Introduction

Marine organisms belonging to the genus Conus are a rich source of pharmacological agents that act on diverse ion channels1, 2, 3, 4, 5. These agents, known as conopeptides, have evolved as selective tools for capture of prey or deterrence of predators. The 500 known different Conus species each contain greater than 100 different components in their venom, leading to an estimated >50 000 different pharmacologically active conopeptides.

The best characterized biologically-active venom components are conotoxins, small, disulfide-rich peptides, most of which are specifically targeted to different voltage- or ligand-gated ion channels3, 6. One rationale for the vast diversity of different conotoxins found in Conus venom is that this is an evolutionary consequence of the remarkable diversity of molecular isoforms of ion channels. Each Conus species uses the equivalent of several drug combinations to rapidly capture its prey. This review focuses on one family of compounds in Conus venom referred to as α-conotoxins, which are nicotinic acetylcholine receptor (nAChR) antagonists.

All Conus species characterized thus far have at least one nAChR antagonist in their venom7. The largest family of such antagonists are the α-conotoxins, small, disulfide rich peptides generally 12–19 amino acids in length8, 9. The α-conotoxins are classified based on their cysteine pattern CC-C-C, with a disulfide connectivity of Cys1-Cys3 and Cys2-Cys4. There is a further sub-classification of α-conotoxins based on the number of residues in their inter-cysteine loops (see Figure 1). The α3/5-conotoxins (3 residues in the first and 5 in the second loop) are selective blockers of the muscle nAChR, whereas the α4/7, α4/4, and α4/3 subfamilies are generally blockers of neuronal nAChRs3, 7, 8. Although this review focuses primarily on neuronally targeted α-conotoxins, the currently known muscle nAChR targeted α-conotoxins are briefly discussed below (also see Table 1).

Figure 1
figure 1

Backbone structure of subfamilies of α-conotoxins. (A) Backbone structure of an α3/5 conotoxin (α-CTx GI), an α4/7 conotoxin (α-CTx MII) and an α4/3 conotoxin (α-CTx ImI). For α-CTx GI and α-CTx MII: green represents the α-helix; yellow, disulfide bonds; purple, side chain of toxin residues. For α-CTx ImI: light blue represents backbone structure; yellow, disulfide bonds; dark blue, side chain of Arg7; green, side chain of Trp10; gray, other “non-essential” toxin residues. (B) Superimposed ribbon diagrams of six α4/7-conotoxins: α-conotoxins PIA (violet), MII (blue), GIC (green), PnIA (yellow), GID (red), and EI (cyan). Although members of each subfamily have similar backbone structures (as shown in B), members across different subfamilies have different structures (shown in A). In A, α-CTx GI and α-CTx MII are from [5], α-CTx ImI is from [136]. Superimposed structures in B is from [137]. All images reproduced by permission from Elsevier, Inc through Rights Link Copyright Clearance Center.

Table Sequence and receptor specificity of α3/5 and other more unusual muscle-specific α-conotoxins. #, amidated C–terminus; ^, free carboxyl C–terminus; γ, γ–carboxyglutamate; O, 4–trans–hydroxyproline; Z, pyroglutamate. The cysteine residues that form the disulfide bridges are in bold.

Muscle nAChR-targeted α-conotoxins

The first α-conotoxins to be purified from Conus venom were α-conotoxins GI, GIA, and GII from the fish-hunting cone snail Conus geographus10. The best characterized is α-CTx GI, which blocks neuromuscular transmission both in- vitro11, 12, 13 and in vivo14, but does not block any of the neuronal nAChR subtypes15, 16. α-CTx MI, from Conus magus, was subsequently isolated and characterized as another blocker of the ACh responses at the neuromuscular junction17. Subsequently, a series of muscle nAChR blocking α-conotoxins were isolated from Conus striatus, α-CTx SI, α-CTx SIA, and α-CTx SII18, 19, 20, Conus consors, α-CTx CnIA, and α-CTx CnIB21 and Conus achatinus, α-CTx Ac1.1a and α-CTx Ac1.1b22.

