bovine chromaffin cells, Aplysia neurons; b, Discovered by peptide activity at nAChRs heterologously expressed in Xenopus oocytes; c, Discovered by peptide physico-chemical characteristics
Trang 1M I N I R E V I E W
Physico-chemical characterization and synthesis of neuronally active a-conotoxins
Marion L Loughnan and Paul F Alewood
Institute for Molecular Bioscience, The University of Queensland, Brisbane, Australia
The high specificity of a-conotoxins for different neuronal
nicotinic acetylcholine receptors makes them important
probes for dissecting receptor subtype selectivity New
sequences continue to expand the diversity and utility of
the pool of available a-conotoxins Their identification
and characterization depend on a suite of techniques
with increasing emphasis on mass spectrometry and
micro-scale chromatography, which have benefited from recent
advances in resolution and capability Rigorous physico-chemical analysis together with synthetic peptide chemistry
is a prerequisite for detailed conformational analysis and to provide sufficient quantities of a-conotoxins for activity assessment and structure–activity relationship studies Keywords: a-conotoxins; Conus; peptide synthesis; post-translational modifications; sulfotyrosine
Classification, primary structure and biology
of a-conotoxins
Cone snails are a group of hunting gastropods that
incapacitate their prey, which consists of worms, molluscs
or fish, by envenomation Conotoxins from the venom of
cone snails are small disulfide-rich peptide toxins that act at
many voltage-gated and ligand-gated ion channels They
can be grouped according to their molecular form into
several superfamilies, each defined by characteristic
arrange-ments of cysteine residues (not necessarily a single pattern),
and characteristic highly conserved precursor signal
sequence similarities Individual conopeptide families
within a superfamily are denoted by Greek letters and
contain peptides that have a particular disulfide framework
and target homologous sites on a particular receptor [1]
Each of the characterized conopeptides is named using
a convention that indicates the activity (Greek letter), the
source species from which the peptide was first isolated
(Arabic letter(s)), the disulfide framework category (Roman
numeral) and the order of discovery within that category
(Arabic capital letter) [1] For example a-AuIB belongs to
the a-conotoxin family and was the second peptide, B, with
that framework, I, isolated and reported from Conus aulicus
[1,2] The names of some conotoxins deviate from this
nomenclature convention because their discovery preceded
its formulation Hence some a-conotoxin names do not conform to the alphabetical identifier system used to indicate order of discovery of peptides with a specified disulfide framework from the venom of any one species The framework identifiers I and II are both used in reference
to disulfide frameworks of the A superfamily without distinction
The A superfamily is so far comprised of the K+channel blocking jA familiy and the a and aA families, which together with the w family act at the nicotinic acetylcholine receptor (nAChR) No aA or w conopeptides have been reported to block neuronal nicotinic receptors with high affinity Rather, they are generally muscle-specific nicotinic receptor antagonists [1] The a-conotoxins fall into two categories depending on whether they act at muscle-type or neuronal-muscle-type receptors The neuronally active a-conotoxins are the focus of this minireview
The known a-conotoxins consist of 12–19 amino acids Most a-conopeptides have four cysteine residues and the general sequence GCCXmCXnC The disulfide connectivity
is between alternate cysteine residues (I-III, II-IV)
The numbers of amino acid residues encompassed by the second and third cysteine residues (m) and the third and fourth cysteine residues (n) are the basis for a further division into several structural subfamilies (a3/5, a4/3, a4/6 and a4/7) [1,3,4] For example a4/6-AuIB belongs to the 4/6 disulfide loop size subgroup of the a-conotoxin family The neuronally active a-conotoxins are typically from the a4/7, a4/6 and a4/3 subfamilies (Table 1) Peptides from the most abundant a4/7subfamily are typically 16 residues in length and range from 1600 to 1900 Da in mass However there have been recent additions to this subfamily in which
Correspondence to P F Alewood, Institute for Molecular Bioscience,
The University of Queensland, Brisbane, QLD 4072, Australia.
Fax: + 61 73346 2101, Tel.: + 61 73346 2982,
E-mail: P.Alewood@imb.uq.edu.au
Abbreviations: c-CRS, c-carboxylation recognition sequence;
nAChRs, nicotinic acetylcholine receptors; RT, retention time;
PTM, post-translational modification; TPST, tyrosyl-protein
sulfotransferase; TCEP, tris(2-carboxyethyl)phosphine; M-biotin,
maleimide-biotin; NEM, N-ethylmaleimide; IAM, iodoacetamide.
(Received 22 January 2004, revised 16 March 2004,
accepted 6 April 2004)
Trang 2there have been extensions at the N-terminus or C-terminus
to a length of up to 19 residues and the mass range has been
extended to almost 2200 Da [5] For example, there are
three additional residues at the N-terminus in the case of
GID [5] Peptides from the a4/3 and a4/6 subfamilies are
typically 12 and 15 residues, respectively Unusually, EI has
the same disulfide framework as a4/7 conotoxins that target
neuronal nAChRs but has been reported to antagonize the
neuromuscular receptor as do the a3/5 and aA conotoxins
[1,6]
There is a conserved proline between the second and third
cysteine residues in almost all a-conotoxins except ImIIA
and ImII [7,8] However, the former has not been confirmed
to be a neuronally active a-conotoxin, despite its sequence
similarity with ImI and ImII There is also a conserved
serine residue between the second and third cysteine residues
in many a-conotoxins The residue N-terminal to the first
cysteine residue of the sequence is in most cases glycine,
although exceptions are the recently isolated peptides GID
and PIA that have c-carboxyglutamic acid and proline,
respectively, in that position (Table 1) [5,9] More generally,
the residues in the first loop tend to fit into defined
categories, whereas the second loop seems to have greater
heterogeneity of residues There appears to be a relationship between selected sequence motifs and receptor subtype specificity and these sequence patterns may be a basis for further defining subclasses within the neuronally active members of the a-conotoxin family [9a]
There are many interesting features of the biology of Conus species and the functional applications of the a-conotoxins in their venom It has been conjectured that
of the estimated 500 Conus species, each appears to make at least one nAChR antagonist [1] However, for some Conus species pharmacological screening of crude venom samples has not shown a-conotoxin activity (A Nicke &
M Loughnan, unpublished results) Nonetheless it has become apparent that in any one species there may be multiple peptides that target nAChRs [1], and it seems likely that the complement of neuronally acting a-conotoxins in one species may cover a range of subtype specificities There are examples of combinations of muscle-type and neuronal-type a-conotoxins in a single species, particularly in the case
of the fish-eating Conus species For example, C.geographus venom contains the muscle-acting a-conotoxins GI and GII together with the neuronally acting a-conotoxins GIC and GID [1,5,10] and C.magus venom contains the
Table 1 Comparison of selected known a-conotoxins from a4/7, a4/6 and a4/3 families, their selectivity for mammalian nAChR subtypes, size, route of discovery, method of synthesis and reference for synthesis Asterisks (*) indicate an amidated C-terminus The letters O and Y denote hydroxyproline and sulfotyrosine, respectively The letter Y
