A 310-helix single turn enforced by crosslinking of lysines with 1,1′-ferrocenedicarboxylic acid

Article abstract of DOI:10.1016/j.jorganchem.2008.11.055

Prior work has shown that covalently linking the side chains of amino acids in the i and i+3 and i and i+4 positions in a peptide will enforce a helical conformation. In this work the ability of an organometallic entity to enforce a helical conformation i

Full text of DOI:10.1016/j.jorganchem.2008.11.055

Journal of Organometallic Chemistry 694 (2009) 902–907
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
A 310-helix single turn enforced by crosslinking of lysines with
1,10-ferrocenedicarboxylic acid
*
Timothy P. Curran , Emma L. Handy
Department of Chemistry, Trinity College, Hartford, CT 06106-3100, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Prior work has shown that covalently linking the side chains of amino acids in the i and i+3 and i and i+4
positions in a peptide will enforce a helical conformation. In this work the ability of an organometallic
entity to enforce a helical conformation in a peptide was explored. The tetrapeptide Boc-Lys-Ala-Val-
Lys-NHCH3 was prepared, then reacted with 1,10-ferrocenedicarboxylic acid chloride. Reaction of the
lysine side chain amines with the diacid chloride resulted in a metallacyclicpeptide (1) in which the
two lysines are crosslinked via the ferrocene. The solution conformation of the metallacyclicpeptide
(1) was studied using CD and NMR spectroscopy. The NMR methods employed were Karplus analysis
of coupling constants, chemical shift changes of NH protons and ROESY data. The results show that the
metallacyclicpeptide (1) adopts a single turn of the 310-helix conformation.
Received 2 September 2008
Received in revised form 22 November 2008
Accepted 24 November 2008
Available online 6 December 2008
Keywords:
Ferrocene
Peptide
Conformation
Bioorganometallic
Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction
metal–ligand interactions [8–11,17,18]. The connection can also
be made by bridging the side chains of two amino acids with a
Helices are an important structural feature of nearly all known
proteins [1]. In addition to defining the overall tertiary structure of
a protein, helices also possess distinct and important biochemical
functions. This is particularly true with respect to biochemical sig-
nalling pathways that are triggered by protein-protein and
protein–DNA interactions [2,3]. As an example, the binding of a
helical peptide to the protein Bcl-XL can induce apoptosis of cancer
cells [4–6].
Because protein helices can mediate important biochemical
pathways, efforts have been made to excise the helical portion of
a protein and use it in place of the whole protein. Unfortunately,
in most cases, excision of a helical segment from a larger protein
yields a peptide that adopts a random coil conformation [7]; from
these results it has been concluded that constraints imposed by
other parts of the protein are needed to maintain the helical shape.
Synthetic chemists have been able to hold short peptides in
helical conformations by introducing non-natural constraints [8–
11]. These constraints can be located at various places in a peptide
chain. In one approach, a constraint that keeps the initial residues
of a peptide in a helical conformation can be added to either the N-
terminus [12,13] or the C-terminus [14] of the peptide; subsequent
amino acids added to the peptide are then positioned so they prop-
agate the helix. In another approach, a constraint can be added by
linking the side chains of two amino acid residues. This link can be
made using an amide bond [8–11,15], a disulfide [8–11,16], or
covalent spacer [8–11,19]. Linking the side chains forces the inter-
nal portion of the peptide to be helical; those parts of the peptide
adjacent to the constraint are then positioned to propagate the he-
lix. In a third approach, the constraint can be made by replacing the
intramolecular hydrogen bond with a covalent spacer [5,20]. As
with the other approaches, this constraint enforces a helical con-
formation on a small part of the peptide; amino acid residues out-
side the constraint are then positioned to extend the helix.
We are interested in developing a helix constraint that is easy to
introduce using both solution and solid phase peptide synthesis
protocols, and that possesses a unique chromophore. These two
properties, which are not present together in the known helix con-
straints, would allow researchers to quickly generate peptide heli-
ces, and then develop UV–Vis spectroscopic methods for analyzing
the ability of these synthetic helices to bind to biochemical targets.
A constraint employing an organometallic entity would meet these
criteria. Herein we report the use of the ferrocene as a side chain
constraint that can enforce a helical conformation.
2. Results and discussion
2.1. Synthesis of 1
The molecule studied was metallacyclictetrapeptide (1)
(Scheme 1), which has lysines at the 1 and 4 positions, and alanine
and valine at positions 2 and 3. Lysine was chosen as the crosslink-
ing amino acid because it has a reactive functional group (an
amine) and its side chain is long enough to allow for cyclization.
