) and
10-12% of Yaa residues are 4(R)-hydroxyproline (Hyp).7 Pro
at the Yaa position is also stabilizing,8 and both (Pro-Hyp-Gly)n
and (Pro-Pro-Gly)n triple helixes have been studied exten-
sively.9-11 Mutations of imino acids cause dissociation of the
collagen triple helix and may cause severe damage in collagen-
rich connective tissues, such as bone, tendon, cartilage, ligament,
skin, and blood vessels.2
chain and the CdO of the Xaa residue (usually Pro) in an
adjacent chain is believed to be one of the major factors in
stabilizing the collagen triple helix.12 Moreover, a specific n-to-
π* stereoelectronic interaction has been found to be an additional
force stabilizing the polyproline type II conformation found in
the collagen triple helix.13
The amide bond has partial double bond character, making
it planar.14 The cis content of the amide bond preceding prolyl
residues is much higher than for other amino acid residues.
About 10-30% of prolyl amide bonds in peptides and unfolded
proteins are in the cis conformation.15 In the collagen triple helix,
all of the amide bonds adopt the trans conformation.3 Previous
studies indicated that prolyl cis-trans isomerization is a major
rate-limiting step during the collagen folding process,16,17 but
the thermodynamic effects of proline isomerization have not
yet been demonstrated. The entropic barrier of bringing three
all-trans chains together has been demonstrated to be significant
in the kinetics of collagen triple-helix folding, indicating a
zipper-like folding mechanism.18 Raines and co-workers showed
that the exo-conformation of the Hyp five-membered ring
increases the trans content of the preceding amide bond, which
improves the stability of the resulting triple helix,19,20 Replacing
Hyp with 4(R)-fluoro- or 4(S)-methylproline at the Yaa position
The collagen triple helix has a requisite Gly residue at every
third position; since Gly lacks a CR side chain, it fits into the
very compact core of the collagen triple helix.6,12 An interchain
hydrogen bond between the NsH of the Gly residue of one
(1) Ramachandran, G. N.; Reddi, A. H. Biochemistry of Collagen; Plenum
Press: New York and London, 1976.
(2) Brodsky, B.; Shah, N. K. FASEB J. 1995, 9, 1537–1546.
(3) Ramachandran, G. N.; Kartha, G. Nature 1954, 174, 269–270.
(4) Ramachandran, G. N.; Kartha, G. Nature 1955, 176, 593–595.
(5) Rich, A.; Crick, F. H. C. Nature 1955, 176, 915–916.
(6) Rich, A.; Crick, F. H. C. J. Mol. Biol. 1961, 3, 483–506.
(7) Ramachandran, G. N.; Bansal, M.; Bhatnagar, R. S. Biochim. Biophys.
Acta 1973, 322, 166–171.
(13) Hinderaker, M. P.; Raines, R. T. Protein Sci. 2003, 12, 1188–1194.
(14) Pauling, L.; Corey, R. B.; Branson, H. R. Proc. Natl. Acad. Sci. U.S.A.
1951, 37, 205–211.
(8) Persikov, A. V.; Ramshaw, J. A. M.; Kirkpatrick, A.; Brodsky, B.
Biochemistry 2000, 39, 14960–14967.
(9) Kobayashi, Y.; Sakai, R.; Kakiuchi, K.; Isemura, T. Biopolymers 1970,
9, 415–425.
(15) Brandts, J. F.; Halvorson, H. R.; Brennan, M. Biochemistry 1975, 14,
4953–4963.
(10) Shaw, B. R.; Schurr, J. M. Biopolymers 1975, 14, 1951–1985.
(11) Feng, Y.; Melacini, G.; Taulane, J. P.; Goodman, M. J. Am. Chem.
Soc. 1996, 118, 10351–10358.
(16) Bächinger, H. P.; Bruckner, P.; Timpl, R.; Engel, J. Eur. J. Biochem.
1978, 90, 605–613.
(17) Bächinger, H. P. J. Biol. Chem. 1987, 262, 17144–17148.
(18) Bachmann, A.; Kiefhaber, T.; Boudko, S.; Engel, J.; Ba¨chinger, H. P.
Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 13897–13902.
(12) Bella, J.; Eaton, M.; Brodsky, B.; Berman, H. M. Science 1994, 266,
75–81.