Binding data on mouse muscle receptors demonstrate that α-CTx GI, α-CTx MI, and α-CTx SIA are preferentially targeted to the α/δ interface of the muscle nAChR (with an affinity >10 000 fold higher than for the α/γ interface23, 24, 25. In contrast, for the Torpedo nAChR, all three peptides display much higher affinity for the α/γ vs the α/δ interface24. Two newly discovered toxins from Conus achatinus, α-CTx Ac1.1a, and α-CTx Ac1.1b, block the mouse α/δ interface with a potency that is >50 000 fold higher than α/γ22. In contrast, α-CTx SI has low affinity for both the α/γ and α/δ interfaces of mouse muscle nAChR, and does not discriminate between the two sites as well as either α-CTx GI or α-CTx MI24. α-CTx SI has less toxicity in vivo than either α-CTx GI or α-CTx MI18. In addition, α-CTx SI does not discriminate between the two binding sites on the Torpedo nAChR26, 27. Structure-activity studies have indicated that Arg at position 9 of α-CTx GI is responsible for the differential affinity of this toxin for the two binding sites of the Torpedo nAChR28; this residue also confers high affinity for the α/γ interface of mouse muscle nAChR27. Instead of Arg, α-CTx SI has a Pro in the homologous position, which may account for its differential pharmacology27, 28. Comparison of solution structures of α-CTx SI with α-CTx GI suggests that it is the loss of the basic charge, rather than differences in backbone structure, that underlies the pharmacological differences between the two toxins29. The residues on the mouse δ subunit that confer high binding affinity to α-CTx MI have been determined and include Ser36, Tyr113, and Ile17830. The α-CTx MI residues that interact with the receptor binding pocket have also been determined30, 31, 32. In addition, a recent study has indicated the importance of a positively charged residue (either an Arg or Lys) at the C-terminus of α-CTx GI, α-CTX SI, and α-CTX SIA in enhancing affinity for both binding sites on the Torpedo nAChR33.

Although all of the “classical” muscle nAChR blocking α-conotoxins belong to the α3/5-conotoxin subfamily, several muscle nAChR blocking α-conotoxins belong to other branches of the α-conotoxin family (Figure 1). α-CTx EI, isolated from the fish-hunting Atlantic species Conus ermineus34, is an α4/7-conotoxin, a subfamily that is usually associated with blocking neuronal nAChRs. The solution structure of α-CTx EI is similar to other α4/7 conotoxins (Figure 1)35. The backbone structure of α4/7 conopeptides is illustrated in Figure 1. Although α-CTx EI does block some neuronal nAChRs36, unlike other α4/7 conotoxins it's also a potent blocker of the muscle nAChR34, 36. Unlike the α3/5 conotoxins that preferentially block the mammalian α/δ interface (see above), α-CTx EI blocks both the α/δ and α/γ interface with similiar affinity34. Moreover, α-CTx EI has higher affinity for the α/δ interface of the Torpedo nAChR, in contrast to the α/γ-interface preferring α3/5-conotoxins34. Recently, two other α4/7-conotoxins that block the muscle nAChR, and to a lesser extent neuronal nAChRs, have been discovered, α-conotoxins SrIA and SrIB from Conus spurius36. Similar to α-CTx EI, α-CTx SrIA and α-CTx SrIB contain post-translational modifications in their native sequences. Unlike other α4/7 conotoxins, however, α-CTx SrIA, and SrIB are reported to have nAChR potentiating activity on the α4β2 subtype36.

Another unusual muscle-specific α-conotoxin is α-CTx PIB, which has an uncommon 4/4 intercysteine loop spacing37. It was purified from the venom of Conus purpurascens and found to block both adult and fetal mouse muscle nAChRs expressed in oocytes with nanomolar potency, with little effect on any neuronal subtypes37.

Neuronal nAChR-targeted α-conotoxins

The α-conotoxins targeting neuronal nAChRs are numerous and have even more exquisite subtype selectivity, probably due to the large diversity of isoforms in this subfamily of receptors. The members of this family of toxins that have been discovered thus far are listed below, in no particular order (also see Tables 2 and 3).