* Corresponding author.
E-mail address: timothy.curran@trincoll.edu (T.P. Curran).
0022-328X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2008.11.055

T.P. Curran, E.L. Handy / Journal of Organometallic Chemistry 694 (2009) 902–907
903
Scheme 1. Synthesis of metallacyclictetrapeptide (1).
In 1 the side chain amines of the two lysines are joined to each
other via amide bonds made to the two carboxyls from 1,10-ferro-
cenedicarboxylic acid. This diacid and its derivatives have been
used by other researchers to constrain peptide conformations
[21–27]. We hypothesized that the ferrocene crosslink would en-
force a helical conformation for the peptide part of 1. From previ-
ous work it is known that helices can be generated by linking
amino acid side chains in the 1 and 4 positions [28–31]. Valine
and alanine were selected because of their demonstrated tendency
to appear in protein helices [1], and because the chemical shifts of
their protons were unlikely to overlap with the amide NH protons
in NMR spectra.
The ferrocene crosslink was made by removal of the two Cbz
groups from 6 using catalytic hydrogenation. The resulting dia-
mine was then reacted with 1,10-ferrocenediacid chloride [32] (un-
der dilute conditions) to yield 1. The metallacyclicpeptide (1) was
purified to homogeneity by flash chromatography and character-
ized by HPLC, 1H NMR spectroscopy, electrospray ionization mass
spectrometry and high-resolution mass spectrometry. Metallacy-
clicpeptide (1) is an amorphous, orange solid. Efforts to crystallize
1 have so far been unsuccessful, so an X-ray crystal structure has
not been obtained.
2.2. Conformational analysis
The synthesis of 1 is outlined in Scheme 1. Commercially avail-
able Boc-Lys(Cbz)-OSu (2) was reacted with N-methylamine to
generate 3. Removal of the Boc group using TFA, followed by reac-
tion of the resulting amine with Boc-Val-OSu yielded dipeptide 4.
Repetition of the same protocol (deprotection of the Boc group fol-
lowed by reaction with Boc-Ala-OSu) generated tripeptide 5. In a
similar way the tetrapeptide 6 was produced by Boc group re-
moval and coupling with 2. In all these reactions the peptide prod-
uct was isolated using an extraction procedure; no further
purification was attempted. The crude products were pure enough
(as judged by TLC) that they could be used in the next step of the
synthesis.
Both NMR and CD spectroscopy were used to determine the
conformation of 1. With CD spectroscopy, it is well known that
helical peptides exhibit spectra in which there are two absorbance
minima located at 208 and 222 nm [33]. To see whether 1 adopts a
helical conformation, the CD spectrum was obtained in CHCl3
(Fig. 1). This spectrum closely matches the classical CD spectrum
for an
a-helix as it shows two absorbance minima at 201 and
220 nm. The overall appearance of the CD spectrum of 1 indicates
that it is helical.
To more firmly establish the helical conformation of 1, proton
assignments were made by analysis of the COSY spectrum of 1 in

904
T.P. Curran, E.L. Handy / Journal of Organometallic Chemistry 694 (2009) 902–907
Fig. 2. Molecular model of compound 1.
angle is ꢀ120° [36]. With both the NMR and modeling data the
dihedral angles are within the range of dihedral angles found for
310-helices seen in protein crystal structures [37].
Fig. 1. The CD spectrum of 1 recorded in CHCl3. The spectrum shows two minima at
201 and 220 nm, which indicates that 1 adopts a helical conformation.
In order to confirm that 1 adopts a 310-helix conformation, the
ROESY spectrum of 1 was obtained in CDCl3. As noted by many
researchers, including Kemp [12,13], Bartlett [14], and Arora [20],
a signature of helices is the observation of NOEs between consec-
utive amide NH protons [dNN(i, i+1)]. The amide NH region portion
of the ROESY spectrum of 1 is shown in Fig. 3. The data shows that
the NH protons in 1 have crosspeaks with the adjacent NH protons.
Thus, the N-terminal lysine NH shows a crosspeak only to the ala-
nine NH. In turn, the alanine NH only shows a crosspeak to the va-
line NH, and the valine NH only shows a crosspeak to the C-
terminal lysine NH, and the C-terminal lysine NH only shows a
crosspeak to the N-methyl amide NH. This pattern of ROESY cross-
Table 1
Chemical shift, JNH–C and dihedral angle (h) data for 1 in CDCl3.
a
H
NH proton
d (ppm)
JNH–CaH (Hz)
ha
hb
Lys (C-terminus)
Val
Ala
7.31
7.11
6.58
5.31
7.5
6.2
3.3
2.8
ꢀ144°
ꢀ137°
ꢀ120°
ꢀ117°
ꢀ154°
ꢀ113°
ꢀ153°
ꢀ93°
Lys (N-terminus)
a
Dihedral angles for 1 calculated from the measured JNH–C
values (see Ref.
