Cone Snail Conotoxin Characterization Service

Why This Case Matters for Drug Discovery

T1-conotoxins are a pharmacologically significant subfamily — typically 9–13 residues with a four-cysteine framework (C–C–C–C), targeting neurotensin receptors, opioid receptors, and sodium channels. They are prime scaffolds for non-opioid pain therapeutics and neurological disease programs. But a critical challenge in conotoxin-based drug discovery is this: the primary amino acid sequence does not uniquely determine the biological activity. The disulfide connectivity — which cysteine pairs form bonds — defines the three-dimensional fold and pharmacophore orientation. Getting it wrong has downstream consequences for every SAR study and analog design decision.

This case illustrates what happens when a conotoxin defies the canonical pattern, and how experimental disulfide mapping — rather than inference from sequence alone — changes the picture entirely.

The Challenge

Standard conotoxin characterization pipelines face two systematic failure modes when encountering novel T1-conotoxins:

• Sequence misassignment in hypervariable loop regions. Generic peptide ID algorithms discard sequences that don't match existing database entries. In novel T1-conotoxins, the 4-residue hypervariable loop (positions 4–7 in a 10-residue peptide) is precisely the region that determines target selectivity — and precisely the region most likely to be misread or dropped.
• Incorrect disulfide connectivity inference. Most automated pipelines assume the canonical C1–C3/C2–C4 pattern for four-cysteine conotoxins, because that is overwhelmingly the most common. But conotoxin superfamilies can harbor non-canonical connectivities that change the fold entirely. No automated disulfide predictor can reliably detect this.

In this study, previous characterization attempts on related species had either failed to resolve disulfide connectivity, or had assumed the canonical pattern — leading to structural models that were subsequently revised.

Analytical Approach

Venom ducts from Conus bandanus (a molluscivorous species found in Vietnamese waters) were dissected and extracted with 0.1% TFA. The extract was fractionated by RP-HPLC on a C18 column. MALDI-TOF/TOF MS (AB Sciex 4800) was used to screen fractions for conotoxin candidates based on mass range. Fractions containing candidate peptides were selected for MS/MS at 8 keV CID. Edman degradation was run in parallel to confirm the N-terminal sequence independently.

For disulfide connectivity, the rapid partial reduction-alkylation method was applied: TCEP treatment at pH 3.0 for 2.5 minutes (partial reduction), followed immediately by IAA alkylation at pH 8.0. This generates a diagnostic mixture of non-reduced, mono-alkylated, and di-alkylated species. MS/MS of each species identifies which cysteine residues carry the alkyl tag — and therefore which were reduced — revealing the native connectivity.

Key Findings

A novel T1-conotoxin, designated Bn5a, was isolated:

The most critical finding: despite clearly belonging to the T1-superfamily (87% signal sequence identity), Bn5a exhibits C1–C4/C2–C3 connectivity — the same pattern seen in T2-conotoxins and the U-superfamily toxin MrIA. This is structurally distinct from the canonical T1 pattern (C1–C3/C2–C4). Edman degradation confirmed all 10 residues including isoleucine at position 5, which had been misassigned as leucine in homologs from other species. The C-terminal residue was confirmed as a free cysteine rather than an amide — a finding with implications for synthetic analog design, since amidated and non-amidated forms can differ in protease stability and receptor affinity.

What This Means for Your Conotoxin Drug Discovery Program

• Sequence alone is not enough. The same 10-residue sequence with a different disulfide connectivity folds into a completely different 3D structure. Any SAR study based solely on sequence — without experimental connectivity confirmation — risks drawing wrong conclusions.
• Non-canonical connectivities are not rare exceptions. Within the T1-superfamily alone, multiple non-canonical patterns have been documented across species. When characterizing a novel conotoxin from a non-model Conus species, assuming the canonical pattern is a material risk for drug discovery programs.
• The C-terminal status matters for synthesis. Free C-terminal cysteine versus amidation changes the synthetic route and affects downstream activity. Bn5a's free C-terminus was only confirmed by MS/MS of the unmodified peptide — it would have been missed by any workflow that only analyzed the amidated form.
• Signal sequence identity does not guarantee canonical connectivity. Bn5a's 87% signal sequence match to T1-superfamily precursors did not predict its non-canonical connectivity. Only the experimental rapid partial reduction-alkylation protocol revealed it.

Conclusion

Accurate determination of disulfide connectivity is not an optional add-on in conotoxin-based drug discovery — it is a prerequisite. The C1–C4/C2–C3 connectivity in Bn5a would have been misassigned by any automated pipeline relying on the canonical T1-conotoxin pattern, and any downstream analog design or SAR study built on that incorrect structure would have been compromised from the start. This case underscores why experimental disulfide mapping, combined with expert manual annotation, is essential before proceeding to any conotoxin lead optimization program.

Read the full paper (DOI: 10.1590/1678-9199-JVATITD-2019-0095)