9
13728 J. AM. CHEM. SOC. 2009, 131, 13728–13732
10.1021/ja904177k CCC: $40.75  2009 American Chemical Society

Proline Isomerization Effects on Collagen Triple-Helix Stability
A R T I C L E S
Figure 1. Pro-trans-Pro (E)-alkene isostere containing collagen host-guest peptide 2, control 3, Pro-trans-Gly isostere peptide 4,22 Gly-trans-Pro isostere
peptide 5, and its control 6 (not shown).23 The (E)-alkene isosteres and interchain hydrogen-bonding patterns (---), or lack thereof (X), are shown in the
structures.
a 5 mL polypropylene tube, swelled in CH2Cl2 (2 mL) for 1.5 h,
capped with MeOH (2 mL), and dried in vacuo for 14 h. A sample
of the resin (10 mg) was dried in vacuo for 16 h, stirred in 30%
piperidine in DMF (0.5 mL) for 30 min, and then diluted with 19.5
mL of ethanol. UV analysis showed that the loading was 0.67
mmol/g (56%). The capped 2-chlorotrityl chloride resin was
separated into two polypropylene tubes (110 mg, 0.074 mmol). To
couple single amino acids (Gly and Pro), HATU (85 mg, 0.22
mmol), HOAt (30 mg, 0.22 mmol), Fmoc-amino acid (0.22 mmol),
and DIEA (0.075 mL, 0.43 mmol) were dissolved in DMF (2 mL),
added to the resin, and shaken at 30 °C for 30 min, and the coupling
was repeated. Fmoc-Gly-Pro-Pro-OH (74 mg, 0.15 mmol), HATU
(57 mg, 0.15 mmol), HOAt (20 mg, 0.15 mmol), and DIEA (0.050
mL, 0.29 mmol) were dissolved in DMF (2 mL), and the mixture
was shaken at 30 °C for 2 h. Fmoc-Pro-Ψ[(E)CHdC]-Pro-OH (33
mg, 0.079 mmol), HATU (57 mg, 0.15 mmol), and HOAt (20 mg,
0.15 mmol) were dissolved in DMF (2 mL), and the mixture was
shaken at 30 °C for 2.5 h. The coupling was monitored by reverse-
phase HPLC. After each coupling, the peptide was capped with
10% Ac2O and 10% DIEA in CH2Cl2 (2.5 mL) for 15 min. The
Fmoc group was removed with 30% piperidine in DMF for 20 and
15 min. The peptide was cleaved from the resin by treatment with
2% H2O and 3% triisopropyl silane in TFA (5 mL) for 3.5 h. After
removal of the resin by filtration and concentration of the solution
in vacuo, Et2O (10 mL) was added, and the white solid that
precipitated was collected. Preparative HPLC: 10 µm protein C4
22 × 250 mm column, (A) 0.1% TFA/H2O, (B) 0.1% TFA/CH3CN,
10% to 95% B gradient over 20 min, flow rate 10 mL/min, λ )
210 nm. The pure peptides were obtained as white solids and stored
at -20 °C under N2. Analytical HPLC (Supporting Information):
5 µm protein C4, 4.6 × 250 mm column, 10% to 95% B gradient
over 20 min, flow rate 1.0 mL/min, λ ) 210 nm. Peptide 2 retention
time 8.3 min. Peptide 3 retention time 7.9 min. LC-MS was
performed with a C18, 2.1 × 100 mm column, 5% to 95% CH3CN/
H2O over 35 min, positive electrospray, and a triple-quadrupole
mass separator. The molecular ion peaks were not available because
of the1500 Da cutoff of the detector. Isostere peptide 2: retention
time 8.3 min; MS m/z [M + 2H]2+/2 calcd for C121H175N29O30/2
increased the trans content of the resulting peptides and also
formed very stable triple helixes.20,21 Thus, we hypothesized
that the entropy loss of locking cis-trans isomerization would
thermodynamically stabilize the collagen triple helix.
