Synthesis and conformational behavior of metallacyclicdipeptides derived from coordination of side chain alkynylamino acids to tungsten

Article abstract of DOI:10.1016/j.jorganchem.2016.01.023

Three dipeptides bearing alkynes on their side chains (9 (derived from dilysine), 14 (derived from dicysteine) and 17 (derived from diglycine)) were prepared and reacted with W(CO)₃(dmtc)₂ [dmtc = dimethyldithiocarbamate] to afford, respectively, three metallacyclicpeptides, 18, 19 and 20. The metallacyclicpeptides were characterized by HPLC, ES-MS, and ¹H NMR. The conformational behavior of the alkynes about the tungsten center was assessed using ¹H NMR. It was found that all three metallacyclicpeptides adopt multiple conformations of the alkynes relative to the tungsten. Both 18 and 19 appear to adopt all 8 possible conformations, while 20 adopts a limited number of conformations. The ability of the alkynes to equilibrate between the syn and anti conformations was assessed by examining the alkyne hydrogen resonances using variable temperature ¹H NMR. It was found that the alkyne ligands in 18 and 19 will equilibrate between the syn and anti conformations. The alkyne hydrogen resonances in 18 coalesce to one signal around 343 K, while the alkyne hydrogen resonances in 19 do not completely coalesce even by 360 K. Complex 18 has a larger ring than complex 19, and the higher temperature of coalescence for 19 is attributed to its smaller ring size. In contrast, complex 20, which has the smallest ring size, cannot equilibrate between the syn and anti conformations, even at elevated temperatures. The results show that cyclic tungsten-bis(alkyne) complexes will form ring systems with ring sizes of approximately 10 atoms, that ring sizes of approximately 10 atoms are rigid, and that rigidity is lost as the ring size is increased.

Full text of DOI:10.1016/j.jorganchem.2016.01.023

Journal of Organometallic Chemistry 806 (2016) 12e21
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Synthesis and conformational behavior of metallacyclicdipeptides
derived from coordination of side chain alkynylamino acids to
tungsten
Timothy P. Curran*, T. Andrew McTeague, Vu D. Nguyen, Craig J. Yennie, Paul R. Handali,
Joseph P. Sanderson-Brown, Zephyr D. Dworsky
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:
Three dipeptides bearing alkynes on their side chains (9 (derived from dilysine), 14 (derived from
dicysteine) and 17 (derived from diglycine)) were prepared and reacted with W(CO)3(dmtc)2
[dmtc ¼ dimethyldithiocarbamate] to afford, respectively, three metallacyclicpeptides, 18, 19 and 20. The
metallacyclicpeptides were characterized by HPLC, ES-MS, and 1H NMR. The conformational behavior of
the alkynes about the tungsten center was assessed using 1H NMR. It was found that all three metal-
lacyclicpeptides adopt multiple conformations of the alkynes relative to the tungsten. Both 18 and 19
appear to adopt all 8 possible conformations, while 20 adopts a limited number of conformations. The
ability of the alkynes to equilibrate between the syn and anti conformations was assessed by examining
the alkyne hydrogen resonances using variable temperature 1H NMR. It was found that the alkyne li-
gands in 18 and 19 will equilibrate between the syn and anti conformations. The alkyne hydrogen res-
onances in 18 coalesce to one signal around 343 K, while the alkyne hydrogen resonances in 19 do not
completely coalesce even by 360 K. Complex 18 has a larger ring than complex 19, and the higher
temperature of coalescence for 19 is attributed to its smaller ring size. In contrast, complex 20, which has
the smallest ring size, cannot equilibrate between the syn and anti conformations, even at elevated
temperatures. The results show that cyclic tungstenebis(alkyne) complexes will form ring systems with
ring sizes of approximately 10 atoms, that ring sizes of approximately 10 atoms are rigid, and that rigidity
is lost as the ring size is increased.
Received 10 September 2015
Received in revised form
15 January 2016
Accepted 16 January 2016
Available online 22 January 2016
Keywords:
Bioorganometallic
Tungsten
Alkyne
Peptide
Side chain
Conformation
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
these alkynylamino acids and alkynylpeptides to tungsten [2]. The
data showed that the two coordinated amino acids or peptides (for
Unlike other transition metals, tungsten will coordinate multi-
ple alkyne ligands and form air stable complexes. The documented
ability of W(CO)3(dmtc)2 to quickly and easily react with two al-
kynes to form air stable bis(alkyne) complexes [1] attracted our
attention, and in ongoing work from our laboratory we have been
exploring the use of this chemistry to bring together [2,3], and in
some cases constrain, peptides to certain secondary structures
[4e6]. In this regard we developed methods for adding alkynes to
the N- and C-termini of amino acid residues, and have coordinated
example 1) do not interact via the formation of hydrogen bonds or
other intramolecular interactions [2]. However, in another set of
studies, peptides bearing an alkyne at both the N-terminus and the
C-terminus will react with W(CO)3(dmtc)2 to yield metallacyclic
species in which the two ends of the peptide are tied together via
the tungstenealkyne coordination [4,5]. These were the first re-
ported cyclic tungsten bis-alkyne complexes. Data showed that in 2
(which has a dialkynyltetrapeptide coordinated to tungsten) the
peptide part of the metallacyclicpeptide adopted a turn confor-
mation [5].
* Corresponding author.
E-mail address: timothy.curran@trincoll.edu (T.P. Curran).
http://dx.doi.org/10.1016/j.jorganchem.2016.01.023
0022-328X/© 2016 Elsevier B.V. All rights reserved.

T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
13
2. Results
2.1. Selection of alkynylamino acids
At the start of this study we sought to explore several different
alkynylamino acids. The primary consideration in choosing which
alkynylamino acids to use was distance of the alkyne from the
peptide main chain. Given our past experience with dia-
lkynylpeptides where the alkynes were located at the N- and C-
termini [4,5], we were concerned that having the alkyne too far
from the main chain would not confer enough conformational
constraint to enforce a distinct conformation. On the other hand,
having the alkyne too close to the main chain might prevent for-
mation of the desired metallacycle. Thus, we studied three alky-
nylamino acids, 3e5, that place the alkyne at varying distances
from the amino acid Ca carbon. Acylation of the side chain amine of
lysine with propargyl chloroformate will generate alkynylamino
acid 3, which positions the alkyne 9 atoms from the Ca carbon.