~~ denotes sulfotyrosine identified after original sequence was published The symbol c denotes c-carboxyglutamic acid Dashes indicate gaps in the sequence alignment Mass (monoisotopic) in daltons, given for disulfide-bonded form Conserved cysteine residues are shown in red and a highly conserved proline in the first loop is shown in green The peptides Vc1.1, GIC, ImII and ImIIA were identified by prediction from the nucleic acid sequence Other peptides were identified by isolation of the peptides from the venom ducts
in response to activity assays or by physico-chemical characteristics Im peptides are from C.imperialis, Au from C.aulicus, M from C.magus, Ep from C.episcopatus, Pn from C.pennaceus, P from C.purpurascens, G from C.geographus and E from C.ermineus Prey groups are denoted by p,
m, v for piscivores, molluscivores and vermivores, respectively Discovery and synthesis methods were as follows: a, Discovered by peptide activity
at nAChRs in native tissues (e.g bovine chromaffin cells, Aplysia neurons); b, Discovered by peptide activity at nAChRs heterologously expressed
in Xenopus oocytes; c, Discovered by peptide physico-chemical characteristics, and confirmed by synthesis and assay; d, Discovered by gene sequencing with peptide sequence deduced from cDNA library obtained by RT-PCR of cone snail mRNA; e, Synthesised by Fmoc assembly, trifluoroacetic acid cleavage and directed disulfide formation (off-resin); f, Synthesised by Fmoc assembly, modified trifluoroacetic acid cleavage and air oxidation in ammonium bicarbonate for disulfide formation; g, Synthesised by tBoc assembly, HF cleavage and air oxidation in ammonium bicarbonate for disulfide formation; N/A, not available.
ImI G CC SD P R C AWR C * v a7 1350.5 c,a,e [20]
ImII A CC SDRR C RWR C * v a7 1508.6 d,e [8] AuIB G CC SY P P C FATNPD- C * m a3/b4 1571.6 b,e [2] AuIA G CC SY P P C FATNSDY C * m (less active) 1724.6 b,e [2] AuIC G CC SY P P C FATNSGY C * m (less active) 1666.6 b,e [2] AnIA CC SH P A C AANNQDY C * v a3/b2, a7 1673.6 c,f [21] AnIB GG CC SH P A C AANNQDY C * v a3/b2, a7 1787.6 a,f [21] AnIC GG CC SH P A C FASNPDY C * v a3/b2, a7 1805.6 a,f [21] MII G CC SN P V C HLEHSNL C * p a3b2; a6b2b3 1709.7 b,e [13] EpI G CC SD P R C NMNNPDY C * m a3b2/a3b4; a7 1866.6 c,a,f [24] Vc1.1 G CC SD P R C NYDHPEI C * m a3a7b4/a3a5b4 1805.7 d,a,g [22] PnIA G CC SL P P C AANNPDY
[A10L]PnIA G CC SL P P C ALNNPDY
PnIB G CC SL P P C ALSNPDY
GIC G CC SH P A C AGNNQHI C * p a3b2 1608.6 d,e [10] GID IRD c CC SN P A C RVNNOHV C # p a3b2 2184.6 b,g [5] PIA RDP CC SN P V C TVHNPQI C * p a6b2b3 1980.8 d,e [9]
EI RDO CC YH P T C NMSNPQI C * p (muscle-type) 2091.8 a [6] a
Synthesis method and activity refer to the unsulfated peptides.
Trang 3muscle-acting a-conotoxin MI and the neuronally acting
a-conotoxin MII [11–13] A biological interpretation of this
emerging pattern of paired ligand types is that prey capture
might rely on the combination of muscle-acting antagonists
to cause paralysis and neuronally acting antagonists to
inhibit the flight-or-fight response [10,14] Although distinct
peptide complements have been attributed to individual
species [15], there are instances of a single a-conotoxin
sequence occurring in more than one species For example,
GID from C.geographus has also been isolated from
C.tulipavenom [5]
Conus venoms together provide an array of ligands
with selectivity for various neuronal nAChR subtypes
(Table 1, [9a]) Evolutionarily, this diversity of toxins has
been generated by a hypermutation process that allows
protection of conserved cysteine residues and high
substi-tution rates for the intervening residues in the mature
toxin peptides [3,16,17] Each venom peptide is processed
from a prepropeptide and the three defined regions of this
precursor (signal sequence, proregion, mature toxin) have
different rates of divergence [3,16] Proposed
diversifica-tion mechanisms include gene duplicadiversifica-tion and subsequent
diversifying selection, or targeted gene mutation with
some sophisticated molecular regulation, perhaps based
on repair processes or recombination processes acting in
discrete exon regions [16–19] Prey-driven diversifying
selection may be a factor for dominant expressed toxins,
given the feeding specificity of cone snail species [17] It
has been suggested that nicotinic ligands from fish-hunters
are more likely than those from snail and worm-hunters
to target vertebrate nAChRs with high affinity [1] Besides
the piscivores (fish-hunters) C.magus, C.geographus and
C.purpurascens, other species that have also yielded
neuronally active a-conotoxins include the vermivores
(worm-hunters) C.imperialis and C.anemone, and the
molluscivores (mollusc-hunters) C.aulicus, C.victoriae,
C.pennaceusand C.episcopatus (Table 1) [2,5,7–10,13,20–
24] Many more species are represented in a-conotoxin
sequence information contained in patent documents, for
example [25] However these are not within the scope of
this review because their activities have not been reported,
although the patent applications reflect the commercial
interest in this class of conopeptides as potential
candi-dates for drug development
Many of the neuronally active a-conotoxins show a
high conservation of the local backbone conformation
although the surfaces are unique [3,25a] The common
structural scaffold suggests that the hypervariability of the sidechain groups confers peptide specificity for different neuronal nAChR subtypes [3] Although the a-conotoxins are considered to have rigid structures, multiple inter-convertible isomers may exist in solution [1,3] (see also below) This potential heterogeneity is an important issue
in the isolation, analysis and chemical synthesis of these peptides
Post-translational modifications of neuronally active a-conotoxins
A feature of conotoxins in general is that they are relatively richly endowed with a wide spectrum of post-translational modifications (PTMs) and this aspect has been comprehen-sively reviewed elsewhere [14,15,26] The classical published a-conotoxin sequences contained comparatively few mod-ifications apart from disulfide bridges and C-terminal amidation However, more recently isolated peptides have expanded the list of modifications and they now seem comparable with the rest of the conotoxins in this respect (Table 2) A thorough exploration of the significance of these modifications for the function of a-conotoxins is yet to
be completed and reported
Disulfide bridge formation is a basic feature of the a-conotoxins with their defined cysteine spacing and disulfide connectivity Non-native disulfide-bonded cono-peptides are often considered to be inactive but this is not always the case Intriguing results have been obtained in structure-function studies of a synthetic variant of a-AuIB with non-native disulfide bond connectivity, where an enhancement of biological activity was observed [27] Hydroxylation of proline has been observed in several neuronally acting a-conotoxins but of these only GID has been described [5] Hydroxyproline also occurs elsewhere in the A superfamily: in the muscle-specific a-EI, in aA-EIVA, EIVB and PIVA and in w-PIIIE [1] The significance of this modification has not been determined Many other cono-toxins that contain multiple hydroxyproline residues also have naturally occurring under-hydroxylated variants such
as in the example of TVIIA [28] and perhaps the modifi-cation may not be critical However there was no evidence
of a variant of GID with proline in place of the single hydroxyproline residue [5]
Amidation of the C-terminus is a feature of most conotoxins and so far occurs in the majority of the a-conotoxins with the exceptions of GID and the
muscle-Table 2 Post-translational modifications (PTMs) in neuronal a-conotoxins.