H
a
[36]).
b
Dihedral angles for 1 obtained from the computer model of 1 displayed in Fig. 1.
The computer model was generated using Spartan 04.
peaks among the amide NH protons in 1 is consistent with the 310
-
helix conformation shown in Fig. 2.
CDCl3. In this way the NH protons for each individual amino acid in
1 were identified (see Table 1 for selected NH chemical shifts).
There is a wide separation of chemical shifts among the different
NH protons. These chemical shift differences indicate that each
NH is in a distinctly unique environment, as would be expected
if 1 adopts a helical conformation [12,13].
In addition to the detection of close contacts between adjacent
amide NH protons, helical conformations can also be detected
through the observations of NOEs between resonances on amino
acid residues brought into close proximity by the twisting of the
peptide chain. If 1 assumes the conformation shown in Fig. 2, then
the valine residue will be in close proximity to the two end lysine
residues. Fig. 4 shows the crosspeaks to the valine methyl reso-
nance in the ROESY spectrum of 1 in CDCl3. These methyl groups
Having identified each NH resonance, the JNH–C
values for
a
H
each amide proton were measured (Table 1). For a helical peptide,
JNH–C H values usually fall in the range between 3 and 5 Hz [34]. As
show crosspeaks to the valine NH, C H and CbH protons, the NH
a
a
the data shows, the NH from the N-terminal lysine and the NH
from the alanine fall in this range. The valine resonance falls just
outside this range, while the resonance for the C-terminal lysine
is outside the range. The trend is for the JNH–C values to move
and C H protons from the C-terminal lysine, and the C H proton
from the N-terminal lysine. That the valine methyl produces NOEs
a
a
to both lysine C H protons can only be true if 1 adopts a 310-helix
a
conformation.
a
H
away from the 3 to 5 Hz range as the peptide progresses from
the N-terminus to the C-terminus. Using the Karplus equation,
the dihedral angles for the measured coupling constants were cal-
culated, and these values are also given in Table 1 [35].
To gain some insight into the coupling constant data, a model of
1 was constructed using Spartan 04. The calculated dihedral angles
from Table 1 were used to set the starting conformations for the
NH protons in the model. After this, the resulting structure was
minimized. The conformation of 1 that was obtained is shown in
Fig. 2. This structure has the peptide in the conformation of a single
turn of a 310-helix [36].
Further evidence from the ROESY spectrum that supports the
conclusion that 1 adopts a 310-helix conformation comes from
analysis of the crosspeaks of the valine C H protons in the ROESY
a
spectrum of 1. In addition to showing crosspeaks to the valine
methyls, CbH and NH protons, the valine C H also shows cross-
a
peaks to the NH protons from the C-terminal lysine and the N-
methylamide. The presence of an NOE between the valine C H
a
and the N-methylamide NH can only occur if 1 adopts a 310-helix
conformation.
To further confirm that 1 adopts a helical conformation in solu-
tion, the 1H NMR, COSY and ROESY spectra of 1 was recorded in d6-
acetone. The ROESY spectrum of 1 in d6-acetone showed the same
pattern of crosspeaks between the NH protons as was seen in
CDCl3, and the coupling constants for the NH protons were of sim-
ilar magnitudes. Both results show that the conformation of 1 is
unchanged in going from CDCl3 to d6-acetone.
The dihedral angles for each NH–C H in the model of 1 were
a
then measured and are reported in Table 1. In general, the com-
puter model produces slightly larger dihedral angles than those
calculated from the coupling constants and the Karplus equation.
However, the trend in the dihedral angles is the same for the values
derived from the Karplus equation and from the energy minimized
One method that has been successfully used to detect intramo-
lecular hydrogen bonds in peptides is the use of solvent titration
computer model. In an idealized 310-helix, the NH–C H dihedral
a

T.P. Curran, E.L. Handy / Journal of Organometallic Chemistry 694 (2009) 902–907
905
Fig. 3. The amide NH region of the ROESY spectrum of 1 in CDCl3. The chemical shifts of the amide NH protons in 1 determined from the COSY spectrum are: N-methyl
(6.9 ppm), N-terminal Lys (5.3 ppm), Ala (6.6 ppm), Val (7.1 ppm), C-terminal Lys (7.3 ppm), Lys-ferrocene amides (8.0 and 6.2 ppm). The pattern of crosspeaks here is
consistent with a helical conformation for 1.
and N-methyl amide should show little or no change in chemical
shift as the solvent is changed. In contrast, if 1 adopts an a-helical
conformation, then the mainchain NH protons from the N-termi-
nal lysine, the alanine and the valine should show a large chemi-
cal shift change from chloroform to acetone, while only the
mainchain NH protons from the C-terminal lysine and the
N-methyl amide should show little or no change in chemical shift
as the solvent is changed. Thus, the key marker to distinguish an
a
-helix from a 310-helix is the behavior of the valine NH resonance.