Alkylation of the side chain thiol of cysteine with propargyl bro-
mide will generate alkynylamino acid 4, which locates the alkyne 4
atoms from the Ca carbon. Finally, alkylation of glycine with
propargyl bromide followed by resolution of the two enantiomers
will generate alkynylamino acid 5 (propargylglycine) which places
the alkyne only 2 atoms removed from the Ca carbon.
Metallacyclicpeptides, like 2, are constructed by linking a metal
to two or more ligands on the peptide, with the metal becoming
one of the atoms in the ring. Metallacyclicpeptides were reviewed
in 2005 by Albrecht and Stortz [7]. All of the examples described in
the review involved the use of coordination chemistry to link basic
ligands on the peptide to the metal. These ligands included imid-
azoles, pyridines, phosphanes, catechols, hydroxyquinolines, ami-
nodiacetates and crown ethers. The metals employed included Ti,
Fe, Co, Ni, Cu, Zn, Mo, Ru, Rh, Pd, Cd and Cs. Metal-ligand coordi-
nation in these studies yielded peptides constrained to helix, sheet
and turn conformations.
Since publication of the review, newer metallacyclicpeptides
have been described. Dugave and co-workers used Re complexes to
constrain peptides that bind to biological targets [8,9]. Kuroda and
co-workers have reported on the ability of peptide ligands to co-
ordinate to Pd and Pt to constrain peptides to the 310-helical
conformation [10]. Helical peptides were also prepared by Futaki
and co-workers via coordination of dipicolyl ligands (located on
amino acid side chains) to a number of transition metals [11].
Metallacyclicpeptides with defined chirality were obtained by
Vazquez and co-workers via coordination of Co, Ni and Zn to bpy
ligands tethered to peptides [12]. In the most recent report, Miguel
and co-workers have created metallacyclicpeptides by dimerizing
two peptides via metal coordination to pyridyl ligands located at
the N- and C-termini [13].
2.2. Preparation of alkynylamino acids and dialkynylpeptides
In our own work we have chosen to explore metal-
Three alkynylamino acids were examined in this study. The first
was the alkynyllysine, 3. The side chain of lysine possesses a reac-
tive primary amine that can be readily acylated. As shown in
Scheme 1, the commercially available lysine derivatives 6 and 7
were coupled to yield the dipeptide 8. The Cbz groups on 8 were
then removed via catalytic hydrogenation, and the resulting
diamine was acylated with propargylchloroformate to generate the
dialkynyldipeptide, 9. All of the dialkynyldipeptides (9, 14, 17) were
purified by flash chromatography and its identity confirmed by 1H
NMR and ES-MS. Although racemization of the amino acid residues
is unlikely given the reaction conditions utilized here, purification
would have removed any diastereomers that might have formed.
The second alkynylamino acid examined was propargylcysteine.
The synthesis of this amino acid from cysteine was described by
Carson and Boggs [15]. Reaction of cysteine (10) with an excess of
NaOEt and propargylbromide (Scheme 2) will yield prop-
argylcysteine, 4. To meet our needs we modified this literature
procedure; instead of isolating the alkynylamino acid following the
alkylation reaction, the crude propargylcysteine was subsequently
reacted with Boc2O to yield the known urethane 11 [16e18]. Next,
11 was acylated with anisidine using a water-soluble carbodiimide
to yield 12. Removal of the Boc group from 12 yielded the amine salt
13. It was coupled with the acid 11 to produce the dia-
lkynyldipeptide 14.
lacyclicpeptides that feature
p-ligands, rather than basic ligands.
Peptide helices have been generated using a ferrocene linker [14],
and turn structures have been generated using coordination of
tungsten to alkyne ligands located at the N- and C-termini of the
peptide [4,5].
Having studied metallacyclicpeptides prepared by coordination
of tungsten to dialkynylpeptides where the two alkynes are located
at the N- and C-termini, we were also interested in whether met-
allacyclicpeptides could also be obtained from coordination of
tungsten to dialkynylpeptides which have the two alkynes are
located on the amino acid side chains.
This paper details our studies on the coordination of dia-
lkynylpeptides to tungsten, where the two alkynes are located
on amino acid side chains. The aim of the work was to delineate
the conformational behavior of these novel species. First, the
paper describes how alkynes can be located on the side chains of
amino acids, and how these alkynylamino acids can be used in
peptide synthesis. Second, the paper details how the dia-
lkynylpeptides can be coordinated to tungsten via reaction with
W(CO)3(dmtc)2. Finally, the conformational behavior of the
resulting metallacyclicpeptides is explored. The data from this
study and prior studies shed light on the factors responsible for
constraining the conformations of these novel, metallacyclic
molecules.
The final amino acid examined was propargylglycine (5), which

14
T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
W(CO)3(dmtc)2 have given low yields. Higher yields are obtained
using freshly prepared W(CO)3(dmtc)2. All of these cyclization re-
actions were run at 1 mM concentrations of the dialkynyldipeptide
in order to minimize the chance for oligomer formation. All of the
reactions first turned from an orange color to a green color, indi-
cating formation of the monoalkyne complex [2,33]. The green
color then turned to a lemon yellow color over the course of several
hours. The appearance of the yellow color indicated formation of a
bisalkyne complex [2,34,35].
To isolate the complexes 18e20, the crude reaction mixture was
evaporated, and the remaining residue was purified by flash
chromatography. The unoptimized yields for 18, 19 and 20 were
30%, 44% and 60% respectively. In all reactions some unreacted
dialkyne remained behind. Analysis of the crude product by ESI-MS
showed evidence for small amounts of oligomeric species, which
has been seen in previous work [4]. Purity was established by HPLC,
and the structures were confirmed by 1H NMR and ESI-MS. The
distinctive isotope pattern in the MS derived from the different
isotopes of tungsten was used to firmly establish the identity of the
metallacyclicdipeptides [2]. The pure complexes are amorphous
yellow solids. They did not show any inclination to form crystalline
solids.