Disulfide bridge formation Yes all
Sulfation of tyrosine Yes EpI, PnIA, PnIB, AnIA, AnIB, AnIC [24,31,21]
Cyclization of N-terminal Gln No
Bromination of tryptophan No
Isomerization of tryptophan No
Epimerization of other residues (L fi D) No
Trang 4acting SII [1,5] The functional significance of the nature of
the C-terminus in neuronal-type a-conotoxins has not been
established However a recent study of the effects of an
amidated or free carboxyl C-terminus on activity of a
synthetic a-conotoxin (AnIB) at nAChR subtypes expressed
in oocytes showed subtype specific differences in activity [21]
Carboxylation of glutamic acid to c-carboxyglutamic
acid has been reported in GID [5] It has been estimated that
10% of conopeptides contain c-carboxyglutamic acid and
the conantokins (NMDA receptor antagonists) have
mul-tiple c-carboxyglutamic acid residues that are important
for maintaining their three dimensional structure [15,29]
The existence of c-carboxylation recognition sequences
(c-CRSs) in conopeptide precursors from some Conus
species has been established and an enzyme responsible for
glutamic acid carboxylation in conantokin G has been
described [30] The c-CRS regions in conantokin G and
bromosleeper peptides [30] are highly dissimilar suggesting
thatthissequenceinformationcannotnecessarilybeextended
to other conopeptide families c-CRSs in a-conotoxin
precursors have not been described
Sulfation of tyrosine has been observed in EpI, PnIA,
PnIB, AnIA, AnIB and AnIC [21,24,31] The mechanism of
sulfation of tyrosine by an enzyme, tyrosyl-protein
sulfo-transferase (TPST), has not been elucidated It may involve
a recognition sequence in the peptide precursor [15] or a
consensus sequence in the mature peptide [32]
Alternat-ively, secondary structure may be the major determinant of
sulfation and the TPST might broadly recognize any
sufficiently exposed tyrosine residue [33,34] The
sulfotyro-sine-containing a-conotoxins have not previously been
reported to have substantially different activity from the
unmodified variants [15,24] Recent comparisons of EpI
with [Y15]EpI, and AnIB with [Y16]AnIB found about
three-fold and ten-fold reduced activities, respectively, of the
unsulfated forms relative to the sulfated peptides [21,35] In
the case of AnIB, tyrosine sulfation selectively influenced the
binding to the mammalian a7 but not the a3b2 subtype [21]
The effects of substitution of phosphotyrosine for
sulpho-tyrosine in a-conotoxins have not been investigated
Post-translational modifications such as O-glycosylation
of serine or threonine residues, bromination of tryptophan,
isomerization of tryptophan or Lfi D epimerization of
other residues have been observed in conotoxins from other
families [1,15] but not reported for a-conotoxins However,
characterizations of a-conotoxins have not routinely
inclu-ded tests for isomerization and epimerization modifications
Aspects of the biosynthesis of conotoxins in the cone snail
and the mechanisms for incorporation of PTMs have
garnered considerable interest [15] Hypermutation of
amino acid residues is a feature of the mature toxins but
in contrast the prepropeptide precursor sequence,
partic-ularly the signal sequence, seems to be highly conserved for
each family of conotoxins [3,16] Precursor sequences are
available for conotoxins from most families but there have
been relatively few precursors published in journals for the
a-conotoxins (Table 3) Nevertheless it is interesting to
examine the available prepropeptide sequence information
for ImIIA and Vc1.1, ImII, GIC and PIA, each of which
was identified by prediction from a genomic DNA clone
[7–10,22] Comparison of the peptide sequences GID, EI,
ImIIA, GIC and PIA suggests that there are anomalies in Tab
Trang 5relation to GID, EI and PIA, which each have an
N-terminal sequence motif RDX It is tempting to speculate
about the relationship of this motif to the potential dibasic
cleavage sites for generation of the mature peptides from the
prepropeptides, particularly in those cases where the residue
at position X becomes post-translationally modified in the
mature peptide However there are also many other
sequences reported in patent documents that have not been
considered here and for many of those the deduced cleavage
site does not have a dibasic motif [25] There has been
increasing utility of molecular biology techniques for
identification of new a-conotoxin sequences Reliance on
the gene sequence of a conopeptide alone without
verifica-tion from the peptide sequence might miss PTM sites, or
misidentify the cleavage site for generation of the mature
peptide Much further work is necessary to elucidate the
mechanisms of incorporation of PTMs in the biosynthesis
of conopeptides Appropriate methods of analysis need
to be addressed to ensure recognition of these PTMs,
if present, in the course of characterization of native
a-conotoxins
Analysis of neuronally active a-conotoxins
using HPLC and MS, including identification
of post-translational modifications
Isolation and identification
Standard procedures for identification and isolation of
a-conotoxins generally incorporate separations using
reversed-phase HPLC, size exclusion or ion exchange
chromatography in combination with mass-based screening
and functional screening Most of the a-conotoxins
identi-fied so far are relatively hydrophilic and hence tractable
Mass-based screening entails searching by LC/MS or MS
alone for components with a mass in the defined range for
a-conotoxins and two disulfide bonds (identified by partial
reduction and alkylation studies and MS/MS) It may also
include diagnostic LC/MS for recognition of some
post-translational modifications and possibly MS/MS for
recog-nition of conserved sequence motifs The small size range
of a-conotoxins would seem ideal for MS-based sequence
determination However, complete de novo sequencing of
conopeptides by MS/MS is still considered experimental,
and primary sequence information is usually obtained by
Edman degradation sequencing, interpreted in conjunction
with MS data for the intact molecule Efficient sequence
analysis usually requires that the peptides are reduced and
the cysteine residues alkylated in order to verify their
identification Although there are standard procedures for
reduction and alkylation, their application to conopeptide
analysis and characterization is by no means trivial, and
often optimization on a case-by-case basis is required [14]
Sample amounts may be limiting even with the enhanced
sensitivity of current automated sequence analysis
instru-ments
Chromatography and structural heterogeneity
Several a-conotoxins have a characteristic asymmetric peak
under standard reversed-phase HPLC elution conditions
[5,36] The anomalous chromatographic behaviour is seen
particularly with a-CnIA, a-MI, a-GI and a-GID for both native and synthetic forms and persists even after repeated refractionation [5,36,37] The asymmetry may be more pronounced under isocratic elution conditions It presum-ably reflects structural heterogeneity and can be interpreted