α-Conotoxins ImI and ImII

The first α-conotoxin targeting neuronal nAChRs was α-CTx ImI. It was also the first α-conotoxin isolated from the venom of a worm hunting species, Conus imperialis38. This α-conotoxin has the unusual inter-cysteine loop spacing 4/3 (Figure 1A). This peptide was found to be inactive when injected interaperitoneally, whereas intracereberal injection caused seizures and death38. In addition, this toxin blocks nicotinic responses of B cells on frog sympathetic ganglia but not mammalian neuromuscular nAChRs38, 39. Subsequent pharmacological characterization on heterologously expressed nAChRs showed an IC50 of 220 nmol/L on homomeric α7 nAChRs and 1.8 μmol/L on homomeric α9 nAChRs16. Ellison et al, (2004) reported an IC50 of 595 nmol/L on homomeric α7 nAChR, but they also showed potent inhibition of heteromeric α3β2 nAChR (IC50 41 nmol/L). The seizure inducing effects of this toxin are presumably due to its block of α-bungarotoxin-sensitive α7 nAChRs on hippocampal neurons40. Structure-activity studies have suggested a role for residues in the first loop of the toxin, Asp5, Pro6, and Arg7, as well as Trp10 in the second loop, in interacting with the α7 subunit41. Homology modeling using Aplysia AChBP, a structural homolog of the N-terminal binding region of nAChRs, confirms a key role for Arg7 and Trp10 in interaction with residues at the ligand binding site42, 43. Co-crystallization of α-CTx ImI with Aplysia AChBP reveals a more open C-loop to accommodate the α-conotoxin44. The α7 subunit residues that interact with the toxin have also been determined45, 46.

In recent years, attempts have been made to improve the biological stability of α-conotoxins by protecting the disulfide bonds against reduction or scrambling due to exposure to intra- or extracellular environments, such as blood. One of these studies utilized selenocysteines in place of cysteines to yield several selenoconotoxin analogs of α-CTx ImI47. Although similar to wildtype α-CTx ImI in both structure and activity against α7 nAChRs, the selenoconotoxin analogs were more stable than wildtype α-CTx ImI under a variety of chemical and biological reducing conditions47. A second method of improving conformational stability is to replace the cystine bridges with non-reducible dicarba linkages. The dicarba analog of α-CTx ImI had a similar structure to wildtype α-CTx ImI, with a slight difference in the geometry of disulfide vs dicarba bridges. The activity of the analog against α7 nAChR was similar to wildtype α-CTx ImI48.

Another α4/3 conotoxin from the venom of Conus imperialis, α-CTx ImII, was identified using a PCR-based discovery strategy49. Although highly homologous to α-CTx ImI in sequence (9 out of 12 residues are shared), α-CTx ImII, unlike α-CTx ImI, does not compete with α-bungarotoxin for binding to heterologously expressed α7 nAChRs, suggesting a distinct and perhaps novel binding site on the α7 nAChR for α-CTx ImII50. The difference in binding of the two toxins is due to the presence of a Pro at position 6 of α-CTx ImI, which has been shown to be important for interaction of this toxin with α7 nAChRs41. α-CTx ImII has an Arg at this position; mutation of this Arg to a Pro creates an analog that competes with α-bungarotoxin binding49.

α-Conotoxins MII, PIA, OmIA

A second α-conotoxin isolated from Conus magus was α-CTx MII. It belongs to the α4/7 subfamily (Figure 1). α-CTx MII blocks the α3β2 nAChR subtype with a potency 2–4 orders of magnitude higher than most other nicotinic receptors, including the muscle subtype51. However, binding studies in knock-out animals52 as well as functional studies53, have shown that α-CTx MII also potently acts at α6-containing nAChRs. Subsequently, a series of α-CTx MII analogs were made that selectively target the chimeric α6/α3β2β354 vs α3β2 nAChR53. These toxin analogs are the most selective α6* nAChR antagonists reported to date. The wildtype α-CTx MII, as well as a number of α6* (the asterisk indicates the presence of additional subunits) selective α-CTx MII analogs, have been used to characterize nAChR subtypes that modulate dopamine release in rat55, 56, 57, 58, 59, 60, 61, mice62, 63, 64, 65, and monkey striatum66, 67. These studies indicate a role for α6β2* and α6α4β2* nAChRs in the modulation of dopamine release in the striatum and the nucleus accumbens. The contribution of both the α4 and the β3 subunits to α-CTx MII binding sites in dopaminergic neurons has been confirmed in studies using knock-out mice64, 68, 69. Some studies also suggest a selective downregulation of α6* nAChRs upon chronic nicotine exposure70, 71. α-CTx MII has also been used to characterize distinct nAChRs that modulate [3H]NE release in rat72 vs mouse hippocampus73. Table 4 summarizes some of the nAChR-mediated physiological functions characterized using the α-conotoxins.