A large change in its chemical shift would indicate an -helical
310-helical
a
conformation;
conformation.
a small change would indicate a
Given in Table 2 are the chemical shifts for the NH protons in 1
in both CDCl3 and d6-acetone. The three NH protons starting from
the C-terminus (N-methyl, lysine and valine NH protons) exhibit
very small changes in chemical shift (<0.2 ppm) in going from
CDCl3 to d6-acetone. In contrast, the two NH protons starting from
the N-terminus (alanine and lysine) exhibit a very large change in
chemical shift (>1.3 ppm). This data indicates that the three NH
protons starting from the C-terminus, including the valine NH,
are involved in intramolecular hydrogen bonds, while the two
NH protons starting from the N-terminus are not involved in intra-
Fig. 4. A portion of the ROESY spectrum of 1 in CDCl3 showing the crosspeaks to the
valine methyl resonances.
molecular hydrogen bonds. This behavior is consistent with a 310
helix conformation for 1 as shown in Fig. 2.
-
[38–43]. Unlike chloroform, acetone is an aggressive hydrogen
bond acceptor. Accordingly, the chemical shifts of the NH protons
in 1 will differ greatly between chloroform and acetone if they
interact with the solvent. If, however, an NH proton in 1 is pro-
tected from the solvent by being involved in an intramolecular
hydrogen bond, then the chemical shift of that proton will not
change much in going from chloroform to acetone. If 1 adopts
the 310-helix conformation shown in Fig. 1, then the mainchain
NH protons from the N-terminal lysine and the alanine should
show a large chemical shift change from chloroform to acetone,
while the mainchain NH protons from the valine, C-terminal lysine
Table 2
Chemical shift comparisons for 1 in CDCl3 and d6-acetone.
NH proton
d (ppm) CDCl3
d (ppm) d6-acetone
Dd (ppm)
N-Methyl
Lys (C-terminus)
Val
6.92
7.31
7.11
6.58
5.31
7.08
7.46
7.31
8.05
6.64
+0.16
+0.15
+0.20
+1.47
+1.33
Ala
Lys (N-terminus)

906
T.P. Curran, E.L. Handy / Journal of Organometallic Chemistry 694 (2009) 902–907
The combination of CD, JNH–C H, ROESY, solvent titration and
water bath. To the chilled solution was added 8.0 mL TFA. The
resulting solution stirred for 1.5 h while it slowly warmed to
23 °C. The solvents were evaporated to yield a yellow oil. This oil
was dried under high vacuum to remove any last traces of TFA.
Next, the oil was redissolved in 15 mL CH2Cl2. To this solution
was added 1.714 g (5.459 mmol) of Boc-Val-OSu and 22 mL
(132 mmol) of DIEA. The resulting solution stirred at 23 °C for
4 h. The solvent was evaporated and the residue that remained
was redissolved in 75 mL EtOAc. This solution was washed:
3 ꢁ 75 mL 0.1 M HCl, 3 ꢁ 75 mL saturated NaHCO3, and
1 ꢁ 75 mL brine. The organic layer was dried (MgSO4), filtered
and evaporated to yield 1.756 g (68%) of 4 as a white solid: m.p.
155–158 °C; TLC, Rf 0.38 (EtOAc); ESI-MS M+Na ion calculated for
C25H40N4O6Na, 515 m/z; found, 515.0 m/z; 1H NMR (300 MHz,
CDCl3) d 7.40–7.30 (5H, m), 6.75 (1H, d, J = 7.3 Hz), 6.60 (1H, br
s), 5.20–5.00 (4H, m), 4.44 (1H, m), 3.98 (1H, m), 3.20 (2H, m),
2.81 (3H, d, J = 3.4 Hz), 2.20–1.30 (7H, m), 1.45 (9H, s), 0.99 (3H,
d, J = 7.3 Hz), 0.93 (3H, d, J = 7.3 Hz).