as a slow interconversion between two conformers; this conclusion has been supported by the results of structure studies NMR studies have shown the existence of two distinct interconvertible conformers for GI and multiple conformers of CnIA [36,37] This heterogeneity may yet prove to be an important feature of some a-conotoxins in the understanding of structure-function relationships and the interaction of these ligands with the nAChR
Identification of PTMs Methods for the characterization of PTMs in conotoxins have been reviewed elsewhere, with particular emphasis
on the utility of MS [14], but specific aspects relevant to a-conotoxins (identification of C-terminal amidation, sulfotyrosine, hydroxyproline and c-carboxyglutamic acid) are revisited here The identification of the PTM is usually confirmed by synthesis of the modified a-conotoxin and comparison with the natural peptide
Identification of the nature of the C-terminus The identification of C-terminal amidation or a free carboxyl terminus in an a-conotoxin is usually straightforward with the one mass unit difference between the two forms readily apparent from the monoisotopic mass determined by high resolution mass spectrometry There may be difficulties in interpretation of MS data when there are ambiguities arising from, for example, asparagine to aspartic acid, or glutamine to glutamic acid changes [14] The a-conotoxins EpI, PnIA, GIC, GID, AnIA and AnIB contain pairs of asparagine residues [5,10,21,23,24] (Table 1), and deamida-tion may confound MS data for these peptides
Determination of sulfotyrosine Identification of sulfo-tyrosine in peptides is usually by mass spectrometry and the lability of the sulfogroup in mass spectrometry analysis allows recognition of the modification, and differentiation of sulfotyrosine and phosphotyrosine [38] Characterization of the sulfotyrosine-containing a-cono-toxin EpI was undertaken by a combination of mass spectrometry and modified amino acid analysis [24] The conotoxins a-PnIA and a-PnIB from C.pennaceus were initially identified and reported as unmodified sequences although an unidentified mass discrepancy had been recognized [23] The verification of tyrosine sulfation in a-PnIA and a-PnIB and revision of those sequences were made in an investigation of labile sulfo- and phospho-peptides by electrospray MALDI and atmospheric pressure MALDI mass spectrometry [31] The sulfation
of three conotoxins from C.anemone was identified on the basis of LC/MS under different conditions together with the difference between the observed mass and that predicted from primary Edman sequence data [21] The presence of either sulfation or phosphorylation may
be indicated when liquid chromatography/electrospray ionization mass spectrometry shows doubly protonated species of the modified a-conotoxins with additional related
Trang 6ions at +40 m/z (+80 Da) Further examination by
MALDI MS or ESI MS in both positive and negative ion
modes, at high and low energy, or MALDI high-energy
collision-induced dissociation, can be undertaken to confirm
tyrosine sulfation by distinguishing features of ionization
and fragmentation [31,38,39]
Determination of hydroxyproline and c-carboxyglutamic
acid The presence of these modified residues in
a-conotoxins is generally determined by a combination
of mass spectrometry, Edman N-terminal sequencing and
amino acid analysis [14] Hydroxyproline residues can be
reliably identified in the course of N-terminal Edman
sequencing of peptides [14] Mass spectrometry has also
been used for the analysis of peptides containing
hydroxy-proline and bromotryptophan, by resolution,
high-accuracy precursor ion scanning utilizing fragment ions
with mass-deficient mass tags [40] c-Carboxyglutamic acid
residues are readily recognized in electrospray mass
spectrometry under standard conditions because of the
lability of the extra carboxyl group ()44 Da) ([5,41] and as
shown in Fig 1) This residue can not be reliably identified
by standard N-terminal Edman sequencing procedures
because inefficient extraction of the polar derivative
generates a largely blank cycle [14,42], although there
may be residual amounts of glutamic acid Modified
Edman sequencing procedures, modified amino acid
analysis or colorimetric c-carboxyglutamic acid assays
can also be used to confirm this residue
Disulfide linkage determination Identification of closely spaced cystine residues in peptides with complex disulfide linkage patterns is still considered a significant analytical challenge and determination of disul-fide linkages in even the relatively simple a-conotoxins can still pose difficulties There have been several studies on characterization of closely spaced, complex disulfide bond patterns in peptides, usually based on stepwise reduction and differential alkylation of cysteine residues in a sequen-tial manner (Fig 2), although there may be complications because of disulfide shuffling via the thiol-disulfide exchange reaction These studies are of relevance to the characteriza-tion of neuronal a-conotoxins although many of the studies have been validated with other well-characterized peptides (some of them including a-conotoxin SI) and proteins One
of the original comprehensive studies of disulfide bond linkage relevant to conotoxins involved analysis of differ-entially alkylated products by Edman N-terminal sequen-cing and MS [43] A subsequent study specific for conopeptide analysis described differential alkylation fol-lowed by tandem mass spectrometry to determine disulfide bond connectivity, and indicated that reduction and alky-lation under acidic conditions is preferred to avoid condi-tions that promote scrambling of disulfide bonds [44] A further modification has been the use of iodination labelling
of the peptide to obtain better separation of intermediates
in partial reduction and alkylation studies [45] Disulfide
Fig 2 Scheme of disulfide determination Diagram of the general strategy for determination of disulfide linkage of conotoxins by dif-ferential reduction and alkylation of cystine residues and subsequent analysis by LC/ESI MS/MS or by Edman sequencing [43,44,46] (A) Partial reduction of peptide containing disulfide bonds by incubation with a low concentration of tris(2-carboxyethyl)phosphine (TCEP) (0.1–0.5 m M ) at 65 C for 10–15 min, generating partially reduced peptides that were immediately alkylated by N-ethylmaleimide (NEM) (two to five-fold molar excess over the Cys residues of the analyte) present in the TCEP solution (B) Complete reduction with dithio-threitol and alkylation by iodoacetamide (IAM) to label the Cys residues of the peptide that were not reduced by TCEP (C) Analysis of the resulting peptides differentially alkylated with NEM and IAM to identify the disulfide linkages.
