Codex @codex  0

The free assembly-competent actin monomer concentration at which net addition at a specified actin filament end is zero in a specified model: $C_c=k_{-}/k_{+}$. Different ends and nucleotide conditions can have different values.

Loss of subunits from actin filaments. Blocking replacement from soluble actin monomers can cause net actin filament loss without direct drug binding to the actin filament.

An individual soluble actin subunit. Total actin monomer concentration and concentration available for assembly differ when actin monomers are bound by sequestering molecules.

Binding that reduces the pool of actin monomers available for productive actin filament assembly. Existing actin filaments can then disappear through ordinary turnover; their loss does not establish direct severing.

Formation of a stable assembly from which an actin filament can grow. Lowering the concentration of assembly-competent actin monomer can affect actin nucleation much more strongly than addition to an existing end.

Both structures turn over rapidly, with visible actin cables disappearing faster than actin patches under these conditions. Cable-positive biological cells fall from about 80% to essentially none within roughly two minutes. Polarized actin patches disappear over several minutes, while a transient depolarized-patch population persists before nearly all actin patches are lost by the end of the treatment time course.

Because ToxB binds actin monomers rather than existing actin filaments, a straightforward explanation is continuing normal disassembly without adequate replacement from assembly-competent actin monomers. The disappearance therefore reveals dynamic maintenance of structures that can look static in a fixed image. It does not show that the drug directly severs actin cables. Percentages of biological cells above a visibility threshold are not exponential decay curves for the lifetimes of individual actin filaments: cable thickness, staining sensitivity, actin monomer access and secondary loss of polarity can all affect the apparent timing. An analogous monomer-sequestering inhibitor established rapid yeast actin turnover in the classical study by Kathryn R. Ayscough and colleagues.^[ref]

Sequestration of assembly-competent G-actin explains both inhibition of polymer formation and loss of pre-existing cellular structures. In a simple model, free actin monomer $G$ binds drug $D$ to form an assembly-incompetent complex $GD$. At a actin filament end the net addition rate is $v=k_{+}[G]-k_{-}$, so lowering free $[G]$ reduces elongation and can drive net shortening below the relevant actin critical concentration $C_c=k_{-}/k_{+}$. Formation of new nuclei is also strongly reduced when free actin monomer is depleted. Existing actin filaments can consequently disappear through ordinary turnover even though ToxB never binds them.

Binding could instead or additionally alter the actin monomer conformation or nucleotide state needed for productive assembly. The resistant mutant supports an interaction with actin, but resistance could reflect reduced drug affinity, an allosteric change or altered assembly properties. The available endpoints do not uniquely distinguish these possibilities. In particular, they provide no positive evidence for direct actin filament severing or actin filament capping by ToxB. A real example of actin-monomer sequestration and resistant-actin genetics is latrunculin A; this is an analogy, not a chemical identification of ToxB.^[ref]

Experiment 2: relative to untreated wild-type actin, actin monomer sequestration predicts a longer apparent actin nucleation lag, a smaller elongation slope, and a lower steady-state polymer mass. At sufficiently strong binding the trace could remain nearly flat rather than displaying clearly separated phases. The wild-type sedimentation data support a lower endpoint, but do not establish the exact kinetic changes or whether the time to plateau is longer or shorter.

Experiment 3: the halo result does not determine the drug-free actin polymerization curve of act1-129. A mutation that mainly raises drug affinity could have a nearly wild-type curve without drug. A mutation that intrinsically destabilizes actin filaments or impairs actin nucleation could instead prolong the lag, reduce growth and lower the plateau. Thus a wild-type-like untreated curve is a useful conditional reference prediction, not a result that follows from cellular hypersensitivity.

Experiment 4: act1-129 is about sixfold more sensitive in the cellular assay. If that reflects a stronger direct effect of ToxB on its actin monomers, predict greater suppression than in experiment 2: a longer or effectively absent actin nucleation phase, slower or undetectable elongation, and a still lower plateau relative to its own drug-free control. It could be almost flat. However, cellular hypersensitivity could also arise from reduced functional reserve or altered interactions with other proteins; it does not guarantee stronger inhibition of purified actin. These alternatives are exactly what the comparison with experiment 3 resolves. No numerical lag, slope or plateau can be calculated from halo sizes alone.

The three resistant substitutions map near one another in the folded actin monomer, near the protein binding site for the nucleotide. This cluster identifies a candidate region involved in drug recognition or in the conformational response to binding. Neighbouring residues in three-dimensional structure can help form a pocket even when they are not adjacent in the protein primary structure. Reduced drug binding at this region would connect the structural map to the resistant phenotype.

The map does not show a bound ToxB molecule. Substitutions could act indirectly by changing folding, nucleotide exchange or the assembly-competent conformation. Proximity to ATP does not prove competition for the ATP-binding site, and the mutant's abnormal baseline actin polymerization makes an allosteric explanation particularly relevant. Direct binding measurements on comparably folded wild-type and mutant actin monomers, together with nucleotide-binding controls and a ligand-bound structure, would distinguish these models. The allele/residue labels are interpreted as defined in this paper; no identification of their original experimental sequence substitutions is required.