a
molecular modeling data for 1 shows that the ferrocene crosslink
enforces the conformation of a single turn of a 310-helix in the tet-
rapeptide portion of this metallacyclicpeptide. This conclusion is
buttressed by results from other researchers which have shown
that 310-helices can be generated by crosslinking the side chains
of amino acids in the i and i+3 positions in a peptide [30,31]. Since
it is relatively easy using one of the common 20 amino acids to
introduce the ferrocene crosslink, and that the chemical steps to
do so are amenable to solid phase peptide synthesis, and that the
crosslink gives the peptide a unique and useful chromophore, this
new method for generating helices has great potential. Further
studies to determine whether a single turn of an
a-helix can be
generated and to demonstrate that the helix can be propagated be-
yond the single turn are underway.
3. Experimental
3.1. General procedures
3.4. Preparation of Boc-Ala-Val-Lys(Z)-NHMe (5)
Amino acid derivatives were purchased from ChemImpex Inter-
national. Anisole and DMF (dimethylformamide) were purchased
from Sigma–Aldrich. TFA, DIEA, and THF were purchased from Acros
Organics. 1,10-Ferrocenedicarboxylic acid was purchased from
Strem Chemical. CDCl3 and d6-DMSO were purchased from Cam-
bridge Isotope Labs. Silica gel for flash chromatography was pur-
chased from Silicycle. CD spectra were recorded using a Cary-16/
OLIS spectrometer using a 3 mg/mL solution of 1 in CHCl3 in a
quartz cell (0.100 cm pathlength). NMR spectra were obtained on
either a GE Omega 300 MHz instrument or a Bruker AVANCE III
400 MHz instrument. Electrospray mass spectra were obtained on
a LCQ APCI/Electrospray LC MS–MS. Samples for ESI mass spectral
analysis were dissolved in MeOH (approximately 1 mg/mL) in boro-
silicate glass test tubes. High-resolution mass spectra were ob-
A solution of 0.5168 g (1.050 mmol) of Boc-Val-Lys(Cbz)-NHMe
(4) in 10 mL CH2Cl2 and 1.0 mL of anisole was chilled to 5 °C in an
ice water bath. To the chilled solution was added 5.0 mL TFA. The
resulting solution stirred for 1.5 h while it slowly warmed to
23 °C. The solvents were evaporated to yield an oil with a red/or-
ange color. This oil was dried under high vacuum to remove any
last traces of TFA. Next, the oil was redissolved in 10 mL CH2Cl2.
To this solution was added 0.3066 g (1.071 mmol) of Boc-Ala-OSu
and 5.0 mL (30 mmol) of DIEA. The resulting solution stirred at
23 °C for 4 h. The solvent was evaporated and the residue that re-
mained was redissolved in 200 mL CH2Cl2 and 10 mL MeOH. This
solution was washed: 3 ꢁ 200 mL 0.1 M HCl, 3 ꢁ 200 mL saturated
NaHCO3, and 1 ꢁ 200 mL brine. The organic layer was dried
(MgSO4), filtered and evaporated to yield 0.470 g (80%) of 5 as an
off-white solid: TLC, Rf 0.11 (EtOAc); ESI-MS M+Na ion calculated
for C28H45N5O7Na, 586 m/z; found, 585.8 m/z; 1H NMR (300 MHz,
CDCl3) d 7.40–7.30 (5H, m), 6.96 (1H, m), 6.64 (2H, m), 5.10 (2H,
m), 4.92 (2H, m), 4.40 (1H, m), 4.18 (1H, m), 4.06 (1H, m), 3.18
(2H, m), 2.78 (3H, d, J = 4.4 Hz), 2.38 (1H, m), 2.00 (1H, m), 1.90–
1.20 (8H, m), 1.44 (9H, s), 0.98 (3H, d, J = 6.8 Hz), 0.90 (3H, d,
J = 6.8 Hz).
tained on
a JEOL AccuTOF DART mass spectrometer. HPLC
analyses were performed on a Finnigan MAT SpectraSystem HPLC
using a Vydac C18 peptide column. The gradient used started with
80% TFA (0.1%)/20% CH3CN and went to 50% TFA (0.1%)/50% CH3CN
in a 6.0 min span. Next, the gradient went from 50% TFA (0.1%)/50%
CH3CN to 0% TFA (0.1%)/100% CH3CN over the next 6.0 min. The flow
rate was 1.0 mL/min. For the final 8.0 min of the analysis the solvent
remained at 0% TFA (0.1%)/100% CH3CN. The UV–Vis detector
was set at 410 nm. Peptides 3–6 were isolated as crude products
with relatively high purity as judged by TLC. They were not purified;
instead they were taken on to the next step in the synthesis as is.