Fig 1 LC/MS analysis of crude venom from C geographus Example
of experiment approach using LC/ES MS of crude extract of
C.geographus showing complexity of crude venom and showing the
identification of a modified peptide, GID (2184.9 Da) [5] Total ion
chromatogram from positive ion analysis using LC/ES QqTOF mass
spectrometry over a range m/z 500–2000 Chromatography was on a
Zorbax 300SB C3, 2.1 · 150 mm, 5 lm column run at 300 llÆmin)1
with a gradient from 0 to 60% solvent B over 60 min Solvent A was
0.1% formic acid in water and solvent B was 90% acetonitrile, 0.09%
(v/v) formic acid in water Location of the selected component has
been indicated Inset: Electrospray reconstructed mass spectrum of
selected component from main figure revealing the presence of two
components 44 Da apart, indicating the presence of a
c-carboxyglut-amic acid residue Doubly and triply charged ions were observed (data
not shown).
Trang 7linkage determination in peptides and proteins by LC/
electrospray ionization tandem mass spectrometry (LC/ESI
MS/MS) in combination with partial reduction by tris
(2-carboxyethyl)phosphine (TCEP) has been described
recently [46] The general procedure in these studies has been that peptides were treated with TCEP in the presence of an alkylation reagent such as maleimide-biotin (M-maleimide-biotin) or N-ethylmaleimide (NEM), followed
by complete reduction with dithiothreitol and alkylation
by iodoacetamide (IAM) Subsequently, peptides that contained alkylated cysteine were analyzed by capillary LC/ESI MS/MS or other means to determine which cysteine residues were modified with M-biotin/NEM or IAM The presence of the alkylating reagent (M-biotin or NEM) during TCEP reduction was found to minimize the occurrence of the thiol-disulfide exchange reaction [46] In the determination of disulfide connectivity, it is advisable to undertake directed synthesis of mispaired as well as correctly disulfide-bonded conopeptides (as described below), and to compare their elution with the naturally occurring conopep-tide in chromatography studies This is particularly applic-able for a-conotoxins because there are a manageapplic-able number of potential disulfide isomer variants Authenticity
is indicated by chromatography coelution of natural and synthetic peptides after coinjection Ideally, disulfide con-nectivity would be further confirmed by assessing the structure of the naturally occurring peptide However, the limitation of scarcity of most conopeptides usually precludes this The possibility of activity of non-native disulfide-bonded isomers of a-conotoxins [27] precludes reliance on activity for identification of the correctly disulfide-bonded synthetic isomer
Peptide synthesis
Chemical synthesis strategies The small size (10–25 amino acids) of a-conotoxins has made chemical synthesis the preferred route of synthetic access with both Boc and Fmoc chemistry widely employed The incorporation of most post-translationally modified residues such as hydroxyproline, c-carboxyglut-amic acid or pyroglutc-carboxyglut-amic acid into synthetic a-conotox-ins is relatively straightforward through incorporation of the suitably protected amino acid in the chain assembly step By contrast, two areas of a-conopeptide synthesis can pose challenges and deserve particular attention: sulfotyrosine incorporation and disulfide bond formation together with selection of the desired disulfide bond isomer
Disulfide bond formation Synthetic strategies for the preparation of a-conotoxins vary between laboratories and often reflect different scientific
Fig 3 Schemes of directed disulfide bond formation Schemes showing orthogonal strategies for selective disulfide bond formation in synthesis
of a-conotoxins (A) One-step directed disulfide formation [52] (B) Two-step standard directed disulfide formation from Olivera and coworkers [2,8,10,13,20,50,51] (C) Two-step directed disulfide formation, illustrated with discrete mispaired isomer with small disulfide loop closed first [54] (D) Two-step directed disulfide forma-tion, in solution or resin-bound, with small disulfide loop closed first [55] (E) Three possible disulfide isomers of a-conotoxins [4,53].
Trang 8needs or preferences The choice of simultaneous
(non-selective) or selective oxidation (Fig 3) for disulfide bond
formation is a major point of difference Information
described in the following sections concerning detailed
studies of selective disulfide formation with the
muscle-specific a-conotoxin SI can be extended to the neuronal
a-conotoxins
Nonselective disulfide bond formation Several research
groups have taken advantage of the fact that the native form
(I-III, II-IV C-C connectivity) of the a-conotoxins is the
predominant form accessible under simple oxidative
condi-tions from the tetrathiol (i.e fully reduced) peptide
Appropriate solvent conditions may influence the efficiency
of the process and either standard Fmoc or Boc chemistry
may be employed to generate the fully reduced conopeptide
[47,48] Disulfide bond formation is performed via a
one-step procedure usually with 0.02–0.1Maqueous ammonium
bicarbonate, pH 6.7–10, or with minor variations such as
the inclusion of 10–30% (v/v) of isopropanol, acetonitrile or
dimethylsulfoxide where required An exception to the
generally used chain assembly and deprotection procedures
employs a two-step deprotection and cleavage from
methylbenzhydrylamine resin using trifluoroacetic acid
and hydrogen fluoride, prior to nonselective disulfide
formation and a shorter oxidation at higher pH (pH 10)
[49]
Selective disulfide bond formation: off-resin
approa-ches Fmoc methodology for chain assembly together with
trifluoroacetic acid-based deprotection and cleavage of
peptide from resin is the most commonly used Typical
approaches have employed orthogonal cysteine
protec-tion (Trityl, Acetamidomethyl) for a two-step disulfide
bond formation, using reagents such as Tris buffer,
20 mM potassium ferricyanide with 0.1M Tris, pH 7.5,
or 10% dimethylsulfoxide in 0.02M ammonium
bicar-bonate, pH 6.7, for formation of the first disulfide bond,
and iodine for the second disulfide bond (Fig 3B)
[2,8,10,13,20,50,51]
A novel approach by Cuthbertson & Indrevoll employed a
one-pot regioselective formation of the two disulfide bonds
of a-conotoxin SI [52] (Fig 3A) By selecting
temperature-sensitive orthogonal cysteine protecting groups, t-butyl and
4-methylbenzyl, the target molecule was efficiently obtained
Thus the first disulfide bridge was formed directly from the
crude material by simultaneous cleavage and oxidation of
the t-butyl groups in trifluoroacetic acid/dimethylsulfoxide/
anisole (97.9 : 2 : 0.1, v/v/v) at room temperature The
subsequent heating of this solution resulted in the cleavage of
the 4-methylbenzyl groups with simultaneous oxidation
yielding the desired bicyclic product [52]
Selective disulfide bond formation: on-resin
approa-ches Two detailed studies of selective disulfide formation
in the muscle-specific a-conotoxin SI (Fig 3) [49,53],
employed both off and on-resin approaches Syntheses of
all three possible disulfide regioisomers, natural and