We now report a dipeptide isostere, Fmoc-Pro-Ψ[(E)CHdC]-
Pro-OH (1), designed to tease out how much cis-trans isomer-
ization affects the stability of a Pro-trans-Pro alkene isostere
containing peptide 2 compared with its host collagen triple-
helix peptide 3 (Figure 1). In this isostere, the (E)-alkene bond
was locked in a trans-Pro conformation, and one key hydrogen
bond between Gly NsH and Xaa CdO was removed, com-
bining the effects of two previously reported isosteres, 4 and 5
(Figure 1).22,23 Those two studies showed that substituting (E)-
alkenes for amide bonds destabilized the collagen triple helix,
but for different reasons.22,23
Experimental Section
Solid-Phase Peptide Synthesis. H-(Pro-Pro-Gly)4-Pro-Ψ[(E)-
CHdC]-Pro-Gly-(Pro-Pro-Gly)5-OH (2) and H-(Pro-Pro-Gly)10-OH
(3) were synthesized by manual solid-phase peptide synthesis.
2-Chlorotrityl chloride resin (400 mg, 1.2 mmol/g) was placed in
(19) Bretscher, L. E.; Jenkins, C. L.; Taylor, K. M.; DeRider, M. L.; Raines,
R. T. J. Am. Chem. Soc. 2001, 123, 777–778.
(20) Shoulders, M. D.; Hodges, J. A.; Raines, R. T. J. Am. Chem. Soc.
2006, 128, 8112–8113.
(21) Holmgren, S. K.; Taylor, K. M.; Bretscher, L. E.; Raines, R. T. Nature
1998, 392, 666–667.
(22) Jenkins, C. L.; Vasbinder, M. M.; Miller, S. J.; Raines, R. T. Org.
Lett. 2005, 7, 2619–2622.
(23) Dai, N.; Wang, X. J.; Etzkorn, F. A. J. Am. Chem. Soc. 2008, 130,
5396–5397.
9
J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009 13729

A R T I C L E S
Dai and Etzkorn
1257.9, found 1257.8; [M + 3H]3+/3 calcd for C121H176N29O30/3
838.9, found 838.8. Control peptide 3: retention time 7.9 min; MS
m/z [M + 2H]2+/2 calcd for C120H174N30O31/2 1266.4, found 1266.3;
[M + 3H]3+/3 calcd for C120H175N30O31/3 844.6, found 844.6.
CD Analysis. The concentrations of peptides in PBS (0.27 mM
KCl, 13.7 mM NaCl, 0.15 mM KH2PO4, 0.81 mM Na2HPO4 in
H2O, pH 7.4) were determined by weight; peptide 2 was 1.32 mg/
mL, and control peptide 3 was 1.26 mg/mL. The peptide solutions
were incubated at 5 °C for more than 48 h. The CD spectra were
obtained with a spectropolarimeter in 0.5 nm increments, 1 nm
bandwidth, and 0.2 cm path length at a scan speed of 100 nm/min.
The spectra were averaged over four consecutive scans. The
solutions were heated from 2 to 70 °C in 3 °C increments with a
10 min equilibration time at each temperature before measurement.
The temperature was measured in the cell. The ellipticity at 227
nm was monitored at each temperature and averaged over four
consecutive measurements. The unfolding process of each peptide
was measured twice, and the data obtained were processed
separately and averaged to obtain the Tm value (Supporting
Information, Figures S1 and S2). The data were fit to the following
equations using SigmaPlot v.10.0:24,25
phase synthesis to reduce the number of steps in the solution-
phase organic synthesis of the isostere. The dipeptide alkene
isostere 1 was coupled with HATU-HOAt, with no external
base added, to prevent isomerization to the R,ꢀ-unsaturated
activated ester,23 and the coupling was monitored by reverse-
phase HPLC. No double bond migration was observed. There
was only one peak with the desired mass in the LC-MS
chromatogram, and the presence of the alkene proton was
confirmed by NMR (Supporting Information). Peptides 2 and
3 were obtained as white solids after purification by RP HPLC
on a protein C4 column.
CD Analysis. The isostere peptide 2 lacked an interchain
backbone hydrogen bond that was present in the Pro-Pro-Gly
host system, so we expected that peptide 2 would have a lower
Tm value than 3. The melting point would not be measurable in
ordinary phosphate-buffered saline (PBS) if Tm was too low.
