Benjamin Thomas Goult

Talin is a central mechanosensitive adaptor that links integrins to the actin cytoskeleton and converts mechanical force into biochemical signals. Vertebrates express two highly homologous talin isoforms, talin-1 and talin-2, yet the molecular basis for their nonredundant functions remains poorly understood. Here, we investigate how isoform-specific sequence differences alter the mechanical response and ligand-binding capabilities of the talin rod. Using single-molecule force spectroscopy and biochemical binding assays, we show that talin-2 preserves the overall mechanical architecture of talin-1 but exhibits distinct mechanical responses in key rod regions. Talin-2 R7-R9 exhibits a previously unobserved three-domain cooperative unfolding pathway, while an alternative sequential pathway reveals early unfolding of R9 at substantially lower force than in talin-1. Strikingly, talin-2 R4-R6 has acquired two additional mechanically regulated interactions: a previously unrecognized vinculin-binding site in R5 and an A-kinase anchoring protein (AKAP) region within R5-R6 that recruits protein kinase A (PKA). Together with the conserved R9 AKAP site, this establishes two spatially separated, mechanically regulated PKA-anchoring regions within the talin-2 scaffold. These findings establish that isoform-specific sequence divergence not only reshapes the mechanical response of the talin scaffold but also expands its repertoire of force-regulated interactions, providing a mechanism for the assembly of different structural and signaling complexes in an isoform-dependent manner.

Leukocyte adhesion deficiency syndrome type III is characterized by recurrent infections and is caused by FERMT3 gene (encoding kindlin-3) mutations. FERMT3 encodes 2 isoforms, a standard kindlin-3 and a longer splicing variant, which differs by the addition of 4 residues isoleucine-proline-arginine-arginine (IPRR) in the pleckstrin homology (PH) domain (kindlin-3–IPRR). Previous studies suggested that kindlin-3–IPRR was dysfunctional in inducing neutrophil adhesion, attributed to altered phospholipid binding of the IPRR-containing PH domain. Here, we show that kindlin-3–IPRR is fully functional in activating β2 integrins and promotes neutrophil adhesion. The IPRR insert enhances phospholipid binding in vitro and promotes association of kindlin-3 with the plasma membrane. By analyzing RNA-sequencing data sets of neutropoiesis, we show that the kindlin-3 short isoform lacking IPRR is highly expressed in neutrophil precursors but downregulated in mature neutrophils. In contrast, a higher proportion of the long isoform was detected in hematopoietic stem cells (HSCs) and mature neutrophils. Kindlin-3–IPRR supports robust adhesion of HSCs. We propose that kindlin-3–IPRR may be crucial in conditions requiring rapid neutrophil mobilization. Indeed, kindlin-3 splicing is altered in neutrophils from patients undergoing stress myelopoiesis after exposure to granulocyte colony-stimulating factor (G-CSF) or HSC transplantation (HSCT) compared with healthy donors. During stress myelopoiesis induced by G-CSF treatment, the kindlin-3 long isoform is selectively upregulated in mature neutrophils, coinciding with enhanced β2-integrin activation. These findings support a model in which alternative splicing of FERMT3 dynamically fine-tunes integrin activation and leukocyte adhesion during hematopoietic stress, with potential implications for monitoring hematopoietic recovery after HSCT.

Talin is a mechanosensitive scaffold that couples integrins to the actin cytoskeleton and coordinates signaling at integrin adhesion complexes. A second integrin-binding site (IBS2) has been proposed within the R11–R12 region of the talin rod, although the molecular basis of this interaction has remained unclear. Here, we show that talin R11 contains a cryptic aggregation-prone region (APR) embedded within Helix50. Mechanical unfolding of R11 exposes this motif, enabling high-affinity talin–talin interactions and force-dependent self-association. The isolated APR forms amyloid-like fibers in vitro, and APR-disrupting gatekeeper mutations suppress fiber formation, talin self-association, and cellular recruitment. We therefore designate this motif Aggregation-Prone Region 1 (APR1). Mitochondrial-targeting experiments demonstrate that APR1-dependent recruitment occurs in cells independently of focal adhesions and integrins. APR1 overlaps the vinculin-binding site in Helix50, and vinculin disrupts preexisting talin assemblies, revealing a mechano-chaperone function for vinculin. Disruption of APR1-mediated self-association does not impair adhesion formation but instead generates larger, more talin-rich adhesions with altered assembly dynamics and signaling properties, indicating a role for APR1 in regulating adhesion remodeling. Together, these findings identify force-dependent talin self-association as a mechanism in mechanotransduction and suggest that the region previously designated IBS2 functions primarily as an aggregation-prone regulatory element.

