χ-conopeptide pharmacophore development: Toward a novel class of norepinephrine transporter inhibitor (Xen2174) for pain

Article abstract of DOI:10.1021/jm9003413

Norepinephrine (NE) amplifies the strength of descending pain inhibition, giving inhibitors of spinal NET clinical utility in the management of pain. χ-MrIA isolated from the venom of a predatory marine snail noncompetitively inhibits NET and reverses all

Full text of DOI:10.1021/jm9003413

J. Med. Chem. 2009, 52, 6991–7002 6991
DOI: 10.1021/jm9003413
χ-Conopeptide Pharmacophore Development: Toward a Novel Class of Norepinephrine Transporter
Inhibitor (Xen2174) for Pain^
Andreas Brust,*,†, Elka Palant,†, Daniel E. Croker,† Barbara Colless,† Roger Drinkwater,† Brad Patterson,†
Christina I. Schroeder,‡ David Wilson,† Carsten K. Nielsen,§ Maree T. Smith,§ Dianne Alewood,† Paul F. Alewood,‡ and
Richard J. Lewis*,†,‡
§
†Xenome Ltd., 120 Meiers Road, Indooroopilly, Queensland, Australia 4068, ‡Institute of Molecular Biosciences and School of Pharmacy,
The University of Queensland, Brisbane, Queensland, Australia. Equal first authors.
Received March 17, 2009
Norepinephrine (NE) amplifies the strength of descending pain inhibition, giving inhibitors of spinal
NET clinical utility in the management of pain. χ-MrIA isolated from the venom of a predatory marine
snail noncompetitively inhibits NET and reverses allodynia in rat models of neuropathic pain. An
analogue of χ-MrIA has been found to be a suitable drug candidate. On the basis of the NMR solution
structure of this related peptide, Xen2174 (3), and structure-activity relationships of analogues, a
pharmacophore model for the allosteric binding of 3 to NET is proposed. It is shown that 3 interacts
with NET predominantly through amino acids in the first loop, forming a tight inverse turn presenting
amino acids Tyr7, Lys8, and Leu9 in an orientation allowing for high affinity interaction with NET. The
second loop interacts with a large hydrophobic pocket within the transporter. Analogues based on the
pharmacophore demonstrated activities that support the proposed model. On the basis of improved
chemical stability and a wide therapeutic index, 3 was selected for further development and is currently
in phase II clinical trials.
Introduction
MrIA-OH (NGVCCGYKLCHOC-OH, aka mr10a and
CMrVIB) and MrIB-OH (VGVCCGYKLCHOC-OH) are
two closely related peptides that were discovered in the venom
of the molluscivorous Conus marmoreus.9,13,14 Sharpe et al.
classified these peptides as χ-conopeptides based on activity
displayed at the neuronal norepinephrine transporter (NETa)
Marine natural products remain an important source of new
chemical diversity for drug discovery.1 The venom of the genus
Conus (marine cone snails) provides an exceptionally rich source
of highly evolved pharmacologically active compounds. These
predatory mollusks are equipped with a venom apparatus that
produces a species-specific cocktail of peptides to immobilize
prey comprising fish, mollusks, and worms. It is estimated that
the venom of a single Conus species may contain between 50 and
200 different venom components,2 translating to more than
50 000 active peptides across the genus. Similar to spider and
scorpion venoms, conopeptides are highly specific for their
evolved target and frequently possess constrained structures
containing multiple disulfide bonds. The disulfide framework
is interspersed by loops of amino acids that define the binding
motifs. The potency, selectivity, and chemical stability of venom
peptides, particularly conopeptides, make these compounds
attractive starting points for drug development.3-6 In fact,
several conotoxins are currently in clinical trials,5 with one
member of the ω-conotoxin family, MVIIA (ziconotide), on
the market for the treatment of chronic pain.7,8 On the basis of
this potential, Xenome has developed a synthetic conopeptide
library from nucleotide derived venom peptide sequences as
a discovery platform for novel drug lead discovery.9-12
a Abbreviations: NET, norepinephrine transporter (aka noradrena-
line transporter); NE, norepinephrine (aka noradrenaline); COO, ester
bond;
HBTU,
N,N,N0,N0-tetramethyl-O-(1H-benzotriazol-1-yl)-
uranium hexafluorophosphate; MSNT, 1-(2-mesitylensulfonyl)-3-
nitro-1H-1,2,4-triazole; EDT, ethanedithiol; TIPS, triisopropylsilane;
TFA, trifluoroacetic acid; DCC, N,N0-dicyclohexylcarbodiimide;
NEM, N-ethylmorpholine; DMAP, 4-dimethylaminopyridine; DMSO,
dimethyl sulfoxide; Acm, acetamidomethyl side chain protecting group
for cysteine; RP-HPLC, reversed-phase high-performance liquid chro-
matography; LC-MS, liquid chromatography coupled mass spectro-
metry; NMR, nuclear magnetic resonance; NOESY, nuclear overhauser
enhancement spectroscopy; TOCSY, total correlated spectroscopy;
FIDs, free induction decays; DQF-COSY, double quantum filtered
correlation spectroscopy; WATERGATE, water suppression by gradi-
ent-tailored excitation; E-COSY, exclusive correlation spectroscopy;
GPCR, G-protein-coupled receptors; CCI, chronic constriction injury;
IT, intrathecal; SAR, structure-activity relationship; SI, Supporting
Information. Natural occurring amino acids are abbreviated to standard
single or standard three letter codes: U represents L-pyroglutamic acid;
O, L-trans-hydroxyproline. Capital letters represent L-amino acids, and
lower case letters represent D-amino acids. Unnatural amino acids are
indicated using a three letter code in brackets such as the following:
[NAL], L-1-napthylalanine; [DMF], L-3,4-dimethoxyphenylalanine;
[DPA], L-β,β-diphenylalanine; [HTY], L-homotyrosine; [HFE], L-homo-
phenylalanine; [MEY], L-O-methyltyrosine; [CHA], L-cyclohexylala-
nine; [DMK], dimethyllysine; [ORN], ornithine; [HLY], L-homolysine;
[NLE], L-norleucine; [CIT], L-citrulline; [NML], N-methyl-L-leucine;
[FLA], L-2-furylalanine; [THI], L-2-thienylalanine; [PYA], L-3-pyridyl-
alanine; [BTA], L-3-benzothienylalanine; [TIC], 1,2,3,4-L-tetrahydro-
isoquinoline-3-carboxylic acid; [HCY], L-homocysteine; [BHK],
β-homolysine; [ABZ], 2-aminobenzoyl.
^ This publication is dedicated to Dianne Alewood: gone too early
and dearly missed by us all.
*To whom correspondence should be addressed. For A.B.: (phone)
þ617 3720 8055; (fax) þ617 3720 8388; (e-mail) andreas.brust@xenome.
com. For R.J.L.: (address) Xenome Ltd., 120 Meiers Road, Indooroo-
pilly, Queensland, Australia; (phone) þ61 7 3720 8055; (fax) þ61 7 3720
8388; (e-mail) richard.lewis@xenome.com.
r
2009 American Chemical Society
Published on Web 10/27/2009
pubs.acs.org/jmc