Trimethylamino N-oxide (TMAO) has been used to study the
unfolding stability of peptides and proteins.22,28,29 TMAO is a
natural osmolyte, which can thermodynamically stabilize folded
proteins against the denaturation process by reducing the degree
of backbone solvation.29-32 Beck et al. found that the Tm value
of unstable collagen peptides was linearly dependent on the
concentration of TMAO.28 Jenkins et al. used the same method
to measure the melting temperatures of an ester isostere and
Pro-trans-Gly isostere 4 collagen peptides by extrapolating the
data to 0 M TMAO.22
For optimal comparison, we measured Tm under exactly the
same conditions that were reported for the Pro-trans-Gly isostere
peptide 4.22 Both the Pro-trans-Pro isostere peptide 2 and control
peptide 3 were dissolved in 9.6 mM PBS buffer pH 7.4 with
different concentrations of TMAO. For the Pro-trans-Pro
isostere peptide 2, the full-range CD spectrum showed a
maximum near 225 nm when the TMAO concentration was
higher than 2.5 M, which indicated the presence of the PPII
helix, and the ellipticity of the peptide increased with increasing
TMAO concentration (Figure 2). When the TMAO concentra-
tion was greater than 2.5 M, peptide 2 showed cooperative
denaturation with increasing temperature, indicating the presence
of a collagen triple helix (Figure 3). Peptide 3 showed
cooperative denaturation with increasing temperature at all
TMAO concentrations (Figure 3).
The Tm values at each TMAO concentration were determined
by fitting the data to a two-state model. Extrapolating the data
to 0 M TMAO gave a melting point of -22.0 ( 1.9 °C for
isostere peptide 2 (Figure 4 and Table 1). The melting point of
control peptide 3, obtained by extrapolating the Tm value to 0 M
TMAO, was determined to be 31.6 °C (Figure 4 and Table 1).
This Tm value was very close to the measured Tm value of
31.5 °C in 9.6 mM PBS buffer with no TMAO present. It was
also close to the literature Tm value of 32.8 °C determined in
the same PBS buffer.33
-∆G0U(T)
(a + b T) + (a + b T) exp
(
)
n
n
d
1 + exp
T
d
RT
Θ )
-∆G0U(T)
(
)
RT
T
∆G0U(T) ) ∆H0(Tm) 1 -
- ∆C0p (Tm - T) + T ln
(
)
( )
[
]
Tm
Tm
∆C0p ) 0
Results
Synthesis. The Pro-trans-Pro isostere 1 was designed with
the (S,R)-stereochemistry to mimic the all-L natural amino acids
found in collagen. Due to the challenges of the synthetic
chemistry, and in order to compare peptide 2 directly with Pro-
trans-Gly isostere peptide 4, we chose the same host peptide,
H-(Pro-Pro-Gly)10-OH (3) (Figure 1).22 The Pro-trans-Pro alkene
isostere 2 eliminated cis-trans isomerization by locking one
Pro amide bond in the trans conformation, and it also lacked
an essential Gly NsH to Xaa Pro CdO hydrogen bond.
Our method for the stereoselective synthesis of the Ser-Ψ[(E)-
CHdC]-Pro and Gly-Ψ[(E)CHdC]-Pro isosteres has been
reported.23,26 Dipeptide isostere 1 has a structure similar to that
of these compounds, so we used our well-developed method
for the synthesis. The synthesis of a Pro-trans-Pro (E)-alkene
isostere with slight variations from our method has been
reported, but the key Ireland-Claisen rearrangement was the
same.27 The R-hydroxycarboxylic acid intermediate was reduced
to the alcohol and crystallized to prove the stereochemistry,27
while we used NMR to establish the stereochemistry. The details
of the synthesis and characterization of 1 are included in the
Supporting Information.
Discussion
Peptide 2, which contains the Pro-trans-Pro alkene isostere,
and control peptide 3 were synthesized by coupling Fmoc-Pro-
Pro-Gly-OH tripeptide units in solid-phase peptide synthesis.23
The Gly residue of the isostere triplet was coupled during solid-
The Gly-Pro isostere peptide 5 showed a ∆Tm of -21.2 °C,
the Pro-Pro isostere peptide 2 showed a ∆Tm of -53.6 °C,
and the Pro-Gly isostere peptide 4 showed a ∆Tm of -57.5 °C
compared to each of their control peptides, respectively (Table
(24) Becktel, W. J.; Schellman, J. A. Biopolymers 1987, 26, 1859–1877.