2.4. Conformation of the dialkynyldilysine complex 18
In previous work with metallacyclicpeptides derived from co-
ordination of dialkynylpeptides where the alkynes were located at
the N- and C-termini, we discovered that these cyclic species can
and do adopt both the syn and anti orientations of the two alkyne
ligands around the tungsten center [4,5]. Because the dia-
lkynylpeptides examined here are not symmetrical, and because
there are two ways to arrange the dmtc ligands around the tung-
sten, there are 8 possible isomers for these complexes. Shown in
Fig. 1 are the 8 possible conformations for the metallacyclicpeptide
18. There are two sites where isomerism develops in these species.
First, the dmtc ligands can adopt one of two possible unique ar-
rangements. Isomers 18AeD possess one of the two possible ar-
rangements for the dmtc ligands; isomers 18EeH possess the other
possible arrangement of the dmtc ligands. Second, the alkynes can
be arranged in either a syn or anti arrangement. Isomers 18AeB and
18EeF have the alkyne ligands in the anti arrangement, while
isomers 18CeD and 18GeH have the alkynes in the syn
arrangement.
It is known from past work that in bis(alkyne) complexes the
resonances for the alkyne hydrogens appear in the region around
11 ppm in the 1H NMR spectrum [1]. For each of the eight possible
conformations there are two different alkyne protons. This means
that in the 1H NMR spectrum of 18, each conformation should give
rise to two singlets derived from each alkyne hydrogen. If 18 adopts
all 8 conformations shown in Fig. 1, then the 1H NMR spectrum
should show 16 singlets in the alkyne hydrogen region, which is
around 11 ppm.
Fig. 2 shows the alkyne hydrogen region for complex 18 in CDCl3
at 254 K. The spectrum shows a large number of singlets, indicating
that 18 likely adopts all 8 conformations shown in Fig. 1. When the
spectrum of 18 is taken at 274 and 294 K in CDCl3 the singlets begin
to coalesce, indicating that rotation about the metal-alkyne bond at
room temperature readily occurs (see Fig. S1 in the Supplementary
Data). When the spectrum of 18 is taken in d6-DMSO at 294 K, two
broad absorbances are observed (Fig. 3). Like the NMR spectrum of
18 in CDCl3, the spectrum of 18 in d6-DMSO indicates a system
where rotation about the metal-alkyne bonds is taking place.
Because DMSO is an aggressive hydrogen bond acceptor and CDCl3
is not, the NMR spectra of molecules that can form hydrogen bonds
(like 18) can often show differences in chemical shifts between the
Scheme 1. Synthetic route to dialkynyldipeptide 9.
can be obtained either from commercial vendors or prepared using
literature procedures. We chose to purchase 5, which is easily
converted into the methyl ester hydrochloride 15 and the N-Boc
derivative 16 [19e31]. Coupling of 15 and 16 using the coupling
reagent HATU produced the dialkynyl dipeptide 17 (Scheme 3).
2.3. Coordination of dialkynylpeptides to tungsten
The dialkynyldipeptides 9, 14, and 17, were reacted with
W(CO)3(dmtc)2 [32] in refluxing methanol to produce, respectively,
the metallacyclicdipeptides 18, 19, and 20 (Scheme 4). We have
performed these reactions two different ways, either by preparing a
large stock of W(CO)3(dmtc)2 and using it when needed, or pre-
paring a small quantity of W(CO)3(dmtc)2 and using it immediately.
In our hands, W(CO)3(dmtc)2 begins decomposing shortly after it
has been prepared, so reactions using
a stock supply of

T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
15
Scheme 2. Synthetic route to dialkynyldipeptide 14.
two solvents. This is the likely reason why the NMR spectra of 18 in
CDCl3 and d6-DMSO at 294 K (Figs. 2 and 3) are not identical.
That the alkynes rotate about the tungsten in 18 was confirmed
by the behavior of the DMSO sample when the temperature was
raised (Fig. 3). The two broad peaks coalesced into one peak around
343 K. When the sample was returned to 294 K the spectrum was
identical to what it had been before heating, indicating that heating
did not induce decomposition. This shows that the various con-
formers (unknown in number) are rapidly equilibrating at this
temperature. In previous work we have found that acyclic bis(al-
kyne) complexes derived from peptides and amino acids also coa-
lesce around 338 K [2]. Thus, the behavior of 18 closely mirrors the
behavior of these acyclic complexes, indicating that there is a great
deal of conformational flexibility for the two alkynes in this met-
allacycle to equilibrate between conformers. This is not surprising,
given that the macrocyclic ring formed by coordination of the
tungsten to the two alkynes comprises 23 atoms.
ring found in 18; it is comprised of 13 atoms, including the tung-
sten. If 19 is as flexible as 18, then it too should adopt all 8 con-
formations shown in Fig. 4. This flexibility can be detected by the
presence of many singlets for the alkyne hydrogens in the 1H NMR
spectrum, and by their ability to coalesce to a single peak at
elevated temperatures. This is what the data shows. The 1H NMR
spectrum of 19 in CDCl3 or d6-DMSO in the region around 11 ppm
shows numerous, overlapping singlets for the alkyne hydrogens.
When the d6-DMSO sample was heated to 360 K, the alkyne
hydrogen singlets begin to coalesce to a single peak, but unlike 18,
they must coalesce at a higher temperature (Fig. 5). This data in-
dicates that 19, like 18, is conformationally flexible and likely
adopts all 8 of the possible orientations shown in Fig. 4. The higher
temperature for coalescence of the alkyne hydrogen peaks in 19
relative to 18 suggests that the smaller ring size in 19 makes it more
difficult for the syn and anti conformers to equilibrate.
The last complex examined here was 20. In this molecule the
tungsten is linked to the two alkynes in a dipeptide of two prop-
argylglycine residues. This complex has the least number of atoms
in the ring completed by coordination of the tungsten, 10 atoms.