disulfide-mispaired, with the sequence of a-conotoxin SI
were described [53] (Disulfide isomers: natural I-III, II-IV;
mispaired nested I-IV, II-III and mispaired discrete I-II,
III-IV, Fig 3E) It was possible to achieve the desired
alignments with either order of loop formation (smaller loop before larger, or vice versa) The highest overall yields were obtained when both disulfides were formed in solution, while experiments where either the first or both bridges were formed while the peptide was on the solid support revealed lower overall yields and poorer selectivities towards the desired isomers This and further studies with a-conotoxin
SI illustrated novel protection schemes and oxidation strategies ([53–55] Fig 3C,D)
Synthesis of sulfated a-conotoxins Sulfopeptides can be prepared by chemical assembly with the incorporation of sulfated residues or less commonly
by global modification of the completed peptide using enzymic or chemical methods of sulfation The chemical synthesis of peptides containing O-sulfated hydroxy amino acids is still considered a difficult, delicate and laborious task for peptide chemists because of the intrinsic acid-lability of the sulfate moiety [50,56,57] An efficient cleavage/deprotection procedure without loss of the sulfate remains to be elucidated for Fmoc-based solid-phase synthesis of sulfopeptides [24,50,57] There have been a few reports of the solid phase synthesis of sulfotyrosine-containing a-conotoxins including EpI, PnIA, PnIB, AnIA, AnIB and AnIC and some ana-logues [21,24,56] Most of the reported syntheses of PnIA, PnIB and the [A10L]PnIA analogue have been of the unsulfated forms [51,58–61] A modified trifluoro-acetic acid-based protocol including low temperature steps and exclusion of thiol-containing scavengers, for example 95% (v/v) trifluoroacetic acid/triisopropylsilane
or 90% (v/v) aqueous trifluoroacetic acid (0C, 8 h), is generally used for cleavage of sulfopeptides assembled with Fmoc chemistry [24,56] Desulfation rate is strongly temperature-dependent whereas sidechain deprotections are less temperature-dependent and effective deprotection protocols can be developed accordingly The use of tetrabutylammonium salts (rather than sodium or barium salts) of O-sulfated hydroxy amino acids minimized desulfation during Fmoc-based assembly, room tempera-ture trifluoroacetic acid cleavage and reversed-phase HPLC purification in applications for synthesis of cholecystokinin-12 and bradykinin containing tyrosine sulfate [57], and could perhaps be applicable to synthesis
of a-conopeptides
Synthesis of variants of a-conotoxins: alanine scans and loop replacements
In addition to replication of natural conotoxins, the standard synthesis procedures described above have been applied for strategies requiring synthesis of specific variants of conotoxins The synthesis of alanine scan peptide variants containing systematic alanine replace-ments of the amino acids [62], and assessment in structure
or function screening procedures, may reveal crucial amino acids or regions that are essential binding and structural determinants A parallel of the alanine scan approach is the synthesis of a-conotoxin variants or
chimeras in which whole loop regions have been swapped with those from other a-conotoxins with different
Trang 9attrib-utes in an attempt to confer a different specificity or
conformation [1,48,63]
Strategies for synthesis of multiple conotoxins in parallel
Synthesis and evaluation of many different peptide
mole-cules may be required for structure-function studies of
a-conotoxins in the search for optimum synthetic variants
The rate-limiting step in these studies is often the time and
effort required for peptide synthesis Options for synthesis
of multiple peptides in parallel have included synthesis on
polyethylene pins, in tea bags and on chip or memb rane
supports such as in the spot synthesis technique [64–66] A
study of [A10L]PnIA variants synthesized in parallel in a 96
well plate has been described [67] However there have been
few other references to the application of these techniques to
the synthesis of a-conotoxins and their analogues, and
quantities may be insufficient for conventional
structure-function studies There may be increasing scope for
applications of parallel synthesis, particularly
membrane-anchored synthetic peptide libraries, for structure-function
analysis by in situ screening using binding assays or
antibody recognition
Concluding remarks
Comprehensive characterization of novel natural
a-cono-toxin peptides has relied on a combination of analyses
These have included Edman N-terminal sequencing, MS
and tandem MS for determination of the primary
sequence, determination of disulfide linkages after
differ-ential reduction and alkylation, and confirmation of the
primary structure Amino acid analysis in conjunction
with Edman sequence analysis and diagnostic ion MS has
been used to assess the possibility of other PTMs
Characterization would ideally include NMR analysis
for proof of structure although this has often been
precluded by scarcity of material It may be the only
means to assure certainty in determination of novel
disulfide linkages The primary structure of the peptide
and the identified PTMs have usually been verified by
subsequent synthetic chemistry and comparison with the
natural peptide The synthesized disulfide-bonded peptide,
with minimum purity of 95–98% determined by
reversed-phase HPLC, has been compared with the natural
peptide, if available, by coinjection of the synthesized
peptide and the naturally occurring peptide, and
authen-ticity has been indicated by chromatography coelution
The correctness of the disulfide linkage has usually been
confirmed by NMR spectroscopy Newly identified
pep-tides that clearly fit previously defined categories are often
not subjected to such rigorous analysis For variants such
as residue replacement peptides and chimeras, where there is
no corresponding natural peptide available for comparison,
verification of authenticity, particularly correct disulfide
linkage, has been reliant on NMR structure analysis Many
nonselective syntheses of a-conotoxins yield a single
pre-dominant disulfide-bonded isomer, although in cases where
multiple isomers are generated in substantial amounts,
further characterization may be warranted Thorough
assessment of novel a-conotoxin folds may generate
import-ant structure-function information
In conclusion, the physico-chemical characterization of the native peptides and chemical synthesis of neuronal a-conotoxins provide an important basis for the pharma-cology, structure and modelling studies that are the subject
of further minireviews in this series
After this manuscript had been submitted for publication
a newly published paper reported 16 conotoxin precursors
of the A superfamily, from six Conus species, defining the A conotoxin gene superfamily [68]
Acknowledgements
We thank Annette Nicke, David Craik, Gene Hopping, Alun Jones and Richard Lewis for their input The LC/MS analysis shown in Fig 1 was run b y Alun Jones.