6992 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22
Brust et al.
in a noncompetitive (allosteric) manner.9,15 NET reuptake
inhibitors have potential in the treatment of neurological
disorders such as depression,16 schizophrenia,17 anxiety,18
chronic pain,19,20 and other psychotic disorders.21 The most
prominent reuptake inhibitors are the tricyclic antidepres-
sants (e.g., amitriptyline and imipramine), which inhibit both
the serotonin and norepinephrine reuptake and are used
systemically to treat major depressive disorders, generalized
anxiety disorder, and neuropathic pain.22
interactions of these amino acids with the transporter were
important for high affinity interactions. In this study we
conducted a detailed SAR analysis of χ-conopeptides and
generated a pharmacophore model for their allosteric binding
to NET. On the basis of superior chemical stability and its
efficacy and side effect profile, 3 (Xen2174) was developed for
clinical use.
Results and Discussion
The χ-conopeptides have two disulfide bonds connecting
Cys1-Cys4 and Cys2-Cys3.9 This arrangement of disulfide
bonds is termed a “ribbon” fold and distinguishes the
χ-conopeptides from other classes of conopeptides that pos-
sess a similar cysteine spacing but different cysteine connec-
tivity as seen in the “globular” folded R- and F-conopeptides
(Cys1-Cys3, Cys2-Cys4).23 Amidated 1 (MrIA-NH2) has
slightly higher affinity at NET compared to the native peptide
(MrIA-OH)9 but maintains the same folding arrangement as
determined by selective synthesis13 and NMR spectroscopy.24
The 3D NMR solution structure of 1 revealed that the rigid
ribbon-framework formed by the disulfide bonds induces a
β-hairpin with two antiparallel strands connected by an
inverse γ-turn, the same motif originally observed in the
NMR structure of 2 (MrIB).9 On the basis of their higher
affinity, all peptides investigated in this study were amidated
at the C-terminus.
A systematic alanine scan of 1 revealed that Gly6, Tyr7,
Lys8, and Leu9 which form the first intercysteine loop, and
His11 in the second loop, are important for binding to NET.15
With the use of NMR chemical shift analysis of 1 and
comparison with the alanine scan mutants, a perturbation
of the backbone structure was seen only upon replacement of
Gly6. This suggested that the overall structure did not change
except for the Gly6 analogue, indicating the side chain
Stability Assessment of 1. It is well-known that peptides
containing an Asn residue at the N-terminus can undergo
cyclization involving the β-carboxamide group. Figure 1A
depicts the proposed mechanism of aspartimide formation
with elimination of ammonia and subsequent hydrolytic
ring-opening to yield the R- or β-aspartyl peptide, respec-
tively, and a mass increase of 1 amu. The tendency toward
deamidation is dependent on the primary and secondary
structures of the peptide and the solvent.25-27 It was reported
that Asn-Gly and Asn-Ser sequences are far more prone to
this reaction than sequences containing combinations of Asn
with bulkier amino acids. 1 contains the reactive N-terminal
Asn-Gly sequence with aspartimide conversion being ob-
served during disulfide bond formation (pH 8.0 and pH 6.0)
and upon storage in an acetate buffer (pH 5.4). Because of
the difficult separation of 1 and the related R/β-aspartyl-
peptides by HPLC, an exact time course of the conversion
was difficult to determine. Mass spectrometry analysis of 1,
stored in acetate buffer (pH 5.4), monitored the aspartimide
formation by measuring the decrease of peak height for the
native peptide (MW = 1407.7) and the increase of the
aspartimide analogue (MW = 1408.7) (Figure 1B).
With the observed N-terminal instability of 1, analogues
replacing the first residue were synthesized and tested to
determine the importance of this residue on binding to NET.
Figure 1. Aspartimide formation: (A) schematic representation of aspartimide formation in peptides containing N-terminal asparagine
(Asn, N); (B) accurate mass measurement for 1 in acetate buffer (pH 5.4) at t = 0 days and t = 6 days. The conversion of native peptide (1) into
the R/β-aspartyl analogues of 1 can be seen. [MHþ 1407.6 amu (1); MHþ 1408.6 amu (R/β-aspartyl-1)].

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22 6993
Selection of 3 as a Preclinical Candidate. On the basis of
obtained NET binding data (Figure 2), ease of synthesis, and
peptide stability in buffer at pH7.4 (refer to SI), peptides 3, 5,
9, 18, 21, and 22 were selected for assessment to establish
efficacy, side effect profile, therapeutic window, and dura-
tion of action in an in vivo rat model of neuropathic pain
(Figure 3). The unilateral chronic constriction injury (CCI)
of the sciatic nerve rat model of neuropathic pain (CCI-
rats)28 was used to assess the ability of the analogues to
attenuate mechanical allodynia, a hallmark symptom of
neuropathic pain, in the ipsilateral (injured side) hindpaw.
At 14 days after CCI surgery, each peptide of interest was
administered as a bolus injection via a chronically implanted
intrathecal (IT) cannula, with the positive and negative
comparators in this study being morphine and vehicle res-
pectively. In Figure 3 panels A-D summarize observed paw
withdrawal threshold efficacy and time course for each tested
compound, while panels E-M summarize compound rela-
ted behavioral observations from assessments in CCI rats.
From the animal data it is clear that 1 nmol of 1 produced
similar efficacy as the lowest maximum efficacious dose
of morphine (17.5 nmol) but fewer side effects. Compared
with 1, peptides 5, 21, and 22 produced greater side effects, 9
produced a similar extent of side effects, and 3 and 18
produced fewer side effects. Interestingly, 21 produced tran-
sient efficacy in the ipsilateral hindpaw, whereas the effects
on the contralateral side were almost maximal and sustained,
indicating a preferential effect on the undamaged side. Of the
peptides producing similar or fewer side effects, efficacy was
transient or of relatively short duration for 9 and 18,
respectively, whereas 3 produced a sustained response for
3 h that was equivalent to 1 at 3 h (albeit with a reduced
maximal response). Dose-response studies confirmed that
higher doses of 3 more than completely reversed allodynia
and produced sustained efficacy beyond 24 h with little
change in the side effect profile observed.10 This superior
side effect profile and efficacy, combined with the improved
plasma stability of 3, 5, and 18 compared to 1 (refer to SI) has
finally lead to selection of 3 as the most promising candidate
for clinical development. A multiple target screen involving
dopamine and serotonin transporters, opioid receptors,
muscarinic receptors, and N-type calcium channel also
indicated the high NET selectivity of 3 (refer to SI). Func-
tional assays performed using human NET confirmed 3 as
a norepinephrine uptake inhibitor (log IC50 = -6.7; refer
to SI).
Figure 2. N-Terminal mutations of 1. Shown are a series of analo-
gues of 1 in which the N-terminal residues are replaced by various
amino acids and assayed for inhibition of [3H]nisoxetine binding to
the human norepinephrine transporter (hNET) and corresponding
affinity of the analogues compared to 1. Amino acids in bold signify
the point of difference from 1 and X... refers to the following
sequence: GYKLCHOC. All peptides represented were synthesized
as the C-terminal amide. Single letter abbreviations are used for
standard L- or D-amino acids with unusual amino acids abbreviated
with three letter codes and definitions available in the Abbreviations
footnote. Bars represent the mean of two to five experiments
performed in triplicate.
Improving the stability of this residue was seen to be critical
for the further development of the peptide as a drug candi-
date. Extensive libraries of Asn1 mutants of 1 were produced
by Fmoc solid phase chemistry. Random oxidation of
purified reduced peptides was performed under the same
conditions used for the synthesis of the native peptide with all
peptides displaying a similar oxidation profile with the
desired disulfide arrangement (Cys1-Cys4, Cys2-Cys3)
corresponding to the second eluting isomer using HPLC
conditions described. Figure 2 displays the binding affinities
of the Asn-1 analogues of 1 for NET, presented as pIC50
values. All disulfide bond isomers generated during the
synthesis process were submitted for assay. In all cases the
second eluting main isomer displayed the highest binding
affinity, and these isomers are compared and discussed.
The corresponding binding data of [Asn1]-1 analogues,
4-20 (Figure 2), indicate that a number of amino acids are
accommodated in the first position of the peptide with little
compromise on affinity. With this in mind, further analogues
were synthesized that extended the length of the chain by one
amino acid to explore the pocket available for binding. From
the in vitro binding data it can be seen that the addition of
tryptophan and homophenylalanine, 21 and 22, respectively,
display enhanced affinity for NET.
NMR Structure of 3. A detailed SAR analysis was under-
taken to help describe a pharmacophore model representing
the interaction of 3 with NET. Previous alanine scan data on
1
15 highlighted residues in loops 1 and 2 critical for binding to
NET. To investigate these residues further, amino acids 6-9
in loop 1 and amino acids 11 and 12 in loop 2 of 3 were
individually replaced with various amino acids sampling a
wide range of side chain chemistry. Following the first
generation of analogues, the more potent peptides from the
series were further derivatized using closely related amino
acids to fine-tune affinity and gain more detailed SAR
information for the development of the pharmacophore
model. A basic requirement to generate an interpretable
pharmacophore model is a high resolution NMR structure.
The NMR structure of 3 (Figure 4A) shows that the peptide
adopts a tertiary structure very similar to the structures of 124
and 2,9 as predicted from the HR chemical shift comparison
(Figure 4B). The structure of 3 comprised a ribbon-like fold