The 1H NMR spectrum of 20 in CDCl3 in the region around 11 ppm
shows 8 singlets. Although it is not limited to one solution
conformation, 20 is different from 18 and 19; it only adopts 4 of the
possible 8 conformations shown in Fig. 6. The spectrum also
2.5. Conformation of the dipropargylcysteine complex 19
Since 18 was conformationally flexible, we next examined
complex 19, which was made by coordination of tungsten to the
two alkynes in
a dipeptide constructed from two prop-
argylcysteines. The ring formed in this complex is smaller than the

16
T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
indicates that there is a decided preference for one of the 4 possible
conformations.
Like the other complexes, 20 was also examined by variable
temperature NMR. The results are shown in Fig. S2 (in the Sup-
plementary Data). In this experiment the temperature of a d6-
DMSO solution of 20 was varied between 293 K and 343 K. Unlike
18 and 19, where the alkyne hydrogen resonances coalesce as the
temperature is raised, with 20 the alkyne hydrogens show no
inclination to coalesce; rather in some cases they move apart and
become more distinct as the temperature is raised. Also, the relative
integrations of the singlets do not change as the temperature is
raised. This indicates that the alkyne groups in 20 are unable to
rotate about the tungsten center, and so cannot equilibrate between
the various conformations shown in Fig. 6.
3. Discussion
The results of this work show that cyclic tungsten bis(alkyne)
complexes can be made by reaction of dialkynylpeptides with
W(CO)3(dmtc)2 where the alkynes reside on the amino acid side
chains. The cyclic complexes can be made with long (18), medium
(19) and short dialkynes (20).
The conformational mobility of 18e20 reinforces our earlier
findings that most cyclic tungsten bis(alkyne) complexes adopt
conformations where the alkyne ligands are both syn and anti to
each other, and that often the two alkyne ligands can equilibrate
between the syn and anti conformations. This is how complexes 18
and 19 behave; they likely adopt all 8 possible conformations (see
Figs. 1 and 4) and will equilibrate between the syn and anti con-
formers. In contrast, the data for the smallest cyclic structure in this
study, complex 20 derived from propargylglycine, shows that this
molecule exists in a limited number of the conformations shown in
Fig. 6, and that each conformation adopted is fixed. Complex 20
Scheme 3. Synthetic route to dialkynyldipeptide 17.
Scheme 4. Syntheses of metallacyclicdipeptides 18e20.

T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
17
Fig. 1. The 8 possible conformational isomers for complex 18.
cannot equilibrate between the syn and anti conformations, even at
elevated temperatures. The conformation obtained when the sec-
ond alkyne coordinates is the conformation of the complex. We
found similar behavior with a dialkynylderivative of alanine, 21,
which adopts a number of different conformational isomers, but
because of the rigid nature of the ring, the conformational isomers
cannot interconvert [4]. Complexes 20 and 21 have similar ring
sizes. There are 11 atoms in the ring (including tungsten) for 21,
while in 20 there are 10 atoms in the in the ring (including tung-
sten). To date, 20 is the smallest cyclic tungsten bis-alkyne complex
that we have been able to make, isolate and characterize. Attempts
to make a smaller cyclic tungsten bis-alkyne complex, 22 (which
would have a ring size of only 8 atoms), only produced oligomeric
species [4]. Thus, our data indicate that a ring of 10 atoms may be
the lower limit for the possible ring sizes possible for cyclic tung-
sten bis-alkyne complexes.
There were two goals for this study. First, to gain a further un-
derstanding of the conformational behavior of cyclic tung-
stenebis(alkyne) complexes, and second, to uncover ways to use
tungstenealkyne coordination to constrain peptide conformation.
With regards to the first goal, the data presented here and in
Fig. 2. The alkyne hydrogen region (11.4e10.7 ppm) of 18 in CDCl3 at 254 K.

18
T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
were purchased from Aldrich Chemical. Anisole, trifluoroacetic acid
(TFA), propargyl bromide, tetrahydrofuran (THF), di-tert-butyl
dicarbonate (Boc2O), 1-ethyl-3-[3-dimethylaminopropyl]carbodii-
mide hydrochloride (EDC), and triethylamine were purchased from
Acros Organics. Diisopropylethylamine (DIEA), methanol, ethanol,
hexanes, ethyl acetate, methylene chloride, toluene, ethyl ether,
hydrochloric acid, sodium hydroxide, magnesium sulfate, and
iodine were purchased from Fisher Scientific. Deuterated solvents
were purchased from Cambridge Isotope Laboratories, Inc. The
propargylglycine derivatives were purchased from Bachem. Other
amino acid derivatives, PyBOP, and piperidine were purchased from
Chem Impex International. Silica gel for flash chromatography was
purchased from Silicycle. NMR spectra were obtained on either
either a Bruker Avance III 400 MHz instrument or a GE Omega 300
instrument. Electrospray mass spectra were obtained on a LCQ
APCI/Electrospray LC MSeMS. Samples for mass spectral analysis
were dissolved in MeOH (approximately 1 mg/mL) in borosilicate
glass test tubes. Theoretical mass spectral isotope patterns were
calculated using the online isotope pattern calculator provided on
the internet by the Swiss Federal Institute of Aquatic Science and
Technology (Eawag) (http://www.envipat.eawag.ch/index.php)
[36]. HPLC analyses were performed on an Hitachi Elite LaChrom
HPLC system equipped with L-2400 detector, an L-2200 autosam-
pler and an L-2130 pump. Each run was monitored at 300 nm,
which is the wavelength of maximum absorbance for the mono-
alkyne complexes. A Phenomenex Luna 250 ꢀ 4.6 mm column was
used as the stationary phase. The mobile phase involved a linear
gradient program using two solvents, 0.1% trifluoroacetic acid and
acetonitrile. The gradient program started at a 20:80 mixture of
acetonitrile and trifluoroacetic acid and changed to 100% acetoni-
trile over the course of 12 min. The solvent was then held at 100%
acetonitrile for an additional 2 min.