3.5. Preparation of Boc-Lys(Z)-Ala-Val-Lys(Z)-NHMe (6)
A solution of 0.988 g (1.75 mmol) of Boc-Ala-Val-Lys(Cbz)-
NHMe (5) in 30 mL CH2Cl2 and 1.0 mL of anisole was chilled to
5 °C in an ice water bath. To the chilled solution was added
15 mL TFA. The resulting solution stirred for 1.5 h while it slowly
warmed to 23 °C. The solvents were evaporated to yield an oil
with a red color. This oil was dried under high vacuum to remove
any last traces of TFA. Next, the oil was redissolved in 10 mL
DMF. To this solution was added 0.745 g (1.56 mmol) of Boc-
Lys(Cbz)-OSu (2) and 6.0 mL (36 mmol) of DIEA. The resulting
solution stirred at 23 °C. After 4 h the reaction solution was
poured into 50 mL CH2Cl2 and was washed: 3 ꢁ 100 mL 0.1 M
HCl, 3 ꢁ 100 mL saturated NaHCO3, and 1 ꢁ 100 mL brine. The or-
ganic layer was dried (MgSO4), filtered and evaporated to yield
0.968 g (67%) of 6 as an off-white solid: m.p. 185–190 °C; ESI-
MS M+Na ion calculated for C42H63N7O10Na, 848 m/z; found,
847.6 m/z; 1H NMR (300 MHz, d6-DMSO) d 8.00–7.70 (4H, m),
7.50–7.20 (12H, m), 6.92 (1H, m), 5.02 (4H, m), 4.36 (1H, m),
4.14 (2H, m), 3.87 (1H, m), 2.98 (4H, m), 2.57 (3H, d, J = 4.4 Hz),
1.98 (1H, m), 1.80–1.00 (15H, m), 1.44 (9H, s), 0.82 (3H, d,
J = 7.3 Hz), 0.78 (3H, d, J = 6.8 Hz),
3.2. Preparation of Boc-Lys(Cbz)-NHMe (3)
To a solution of 3.100 g (6.492 mmol) of Boc-Lys(Cbz)-OSu (2)
dissolved in 50 mL CH2Cl2 was added 10 mL of 2.0 M CH3NH2 in
THF. The resulting solution stirred at 23 °C for 4 h. The solvent
was evaporated and the residue that remained was redissolved
in 50 mL EtOAc. This solution was washed: 3 ꢁ 50 mL 0.1 M HCl,
3 ꢁ 50 mL saturated NaHCO3, and 1 ꢁ 50 mL brine. The organic
layer was dried (MgSO4), filtered and evaporated to yield 2.079 g
(82%) of 3 as a white solid: m.p. 64–66 °C; TLC, Rf 0.45 (EtOAc);
ESI-MS M+Na ion calculated for C20H31N3O5Na, 416 m/z; found,
415.7 m/z; 1H NMR (400 MHz, CDCl3) d 7.40–7.30 (5H, m), 6.12
(1H, m), 5.11 (3H, m), 4.85 (1H, m), 4.03 (1H, m), 3.20 (2H, m),
2.82 (3H, d, J = 4.9 Hz), 2.0–1.3 (6H, m), 1.44 (9H, s).
3.3. Preparation of Boc-Val-Lys(Cbz)-NHMe (4)
A solution of 2.079 g (5.290 mmol) of Boc-Lys(Cbz)-NHMe (3) in
15 mL CH2Cl2 and 1.0 mL of anisole was chilled to 5 °C in an ice

T.P. Curran, E.L. Handy / Journal of Organometallic Chemistry 694 (2009) 902–907
907
[3] A. Klug, FEBS Lett. 579 (2005) 892–894.
3.6. Preparation of metallacyclicpeptide (1)
[4] M. Sattler, H. Liang, D. Nettesheim, R.P. Meadows, J.E. Harlan, M. Eberstadt, H.S.
Yoon, S.B. Shuker, B.S. Chang, A.J. Minn, C.B. Thompson, S.W. Fesik, Science 275
(1997) 983–986.