Structure-activity studies have identified the amino acid residues on the α3 and the β2 nAChR subunits that interact with α-CTx MII. These include Lys185 and Ile188 on the α3 subunit and Thr59, Val109, Phe117 and Leu119 on the β2 subunit74, 75. In addition, α-CTx MII[S4A; E11A; L15A] has been used to identify the nAChR subunit amino acid residues that interact with, and confer selectivity, for the α6 vs the α3 subunit76. Notably, these nAChR subunit residues are distinct from those that interact with wildtype α-CTx MII74.

Radiolabeling of α-CTx MII has allowed direct measurement of the binding sites for this α-conotoxin within the brain77. Fluorescently-labeled analogs of α-CTx MII have also recently been synthesized78. Using the radiolabeled α-CTx MII, Quik and co-workers have found a selective downregulation of α-CTx MII binding sites within rodent and monkey striatum after nigorstriatal damage79, 80, 81, 82 as well as in humans with Parkinson's82. Direct measurement of α-CTx MII sites also shows preferential recovery of these sites in monkeys allowed to recover from nigrostriatal damage66. Binding studies using an analog of α-CTx MII, α-CTx MII[E11A], have shown a selective loss of a specific subtype of nAChR (α6α4β2*) in rodent and monkey models of Parkinson's disease. These studies were replicated in post-mortem tissue from humans with Parkinson's83. Thus, this α-CTx MII analog has a nAChR subtype selectivity that has potential in the early diagnosis of Parkinson's disease.

A number of attempts have been made to enhance the stability of α-CTx MII against proteolysis as well to improve its lipophilicity, with the aim of increasing the oral bioavailability of this toxin. Backbone cyclization by use of linker placement and joining of the N- and C-termini resulted in several cyclic α-CTx MII analogs. Two of these analogs had structures similar to native α-CTx MII, although only one of these analogs-the one with the longest linker-retained activity similar to native α-CTx MII. The cyclic analog was much more stable than the native toxin in human plasma84. To enhance the lipophilicity of α-CTx MII, the toxin was conjugated with 2-amino-D,L-dodecanoic acid (Laa) at the N-terminus. This terminally-conjugated α-CTx MII analog had similar structure and activity compared to the parent peptide85, but displayed significantly improved permeability across Caco-2 cell monolayers85. Although the Laa-conjugated analog did not cross the blood brain barrier to any great extent, its absorption through the GI tract after oral administration was greater than the parent α-CTx MII86.

A number of other α-conotoxins have selectivity either towards α3β2 or α6β2β3 nAChRs. α-CTx PIA, discovered through PCR-based cloning of cDNA from venom duct of Conus purpurascens, has about 70-fold lower IC50 for chimeric α6/α3β2β3 vs α3β2 nAChRs87. In contrast, α-CTx OmIA, purified from venom of Conus omaria, has a preferential (20-fold) selectivity for α3β2 vs α6/α3b2b3 nAChRs88, 89. It is also a potent inhibitor of α7 nAChR89.

α-Conotoxins GIC and GID

Two α-conotoxins specific for neuronal nAChRs have been purified from the venom of Conus geographus. α-CTx GIC, an α4/7 conotoxin identified from the genomic DNA of Conus geographus, has low nanomolar potency for α3β2 and α6β2 nAChRs vs muscle, α3β4 and α4β2 nAChRs90, 91. α-CTx GID, although similar to α-CTx GIC in its inter-cysteine loop spacing (α4/7), is structurally different from α-CTx GIC and other neuronal nAChR targeted α-conotoxins in several ways. First, it has additional N-terminal residues compared to α-CTx GIC and most other previously identified α-conotoxins. Second, it has two posttranslational modifications: a γ-carboxyglutamic acid residue just before the first Cys residue and a hydroxyproline at position 16. Third, there is a positively charged residue, Arg, at position 12, whereas α-CTx GIC and most other α-conotoxins have either a hydrophobic (Ala or Phe) or uncharged (Asn) residue. Interestingly, a positive charge at this position seems to be responsible for α-CTx GID's relatively high affinity for α4β2 nAChRs compared to other α-conotoxins lacking this positive residue92, 93. It is worth noting here that the two muscle-specific α-conotoxins-SrIA and SrIB-that have potentiating effects on the α4β2 nAChR also have an Arg at the homologous position (Table 1). An 'Ala walk' of α-CTx GID also indicates important roles for most residues of the toxin, with the exception of Val13, in interaction with α4β2 nAChR93. Similar to α-CTx GIC, α-CTx GID also potently blocks α3β2 nAChR92, 93. In contrast to α-CTx GIC, however, α-CTX GID also blocks α7 nAChR92, 93. Structure-activity studies indicate the most important residue for interaction with both the α7 and the α3β2 nAChRs is Pro9, with smaller contribution by Asp3 and Arg12. Asn14 seems to be important for interaction with α7 nAChR only93. The β2 subunit residues conferring the high potency of α-CTx GID on α3β2 nAChR are the same residues that also confer high potency of α-CTx MII and α-CTx PnIA for this receptor subtype75.