6994 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22
Brust et al.
Figure 3. In vivo data (CCI rat pain model). Concentrations chosen for testing were based on binding affinity, with peptides 21 and 22 having
higher binding affinity and thus tested at lower doses. The dose chosen for morphine is an efficacious dose. Peptides were administered as a
single bolus injection via a cannula chronically implanted into the intrathecal space. Antinociception was assessed using calibrated von Frey
filaments to determine changes in paw withdrawal thresholds in the hindpaw on the nerve damaged (ipsilateral) side relative to the
corresponding measurements determined just prior to peptide administration. (A-D) Data for morphine,10 vehicle,10 and peptides 1, 3,10
5, 9, 18, 21, 22; N = 4-6. Solid line represents data from the ipsilateral side and the dashed line from the contralateral (undamaged) side. The
response to vehicle indicates the baseline sensitivity for the ipsilateral and contralateral hindpaws. (E-M) Observations from in vivo testing of
selected peptides showing side effect profile using the CCI rat model of neuropathic pain. > indicates effects seen beyond 3 h time frame.
and a β-hairpin turn formed by residues 3-12, and an inverse
γ-turn at residues 6-8 that connects the strands from
residues 3-5 and 10-12. The observed hydrogen bonds in
the structure of 3 were identified on the basis of distances and
angles (Gly6-HN-Leu9-CO, Gly6-CO-Lys8-NH, Gly6-
CO-Leu9-NH, His11-HN-Cys4-CO, and Cys4-HN-
His11-CO). These identified hydrogen bonds corresponded
to the slowly exchanging amide protons observed by NMR.
Secondary HR chemical shifts of the various analogues of
3 were compared with that of 1 and 3 to determine if the
differences seen in binding affinity were due to interactions
within the binding site or to distortions in the backbone
leading to changes in the overall fold of the peptide. To
perform this study, secondary HR chemical shifts were
compared with random coil values29 of individual amino
acids within a peptide to gain a sensitive measure of back-
bone conformation.30-32 For a series of structurally related
peptides, secondary HR chemical shift data can be used to
identify the location but not the nature of local changes in
backbone conformation.33 Comparison of secondary HR
chemical shift data for 1 and 3 indicated that the backbone
structures of both peptides are essentially identical
(Figure 4B). In contrast to the disulfide-constrained portion
of 1 and 3, NMR analysis failed to define the N-terminal tail
structure, indicating this region was likely poorly con-
strained in solution. Despite this region being less structured,
most amino acid substitutions or additions to the N-termi-
nus could increase binding affinity in particular for the
N-terminal extended analogues 21 and 22 (Figure 2). Un-
fortunately, these changes increased the extent of side effects
seen in the in vivo CCI rat model of neuropathic pain
(Figure 3) believed to arise from off-target interactions.
These peptides were not further investigated, as these
changes in 21 and 22 represent an altered, extended pharma-
cophore accessing additional binding sites on NET and are
potentially responsible also for off-site-related unwanted
side effects, and hence, further analogues involving the
N-terminal tail were not synthesized or further investigated.

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22 6995
Figure 4. Structure of 3: (A) superimposition of the 20 lowest energy structures for 3 calculated using high resolution NMR; (B) comparison of
secondary HR chemical shift data obtained for 1 and 3 indicating an identical backbone structure.
H-Bond Turn Stabilization. In previous work performed
by Sharpe et al.15 the [G6A]-1 substitution was shown to
have significantly reduced pharmacological activity at NET
that appeared to correlate with a significant structural
perturbation identified from the secondary HR chemical
shift data. The structural importance of this residue is further
reflected, as it is conserved throughout the χ-conopeptide
family. We believe that the Gly residue allows H-bond
formation between Gly6-HN-Leu9-CO, Gly6-CO-Lys8-
NH, and especially Gly6-CO-Leu9-NH. No Gly6 replace-
ment with standard amino acids or D-Ala was able to
maintain activity at NET (data not shown), and thus, no
further mutations around this residue were pursued. To
further probe the importance of loop stabilizing H-bonds
in the peptide, in particular Gly6-CO to Leu9-NH, backbone
modifications were introduced. To achieve this, the amide
bond between Lys8 and Leu9 was replaced by an ester bond
(23) or by N-methylation of the amide bond (24) (Figure 5).
These modifications prevent the formation of the intramo-
lecular Gly6-CO to Leu9-NH hydrogen bond due to a lack of
a H-donor residue. The removal of the turn stabilizing
H-bond resulted in total loss of activity of the depsi-peptide
analogue (23) as measured by binding to NET. Furthermore,
N-methylation of the Leu9-NH amide (24) removed the
ability of this amide bond to act as a H-donor, resulting in
loss of affinity at NET, along with pronounced backbone
structure perturbation as seen in the secondary HR chemical
shift data for 24 (Figure 6). This led to the conclusion that the
backbone H-bonds identified in 3 are essential for the overall
stabilization of the hairpin scaffold in 3 and the ability to
bind with high affinity to NET.
29, which maintained comparable potency to 3. On compar-
ison of the secondary HR chemical shift data for selected
Tyr7 mutants with 3 (Figure 6), some significant local
perturbation for Lys8 was observed. For the analogue 36,
this was expected, as substitution with a proline has a
significant structural effect on the backbone. However, the
effects of Tyr7[HTY]-3 (27) and Tyr7[CHA]-3 (30) on the HR
proton of Lys8 were unexpected but may be explained by an
interaction of the Tyr7 phenolic-OH and the Lys8-NH2 as an
additional backbone stabilizing factor. The inhibition of
[3H]nisoxetine from NET membrane for the Tyr7 analogues
was reduced by even small changes such as the removal of the
phenol function in tyrosine, Tyr7Phe-3 (28), or the extension
of the side chain by one CH2 group, Tyr7[HTY]-3 (27). There
was also no binding for analogues with fused aromatic ring
systems, Tyr7W-3 (31) or Phe with a double methoxy sub-
stitution at the third and fourth position, Tyr7[DMF]-3 (25),
clearly highlighting a restricted binding pocket for Tyr7.
NMR analysis of the 20 lowest energy conformations of 3
revealed two thermodynamic preferred orientations of the
aromatic tyrosine moiety. Although a low energy barrier
separates the preferred orientations, considerations were
given to investigate if both orientations are able to interact
with NET. Diphenylalanine [DPA] was chosen to mimic
simultaneous populations of both preferred orientations;
however, this analogue, 26, did not display a similar binding
affinity to 3. This indicated that the preferred orientation
required for the tyrosine side chain at position 7 lies in a
narrow spatial window.
Lys8 Residue Analogues. Analysis of position 8 in the
NMR solution structure of 3 (Figure 4) revealed that the
aliphatic linear aminobutyl side chain of Lys8 has a high
degree of structural freedom. It is interesting to note that
truncation or extension of the lysine side chain, 3 f 43 or 3 f
44, respectively, abolished NET affinity. Even though the
Lys side chain appears to have a high degree of structural
freedom, the length of the freely rotating Lys chain is of
greatest importance in regard to affinity displayed at NET.
Further chemical diversity of this position was probed.
However, none of the conservative substitutions (42-45)
Tyr7 Residue Analogues. Analogues of 3 in which Tyr7 was
replaced with a variety of amino acids were synthesized and
assayed for NET binding (Figure 5, compounds 25-41).
These analogues included conservative substitutions such as
phenylalanine (28), methyltyrosine (29), homotyrosine (27),
and 2,4-dimethoxyphenylalanine (25), as well as nonconser-
vative substitutions with acidic, basic, and hydrophobic
amino acids. No inhibition was observed with these analo-
gues with the exception of the O-methyltyrosine analogue,

6996 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22
Brust et al.
Figure 5. Analogues of 3 replacing various residues. The top bar represents 3 with the substituted residue in the bars below. Peptides were
assayed for inhibition of [3H]nisoxetine binding to human norepinephrine transporter and represented as pIC50 values. The potency of
analogues with bars marked “<” represents estimates based on the lack of any inhibition detected at concentrations up to 100 μM. All peptides
represented were synthesized as the C-terminal amide. Single letter abbreviations are used for standard L- or D-amino acids with unusual amino
acids abbreviated with three letter codes, and definitions are available in the Abbreviations footnote. Bars represent the mean of two to five
experiments performed in triplicate.
or the nonconservative substitutions (46-52) displayed ac-
tivity at the NET. Secondary HR chemical shift comparison
of some Lys8 mutants showed an unchanged overall peptide
backbone structure with the exception of a surprising per-
turbation of the backbone for the homolysine (44) analogue
(Figure 6). This effect may be linked to the previously
surmised H-bond interaction with the phenolic-OH function
of Tyr7 acting to stabilize the structure.
Leu9 Residue Analogues. The aliphatic side chain of Leu9
displays a well-defined orientation projected orthogonally
outward from the β-hairpin plane (Figure 4). It is apparent
that the size of the aliphatic substitution plays a significant
role in the affinity of 3 for NET, since the valine analogue 53
was unable to displace [3H]nisoxetine while bulkier aliphatic
residues including norleucine (54), homoleucine (55), or
cyclohexylalanine (56) had increased affinity compared with