(25) Horng, J. C.; Hawk, A. J.; Zhao, Q.; Benedict, E. S.; Burke, S. D.;
Raines, R. T. Org. Lett. 2006, 8, 4735–4738.
(28) Beck, K.; Chan, V. C.; Shenoy, N.; Kirkpatrick, A.; Ramshaw, J. A.;
Brodsky, B. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 4273–4278.
(29) Baskakov, I.; Bolen, D. W. J. Biol. Chem. 1998, 273, 4831–4834.
(30) Liu, Y.; Bolen, D. W. Biochemistry 1995, 34, 12884–12891.
(31) Wang, A.; Bolen, D. W. Biochemistry 1997, 36, 9101–9108.
(32) Bolen, D. W.; Baskakov, I. V. J. Mol. Biol. 2001, 310, 955–963.
(26) Wang, X. J.; Xu, B.; Mullins, A. B.; Neiler, F. K.; Etzkorn, F. A.
J. Am. Chem. Soc. 2004, 126, 15533–15542.
(27) Bandur, N. G.; Harms, K.; Koert, U. Synlett 2005, 2005, 773–776.
9
13730 J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009

Proline Isomerization Effects on Collagen Triple-Helix Stability
A R T I C L E S
Figure 2. Full-range CD spectra. (a) Isostere peptide 2 (0.99 mg/mL in 3.25 M TMAO, 1.3 mg/mL for the rest) in PBS buffer (9.6 mM, pH 7.4) at 3 °C.
Blue, 2.5 M TMAO; red, 3.0 M TMAO; yellow, 3.25 M TMAO; green, 3.5 M TMAO. (b) Control peptide 3 (1.3 mg/mL) in PBS buffer (9.6 mM, pH 7.4)
at 3 °C. Green, 0 M TMAO; red, 0.5 M TMAO; blue, 1.0 M TMAO.
Figure 3. Measurement of Tm: CD molar ellipticity-temperature dependence in different concentrations of TMAO. (a) Isostere peptide 2. Yellow, 3.5 M
TMAO (1.3 mg/mL); blue, 3.25 M TMAO (0.99 mg/mL); red, 3.0 M TMAO (1.3 mg/mL); green, 2.5 M TMAO (1.3 mg/mL). (b) Control peptide 3
(concentration 1.3 mg/mL). Red, 1.0 M TMAO; blue, 0.5 M TMAO; green, 0 M TMAO.
The Pro-Gly secondary amide bond naturally exists almost
exclusively (99.9+%) in the trans conformation, so the Pro-
trans-Gly alkene isostere 4 did not have any effect on cis-trans
isomerization (Figure 1).22,34 This isostere does not contain the
N-H of Gly, so one interchain backbone hydrogen bond was
missing.22 The Pro-trans-Pro isostere 2 locked one prolyl amide
bond in the trans conformation and showed a 3.9 °C higher
∆Tm value than for 4 (Table 1). This difference in stability is
close to the ∆Tm value of +3.5 °C for replacing a Pro residue
with a Hyp in the middle of the host-guest peptide, H-(Pro-
Pro-Gly)9-OH, in which the trans isomer content was increased
by the exo-puckering of Hyp.35 We believe this +3.9 °C
difference can be attributed to the elimination of cis-trans
Figure 4. Dependence of Tm of peptide isostere 2 and control peptide 3
on TMAO concentration. Blue diamonds, Pro-trans-Pro isostere peptide 2;
red squares, control peptide 3; red circle, measured Tm (not included in
linear fit).
isomerization by locking the prolyl amide bond at the Xaa
position to the trans conformation. This result indicates that
locking one prolyl amide in the trans conformation has a small
contribution to the overall collagen triple-helix stability. In the
R1 chain of human type I collagen, there are 278 imino acid
residues; thus, the cumulative effects of cis-trans isomerization
may be quite significant.36,37
Table 1. Melting Point Comparison of Alkene Isostere Peptides
and Their Controls
peptide
Tm (°C)
∆Tm (°C)
-53.6
Both Gly-trans-Pro (5)23 and Pro-trans-Pro (2) isostere
peptides eliminated a cis-trans isomerization by locking one
prolyl amide bond in the trans conformation with an (E)-alkene
bond (Figure 1). Peptide isostere 5 retained all interchain
Pro-trans-Pro peptide 2
Pro-Pro-Gly control 3
-22.0a
+31.6a
-24.7a
+28.3b
+50.0b
Pro-trans-Gly peptide 422
Gly-trans-Pro peptide 523
Pro-Hyp-Gly control 623
-57.5
-21.7
a 9.6 mM PBS pH 7.4, varying [TMAO]. Tm value obtained by
(33) Jenkins, C. L.; Bretscher, L. E.; Guzei, I. A.; Raines, R. T. J. Am.