α-Conotoxin EpI

α-CTx EpI, an α4/7-conotoxin purified from the venom of Conus episcopatus, is another example of a post-translationally modified conotoxin. This conotoxin is different from previously characterized α-conotoxins in that it has a sulfated Tyr at position 1594. However, both the native sulfated and synthetic non-sulfated peptide are similar in activity and inhibit α-bungarotoxin-resistant nAChR responses in adrenal chromaffin cells and rat intracardiac ganglia parasympathetic neurons, but do not inhibit muscle nAChR responses94. Therefore, this toxin was designated as a specific blocker of α3β2 and α3β4, but not α7, nAChRs. However, studies with heterologously expressed nAChRs in oocytes showed the inverse selectivity, potent inhibition of α7 nAChRs with little effect on α3β4 or α3β2 nAChRs95. The inhibition of α7 nAChRs by α-CTx EpI is not surprising considering that α-CTx EpI is identical to α-CTx ImI in its first loop, the region containing residues shown to be important for interaction of α-CTx ImI with α7 nAChRs41.

α-Conotoxins AnIA, AnIB, and AnIC

Three peptides isolated from the venom of Conus anemone add to the list of sulfated α-conotoxins96. α-CTx AnIA, α-CTx AnIB, and α-CTx AnIC all have a sulfotyrosine at position 16. The synthetic α-CTx AnIB had subnanomolar potency at α3β2 nAChRs, with about 200-fold lower potency at α7 nAChR, and little or no activity on muscle and other tested heteromeric nAChRs. The non-sulfated peptide retained activity at α3β2 nAChRs, whereas it was 10-fold less potent on α7 nAChR. Removal of the two N-terminal glycines (to yield the same toxin as α-CTx AnIA) negatively affected binding kinetics, and as a result potency, of the toxin on α3β2 nAChRs96.

α-Conotoxins AuIA, AuIB, and AuIC

Three α-conotoxins isolated from the venom of Conus aulicus are α-CTx AuIA, AuIB, and AuIC. Similar to most other neuronal nAChR blocking α-conotoxins, α-CTx AuIA and AuIC belong to the α4/7 family, whereas α-CTx AuIB has the unusual 4/6 inter-cysteine loop spacing72. All three are selective blockers of the neuronal subtype, α3β4, with α-CTx AuIB being the most potent72. This α-conotoxin has been used to show involvement of distinct nAChR subtypes in modulation of [3H]NE, [3H]ACh, and [3H]DA release from hippocampal, interpeduncular nucleus and striatal synaptosomes, respectively72, 97.

α-Conotoxins PnIA and PnIB

α-CTx PnIA and PnIB were purified from the venom of Conus pennaceous98. Both toxins have a post-translational modification, sulfotyrosine at position 1599; however, the functional characterizations were all performed with non-sulfated synthetic toxins. Although they only differ in two amino acids, α-CTx PnIA is a potent blocker of α3β2 nAChR, whereas α-CTx PnIB blocks α7 nAChR more potently100.

Substitution of Ala10 in α-CTx PnIA with Leu, found in the homologous position in α-CTx PnIB, shifts selectivity of α-CTx PnIA[A10L] towards α7, with a potency that is even greater than α-CTx PnIB100, 101. The systematic truncation of the second loop affects the potency of α-CTx PnIA[A10L] for both the α7 nAChR and AChBP due to the loss of several hydrogen bonds between toxin and receptor upon toxin truncation102. Replacing Asn11 with Ser in α-CTx PnIA, the other residue different between α-CTx PnIA and α-CTx PnIB, caused loss of potency for both α3β2 and α7 nAChRs100.