Direct visualisation of drug-target engagement within living cells remains a major challenge. Here we develop a cell-permeable fluorogenic analogue of the phosphodiesterase 4 (PDE4) inhibitor rolipram (MAN193), generated by conjugation to fluorescein diacetate. Following intracellular activation, the probe functions as a near-neutral PDE4 antagonist and enables direct visualisation of endogenous PDE4 populations. Using advanced fluorescence spectroscopy imaging approaches, including molecular brightness analysis and fluorescence anisotropy imaging, we demonstrate rolipram-displaceable binding of the analogue to cytosolic PDE4 and resolve thesubcellular distribution of binding sites. We further show that the probe enables visualisation of untagged PDE4 at focal adhesion complexes in cardiomyocyte-like cells. Together these findings establish fluorogenic drug conjugation combined with quantitative spectroscopy imaging as a generalisable strategy to map intracellular drug-target engagement with subcellular resolution, providing an effective framework for interrogating the spatial pharmacology of small molecules in living systems.

SHANK proteins form core postsynaptic density (PSD) scaffolds that organise synaptic signalling complexes through multiple protein-protein interaction domains, including PDZ domains that typically recognise C-terminal peptide motifs. Here, we identify an internal PDZ recognition mechanism that links the synaptic protein Densin-180 to SHANK and promotes SHANK scaffold assembly. We map SHANK binding to an internal PDZ-binding motif in Densin-180 (residues 843-863) and show by NMR spectroscopy and fluorescence polarisation that this motif binds SHANK1-3 PDZ domains with high affinity and specificity, competing with canonical C-terminal ligands. The crystal structure of the Densin-180-SHANK complex reveals that Phe858 inserts into the hydrophobic pocket of the PDZ domain despite the absence of a terminal carboxylate. In neurons, this interaction mediates Densin-180 recruitment to dendritic spines and drives activity-dependent reorganisation of postsynaptic SHANK3 assemblies, whereas mutation of Phe858 disrupts both processes. Disrupting the Densin-180-SHANK interaction using a Densin-180-derived peptide impairs activity-dependent structural plasticity of spines, accompanied by defects in PSD organisation and actin cytoskeleton remodelling. These findings define an internal mode of PDZ recognition and reveal how Densin-180 couples neuronal activity to the remodelling of postsynaptic SHANK scaffolds.

Cells transmit force between the extracellular matrix and the actin cytoskeleton through integrin adhesion complexes centred on talin and vinculin. Vinculin binds talin through α-helical vinculinbinding sites (VBS) that are exposed when talin rod domains unfold under force. Dissecting the significance of this interaction has relied heavily on the A50I mutation in vinculin, which has been widely used as a talin-binding-null mutant. Here we show that although the A50I mutation abolishes binding to the α-catenin VBS, it retains nanomolar-affinity binding to multiple talin VBS. We therefore designed an improved mutant, I12K/A50I, that eliminates this residual talin binding. Biochemical assays and single-molecule stretching experiments demonstrate that I12K/A50I VD1 fails to bind talin even when VBS are exposed by force. Using vinculin tension and conformation sensors, we show that talin binding is required for efficient recruitment of vinculin to focal adhesions and for establishing spatial gradients of vinculin tension. However, vinculin can still experience mechanical load in the absence of talin binding. These results demonstrate that A50I is not a talin-binding-null and reveal that, while talin is not required for vinculin loading, it is essential for organising the spatial distribution of mechanical load within adhesion complexes.