A solution of 0.210 g (0.254 mmol) of 6 in 50 mL MeOH was
placed in a Parr hydrogenation vessel. To this solution was added
0.102 g of 5% Pd–C. The resulting mixture was hydrogenated at
30 psi for 24 h. The catalyst was removed by vacuum filtration
through celite, and the filtrate was evaporated. The solid that re-
mained was redissolved in 100 mL anhydrous DMF and 200 mL
anhydrous CH2Cl2 and was placed in a flask equipped with an addi-
tion funnel. To the flask was added 0.40 mL DIEA (2.4 mmol); to the
addition funnel was added a solution of 0.091 g (0.29 mmol) of 1,10-
ferrocenediacid chloride in 30 mL CH2Cl2. The solution in the addi-
tion funnel was added to the flask dropwise over 1 h. The resulting
solution stirred for 18 h. Evaporation of the solvents yielded a black
solid. Flash chromatography (9:1 EtOAc/MeOH) produced 0.041 g
(20%) of pure 1 as an orange solid: TLC, Rf 0.46 (9:1 EtOAc/MeOH);
HPLC, Rt 6.6 min; ESI-MS M+Na ion calculated for C38H57N7O8FeNa,
818 m/z; found, 818.0 m/z; HRMS M+H ion calculated for C38H58N7-
O8Fe, 796.369705 m/z; found, 796.369385 m/z; 1H NMR (400 MHz,
CDCl3) d 8.02 (1H, m), 7.31 (1H, d, J = 7.5 Hz), 7.10 (1H, d, J = 6.2 Hz),
6.92 (1H, m), 6.58 (1H, d, J = 3.3 Hz), 6.25 (1H, t, J = 5.6 Hz), 5.31 (1H,
d, J = 2.8 Hz), 4.54–4.32 (8H, m), 4.20 (1H, m), 4.14 (2H, m), 3.96
(1H, m), 3.60 (1H, m), 3.52 (1H, m), 3.32 (1H, m), 3.17 (1H, m),
2.80 (3H, d, J = 4.6 Hz), 2.42 (1H, m), 2.20–1.35 (12H, m), 1.47
(9H, s), 1.27 (3H, d, J = 1.6 Hz), 1.05 (6H, d, J = 7.0 Hz).
[5] X. Liu, S. Dai, Y. Zhu, P. Marrack, J.W. Kappler, Immunity 19 (2003) 341–352.
[6] L.D. Walensky, A.L. Kung, I. Escher, T.J. Malia, S. Barbuto, R.D. Wright, G.
Wagner, G.L. Verdine, S.J. Korsmeyer, Science 305 (2004) 1466–1470.
[7] M.J. Kelso, D.P. Fairlie, Molecular pathomechanisms and new trends, in: G.
Keri, I. Toth (Eds.), Drug Research, Taylor and Francis, London, 2003, pp. 578–
598.
[8] D.S. Kemp, Trend. Biotechonol. 8 (1990) 249–255.
[9] J.P. Schneider, J.W. Kelly, Chem. Rev. 95 (1995) 2169–2187.
[10] M.J.I. Andrews, A.B. Tabor, Tetrahedron 55 (1999) 11711–11743.
[11] J. Garner, M.M. Harding, Org. Biomol. Chem. 5 (2007) 3577–3585.
[12] D.S. Kemp, T.P. Curran, J.G. Boyd, T.J. Allen, J. Org. Chem. 56 (1991) 6683–6697.
[13] G.E. Job, R.J. Kennedy, B. Heitmann, J.S. Miller, S.M. Walker, D.S. Kemp, J. Am.
Chem. Soc. 128 (2006) 8227–8233.
[14] R.E. Austin, R.A. Maplestone, A.M. Sefler, K. Liu, W.N. Hruzewicz, C.W. Liu, H.S.
Cho, D.E. Wemmer, P.A. Bartlett, J. Am. Chem. Soc. 119 (1997) 6471–6472.
[15] N.E. Shepherd, H.N. Hoang, V.S. Desai, E. Letouze, P.R. Young, D.P. Fairlie, J. Am.
Chem. Soc. 128 (2006) 13284–13289.
[16] A.K. Galande, K.S. Bramlett, J.O. Trent, T.P. Burris, J.L. Wittliff, A.F. Spatola,
Chembiochem 6 (2005) 1991–1998.
[17] M. Matzapetakis, V.L. Pecoraro, J. Am. Chem. Soc. 127 (2005) 18229–18233.
[18] R.L. Beyer, H.N. Hoang, T.G. Appleton, D.P. Fairlie, J. Am. Chem. Soc. 126 (2004)
15096–15105.
[19] J.A. Ihalainen, J. Bredenbeck, R. Pfister, J. Helbing, L. Chi, I.H. van Stokkum, G.A.
Woolley, P. Hamm, Proc. Natl. Acad. Sci. USA 104 (2007) 5383–5388.
[20] D. Wang, K. Chen, G. Dimartino, P.S. Arora, Org. Biomol. Chem. 4 (2006) 4074–
4081.