4.2. Preparation of Boc-Lys(Cbz)-Lys(Cbz)-OMe, 8
To a solution of 505 mg (1.06 mmol, 1 equiv.) of Boc-Lys(Z)-OSu
(6) and 363 mg (1.10 mmol, 1.03 equiv.) of H-Lys(Z)-OMe (7) in
20 mL CH2Cl2 was added 1.0 mL (5.7 mmol, 5 equiv.) of DIEA. The
resulting clear solution stirred at 23 ꢁC for 18 h. The solvents were
evaporated and the remaining oil was redissolved in 100 mL EtOAc
and washed: 3 ꢀ 25 mL 1 M HCl, 3 ꢀ 25 mL saturated NaHCO3 and
1 ꢀ 25 mL brine. The EtOAc was dried (MgSO4), filtered, evaporated
and dried under high vacuum to yield 648 mg (93%) of pure 8 as a
white foam: TLC (2:1 EtOAc/hexanes), Rf 0.46; 1H NMR (400 MHz,
CDCl3): 7.34 (10H, m, 2 ꢀ C6H5), 6.70 (1H, d, J ¼ 7.2 Hz, amide NH),
5.25e4.90 (7H, m, 3 ꢀ urethane NH þ 2 ꢀ OCH2), 4.55 (1H, m, CaH),
4.09 (1H, m, CaH), 3.69 (3H, s, OMe), 3.17 (4H, m, 2 ꢀ CH2N),
1.97e1.27 (12H, m, 2 ꢀ CH2CH2CH2), 1.43 (9H, s, OtBu); ESMS,
M þ Na ion pattern calculated for C34H48N4O9Na: 679 (100), 680
(39.2), 681 (9.3); found 679 (100), 680 (36.5), 681 (10.8).
Fig. 3. The 1H NMR spectra of the alkyne hydrogens (11.15e10.85 ppm) in complex
18 at temperatures ranging from 294 K to 353 K in d6-DMSO.
previous work [4,5] has shown that metallacyclic species employ-
ing tungstenealkyne coordination are unlikely to have ring sizes
less than 10 atoms. Further, rings comprising 10 or 11 atoms are
rigid, the alkyne ligands are unable to rotate about the metal; un-
fortunately these species do not adopt one single conformation in
solution. Finally, ring sizes larger than 10 or 11 atoms possess the
ability to interconvert and equilibrate between syn and anti
conformers.
4.3. Preparation of Boc-Lys(Poc)-Lys(Poc)-OMe, 9
With regards to the second goal, the ability of the tungsten
alkyne bonds to interconvert between they syn and anti con-
formers has so far prevented the discovery of a system that con-
strains a peptide to one discreet conformation. The results of prior
work [4,5] and this work, however, indicate that such a system
must encompass a ring of around 10 atoms. Explorations in this
area are ongoing.
To a solution of 624 mg (0.953 mmol, 1.0 equiv.) of 8 in 50 mL
EtOH was added 50 mg of 10% PdeC. The resulting mixture was
hydrogenated on a Parr apparatus for 2 h at 25 ꢁC. After 2 h the
catalyst was removed by filtration through celite and the filtrate
evaporated. The remaining oil was redissolved in 20 mL CH2Cl2, and
to this solution was added 0.5 mL (2.9 mmol, 3 equiv.) of DIEA and
232
mL (2.38 mmol, 2.5 equiv.) of propargylchloroformate. After
4. Experimental
stirring at 23 ꢁC for 72 h the solvents were evaporated. The residue
that remained was redissolved in 50 mL EtOAc and washed:
3 ꢀ 25 mL 1 M HCl, 3 ꢀ 25 mL saturated NaHCO3 and 1 ꢀ 25 mL
brine. The EtOAc was dried (MgSO4), filtered, evaporated and dried
under high vacuum to yield 226 mg (43%) of 9 as a clear oil: TLC (4:1
4.1. General procedures
Anisidine, L-cysteine hydrochloride and propargylchloroformate

T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
19
Fig. 4. The 8 possible conformational isomers for complex 19.
EtOAc/hexanes) Rf 0.63; 1H NMR (400 MHz, CDCl3): 6.88 (1H, d,
J ¼ 7.7 Hz, amide NH), 5.35 (2H, m, 2 ꢀ urethane NH), 5.19 (1H, t,
J ¼ 5.2 Hz, urethane NH), 4.75e4.60 (4H, m, 2 ꢀ OCH2), 4.56 (1H, m,
CaH), 4.17 (1H, m, CaH), 3.76 (3H, s, OMe), 3.31e3.09 (4H, m,
2 ꢀ CH2N), 2.51 (1H, t, J ¼ 2.5 Hz, CCH), 2.49 (1H, t, J ¼ 2.4 Hz, CCH),
1.85e1.25 (12H, m, 2 ꢀ CH2CH2CH2), 1.45 (9H, s, OtBu); ESMS,
M þ Na ion pattern calculated for C26H40N4O9Na: 575 (100), 576
(30.4), 577 (6.3); found 575 (100), 576 (28.1), 577 (11.5).
evaporated and the residue redissolved in 25 mL EtOAc and
washed: 3 ꢀ 25 mL 0.1 M HCl, 3 ꢀ 25 mL saturated NaHCO3 and
1 ꢀ 25 mL brine. The organic layer was dried (MgSO4), filtered and
evaporated. The oil that remained was purified by flash chroma-
tography (2:1 EtOAc:hexanes) to afford 0.258 g of pure 12 (37%) as
an oil: TLC (1:1 EtOAc/MeOH), Rf 0.67; 1H NMR (400 MHz, CDCl3):
8.17 (1H, s, ArNH), 7.43 (2H, d, J ¼ 8.9 Hz, 2 ꢀ ArH), 6.87 (2H, d,
J ¼ 9.0 Hz, 2 ꢀ ArH), 5.44 (1H, d, J ¼ 6.6 Hz, urethane NH), 4.49 (1H,
q, J ¼ 6.8 Hz, CaH), 3.80 (3H, s, OMe), 3.34 (2H, d, J ¼ 2.6 Hz, SCH2),
3.15 (2H, m, b-CH2), 2.32 (1H, s, CCH), 1.48 (9H, s, OtBu); ESMS,
4.4. Preparation of Boc-Cys(CH2CCH)eOH, 11 [8e10]
M þ Na ion pattern calculated for C18H24N2O4Na: 387 (100), 388
(21.5), 389 (7.4); found 387 (100), 388 (19.1), 389 (6.3).