Chem. Soc. 2003, 125, 6422–6427.
extrapolation to 0 M TMAO. b 10 mM PBS pH 7.0.
(34) Scherer, G.; Kramer, M. L.; Schutkowski, M.; Reimer, U.; Fischer,
G. J. Am. Chem. Soc. 1998, 120, 5568–5574.
1). This order of melting point difference from each control
(∆Tm) is in accord with the entropic cost of folding due to
cis-trans isomerization and the loss of interchain hydrogen
bonds.
(35) Jiravanichanun, N.; Hongo, C.; Wu, G.; Noguchi, K.; Okuyama, K.;
Nishino, N.; Silva, T. ChemBioChem 2005, 6, 1184–1187.
(36) Fietzek, P. P.; Kuhn, K. Mol. Cell. Biochem. 1975, 8, 141–157.
(37) Kadler, K. Protein Profile 1995, 2, 491–619.
9
J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009 13731

A R T I C L E S
Dai and Etzkorn
backbone hydrogen bonds, while peptide isostere 2 eliminated
the CdO of the Xaa Pro residue that is required for a key
interchain backbone hydrogen bond with the N-H of Gly in
another chain. Peptide 2 was less stable than 5 by 31.9 °C, which
we attribute to the missing interchain backbone hydrogen bond
(Table 1).23 Both peptides 2 and 5 were less stable than their
control peptides, 3 and 6, respectively.23 This shows that intrinsic
properties of the amide bond, including n-to-π* interactions,13
have a major influence on the stability of the collagen triple
helix.
Jenkins et al. also studied the peptide with an ester bond
replacing one Pro-Gly amide bond.22 This ester isostere also
removed one interchain backbone hydrogen bond due to the
removal of a Gly N-H. The Tm value of the ester peptide was
22.1 °C lower than that of the control peptide (Pro-Pro-Gly)10.22
By comparison with the -57.5 °C ∆Tm value of the Pro-Gly
isostere peptide 4, the importance of the hydrogen-bonded
carbonyl groups is shown to be significant. Our results confirm
that the backbone interchain hydrogen bond is a major factor
in stabilizing the collagen triple helix.
lower than its control peptide 3. Comparison with our previously
reported Gly-trans-Pro isostere peptide 5 confirmed that the
backbone interchain hydrogen bond between Gly NsH and Xaa
CdO of the adjacent chain is one of the major forces in
stabilizing the collagen triple helix.23 Because alkene isosteres
are all destabilizing, intrinsic properties of the amide bond,
including n-to-π* interactions, are also significant.13 The Pro-
trans-Gly isostere peptide 4 reported by Jenkins et al.22 was
3.9 °C less stable than our Pro-trans-Pro isostere peptide 2. Thus,
cis-trans isomerization appears to have a small effect on triple-
helix stability. Even so, the small contribution from cis-trans
isomerization may be cumulatively significant, considering the
large number of prolyl residues in full-length natural collagen.
Acknowledgment. We thank Dr. Matthew Shoulders (Univer-
sity of Wisconsin, Madison) for helpful discussions, Dr. Mehdi
Ashraf-Khorasani for LC-MS data acquisition, Virginia Tech
Department of Biochemistry for the use of the CD instrument, ACS-
PRF for grant 44314-AC7, and the NSF for grant CHE-0749061.
Supporting Information Available: Experimental procedures
and characterization for the synthesis of 1; NMR spectra and
HPLC chromatograms for 1-3. This material is available free
of charge via the Internet at http://pubs.acs.org.
Conclusions
Replacing any amide within collagen triple-helix peptides
with an (E)-alkene bond causes a significant decrease in stabil-
ity.22,23 We replaced one Pro-trans-Pro in a collagen peptide
with an (E)-alkene isostere and found that it melted 53.6 °C
JA904177K
9
13732 J. AM. CHEM. SOC. VOL. 131, NO. 38, 2009