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22 6997
chemical shift data it can be seen that the larger fused ring
analogues showed some distortions in the backbone. How-
ever, these were not in the important turn region between
Gly6 and Leu9 but located toward the N-terminal tail region
(Figure 3). The substitution of [BTA], L-3-benzothienylala-
nine (69), at position 11 did lead to perturbations in the turn
backbone being reflected in loss of NET affinity (Figures 5
and 6).
Hyp12 Residue Analogues. Previous work15 demonstrated
that hydroxyproline (O) was not essential for χ-conopeptide
binding to NET. This is surprising, as proline analogues
usually have a strong structural effect, which often initiate
turns because of the formation of a kink in the backbone
forced by the cyclic structure. It was therefore unexpected
that the overall fold of 3 is sufficiently stable that replace-
ment of the turn-inducing cyclic Hyp (O) by an Ala residue is
tolerated without significant loss of activity when compared
to 3. Furthermore, a variety of Hyp replacements (75-80),
with the exception of aspartic acid (79), are tolerated without
loss of binding affinity to NET. The SAR results in relation
to the His position have shown that both basic and large
aromatic residues are tolerated in this position. With this
knowledge, replacements were made in 3 of the neighboring
Hyp residue to the basic Lys (75) and aromatic amino acids
like Tyr (77) or methyltyrosine (78). All these replacements
maintained binding affinity at NET. Thus, similar to the
neighboring His site, position 12 is able to accommodate a
variety of residues including larger aromatic residues. This
indicates that a large hydrophobic binding pocket can be
accessed by loop 2 amino acids, framed by Cys3 and Cys4, in
3. Secondary HR shift analysis of 78 indicated that the
mutation of hydroxyproline (3) to methyltyrosine (78) in-
duced some minor backbone structural changes (Figure 6).
However, these do not strongly influence the orientations of
the key binding residues of loop 1, and thus, affinity is little
altered.
Multiple Residue Replacement Analogues. The ribbon
cysteine framework (1-4, 2-3) of the χ-conopeptides ap-
pears to be very stable, accommodating single amino acid
mutations with little disruption to the backbone. The excep-
tion to this is the substitution of residues within the first loop
with bulky residues, which creates minor distortions to the
overall backbone observed in the secondary HR chemical
shift data presented above. It is anticipated that a change
in the cysteine stereochemistry with the introduction of
D-enantiomers would result in a disruption to the disulfide
framework obtained after oxidation. Not surprisingly, re-
placement of the four L-Cys residues by D-Cys (81) caused
complete loss of NET affinity. However, pairwise replace-
ments with D-Cys produced unexpected effects. Cys1 and
Cys4 replacement with D-Cys (84) was well tolerated,
whereas the double D-Cys2 and D-Cys3 mutant (82) lost all
affinity for NET. This finding suggests that Cys2 and Cys3
are critical to the stabilization of the inverse turn orientation
of residues 6-9. It is hypothesized that a stereochemical
variation in this scaffold bridge alters the topology of key
binding residues, this being further supported by the single
D-Cys mutants 83, 85, and 86. When the four Cys residues
were replaced with the methylene extended homocysteine
[HCY], 87 was also not tolerated, perhaps because it
increases torsional freedom of the disulfide scaffold, result-
ing in an altered backbone.
Figure 6. Secondary HR chemical shifts (ppm) showing significant
perturbations of various analogues compared to parent compound
3 represented in each of the panels by a closed circle. Panel A
displays secondary HR chemical shift (ppm) data for Tyr7 analo-
gues, namely, Y7[HTY]-3 (27), Tyr7[CHA]-3 (30), and Y7P-3 (36).
Panel B displays secondary HR chemical shift (ppm) data for Lys8
and Leu9 analogues, namely, K8[HLY]-3 (44), L9[NML]-3 (24),
and L9I-3 (58). Panel C displays secondary HR chemical shift (ppm)
data for His11 and Hyp12 analogues, namely, H11[NAL]-3 (68),
H11[BTA]-3 (69), and O12[MEY]-3 (78).
3, indicating that the aliphatic-hydrophobic binding patch
is large enough to accommodate bulky aliphatic residues. No
backbone conformational changes were observed in the
secondary HR chemical shift analysis for the Leu9 analogues
(Figure 6), again highlighting the structural stability of the
cysteine stabilized framework.
His11 Residue Analogues. The histidine imidazole moiety
adopts an orientation parallel to Leu9, displaying essentially
two main side chain conformations that project orthogonally
out from the β-hairpin plane (Figure 4). This orientation is
not maintained by the D-enantiomer analogue, 70, resulting
in decreased binding affinity compared to 3. Screening of
analogues 62-74 revealed that contribution of the histidine
amino acid to affinity at NET is influenced by aromatic
interactions as well by H-bond interactions from donor-
acceptor sites. The five-ring aromatic substituted alanine,
[FLA] (64), displayed similar binding affinity to 3. Similar
affinities to 3 were seen for six-ring aromatic analogues such
as 63, 65, and 66 along with fused ring systems, such as 67 and
68 and the arginine analogue 62. From the secondary HR
The information gained from the SAR analysis aided the
design of analogues displaying greater affinity for NET

6998 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22
Brust et al.
Figure 7. Proposed pharmacophore model of 3. Important interaction sites are highlighted in color, with the framed turn region being crucial
for activity at NET. The effect on potency of side chain modifications that drove the pharmacophore development compared with 3 is
indicated, along with representative side chain substitutions. Gly6 has no direct pharmacophore contribution but immense importance for
topology of the turn region (framed). The gray area represents the Tyr7 position interacting with NET via aromatic interactions as well as
H-bond acceptor interaction. The blue area represents the Lys8 position with this being an essential basic residue for NET binding. The brown
area represents the Leu9 position, which is able to assess a large hydrophobic patch on the norepinephrine transporter. The large purple area
(His11-Hyp12) represents a site that can bind to the transporter via aromatic or basic interactions.
compared with 3. Single amino acid substitutions such as
Y7[MEY]-3 (29) resulted in a slight increase in affinity, while
larger aliphatic substitutions at position 9, for example,
L9[CHA]-3 (56), significantly enhanced affinity for the
transporter in comparison with 3. Other improvements in
affinity were seen when His11 was substituted with larger
aromatic residues such as tryptophan (67). To determine if
these individual effects are additive, combinations of multi-
ple amino acid analogues were designed. Both Y7[MEY]-
3 (29) and H11W-3 (67) increased affinity by a small incre-
ment compared with 3. Interestingly, when both of these
substitutions were combined to form peptide Y7[MEY]-
H11W-3 (88), a 10-fold increase in affinity for the transpor-
ter was noted. The further substitution forming Y7[MEY]-
L9[CHA]-H11W-3 (89) increased binding affinity 20-fold
compared with 3. Binding affinity results for 88 clearly
demonstrate that these multiple substitutions are more than
additive. However, comparing peptide 89 with the single
substitution of L9[CHA]-3 (56) indicates that Y7[MEY] and
H11W substitutions are not contributing significantly to the
affinity for NET. It is not known if these changes in affinity
are due to structural changes within the molecule and were
not further investigated.
rigid hairpin-like conformation was generally maintained for
those peptides, thus preserving activity at the transporter,
despite the introduction of bulky residues at specific posi-
tions within the sequence. From these observations, it is
believed that the action of 3 at the norepinephrine transpor-
ter does not require a conformational change to the back-
bone structure of the peptide to access the binding pocket.
This allows for the development of a pharmacophore model
based on the solution NMR structure in combination with
NET affinity data (Figure 7).
Results from the SAR of 3 can be divided into three areas,
each focusing on particular sections within the sequence. The
first section relates to the N-terminal tail region of the
peptide, which when modified did arrive at peptides (21,
22) with higher binding affinity at the transporter compared
with the parent molecule 3. However, these modifications
appeared to increase undesirable side effects without a
concomitant increase in pain-relieving potency in the CCI
rat model of neuropathic pain. It is hypothesized that the side
effects are due to off-site effects. Increased binding to NET is
caused when the additional introduced residues extend-
ing the pharmacophore and interacting with regions on the
transporter are not accessible by the minimal pharmaco-
phore of 3. Replacement of the N-terminal Asn with pyro-
glutamic acid led to 3 with improved chemical stability as
well as good efficacy and high potency when administered by
the IT route in a widely utilized rat model of neuropathic
pain at doses that were well-tolerated. On the basis of these
improvements, 3 was progressed into preclinical and clinical
development.
Pharmacophore Model of 3. By definition, χ-conopeptides
contain two disulfide bonds with the connectivity being
Cys1-Cys4 and Cys2-Cys3. In combination with the loop
sizes of four amino acids in the first loop and two amino acids
in the second loop, this disulfide bond arrangement and the
multiple H-bonds result in a very stable core structure. As
indicated from secondary HR chemical shift experiments, the