KANK proteins link integrin adhesions to the cortical microtubule stabilising complex (CMSC) through interactions with the adhesion adaptor talin. However, how KANK proteins are regulated remains unclear. Here we show that the KN domain of KANK proteins contains separable regions that mediate talin binding and a conserved intramolecular interaction. Using fluorescence polarisation, NMR spectroscopy and structural analysis, we map an interaction between the N-terminal KN domain and the C-terminal ankyrin repeat domain and identify residues 60-68 of the KN domain as required for this intramolecular interaction. In contrast, the canonical LD motif within residues 30-60 mediates binding to talin. Deletion of residues 60-68 disrupts the intramolecular interaction while preserving talin binding, demonstrating that the KN domain contains distinct modules for talin engagement and intramolecular regulation. This regulatory architecture is conserved across the KANK family, although sequence variation modulates the strength of the intramolecular interaction. Together, these findings identify a modular organisation within the KANK KN domain that separates talin recognition from intramolecular regulation and is consistent with an autoinhibitory mechanism.

Cells transmit force between the extracellular matrix and the actin cytoskeleton through integrin adhesion complexes centred on talin and vinculin. Vinculin binds talin through α-helical vinculin-binding sites (VBS) that are exposed when talin rod domains unfold under force. Dissecting the significance of this interaction has relied heavily on the A50I mutation in vinculin, which has been widely used as a talin-binding-null mutant. Here we show that although the A50I mutation abolishes binding to the α-catenin VBS, it retains nanomolar-affinity binding to multiple talin VBS. We therefore designed an improved mutant, I12K/A50I, that eliminates this residual talin binding. Biochemical assays and single-molecule stretching experiments demonstrate that I12K/A50I VD1 fails to bind talin even when VBS are exposed by force. Using vinculin tension and conformation sensors, we show that talin binding is required for efficient recruitment of vinculin to focal adhesions and for establishing spatial gradients of vinculin tension. However, vinculin can still experience mechanical load in the absence of talin binding. These results demonstrate that A50I is not a talin-binding-null and reveal that, while talin is not required for vinculin loading, it is essential for organising the spatial distribution of mechanical load within adhesion complexes.

The tight regulation of the integrin family of extracellular matrix (ECM) receptors is essential for the coordination of most cellular processes. Talin regulates integrin function by binding to the cytoplasmic tail of the integrin beta subunit. Two integrin-binding sites (IBS) in talin have been reported, one in the talin head domain (IBS1), and a second in the C-terminal rod region of talin (IBS2), mapped to the R11-R12 domains. Whilst the structural details of integrin binding to IBS1 are well understood, IBS2’s mode of binding integrin is less clear, and the biochemical details of this site have been elusive. Here we report that talin R11 contains a cryptic high affinity talin-binding site. We show that mechanical unfolding of R11 exposes an Aggregation Prone Region (APR) in Helix50 that oligomerizes with other talin R11s. Our data support a model whereby the process of mechanically unfolding one R11, exposes and maintains the APR in a high affinity conformation that drives unfolding of other R11 domains and oligomerisation. Atomic Force Microscopy confirms that the APR region alone forms micron long amyloid-like fibres, via a beta-sheet amyloid arrangement which can be abolished using a canonical “gatekeeper” point mutation, V2078K, that prevents APR interactions. Introducing this gatekeeper mutant in talin eliminates recruitment of talin IBS2 to integrin adhesion sites. The APR region in talin overlaps with a vinculin-binding site in Helix50, and we show a novel role for vinculin as a mechano-chaperone for talin, able to resolve the talin large species by binding to the VBS in Helix50. In light of this new information, we propose that the ability of R11-R12 to target to integrin adhesion complexes and enhance integrin activation can be better explained by a model where functional aggregation of talin serves to recruit and activate more talin molecules at the adhesion site leading to enhanced integrin activation. Whilst further work is required to fully exclude integrin binding to R11 we suggest that the name IBS2 might be a misnomer and propose the name Aggregation Prone Region 1 (APR1) for this talin oligomerisation motif.