To 60 mL of absolute EtOH under N2 was added 1.030 g
(44.8 mmol, 4.0 equiv.) of Na metal. Once all the Na had reacted,
1.956 g (11.1 mmol, 1.0 equiv.) of L-cysteine hydrochloride hydrate
and 4.90 mL (44.8 mmol, 4.0 equiv.) of propargyl bromide were
added to the solution. After stirring for 2 h, the solvent was evap-
orated. The remaining residue was redissolved in 15 mL THF. To this
solution was added a solution of 4.865 g (22.3 mmol, 2.0 equiv.) of
Boc2O dissolved in 12.5 mL THF, followed by 5.0 mL (55 mmol,
5 equiv.) of Et3N and 25 mL of H2O. After stirring for 18 h the THF
was evaporated, and the remaining aqueous layer was brought to
pH > 11 by the addition of 1 M NaOH. The alkaline aqueous layer
was washed 3 ꢀ 25 mL EtOAc, then acidified to pH 2 using
concentrated HCl. The resulting acidic aqueous layer was extracted
3 ꢀ 25 mL EtOAc. The combined EtOAc extracts were dried
(MgSO4), filtered and evaporated to yield 1.96 g of 11 [16e18] as a
tan oil: TLC (1:1 EtOAc/MeOH), Rf 0.67; 1H NMR (400 MHz, CDCl3):
6.36 (1H, br s, OH), 5.39 (1H, d, J ¼ 7.7 Hz, NH), 4.62 (1H, q,
J ¼ 5.9 Hz, CaH), 3.33 (2H, dd, J ¼ 8.8 and 2.7 Hz, SCH2), 3.27 (2H, dd,
4.6. Preparation of Boc-Cys(CH2CCH)- Cys(CH2CCH)eNHC6H4OMe,
14
To a solution of 0.100 g (0.274 mmol, 1.0 equiv.) of 12 in 1 mL
CH2Cl2 and 50 m
L of anisole at 5 ꢁC was added 1 mL of trifluoroacetic
acid. The resulting solution slowly warmed to 23 ꢁC over the course
of 4 h. The solvents were evaporated and the crude amine salt (13)
was dried under high vacuum for 14 h. Crude 13 was then redis-
solved in 1 mL CH2Cl2. To this solution was added 0.098 g
(0.40 mmol, 1.5 equiv) of 16 and 0.29 mL (2.1 mmol, 7.6 equiv.) of
triethylamine. After stirring at 23 ꢁC for 4 h the solvents were
evaporated. The residue was redissolved in 30 mL EtOAc and
washed 3 ꢀ 10 mL 0.1 M HCl, 3 ꢀ 10 mL saturated NaHCO3 and
1 ꢀ 10 mL brine. The organic layer was dried (MgSO4), filtered and
evaporated to yield 0.042 g (30%) of 14 as an oil: TLC (3:2 EtOAc/
hexanes), Rf 0.60; 1H NMR (400 MHz, CDCl3): 8.71 (1H, s, ArNH),
7.45 (2H, d, J ¼ 9.1 Hz, 2 ꢀ ArH), 6.19 (2H, d, J ¼ 9.0 Hz, 2 ꢀ ArH), 5.62
(1H, d, J ¼ 6.4 Hz, amide NH), 5.52 (1H, d, J ¼ 7.6 Hz, urethane NH),
4.87 (1H, m, CaH), 4.45 (1H, q, J ¼ 5.9 Hz, CaH), 3.77 (3H, s, OMe),
J ¼ 13.3 and 4.8 Hz,
b-CH2), 2.32 (1H, s, CCH), 1.49 (9H, s, OtBu).
4.5. Preparation of Boc-Cys(CH2CCH)eNHC6H4OMe, 12
To a solution of 0.500 g (1.93 mmol, 1.0 equiv.) of 11 in 2.0 mL
CH2Cl2 at 5 ꢁC was added 0.261 g (2.12 mmol, 1.1 equiv.) of anisidine
and 0.443 g (2.31 mmol, 1.2 equiv.) of EDC. The resulting solution
warmed to 23 ꢁC over the course of 4 h. The solvents were
3.31 (4H, m, 2 ꢀ SCH2), 3.12 (4H, m, 2 ꢀ
b-CH2), 2.33 (2H, m,
2 ꢀ CCH), 1.44 (9H, s, OtBu); ESMS, M þ Na ion pattern calculated
for C24H31N3O5Na: 528 (100), 529 (29.2), 530 (9.3), 531 (2.7); found
528 (100), 529 (25.9), 530 (11.5), 531 (2.5).

20
T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
(65%) of 17 as a clear oil: TLC (Et2O), Rf 0.71; 1H NMR (400 MHz,
CDCl3): 7.17 (1H, s, amide NH), 5.39 (1H, s, urethane NH), 4.67 (1H,
m, CaH), 4.31 (1H, m, CaH), 3.72 (3H, s, OMe), 2.72e2.56 (4H, m,
2 ꢀ CH2), 2.05 (1H, s, CCH), 1.98 (1H, s, CCH), 1.40 (9H, s, OtBu);
ESMS, M þ Na ion pattern calculated for C16H22N2O5Na: 345 (100),
346 (17.9), 347 (2.7); found 345 (100), 346 (15.5), 347 (2.0).