Article
The second section relates to the second loop of the
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22 6999
acetonitrile, and methanol were supplied by Sigma Aldrich. The
resin used was Fmoc-Rink resin (0.65 mmol/g) supplied by
Auspep P/L, and Wang resin (0.81 mmol/g) was supplied by
Novabiochem. Ethane dithiol (EDT) was supplied by Merck.
Synthetic Strategy. Peptide amides were synthesized on an
Advanced ChemTech (ACT-396) automated peptide synthesi-
zer using Rink amide resin (0.05 nmol). Peptide acids were
synthesized on Wang resin loaded with the C-terminal
Fmoc-Cys(Trt)-OH using MSNT/1-methylimidazole activation
(4 equiv excess).34 Continued assembly for the peptide acids as
well as the complete assembly of peptide amides was performed
using HBTU in situ activation protocols35 to couple the Fmoc-
protected amino acid to the resin (4 equiv excess). After chain
assembly and final Fmoc deprotection the reaction block was
transferred onto an ACT-LABTECH shaker and peptides were
cleaved from the resin at room temperature (RT) in TFA/H2O/
TIPS/EDT (87.5:5:5:2.5) for 3 h. Cold diethyl ether (30 mL) was
then added to the filtered cleavage mixture and the peptide
precipitated. The precipitate was collected by centrifugation and
subsequently washed with further cold diethyl ether to remove
scavengers. The final product was dissolved in 50% aqueous
acetonitrile and lyophilized to yield a fluffy white solid. The
crude, reduced peptide was examined by reversed-phase HPLC
for purity and the correct molecular weight confirmed by
electrospray mass spectrometry (ESMS).
Synthesis of Ester Bond Containing Peptide 23. Peptide 23 was
assembled using the above standard Fmoc protocol with the
following exception. L-R-Hydroxyisovaleric acid was used
instead of Fmoc-L-Leu-OH and coupled using DCC/NEM
activation.36 The ensuing formation of the ester bond bet-
ween Fmoc-Lys(Boc)-OH and the hydroxy-peptide-resin was
achieved after extended reaction times of 48 h using DCC/
NEM/DMAP activation. Further continuation of assembly
was performed using the described standard Fmoc protocol.
Cleavage as well as oxidation was performed for peptide 23
under identical conditions to other peptides.
peptide corresponding to residues His11 and Hyp12. It was
noted from the Ala scan data15 that substitution of His11
with Ala reduced potency at the transporter whereas the
substitution of Hyp12 with Ala did not significantly change
potency compared to 1. His11 and Hyp12 could be replaced
by larger aromatic residues, suggesting that the binding
pocket available to this section of 3 is somewhat larger.
From the secondary HR chemical shift data it would appear
that the topology of these two residues is not critical to
binding, as distortions in this region were tolerated. This can
also be said for the two cysteines flanking the second loop,
Cys4 and Cys13, where L-Cys to D-Cys changes were well
tolerated.
The third section relates to the sensitive loop 1 turn region
represented by the four amino acids, GYKL, which form the
region of the peptide that is the most sensitive to topological
alterations. From the Ala scan data it was seen that sub-
stituting any one of these amino acids with Ala abolished the
activity at NET compared with the parent compound, 1. The
topology of these residues is restrained by the cysteine
framework as well as an intraloop network of H-bonds.
Tyr7 and Lys8 appear to be fundamental for binding to NET
with limited allowable substitutions for Tyr7 ([MEY]) and
no allowable substitutions found for Lys8, whereas Leu9
provided additional opportunities for improving affinity,
with the introduction of larger aliphatic residues at this
position enhancing potency.
Examination of the pharmacophore model (Figure 7)
reveals that χ-conopeptides interact with the transporter
from only one side of the hairpin-like turn, mainly residues
Leu9 and His11 which protrude orthogonally outward. The
other side of the peptide contains the disulfide bridges, which
do not appear to be involved in binding to the transporter.
Interestingly, all analogues with significantly higher affinity
than 1 or 3 required the introduction of unnatural amino
acids. Unfortunately, analogues at the N-terminus that had
improved affinity for the transporter produced undesirable
side effects and/or had reduced efficacy in the rat model of
neuropathic pain. This study demonstrates that peptides
found within venom can be optimized for therapeutic use.
1 identified in the venom of Conus marmoreus provided a
suitable starting point for SAR studies and lead optimiza-
tion. The resulting optimized clinical lead, 3, has now entered
phase II clinical trials.
Random Disulfide Bond Formation in Peptides 1-89. Pure,
reduced peptides (1 mg/mL) were oxidized by stirring at RT in
30% DMSO/0.1 M NH4OAc, pH 6.0, for 16 h. The solutions
were subsequently diluted to a DMSO concentration of <5%,
prior to RP-HPLC purification and lyophilization.
Selective Disulfide Bond Formation. Selectively folded MrIA-
OH and MrIA-NH2 were assembled on Wang (acid) and Rink
(amide) resin, respectively. For each of the different isomers one
pair of cysteine residues was incorporated into the sequence,
using Cys(Trt) protection, while the other Cys pair used the
orthogonal protection of Cys(Acm). After cleavage and purifi-
cation of the reduced peptide the first disulfide bond was formed
using the standard 30% DMSO/0.1 M NH4OAc buffer (pH 6)
method described above.
The formation of the second disulfide bond was performed by
dissolving the purified, singly oxidized di-(Acm) peptide at
1 mM in 80% acetic acid/water. Iodine (10 equiv) dissolved in
a small volume of ethyl acetate was added and the mixture
stirred at RT. The progress of the Acm-deprotection and
subsequent disulfide bond formation were monitored by direct
injection ESMS at regular intervals until completion (approx
30-90 min). The reaction was then quenched by addition of
ascorbic acid solution (10 mg/mL), resulting in a decolorization
of the solution. After dilution with five times the volume of
water and adjusting to pH 3, the fully oxidized peptide was
loaded onto a RP-HPLC column and purified as described
below.
HPLC Analysis and Purification. Analytical HPLC runs were
performed using a Shimadzu HPLC system LC10A with a dual
wavelength UV detector set at 214 and 254 nm. A reversed-
phase C-18 column (Zorbax 300-SB C-18; 4.6 mm ꢀ 50 mm)
with a flow rate of 2 mL/min was used. Gradient elution was
performed with the following buffer systems: (solvent A) 0.05%
TFA in water and (solvent B) 0.043% TFA in 90% acetonitrile
Experimental Section
Materials. Protected Fmoc-amino acid derivatives were pur-
chased from Novabiochem or Auspep P/L. The following side
chain protected amino acids were used: Cys(Acm), Cys(Trt),
His(Trt), Hyp(tBu), Tyr(tBu), Lys(Boc), Trp(Boc), Arg(Pbf),
Asn(Trt), Asp(OtBu), Glu(OtBu), Gln(Trt), Ser(tBu), Thr(tBu),
Tyr(tBu). All other Fmoc amino acids were unprotected. Unu-
sual Fmoc amino acids [HTY], [ORN], [HLY], [CIT], and
[BHK] were used with following side chain protections: -OH
as -OtBu, -NH2 as -NH(Boc), -COOH as -COO(tBu), and
-SH as -S(Trt). All other unusual Fmoc amino acids were
used unprotected. Unusual amino acids were sourced from
Novabiochem, Auspep, CSPS-Pharmaceuticals, or NeoMPS.
L-R-Hydroxyisovaleric acid was used unprotected. Dimethyl-
formamide (DMF), dichloromethane (DCM), diisopropylethy-
lamine (DIEA), and trifluoroacetic acid (TFA) were supplied by
Auspep P/L (Melbourne, Australia) as peptide synthesis grade.
2-(1H-Benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexaflu-
orophosphate (HBTU), triisopropylsilane (TIPS), HPLC grade