[21] S. Kirin, H.-B. Kraatz, N. Metlzer-Nolte, Chem. Soc. Rev. 35 (2006) 348–354.
[22] D.R. van Staveren, N. Metzler-Nolte, Chem. Rev. 104 (2004) 5931–5985.
[23] S. Chowdhury, K.A. Mahmoud, G. Schatte, H.-B. Kraatz, Org. Biomol. Chem. 3
(2005) 3018–3023.
[24] S.I. Kirin, U. Schatzschneider, X. de Hatten, T. Weyhermüller, N. Metzler-Nolte,
J. Organomet. Chem. 691 (2004) 3451–3457.
Acknowledgements
[25] T. Moriuchi, T. Nagai, T. Hirao, Org. Lett. 8 (2006) 31–34.
[26] T. Moriuchi, A. Nomoto, K. Yoshida, A. Ogawa, T. Hirao, J. Am. Chem. Soc. 123
(2001) 68–75.
[27] R.S. Herrick, R.M. Jarret, T.P. Curran, D.R. Dragoli, M.B. Flaherty, S.E. Lindyberg,
R.A. Slate, Tetrahedron Lett. 37 (1996) 5289–5292.
[28] M. Pellegrini, M. Royo, M. Chorev, D.F. Mierke, J. Pept. Res. 49 (1997) 404–414.
[29] W. Zhang, J.W. Taylor, Tetrahedron Lett. 37 (1996) 2173–2176.
[30] E. Schievano, A. Bisello, M. Chorev, A. Bisol, S. Mammi, E. Peggion, J. Am. Chem.
Soc. 123 (2001) 2743–2751.
[31] A.K. Boal, I. Guryanov, A. Moretto, M. Crisma, E.L. Lanni, C. Toniolo, R.H. Grubbs,
D.J. O’Leary, J. Am. Chem. Soc. 129 (2007) 6986–6987.
[32] E.W. Knobloch, W.H. Rauscher, J. Polym. Sci. 54 (1961) 651–656.
[33] N. Greenfield, G.D. Fasman, Biochemistry 8 (1969) 4108–4116.
[34] A. Pardi, M. Billeter, K. Wuthrich, J. Mol. Biol. 180 (1984) 741–751.
[35] V.F. Bystrov, Prog. NMR Spectrosc. 10 (1976) 41–82.
[36] C. Toniolo, E. Benedetti, Trend. Biochem. Sci. 16 (1991) 350–353.
[37] D.J. Barlow, J.M. Thornton, J. Mol. Biol. 201 (1984) 601–619.
[38] S. Hanessian, G. Papeo, K. Fettis, E. Therrien, M.T.P. Viet, J. Org. Chem. 69 (2004)
4891–4899.
[39] S.K. Maji, D. Haldar, D. Bhattacharyya, A. Bannerjee, J. Mol. Struct. 646 (2003)
111–123.
[40] S. Vijayalakshmi, R.B. Rao, I.L. Karle, P. Balaram, Biopolymers 53 (2000)
84–98.
This work was supported by summer undergraduate research
fellowships to ELH by Pfizer and the Connecticut Business and
Industry Association (CBIA), and by Bristol-Myers Squibb. This
work was also supported by an NSF-MRI Award (CHE-0619275)
for the NMR spectrometer used in these experiments. We thank
JEOL for their assistance in obtaining a high-resolution mass spec-
trum of 1; Whitney Smith, Jon Weiss, Andrew Rosenau, Neena
Chakrabarti, Peter Hendrickson and Julianne Boccuzzi for their
assistance in obtaining routine 1H NMR spectra; and Professor
Richard Prigodich (Trinity College) for his assistance obtaining
the CD spectrum. Finally, we thank the Trinity College Faculty Re-
search Committee for financial support.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jorganchem.2008.11.055.
[41] R.M. Jain, K.R. Rajashankar, S. Ramakumar, V.S. Cahuhan, J. Am. Chem. Soc. 119
(1997) 3205–3211.
[42] I.L. Karle, A. Pramanik, A. Bannerjee, S. Bhattacharyya, P. Balaram, J. Am. Chem.
Soc. 119 (1997) 9087–9095.
References
[1] T.E. Creighton, Proteins Structures and Molecular Properties, W.H. Freeman
and Company, New York, 1993.
[43] T.P. Curran, K.A. Marques, M.V. Silva, Org. Biomol. Chem. 3 (2005) 4134–
4138.
[2] R. Lavery, Q. Rev. Biophys. 38 (2005) 339–344.