4.8. Preparation of cyclic complex 18
To a refluxing solution of 205 mg (0.371 mmol, 1 equiv.) of 9 in
250 mL degassed MeOH under N2 was added dropwise over 1 h a
solution of 189 mg (0.371 mmol, 1 equiv.) of W(CO)3(dmtc)2 [32]
dissolved in 30 mL of degassed CH2Cl2. After 2 h the solution had
taken on a lemon yellow color. Reflux was stopped the solvents
evaporated. Flash chromatography (EtOAc) was used to isolate the
product. Obtained 110 mg (30%) of 18 as an amorphous yellow
solid: TLC (EtOAc), Rf 0.53; HPLC, Rt 7.44 min; 1H NMR (400 MHz,
CDCl3):11.2e10.8 (2H, ms, 2 ꢀ CCH), 7.6e6.7 (1H, m, amide NH),
6.2e4.0 (9H, m, 2 ꢀ OCH2 þ 3 ꢀ urethane NH þ 2 ꢀ CaH), 3.8e3.0
(19H, m, OMe þ 2 ꢀ CH2N þ 4 ꢀ dmtc Me), 1.9e1.1 (21H, m,
2 ꢀ CH2CH2CH2 þ OtBu); ESMS, M þ Na ion pattern calculated for
WC32H52N6O9S4Na: 997 (60.3), 998 (57.3), 999 (100), 1000 (43.1),
1001 (89.3), 1002 (34.3), 1003 (20.7), 1004 (6.4), 1005 (2.2); found
997 (62.3), 998 (63.4), 999 (100), 1000 (40.9), 1001 (91.6), 1002
(30.5), 1003 (18.9), 1004 (4.9), 1005 (1.7).
4.9. Preparation of cyclic complex 19
To a refluxing solution of 0.504 g (0.992 mmol, 1 equiv.) of
W(CO)3(dmtc)2 [32] in 1.1 L of degassed MeOH under N2 was added
via syringe a solution of 0.504 g (0.997 mmol, 1 equiv.) of 14 dis-
solved in 7 mL of degassed MeOH. The solution immediately took
on a dark green color. The color turned to lemon yellow after 45 min
of reflux. The methanol was evaporated and the crude product
purified by flash chromatography (1:1 ethyl acetate/hexanes). A
total of 0.408 g (44%) of pure 19 was obtained as an amorphous
yellow solid. TLC (1:1 EtOAc/hexanes), Rf 0.20; HPLC, Rt 8.42 min;
1H NMR (400 MHz, CDCl3): 11.5e11.0 (2H, ms, 2 ꢀ CCH), 8.8e8.2
(1H, ms, ArNH), 7.7e7.4 (3H, m, amide NH þ 2 ꢀ ArH), 6.9e6.7 (2H,
mm 2 ꢀ ArH), 5.6e5.1 (1H, m, urethane NH), 5.0e4.0 (6H, m,
2 ꢀ CaH þ 2 ꢀ SCH2), 3.80 (3H, ms, OMe), 3.7e2.2 (16H, m, 4 ꢀ dmtc
Me þ 2 ꢀ -CH2), 1.45 (9H, ms, OtBu); ESMS, M þ Na ion pattern
b
calculated for WC30H43N5O5S6Na: 950 (57.9), 951 (54.3), 952 (100),
953 (44.6), 954 (93.0), 955 (35.2), 956 (26.5), 957 (8.5); found 950
(67.1), 951 (56.3), 952 (100), 953 (44.3), 954 (90.2), 955 (37.8), 956
(31.8), 957 (10.1).
4.10. Preparation of cyclic complex 20
To a refluxing solution of 91.5 mg (0.180 mmol, 1.05 equiv.) of
W(CO)3(dmtc)2 [32] in 200 mL of degassed MeOH under N2 was
added via syringe a solution of 55 mg (0.17 mmol, 1 equiv.) of 17
dissolved in 5 mL of degassed MeOH. The solution immediately
took on a dark green color. The color turned to lemon yellow after
2 h of reflux. The methanol was evaporated and the crude product
purified by flash chromatography (ethyl acetate). A total of 80 mg
(63%) of pure 20 was obtained as an amorphous yellow solid. TLC
(1:1 EtOAc/hexanes), Rf 0.16; HPLC, Rt 7.02 min; 1H NMR (400 MHz,
CDCl3): 11.5e11.2 (2H, ms, 2 ꢀ CCH), 5.5e5.0 (1H, m, amide NH),
4.88 (1H, brs, urethane NH), 4.66e4.40 (1H, m, CaH), 4.35e3.55 (7H,
Fig. 5. The 1H NMR spectra of the alkyne hydrogens (11.5e10.7 ppm) in complex 19 in
d6-DMSO at temperatures ranging from 294 K to 360 K.
4.7. Preparation of Boc-Prg-Prg-OMe, 17
To a solution of 1.60 g (7.51 mmol, 1 equiv.) of 16 in 50 mL CH2Cl2
was added 1.23 g (7.50 mmol, 1 equiv.) of 15, followed by 1.5 mL
(8.6 mmol, 1.1 equiv.) of DIEA and 1.63 g (8.49 mmol, 1.1 equiv.) of
EDC. After stirring at 23 ꢁC for 18 h an additional 50 mL of CH2Cl2
was added, and the diluted solution was washed: 4 ꢀ 25 mL 0.1 M
HCl, 3 ꢀ 25 mL saturated NaHCO3 and 1 ꢀ 25 mL brine. The organic
layer was dried (MgSO4), filtered and evaporated to yield 1.57 g
m, CaH
2 ꢀ CH2
þ
OMe
þ
1 ꢀ dmtc Me), 3.54e3.24 (7H, m,
þ 1 ꢀ dmtc Me), 3.23e3.00 (6H, m, 2 ꢀ dmtc Me),
1.52e1.30 (9H, ms, OtBu); ESMS, M þ Na ion pattern calculated for
WC22H34N4O5S4Na: 767 (64.7), 768 (53.8), 769 (100), 770 (33.5),
771 (90.4), 772 (25.5), 773 (17.8), 774 (4.3); found 767 (65.0), 768

T.P. Curran et al. / Journal of Organometallic Chemistry 806 (2016) 12e21
21
Fig. 6. The 8 possible conformational isomers for complex 20.
[10] N. Ousaka, N. Tani, R. Sekiya, R. Kuroda, Chem. Commun. (2008) 2894e2896.