7000 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22
Brust et al.
in water, from 0% B to 80% B in 20 min. The crude peptides and
oxidized peptides were purified by semipreparative HPLC on a
Shimadzu HPLC system LC8A associated with a reversed-
phase C-18 column (Vydac C-18, 25 cm ꢀ 10 mm) running at
a flow rate of 5 mL/min with a 1%/min gradient of 0% B to 40%
B. The purity of the final product was evaluated by analytical
HPLC (Zorbax 300SB C-18, 4.6 mm ꢀ 100 mm) with a flow rate
of 1 mL/min and a 1.67%/min gradient of B (5-45%).
Purities of synthesized peptides (1, 3, 5-8, 10, 12, 17, 20,
22-24, 30, 35, 43, 45, 49, 53, 54, 56, 59, 60-62, 65-67, 71, 72,
74-76, 81, 83-88) were all greater than 95% with the exceptions
listed in the Support Information. The minor compounds in
peptide samples consisted of other disulfide connectivity iso-
mers that closely eluted to the desired 1-4, 2-3 connectivity
isomer. Isolated minor isomers were also tested in the NET in
vitro assay with none displaying affinity for the transporter.
Peptide Concentration Assessment. Peptide concentrations
used in in vitro screening were calculated on the basis of peak
size detected at 214 nm by HPLC. Peak size was calibrated using
a peptide standard (in this case 1) with known peptide content
established by amino acid analysis. Molecular extinction coeffi-
cients are calculated for the standard, and the peptide of interest
applying increments was established by Buck et al.37 By use of
the Lambert-Beer law, the peptide concentration is calculated
on the basis of absorptions of standard and sample using
calculated extinction coefficients.
Electrospray Mass Spectrometry (ESMS). Electrospray mass
spectra were collected inline during analytical HPLC runs on an
Applied Biosystems API-150 spectrometer operating in the
positive ion mode with an OR of 20, Rng of 220, and Turbos-
pray of 350ꢀ. Masses between 300 and 1800 amu were detected
(step 0.2 amu, dwell 0.3 ms). Accurate mass measurement of the
aspartimide formation of 1 (Figure 1B) was achieved by mass
detection between 1402 amu and 1414 amu (step 0.1 amu, dwell
30 ms).
NMR Structure Evaluation of 3. Samples prepared for NMR
spectroscopy contained approximately 7-8 mg of purified 3
dissolved in the following solvent systems: 95% H2O/5% D2O
(v/v) at pH 3.38 and 99.96% D2O. D2O (99.96%) was obtained
from Cambridge Isotope Laboratories, Woburn, MA. pH
values were measured at 298 K, and no corrections for deuter-
ium isotope effects were made.
Spectra were recorded on a Bruker ARX-500 or Bruker
Advance 600 spectrometer at 290 and 298 K. The carrier
frequency in all experiments was set at the center of the spectra
on the water resonance frequency, and quadrature detection was
used in both dimensions. 2D spectra were acquired in phase
sensitive mode using time-proportional phase incrimination for
quadrature detection in the t1 dimension.38 Exclusive homo-
nuclear 2D experiments consisted of a DQF-COSY,39 TOCSY40
using a MLEV-17 spin lock sequence41 with an isotropic mixing
time of 80 ms, E-COSY,42 and NOESY43 with mixing times of
200 and 300 ms. Water suppression was achieved using a
modified WATERGATE44 sequence. All 2D spectra were col-
lected with over 4096 data points in the F2 dimension and 512 or
600 increments in the F1 dimension. The presence of slowly
exchanging amide protons were analyzed through a series of 1D
and TOCSY spectra run immediately after dissolving the fully
protonated sample in D2O.
Spectra were processed on a Silicon Graphics workstation
using XWINNMR (Bruker) software. The t1 dimension was
zero-filled to 2048 real data points, and 90ꢀ phase-shifted sine
bell window functions were applied prior to Fourier transfor-
mation. Chemical shifts were referenced to internal sodium
2,2-dimethyl-2-silapentane-5-sulfonate (DSS). Processed spec-
tra were analyzed and assigned within the program Sparky.45
Assignment procedure followed the well-established spin system
identification and sequential assignment technique.46
were integrated and calibrated in Sparky45 and distance con-
straints derived using CYANA.47 Corrections for pseudoatoms
were added to distance constraints where needed.48 Backbone
3
dihedral angle restraints were derived from JHN-HR coupling
constants measured from a one-dimensional spectrum or from
line shape analysis of antiphase cross-peak splitting in the DQF-
COSY spectrum. Angles were restrained following the standard
parametrization of the Karplus relationship46 to -120 ( 30ꢀ for
3JHN-HR = 8.0-9.5 Hz (Val3, Cys4, Cys5, Leu9, Cys10, His11)
3
and -65 ( 15ꢀ for JHN-HR = 3.0-5.8 Hz (Cys13). An addi-
tional φ angle restraint of -100 ( 80ꢀ was applied to Tyr7 where
the intraresidual HR-HN NOE was clearly weaker than the
NOE between HN and the HR of the preceding residue.49 The
intense intraresidue HR-HN NOE for Lys8, combined with a
3JHN-HR value of ∼7 Hz, allowed its φ angle to be restrained to
þ50 ( 40ꢀ.50,51
χ1 dihedral angles were derived for three residues (Cys4,
3
Cys10, Cys13) from JRβ coupling constants measured from
E-COSY spectra in combination with HN-Hβ2, HN-Hβ3
,
HR-Hβ2, and HR-Hβ3 NOE peak intensities.52 The χ1 angles
were restrained to þ180 ( 30ꢀ (Cys4, Cys10) or -60 ( 30ꢀ
(Cys13).
Preliminary structures were calculated using a torsion angle
simulated annealing within CYANA.47 Final structures were
calculated using simulated annealing and energy minimization
protocols within CNS, version 1.1.53 The starting structures
were generated using random (φ, ψ) dihedral angles and energy
minimized to produce structures with the correct local geome-
try. A set of 50 structures were generated by a torsion angle
simulated annealing protocol.54,55 This protocol involves a
high-temperature phase comprising 4000 steps of 0.015 ps of
torsion angle dynamics, a cooling phase with 4000 steps of
0.015 ps of torsion angle dynamics during which the tempera-
ture is lowered to 0 K, and finally an energy minimization phase
comprising 500 steps of Powell minimization. Structures con-
sistent with restraints were subjected to further molecular
dynamics and energy minimization in a water shell, as described
by Linge and Nilges.56 The refinement in explicit water involves
the following steps: first, heating to 500 K via steps of 100 K,
each comprising 50 steps of 0.005 ps of Cartesian dynamics;
second, 2500 steps of 0.005 ps of Cartesian dynamics at 500 K
before a cooling phase where the temperature is lowered in steps
of 100 K, each comprising 2500 steps of 0.005 ps of Cartesian
dynamics. Finally, the structures were minimized with 2000
steps of Powell minimization. Structures were analyzed using
PROMOTIF57 and PROCHECK-NMR.58
Chemical Shift Analysis of Analogues of 3. All spectra were
recorded on a Bruker 600 Advance spectrometer equipped with
an x, y, z gradient unit. Synthetic samples (∼2-11 mM) were
used for HR chemical shift determinations. All samples were
examined in 90% H2O/10% D2O (pH 3.5) except for 72 (His to
[NAL] replacement) which required 20% CD3CN to dissolve.
Restraints acquired for HR chemical shift determinations were
from spectra recorded at 298 K. 1D proton NMR and 2D
TOCSY41 experiments were run for all peptides. For peptides
that displayed differences in the backbone chemical shifts
compared to 315 and therefore could not be unambiguously
assigned, additional NOESY43 experiments were run. These
peptides included 24, 68, 69, and 82. For all the other peptides,
the residue replacements did not alter the backbone conforma-
tion and a TOCSY experiment provided satisfactory data for
assignment of the HR chemical shifts. All spectra were recorded
using Bruker TopSpin in a phase sensitive mode over 12 ppm
with 4096 data points in F2, 90-256 FIDs, 8-64 scans, and a
recycle delay of 2.0 s. Water suppression for TOCSY and
NOESY experiments was achieved using a modified WATER-
GATE sequence. Spectra were processed using XWINMR. The
chemical shifts were referenced internally to DSS at 0.00 ppm
Structure Calculations. Cross-peaks from NOESY spectra
recorded in 95% H2O/5% D2O with a mixing time of 300 ms
and compared to the chemical shifts for amino acids in random
€
46
coil, as described by Wuthrich. Secondary shift changes of