Structure-activity studies have located residues in the α3 subunit that confer affinity for α-CTx PnIA. These include Pro182, Ile188 (also found to interact with α-CTx MII) and Gln198103. All three residues were found to be located within the C-loop of the subunit. The β2 subunit residues that interact with α-CTx PnIA are the same residues that also interact with α-CTx MII and α-CTx GID75. Adding an additional positive charge to the C-terminus of α-CTx PnIA[A10L] to yield the analog α-CTX PnIA[A10L, D14K] enhanced the affinity of the toxin for Lymnaea and Aplysia AChBPs33, 104. A recent study presented the crystal structure of α-CTx PnIA[A10L;D14K] bound to Aplysia AChBP and indicated predominantly hydrophobic and hydrophobic/ aromatic interactions between the analog and the AChBP binding pocket104. Double cycle mutant analysis has also indicated pairwise hydrophobic and aromatic interactions between α-CTx PnIB and the α7 nAChR105.

α-Conotoxin BuIA

α-CTx BuIA was cloned from RNA extracted from the venom of Conus bullatus106. Similar to the muscle nAChR blocking α-CTx PIB, it possesses a 4/4 cysteine loop spacing. Unlike other α-conotoxins, α-CTx BuIA is not specific for a particular subtype of nAChR, blocking almost all neuronal subtypes with nanomolar potency, with exception of α4β2106. However, its differential kinetics can distinguish between nAChRs that contain either a β2 or a β4 subunit: block of β2* nAChRs is rapidly reversed whereas block of β4* nAChRs is only slowly reversed upon toxin washout106. The subunit residues that are critical to these off-rate differences have been determined and, interestingly, are the same residues that interact with a number of other α-conotoxins75, 107. Two studies have solved the structure of α-CTx BuIA108, 109. The latter study has shown that the native globular structure of α-CTx BuIA is highly flexible and maintains multiple conformations in solution, as opposed to some other α-conotoxins that have a rigid globular structure109. This multiple-conformation structural feature may underlie the toxin's promiscuous selectivity profile.

The differential kinetics of α-CTx BuIA were utilized to determine the extent of the participation of the β2 and the β4 subunits in nAChRs on rat and mouse hippocampal noradrenergic terminals. These studies indicated the presence of the β4 subunit in all nAChRs on rat terminals, but its presence in only 60% of mouse terminal nAChRs73.

α-Conotoxins PeIA, RgIA and Vc1.1

Three newly discovered α-conotoxins have been shown to target the α9α10 nAChR, a subtype with highly unusual pharmacology as compared to other nAChRs110, 111. α-CTx PeIA, cloned from the venom of Conus pergrandis, blocks heterologously expressed α9α10 nAChRs, as well as native α9α10 nAChRs in cochlear hair cells with IC50s of about 7 and 4 nmol/L, respectively112. However, this toxin is also a potent blocker of α3β2 and chimeric α6/α3β2β3 nAChRs, with IC50s of about 23 and 30 nmol/L, respectively112.

α-CTx RgIA, cloned from the venom of marine worm-hunting species Conus regius, is another α4/3 conotoxin similar to α-CTx ImI (Figure 1); however, unlike α-CTx ImI, it's a much more potent blocker of α9α10 than α7 nAChRs113 and the most selective α9α10 antagonist reported to date. α-CTx Vc1.1, cloned from the venom of Conus victoriae, is an α4/7 conotoxin originally shown to block nAChRs found in adrenal chromaffin cells114, 115. In addition, this peptide blocks vascular inflammatory responses evoked by electrical stimulation of unmyelinated sensory nerves114. Further pharmacological characterization of this α-conotoxin indicated that it is a potent inhibitor of the α9α10 nAChR116, 117. Unlike α-CTx RgIA, however, native α-CTx Vc1.1 (referred to as Vc1a) contains three post-translational modifications: a hydroxyproline at position 6, a γ-carboxyglutamate at position 14 and an amidated C-terminus117. Both α-CTx RgIA and α-CTx Vc1.1 were shown to be effective analgesic agents in a rat model of nerve injury116, 117, 118 and α-CTx Vc1.1 (drug name ACV1) entered human phase II clinical trials for treatment of neuropathic pain119. However, the role of the α9α10 nAChRs in mediating this analgesic effect has been challenged117, 120.