References
1 McIntosh, J.M., Santos, A.D & Olivera, B.M (1999) Conus peptides targeted to specific nicotinic acetylcholine receptor sub-types Annu.Rev.Biochem.68, 59–88.
2 Luo, S.Q., Kulak, J.M., Cartier, G.E., Jacobsen, R.B., Yoshikami, D., Olivera, B.M & McIntosh, J.M (1998) a-Conotoxin AuIB selectively blocks a3b4 nicotinic acetylcholine receptors and nico-tine-evoked norepinephrine release J.Neurosci.18, 8571–8579.
3 Arias, H.R & Blanton, M.P (2000) a-Conotoxins Int.J.Bio-chem.Cell Biol.32, 1017–1028.
4 Dutton, J.L & Craik, D.J (2001) a-Conotoxins: Nicotinic acetyl-choline receptor antagonists as pharmacological tools and potential drug leads Curr.Med.Chem.8, 327–344.
5 Nicke, A., Loughnan, M., Millard, E., Alewood, P., Adams, D., Daly, N., Craik, D & Lewis, R (2003a) Isolation, Structure and Activity of GID, a novel a-4/7-conotoxin with an extended N-terminal sequence J.Biol.Chem.278, 3137–3144.
6 Martinez, J.S., Olivera, B.M., Gray, W.R., Craig, A.G., Groebe, D.R., Ab ramson, S.N & McIntosh, J.M (1995) a-Conotoxin EI,
a new nicotinic acetylcholine-receptor antagonist with novel selectivity Biochemistry 34, 14519–14526.
7 Zhao, D & Huang, P (1999) Conus imperialis conotoxin ImIIA precursor mRNA EMBL/Genbank/DDBJ Databases accession number Q9U619
8 Ellison, M., McIntosh, J.M & Olivera, B.M (2003) a-Conotoxins ImI and ImII Similar a7 nicotinic receptor antagonists act at different sites J.Biol.Chem.278, 757–764.
9 Dowell, C., Olivera, B.M., Garrett, J.E., Staheli, S.T., Watkins, M., Kuryatov, A., Yoshikami, D., Lindstrom, J.M & McIntosh, J.M (2003) a-Conotoxin PIA is selective for a6 subunit-containing nicotinic acetylcholine receptors J.Neurosci.
23, 8445–8452.
9a Nicke, A., Wonnacott, S & Lewis, R.J (2004) a-Conotoxins as tools for the elucidation of structure and function of neuronal nicotinic acetylcholine receptor subtypes Eur.J.Biochem.271, 2305–2319.
10 McIntosh, J.M., Dowell, C., Watkins, M., Garrett, J.E., Yoshi-kami, D & Olivera, B.M (2002) a-Conotoxin GIC from Conus geographus, a novel peptide antagonist of nicotinic acetylcholine receptors J.Biol.Chem.277, 33610–33615.
11 Cortez, L.M., del Canto, S.G., Testai, F.D & Bonino, M.J.B.D (2002) Conotoxin MI inhibits the a/d acetylcholine binding site of the Torpedo marmorata receptor Biochem.Biophys.Res.Com-mun 295, 791–795.
12 McIntosh, J.M., Cruz, L.J., Hunkapiller, M.W., Gray, W.R & Olivera, B.M (1982) Isolation and structure of a peptide toxin from the marine snail Conus magus Arch.Biochem.Biophys.218, 329–334.
Trang 1013 Cartier, G.E., Yoshikami, D.J., Gray, W.R., Luo, S.Q., Olivera,
B.M & McIntosh, J.M (1996) A new a-conotoxin which targets
a3b2 nicotinic acetylcholine receptors J.Biol.Chem.271, 7522–
7528.
14 Craig, A.G (2000) The characterization of conotoxins J.Toxicol.
Toxin Rev 19, 53–93.
15 Craig, A.G., Bandyopadhyay, P & Olivera, B.M (1999)
Post-translationally modified neuropeptides from Conus venoms Eur.
J.Biochem.264, 271–275.
16 Olivera, B.O., Walker, C., Cartier, G.E., Hooper, D., Santos, A.D.,
Schoenfeld, R., Shetty, R., Watkins, M., Bandyopadhyay, P &
Hillyard, D.R (1999) Speciation of cone snails and interspecific
hyperdivergence of their venom peptides: Potential evolutionary
significance of introns Ann.NY Acad.Sci.870, 223–237.
17 Conticello, S.G., Gilad, Y., Avidan, N., Ben-Asher, E., Levy, Z &
Fainzilber, M (2001) Mechanisms for evolving hypervariability:
The case of conopeptides Mol.Biol.Evol.18, 120–131.
18 Duda, T.F Jr & Palumbi, S.R (1999) Molecular genetics of
ecological diversification: Duplication and rapid evolution of toxin
genes of the venomous gastropod Conus Proc.Natl Acad.Sci.
USA 96, 6820–6823.
19 Duda, T.F Jr & Palumbi, S.R (2000) Evolutionary diversification
of multigene families: Allelic selection of toxins in predatory cone
snails Mol.Biol.Evol.17, 1286–1293.
20 McIntosh, J.M., Yoshikami, D., Mahe, E., Nielsen, D.B., Rivier,
J.E., Gray, W.R & Olivera, B.M (1994) A nicotinic
acetylcholine-receptor ligand of unique specificity, a-conotoxin ImI J.Biol.
Chem 269, 16733–16739.
21 Loughnan, M.L., Nicke, A., Jones, A., Adams, D.J., Alewood,
P.F & Lewis, R.J (2004) Chemical and functional identification
and characterization of novel sulfated a-conotoxins from the cone
snail Conus anemone J.Med.Chem.47, 1234–1241.
22 Sandall, D.W., Satkunanathan, N., Keays, D.A., Polidano, M.A.,
Liping, X., Pham, V., Down, J.G., Khalil, Z., Livett, B.G &
Gayler, K.R (2003) A novel a-conotoxin identified by gene
sequencing is active in suppressing the vascular response to
selective stimulation of sensory nerves in vivo Biochemistry 42,
6904–6911.