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22 7001
>0.1 ppm were considered as significant59 and were used as an
indication of structural perturbation upon amino acid muta-
tions.
peptide with apparent antinociceptive activity. J. Biol. Chem. 2000,
275 (42), 32391–32397.
(14) Balaji, R. A.; Ohtake, A.; Sato, K.; Gopalakrishnakone, P.; Kini,
R. M.; Seow, K. T.; Bay, B.-H. R-Conotoxins, a new family of
conotoxins with unique disulfide pattern and protein folding.
J. Biol. Chem. 2000, 275 (50), 39516–39522.
(15) Sharpe, I. A.; Palant, E.; Schroeder, C. I.; Kaye, D. M.; Adams, D.
J.; Alewood, P. F.; Lewis, R. J. Inhibition of the norepinephrinþe
transporter by the venom peptide chi-MrIA. Site of action, Na
dependence, and structure-activity relationship. J. Biol. Chem.
2003, 278, 40317–40323.
(16) Brunello, N. J.; Mendlewicz, J.; Kasper, S.; Leonard, B.;
Montgomery, S.; Nelson, J.; Paykel, E.; Versiani, M.; Racagni, G.
The role of noradrenaline and selective noradrenaline reuptake inhibi-
tion in depression. Eur. Neuropsychopharmacol. 2002, 12, 461–475.
(17) Arnsten, A. F. Adrenergictargetsforthetreatmentofcognitivedeficits
in schizophrenia. Psychopharmcology (Berlin) 2004, 174, 25–31.
(18) Tanaka, M.; Yoshida, M.; Emoto, H.; Ishii, H. Noradrenaline
systems in the hypothalamus, amygdala and locus coeruleus are
involved in the provocation of anxiety: basic studies. Eur. J.
Pharmacol. 2000, 405, 397–406.
Radioligand Binding Studies. For radioligand binding studies,
membranes from transfected COS-7 cells were incubated in
96-well plates with [3H]nisoxetine (4.3 nM) in the absence or
presence of χ-conopeptide (1 nM to 100 μM, in triplicate) in
buffer for 1 h at RT, as previously described.15 Filter-retained
radioactivity was quantified by liquid scintillation counting.
Curve fitting of concentration-response data was performed
by nonlinear regression using individual data points with
GraphPad Prism 3.0. The equation of Cheng and Prusoff60
was used to convert IC50 values for [3H]nisoxetine displacement
to pKi values, since χ-conopeptides act competitively with this
ligand.9 Statistical analyses were undertaken using the Graph-
Pad Prism software package with a statistical significance
criterion of P < 0.05.
In Vivo Studies. The CCI rat model of neuropathic pain was
performed as described in Nielsen et al.10 Three animals were
used for each analogue tested, with dose selected on the basis of
potency to inhibit the NET transport of NE.
(19) Drummond, P. D. Noradrenaline increases hyperalgesia to heat in
skin sensitised by capsaicin. Pain 1995, 60, 311–315.
(20) Furst, S. Transmitters in antinociception in the spinal cord. Brain
Res. Bull. 1999, 48, 129–141.
(21) Ressler, K. J.; Nemeroff, C. B. Role of norepinephrine in the
pathophysiology and treatment of mood disorders. Biol. Psychia-
try 1999, 46, 1219–1223.
(22) Ziegler, D. Painful diabetic neuropathy: treatment and future
aspects. Diabetes/Metab. Res. Rev. 2008, 24 (S1), S52–S57.
(23) Kang, T. S.; Radic, Z.; Talley, T. T.; Jois, S. D. S.; Taylor, P.; Kini,
R. M. Protein folding determinants: structural features determin-
ing alternative disulfide pairing in R- and χ/λ-conotoxins. Biochem-
istry 2007, 46, 3338–3355.
(24) Nilsson, K. P. R.; Lovelace, E. S.; Caesar, C. E.; Tynngard, N.;
Alewood, P. F.; Johansson, H. M.; Sharpe, I. A.; Lewis, R. J.; Daly,
N. L.; Craik, D. J. Solution structure of χ-conopeptide MrIA, a
modulator of the human norepinephrine transporter. Pept. Sci.
2005, 80, 815–823 (PDB 2ew4) .
Acknowledgment. We thank Norelle L. Daly for assistance
in generation of the NMR structure and related graphics. This
work was supported in part by an AusIndustry START Grant
and an NHMRC Program Grant.
Supporting Information Available: HPLC purity data of all
peptides, details of peptide quantification used in the determi-
nation of concentration used in in vitro assays, buffer and
plasma stability data, receptor selectivity data of 3, and in vitro
functional assay results for 3. This material is available free of
charge via the Internet at http://pubs.acs.org.
(25) Blankemeyer-Menge, B.; Nimtz, M.; Frank, R. An efficient meth-
od for anchoring Fmoc-amino acids to hydoxyl-functionalised
solid supports. Tetrahedron Lett. 1990, 31, 1701–1704.
(26) Schnolzer, M.; Alewood, P.; Jones, A.; Alewood, D.; Kent, S. B. In
situ neutralization in Boc-chemistry solid phase peptide synthesis.
Rapid, high yield assembly of difficult sequences. Int. J. Pept.
Protein Res. 1992, 40, 180–193.
(27) Blankenship, J. W.; Balambika, R.; Dawson, P. E. Probing back-
bone hydrogen bonds in the hydrophobic core of GCN4. Biochem-
istry 2002, 41, 15676–15684.
References
(1) Newman, D. J.; Cragg, G. M. Marine natural products and related
compounds in clinical and advanced preclinical trials. J. Nat. Prod.
2004, 67, 1216–1238.
(2) Jones, R. M.; Bulaj, G. Conotoxins. New vistas for peptide
therapeutics. Curr. Pharm. Des. 2000, 6, 1249–1285.
(3) Armishaw, C. J.; Alewood, P. F. Conotoxins as research tools and
drug leads. Curr. Protein Pept. Sci. 2005, 6, 221–240.
(4) Livett, B. G.; Gayler, K. R.; Khalil, Z. Drugs from the sea:
conopeptides as potential therapeutics. Curr. Med. Chem. 2004,
11, 1715–1723.
(5) Alonso, D.; Khalil, Z.; Satkunanthan, N.; Livett, B. G. Drugs from
the sea: conotoxins as drug leads for neuropathic pain and other
neurological conditions. Mini-Rev. Med. Chem. 2003, 3, 785–787.
(6) Shen, G. S.; Layer, R. T.; McCabe, R. T. Conopeptides: from
deadly venoms to novel therapeutics. Drug Discovery Today 2000,
3, 98–106.
(28) Bennett, G. J.; Xie, Y. K. A peripheral mononeuropathy in rat that
produces disorders of pain sensation like those seen in man. Pain
1988, 33, 87–107.
(29) Stein, E. G.; Rice, L. M.; Brunger, A. T. Torsion-angle molecular
dynamics as a new efficient tool for NMR structure calculation.
J. Magn. Reson. 1997, 124, 154–164.
(30) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J.;
Swaminathan, S.; Karplus, M. CHARMM: a program for macro-
molecular energy, minimization, and dynamics calculations.
J. Comput. Chem. 2004, 4, 187–217.
(7) Garber, K. Peptide leads new class of chronic pain drugs. Nat.
Biotechnol. 2005, 23, 399.
(31) Hyberts, S. G.; Goldberg, M. S.; Havel, T. F.; Wagner, G. The
solution structure of eglin c based on measurements of many NOEs
and coupling constants and its comparison with X-ray structures.
Protein Sci. 1992, 1, 736–751.
(32) Gehrmann, J.; Alewood, P. F.; Craik, D. J. Structure determina-
tion of the three disulphide bond isomers of R-conotoxin GI: a
model for the role of disulphide bonds in structural stability.
J. Mol. Biol. 1998, 278, 401–415.
(33) Nielsen, K. J.; Skjaerbaek, N.; Dooley, M.; Adams, D. A.;
Martensen, M.; Dodd, P. R.; Craik, D. J.; Alewood, P. F.; Lewis,
R. J. Structure-activity studies of conantokins as human
N-methyl-D-aspartate receptor modulators. J. Med. Chem. 1999,
42, 415–426.
(8) Wermeling, D. P.; Berger, J. R. Ziconotide infusion for severe
chronic pain: case series of patients with neuropathic pain. Phar-
macotherapy 2006, 26, 395–402.
(9) Sharpe, I. A.; Gehrmann, J.; Loughnan, M. L.; Thomas, L.;
Adams, D. A.; Atkins, A.; Palant, E.; Craik, D. J.; Adams, D. J.;
Alewood, P. F.; Lewis, R. J. Two new classes of conopeptides
inhibit the alpha 1-adrenoceptor and noradrenaline transporter.
Nat. Neurosci. 2001, 4, 902–907.
(10) Nielsen, C. K.; Lewis, R. J.; Alewood, D.; Drinkwater, R.; Palant,
E.; Patterson, M.; Yaksh, T. L.; McCumber, D.; Smith, M. T. Anti-
allodynic efficacy of the chi-conopeptide, Xen2174, in rats with
neuropathic pain. Pain 2005, 118, 112–124.
(11) Lima, V.; Mueller, A.; Kamikihara, S. Y.; Raymundi, V.;
Alewood, D.; Lewis, R. J.; Chen, X.; Minneman, K. P.; Pupo, A.
S. Differential antagonism by conotoxin rho-TIA of contractions
mediated by distinct alpha1-adrenoceptor subtypes in rat vas
deferens, spleen and aorta. Eur. J. Pharmacol. 2005, 508, 183–192.
(12) Lewis, R. J.; Garcia, M. L. Therapeutic potential of venom
peptides. Nat. Rev. Drug Discovery 2003, 2, 790–802.
(13) McIntosh, J. M.; Corpuz, G. O.; Layer, R. T.; Garrett, J. E.;
Wagstaff, J. D.; Bulaj, G.; Vyazovkina, A.; Yoshikami, D.; Cruz,
L. J.; Olivera, B. M. Isolation and characterization of a novel conus
(34) Blankemeyer-Menge, B.; Nimtz, M.; Frank, R. An efficient meth-
od for anchoring Fmoc-amino acids to hydroxyl-functionalised
solid supports. Tetrahedron Lett. 1990, 31, 1701–1704.
€
(35) Schnolzer, M.; Alewood, P.; Jones, A.; Alewood, D.; Kent, S. B. In
situ neutralization in Boc-chemistry solid phase peptide synthesis.
Rapid, high yield assembly of difficult sequences. Int. J. Pept.
Protein Res. 1992, 40 (3-4), 180–193.
(36) Blankenship, J. W.; Balambika, R.; Dawson, P. E. Probing back-
bone hydrogen bonds in the hydrophobic core of GCN4. Biochem-
istry 2002, 41 (52), 15676–15684.