Structure-activity data with α-CTx RgIA have indicated that this toxin binds to the ACh binding site on the receptor and that toxin residues Asp5, Pro6, Arg7, and Arg9 are important for interaction with α9α10 nAChRs121. The recently published three-dimensional structure of α-CTx RgIA also confirms an important role for Arg9 in interacting with negatively charged residues in the α9α10 nAChR122. Saturation transfer difference NMR studies examining the binding of α-CTx Vc1.1 to Lymnaea stagnalis AChBP, a close structural homolog of the α7 nAChR binding site, indicates a role for Tyr10 in binding to the AChBP123. Interestingly, this residue in α-CTx RgIA does not seem to be important for interaction with α9α10 nAChRs121.

α-Conotoxin TxIA

α-CTx TxIA, from the venom of Conus textile, was recently discovered using a novel approach for identification of new α-conotoxins from crude venom124. In this approach, the venom of Conus textile was screened against the Lymnaea AChBP in a competition binding assay with 125I-α-bungarotoxin. Biochemical characterization indicated that α-CTx TxIA belongs to the α4/7 conotoxin family and has the same cysteine arrangement and disulfide connectivity common to other α-conotoxins in this family. Binding and functional assays indicated that the affinity of this toxin for Lymnaea AChBP (1.7 nmol/L) was higher than other previously identified α-conotoxins, and that α-CTx TxIA had high potency for α3β2 nAChRs. Structure-activity studies, together with co-crystallization of an analog of α-CTxIA, α-CTx TxIA[A10L], with Aplysia AChBP indicated an important role for a long chain hydrophobic residue at position 9 or 10 and the Arg at position 5 for toxin affinity for AChBP and α7, but not α3β2, nAChRs124

α-Conotoxin Lp1.1

α-CTx Lp1.1 was cloned from both the genomic DNA and cDNA of Conus leopardus125. Although it belongs to the α4/7 conotoxin family, its primary sequence is unique in that it lacks the conserved Ser and Pro that is found in the first loop of all known neuronally active α-conotoxins (Table 2). This toxin caused uncoordinated swimming when injected intramuscularly in fish. At higher concentrations, it causes seizure and paralysis126. Interestingly, another α4/7 conotoxin (LeD2) isolated from a different Conus species, Conus litteratus, has the identical sequence to Lp1.1127.

α-Conotoxins ArIA and ArIB

We recently identified two new α-conotoxins from the venom of Conus arenatus, α-CTx ArIA and α-CTx ArIB128. Both belong to the α4/7 conotoxin family and are potent blockers of the α7 nAChR. However, both toxins also block the α3β2 nAChR with nanomolar potency128. Structure-function analysis was used to create two analogs of α-CTx ArIB, α-CTx ArIB[V11L; V16A] and α-CTx ArIB[V11L; V16D], which have high affinity for α7 nAChRs but have comparatively low activity on α3β2 nAChRs128. Compared to α-bungarotoxin, however, the faster off-rate kinetics of the α-CTx ArIB analogs make them useful ligands in equilibrium binding experiments. α-CTx ArIB[V11L; V16D] blocks rat, mouse and human nAChRs128, 129. A radiolabeled version, 125I-α-CTx ArIB[V11L; V16A], has also been developed130.

Concluding remarks

The last quarter of the century has witnessed the discovery of variety of different α-conotoxins targeting various isoforms of nAChRs. The next few years promise even more groundbreaking progress thanks to the recent advancements in phylogeny and exogenomic discovery of novel conotoxins6, 131. In addition, structure-activity studies in combination with homology modeling will lead to better understanding of interactions between α-conotoxins and nAChR ligand binding site, allowing the creation of analogs with improved potency and/or selectivity towards particular subtypes of nAChRs. In view of the important physiological role of nAChRs in pain, inflammation, nicotine addiction, Alzheimer's and Parkinson's disease, specific targeting of the relevant nAChR subtypes is an attractive pharmaceutical strategy, with the α-conotoxins being among the most promising drug development leads.

Table Sequence and receptor specificity of neuronal-specific α-conotoxins. # denotes an amidated C-terminus; ^, free carbaoxyl C-terminus; γ, γ-carboxyglutamate; O, 4-trans-hydroxyproline; Y(S), sulfotyrosine. The cysteine residues that form the disulfide bridges are in bold.
Table Rank order of activity of α-conotoxins on select nAChR subtypes.
Table Some physiological functions mediated by nAChRs and the subtype(s) involved, as determined by pharmacological characterization using the indicated α-conotoxin (s).