[11] Y. Azuma, H. Imai, T. Yoshimura, T. Kawabata, M. Imanishi, S. Futaki, Org.
Biomol. Chem. 10 (2012) 6062e6068.
(61.7), 769 (100), 770 (37.7), 771 (93.7), 772 (26.3), 773 (18.3), 774
(4.6).
ꢀ
ꢀ
ꢀ
[12] G. Rama, A. Arda, J.-D. Marechal, I. Gamba, H. Ishida, J. Jimenez-Barbero,
M.E. Vazquez, M. Vazquez-Lopez, Chem. Eur. J. 18 (2012) 7030e7035.
[13] C.M. Alvarez, R. García-Rodríguez, D. Miguel, Dalton Trans. 45 (2016)
Acknowledgments
ꢀ
963e972.
This work was supported by a grant from the National Science
Foundation (NSF-RUI CHE-0305325), by grants from the Trinity
College Faculty Research Committee, and by Student Research As-
sistant Grants from Trinity College. The Bruker Avance III NMR was
obtained with a National Science Foundation MRI grant (NSF-MRI
CHE-0619275), and some of the experiments were performed in a
laboratory that was renovated with a National Science Foundation
ARI grant (NSF-ARI CHE-0963165). We thank Jack Suitor for assis-
tance with the HPLC analyses.
[14] T.P. Curran, E.L. Handy, J. Organomet. Chem. 694 (2009) 902e907.
[15] J.F. Carson, L. Boggs, J. Org. Chem. 30 (1965) 895.
[16] C. Liu, X. Gu, Y.Z. Zhu, Bioorg. Med. Chem. Lett. 20 (2010) 6942.
[17] C. Liu, W. Guo, X. Shi, M.A. Kaium, X. Gu, Y.Z. Zhu, Eur. J. Med. Chem. 46 (2011)
3996.
ꢁ
[18] A.H. Henseler, C. Ayats, M.A. Pericas, Adv. Synth. Catal. 356 (2014) 1795.
[19] J.-L. Fauchere, O. Leukart, A. Eberle, R. Schwyzer, Helv. Chim. Acta 62 (1979)
ꢁ
1385.
[20] H. Willisch, W. Hiller, B. Hemmasi, E. Bayer, Tetrahedron 47 (1991) 3947.
[21] N.A. Abood, R. Nosal, Tetrahedron Lett. 35 (1994) 3669.
[22] E.M. Wallace, J.A. Moliterni, M.A. Moskal, A.D. Neubert, N. Marcopulos,
L.B. Stamford, A.J. Trapani, P. Savage, M. Chou, A.Y. Jeng, J. Med. Chem. 41
(1998) 1513.
Appendix A. Supplementary data
ꢀ
[23] A. Lopez, R. Pleixats, Tetrahedron Asymmetry 9 (1998) 1967.
[24] K. Lee, S.Y. Hwang, C.W. Park, Bioorg. Med. Chem. Lett. 9 (1999) 1013.
[25] A. Dondoni, P.P. Giovannini, A. Massi, Org. Lett. 6 (2004) 2929.
[26] Y. Ishii, R. Fujimoto, M. Mikami, S. Murakami, Y. Miki, Y. Furukawa, Org.
Process Res. Dev. 11 (2007) 609.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2016.01.023.
[27] D.J. Lee, K. Mandal, P.W.R. Harris, M.A. Brimble, S.B.H. Kent, Org. Lett. 11
(2009) 5270.
[28] N.J. Stanley, D.S. Pedersen, B. Nielsen, T. Kvist, J.M. Mathiesen, H. Brauner-
References
€
Osborne, D.K. Taylor, A.D. Abell, Bioorg. Med. Chem. Lett. 20 (2010) 7512.
[29] X. Li, T. Fekner, M.K. Chan, Chem. Asian J. 5 (2010) 1765.
[30] J. McConathy, D. Zhou, S.E. Shockley, L.A. Jones, E.A. Griffin, H. Lee, S.J. Adams,
R.H. Mach, Mol. Imaging 9 (2010) 329.
[31] G.W. Liechti, E. Kuru, E. Hall, A. Kalinda, Y.V. Brun, M. VanNieuwenhze,
A.T. Maurelli, Nature 506 (2013) 507.
[32] S.J.N. Burgmayer, J.L. Templeton, Inorg. Chem. 24 (1985) 2224.
[33] J.L. Templeton, R.S. Herrick, J.R. Morrow, Organometallics 3 (1984) 535e541.
[34] R.S. Herrick, J.L. Templeton, Organometallics 1 (1982) 842e851.
[35] J.R. Morrow, T.L. Tonker, J.L. Templeton, W.R. Kenan, J. Am. Chem. Soc. 107
(1985) 5004e5005.
[1] J.L. Templeton, Adv. Organomet. Chem. 29 (1989) 1.
[2] T.P. Curran, A.L. Grant, R.A. Lucht, J.C. Carter, J. Affonso, Org. Lett. 4 (2002)
2917.
[3] T.P. Curran, W.E. Smith, P.C. Hendrickson, J. Organomet. Chem. 711 (2012) 15.
[4] T.P. Curran, R.S.H. Yoon, B.R. Volk, J. Organomet. Chem. 689 (2004) 4837.
[5] T.P. Curran, A.B. Lesser, R.S.H. Yoon, J. Organomet. Chem. 692 (2007) 1243.
[6] T.P. Curran, A.N. Boynton, S.M. Berk, E.-M.C. Pedro, J. Organomet. Chem. 782
(2015) 31e36.
[7] M. Albrecht, P. Stortz, Chem. Soc. Rev. 34 (2005) 496e506.
ꢀ
[8] C. Clavaud, M. Heckenroth, C. Stricane, A. Menez, C. Dugave, Bioconjugate
Chem. 17 (2006) 807e814.
[9] M. Aufort, M. Gonera, N. Chaignon, L. Le Clainche, C. Dugave, Eur. J. Med.
Chem. 44 (2009) 3394e3401.
[36] Theoretical isotope patterns were calculated using a program available at a
website provided by Eawag, http://www.envipat.eawag.ch/.