7002 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 22
Brust et al.
(37) Buck, M. A.; Olah, T. A.; Weitzmann, C. A.; Cooperman, B. S.
Protein estimation by the product of integrated peak area and flow
rate. Anal. Biochem. 1989, 182, 295–299.
(50) Omecinsky, D. O.; Holub, K. E.; Adams, M. E.; Reily, M. D.
Three-dimensional structure analysis of mu-agatoxins: fur-
ther evidence for common motifs among neurotoxins with
diverse ion channel specificities. Biochemistry 1996, 35 (9), 2836–
2844.
(51) Ludvigsen, S.; Poulsen, F. M. Positive theta-angles in proteins by
nuclear magnetic resonance spectroscopy. J. Biomol. NMR 1992,
2 (3), 227–233.
€
(38) Marion, D.; Wuthrich, K. Application of phase sensitive two-
dimensional correlated spectroscopy (COSY) for measurements
of 1H-1H spin-spin coupling constants in proteins. Biochem.
Biophys. Res. Commun. 1983, 113 (3), 967–974.
(39) Rance, M.; Sorensen, O. W.; Bodenhausen, G.; Wagner, G.; Ernst,
R. R.; Wuthrich, K. Improved spectral resolution in COSY 1H
NMR spectra of proteins via double quantum filtering. Biochem.
(52) Wagner, G.; Braun, W.; Havel, T. F.; Schaumann, T.; Go, N.;
€
€
Wuthrich, K. Protein structures in solution by nuclear magnetic
Biophys. Res. Commun. 1983, 117 (2), 479–485.
resonance and distance geometry. The polypeptide fold of the basic
pancreatic trypsin inhibitor determined using two different algo-
rithms, DISGEO and DISMAN. J. Mol. Biol. 1987, 196 (3),
611–639.
(40) Braunschweiler, L.; Ernst, R. R. Coherence transfer by isotropic
mixing: application to proton correlation spectroscopy. J. Magn.
Reson. 1983, 53 (3), 521–528.
€
(41) Bax, A.; Davis, D. G. MLEV-17-based two-dimensional homo-
nuclear magnetization transfer spectroscopy. J. Magn. Reson.
1985, 65 (2), 355–360.
(53) Brunger, A. T.; Adams, P. D.; Rice, L. M. New applications of
simulated annealing in X-ray crystallography and solution NMR.
Structure 1997, 5 (3), 325–336.
€
(54) Rice, L. M.; Brunger, A. T. Torsion angle dynamics: reduced
variable conformational sampling enhances crystallographic struc-
ture refinement. Proteins 1994, 19 (4), 277–290.
€
(55) Stein, E. G.; Rice, L. M.; Brunger, A. T. Torsion-angle molecular
dynamics as a new efficient tool for NMR structure calculation.
(42) Griesinger, C.; Sorensen, O. W.; Ernst, R. R. Practical aspects of
the ECOSY technique. Measurement of scalar spin-spin coupling
constants of peptides. J. Magn. Reson. 1987, 75 (3), 474–492.
(43) Jeener, J.; Meier, B. H.; Bachmann, P.; Ernst, R. R. Investigation
of exchange processes by two-dimensional NMR spectroscopy.
J. Chem. Phys. 1979, 71 (11), 4546–4553.
J. Magn. Reson. 1997, 124 (1), 154–164.
(44) Piotto, M.; Saudek, V.; Sklenar, V. Gradient-tailored excitation for
single runtime NMR spectroscopy of aqueous solutions. J. Biomol.
NMR 1992, 2 (6), 661–665.
(56) Linge, J. P.; Nilges, M. Influence of non-bonded parameters on the
quality of NMR structures: a new force field for NMR structure
calculation. J. Biomol. NMR 1999, 13 (1), 51–59.
(57) Hutchinson, E. G.; Thornton, J. M. PROMOTIF, a program to
identify and analyze structural motifs in proteins. Protein Sci. 1996,
5 (2), 212–220.
(45) Goddard, T. D.; Kneller, D. G. SPARKY 3 NMR Software;
University of California: San Francisco, CA.
€
(46) Wuthrich, K. NMR of Proteins and Nucleic Acids; Wiley-Inter-
science: New York, 1986.
(58) Laskowski, R. A.; Rullmannn, J. A.; MacArthur, M. W.; Kaptein,
R.; Thornton, J. M. AQUA and PROCHECK-NMR: programs
for checking the quality of protein structures solved by NMR.
J. Biomol. NMR 1996, 8 (4), 477–486.
€
€
(47) Guntert, P.; Mumenthaler, C.; Wuthrich, K. Torsion angle
dynamics for NMR structure calculation with the new program
DYANA. J. Mol. Biol. 1997, 273 (1), 283–298.
€
(48) Wuthrich, K.; Billeter, M.; Braun, W. Pseudo-structures for the 20
(59) Wishard, D. S.; Sykes, B. D.; Richards, F. M. The chemical shift
index: a fast and simple method for the assignment of protein
secondary structure through NMR spectroscopy. Biochemistry
1992, 31, 1647–1651.
(60) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes 50%
inhibition (I50) of an enzymatic reaction. Biochem. Pharmacol.
1973, 22 (23), 3099–3108.
common amino acids for use in studies of protein conformations by
measurements of intramolecular proton-proton distance con-
straints with nuclear magnetic resonance. J. Mol. Biol. 1983, 169
(4), 949–961.
(49) Clubb, R. T.; Ferguson, S. B.; Walsh, C. T.; Wagner, G. Three-
dimensional solution structure of Escherichia coli periplasmic
cyclophilin. Biochemistry 1994, 33 (10), 2761–2772.