23 Fainzilber, M., Hasson, A., Oren, R., Burlingame, A.L., Gordon,
D., Spira, M.E & Zlotkin, E (1994) New mollusc-specific
a-conotoxins block aplysia neuronal acetylcholine receptors.
Biochemistry 33, 9523–9529.
24 Loughnan, M.L., Bond, T., Atkins, A., Cuevas, J., Adams, D.J.,
Broxton, N.M., Livett, B.G., Down, J.G., Jones, A., Alewood,
P.F & Lewis, R.J (1998) a-Conotoxin EpI, a novel sulfated
peptide from Conus episcopatus that selectively targets neuronal
nicotinic acetylcholine receptors J.Biol.Chem.273, 15667–15674.
25 Watkins, M., Olivera, B.M., Hillyard, D.R., McIntosh, J.M &
Jones, R.M (2000) a-Conotoxin Peptides International Patent
Application WO 00/44776.
25a Millard, E.L., Daly, N.L & Craik, D.J (2004) Structure activity
relationships of a-conotoxins targeting neuronal nicotinic
acetyl-choline receptors Eur.J.Biochem.271, 2320–2326.
26 Olivera, B.M & Cruz, L.J (2001) Conotoxins, in retrospect.
Toxicon 39, 7–14.
27 Dutton, J.L., Bansal, P.S., Hogg, R.C., Adams, D.J., Alewood,
P.F & Craik, D.J (2002) A new level of conotoxin diversity: a
non-native disulfide bond connectivity in a-conotoxin AuIB
reduces structural definition but increases biological activity.
J.Biol.Chem.277, 48849–48857.
28 Hill, J.M., Atkins, A.R., Loughnan, M.L., Jones, A., Adams,
D.A., Martin, R.C., Lewis, R.J., Craik, D.J & Alewood, P.F.
(2000) Conotoxin TVIIA, a novel peptide from the venom of
Conus tulipa 1 Isolation, characterization and chemical synthesis.
Eur.J.Biochem.267, 4642–4648.
29 Hauschka, P.V., Mullen, E.A., Hintsch, G & Jazwinski, S (1988) Abundant occurrence of gamma-carboxyglutamic acid-containing peptides in the marine gastropod family Conidae In Current Advances in Vitamin K Research (Suttie, J.W., ed.), pp 237–243 Science Publishers, New York.
30 Bandyopadhyay, P.K., Colledge, C.J., Walker, C.S., Zhou, L.M., Hillyard, D.R & Olivera, B.M (1998) Conantokin-G precursor and its role in gamma-carboxylation by a vitamin K-dependent carboxylase from a Conus snail J.Biol.Chem.273, 5447–5450.
31 Wolfender, J.L., Chu, F.X., Ball, H., Wolfender, F., Fainzilber, M., Baldwin, M.A & Burlingame, A.L (1999) Identification of tyrosine sulfation in Conus pennaceus conotoxins a-PnIA and a-PnIB: Further investigation of labile sulfo- and phosphopeptides
by electrospray, matrix-assisted laser desorption/ionization (MALDI) and atmospheric pressure MALDI mass spectrometry J.Mass Spectrom.34, 447–454.
32 Huttner, W.B (1987) Protein tyrosine sulfation Trends Biochem Sci 12, 361–363.
33 Nicholas, H.B., Chan, S.S & Rosenquist, G.L (1999) Reevalua-tion of the determinants of tyrosine sulfaReevalua-tion Endocrine 11, 285– 292.
34 Moore, K.L (2003) The biology and enzymology of protein tyrosine O-sulfation J.Biol.Chem.278, 24243–24246.
35 Nicke, A., Samochocki, M., Loughnan, M.L., Bansal, P.S., Maelicke, A & Lewis, R.J (2003) a-Conotoxins EpI and AuIB switch subtype selectivity and activity in native versus recombinant nicotinic acetylcholine receptors FEBS Lett 554, 219–223.
36 Favreau, P., Krimm, I., Le Gall, F., Bobenrieth, M.J., Lamthanh, H., Bouet, F., Servent, D., Molgo, J., Menez, A., Letourneux, Y.
& Lancelin, J.M (1999) Biochemical characterization and nuclear magnetic resonance structure of novel a-conotoxins isolated from the venom of Conus consors Biochemistry 38, 6317–6326.
37 Maslennikov, I.V., Sobol, A.G., Gladky, K.V., Lugovskoy, A.A., Ostrovsky, A.G., Tsetlin, V.I., Ivanov, V.T & Arseniev, A.S (1998) Two distinct structures of a-conotoxin GI in aqueous solution Eur.J.Biochem.254, 238–247.
38 Rappsilber, J., Steen, H & Mann, M (2001) Labile sulfogroup allows differentiation of sulfotyrosine and phosphotyrosine in peptides J.Mass Spectrom.6, 832–833.
39 Nemeth-Cawley, J.F., Karnik, S & Rouse, J.C (2001) Analysis of sulfated peptides using positive electrospray ionization tandem mass spectrometry J.Mass Spectrom.36, 1301–1311.
40 Steen, H & Mann, M (2002) Analysis of bromotryptophan and hydroxyproline modifications by high-resolution, high-accuracy precursor ion scanning utilizing fragment ions with mass-deficient mass tags Anal.Chem.74, 6230–6236.
41 Nakamura, T., Yu, Z.G., Fainzilber, M & Burlingame, A.L (1996) Mass spectrometric-based revision of the structure of a cysteine-rich peptide toxin with gamma-carboxyglutamic acid, TxVIIA, from the sea snail, Conus textile Protein Sci 5, 524–530.
42 Cairns, J.R., Williamson, M.K & Price, P.A (1991) Direct Identification of gamma-carboxyglutamic acid in the sequencing
of Vitamin-K dependent proteins Anal.Biochem.199, 93–97.
43 Gray, W.R (1993) Disulfide structures of highly bridged peptides – a new strategy for analysis Protein Sci 2, 1732–1748.
44 Jones, A., Bingham, J.-P., Gehrmann, J., Bond, T., Loughnan, M., Atkins, A., Lewis, R.J & Alewood, P.F (1996) Isolation and characterization of conopeptides by HPLC combined with mass Rapid Commun.Mass Spectrom.10, 138–143.
45 Shon, K.-J., Olivera, B.M., Watkins, M., Jacobsen, R.B., Gray, W.R., Floresca, C.Z., Cruz, L.J., Hillyard, D.R., Brink, A., Ter-lau, H & Yoshikami, D (1998) l-Conotoxin PIIIA, a new peptide for discriminating among tetrodotoxin-sensitive Na channel sub-types J.Neurosci.18, 4473–4481.