Cell Biology · Unit 5: Cells in tissues

Adhesion and the Extracellular Matrix

Cells in a tissue hold on to two things: each other, through cadherins, and the matrix around them, through integrins. This lesson follows each chain from outside the cell to the cytoskeleton, takes the matrix apart into its working parts, and weighs the evidence that some of these links respond to force. Afterwards you can read an attachment or aggregation assay, correct it for background, and say what it does and does not prove.

Watch first Two Grips, Each Wired to a Cable · 2:51

Take the calcium out of the fluid around a sheet of cultured epithelial cells and the cells round up and let go of one another. Put the calcium back and they knit together again.

The calcium is not glue. It stiffens the outer arms of a family of adhesion proteins called cadherins, so that the arm on one cell can grip the arm on the next. That is one of two ways a cell in a tissue holds on. The other is to grip the extracellular matrix (ECM), the woven material that cells secrete into the space around them. Both grips continue inside the cell, onto the cytoskeleton, so a pull on the outside reaches the cell’s own filaments. This lesson follows each chain, takes the matrix apart, asks which links feel force, and then reads the kind of experiment that identified them.

Two ways to hold on: cadherins and integrins

Cell to cell. A classical cadherin crosses the plasma membrane once. Its outer part is a chain of repeated domains with calcium ions bound at the joints, and it binds the same kind of cadherin on the neighbouring cell. Binding to one’s own kind is called homophilic. These contacts gather at adherens junctions, belts of cadherin that circle the top of an epithelial cell (the lesson on epithelia builds the full sheet).

Inside, the chain runs like this. β-catenin binds the cadherin’s cytoplasmic tail directly. β-catenin recruits αE-catenin, which couples the junction to actin filaments. A third protein, p120-catenin, binds the tail close to the membrane, the juxtamembrane region, and helps keep the cadherin at the surface. Picture a rope (the cadherin) tied through a wall to a hook (β-catenin), with a second line (α-catenin) running from the hook to the cell’s internal cables.

β-Catenin has a second job. Outside junctions it is also the transcriptional co-activator of the Wnt pathway, covered in the Wnt lesson. The same protein serves adhesion and gene control, and cells regulate how much goes to each.

Cell to matrix. Integrins are the main matrix receptors. Each is a heterodimer, one α subunit paired with one β subunit. Vertebrates have 18 α and 8 β subunits, which pair into 24 different integrins, each with its own preferred ligands. Many recognise a short sequence, Arg-Gly-Asp (RGD), found in fibronectin and several other matrix proteins.

Inside the cell, talin binds the integrin’s β tail and links it to actin; vinculin binds talin and actin and reinforces the link. In cultured cells these complexes gather into focal adhesions, plaques at the cell’s underside where integrins hold the matrix. Crawling cells make and break them constantly (see migration and cilia), and the signals they send through Rho GTPases belong to integrins and Rho GTPases.

The exception. Not every integrin reaches actin. In hemidesmosomes, which pin the basal cells of epithelia such as skin to their basement membrane, integrin α6β4 binds laminin-332 outside the cell and connects through plectin to keratin intermediate filaments. Two chains with the same overall shape end on different cables.

Not every cell-cell contact needs calcium either. The Chinese hamster V79 cells that Masatoshi Takeichi studied in 1977 had both a calcium-dependent and a calcium-independent mechanism.

What the matrix is made of, and what each part does

Think of the ECM as reinforced concrete: steel cables for pulling, filler for pushing, and a mesh that holds it together. Each component maps onto one of those jobs.

  • Fibrillar collagens are built from three chains wound into a triple helix (the structure is in protein 3D structure). The helices pack side by side into fibrils, and their arrangement gives bone and tendon their strength under tension. Cells secrete them as soluble precursors and trim them outside, a route traced in the secretory pathway.
  • Elastic fibres let tissues such as artery walls and lung stretch and recoil.
  • Fibronectin forms fibrillar networks that cells grip through integrins, and the network also binds growth factors and other signalling molecules.
  • Basement membranes are thin sheets under every epithelium. Their main noncollagenous proteins are laminins, and their scaffold is type IV collagen, which forms networks rather than cables.
  • Proteoglycans are core proteins carrying long glycosaminoglycan chains, sugar polymers studded with negative charges (see carbohydrates). They draw in water and form hydrated gels that help a tissue such as cartilage resist compression. Hyaluronan, a very long glycosaminoglycan without a protein core, helps keep the matrix hydrated.

The matrix is also a store. Growth factors held in it can be released when the matrix is cut or pulled. Latent TGF-β, held in the matrix until it is freed, is the standard example (see TGF-β and Smads). Neurons use basal lamina too: the agrin that organises the neuromuscular junction is deposited there.

Adhesions that feel force

A cell pulls on its adhesions with myosin II. Do the links themselves respond? Two kinds of evidence say some can, and a third says force is not always required.

Talin unfolds. The talin rod is a stack of 13 helical bundles. In 2009 Armando del Rio and colleagues stretched a purified rod fragment at 12 piconewtons (pN) and saw fluorescent vinculin-head binding events rise from one at most to as many as three. The reading is that force unfolds bundles and exposes vinculin sites hidden inside them. Magnetic-tweezer work on the mouse talin1 rod in 2016 found bundles unfolding and refolding at a few piconewtons, and a model built on those rates suggests the rod can act as a force buffer.

Cadherin-catenin-actin bonds can hold longer under load. In 2014 Craig Buckley, Jiongyi Tan and colleagues held a reconstituted complex, built with zebrafish αE-catenin and β-catenin, against an actin filament in an optical trap. A subpopulation of bonds lasted up to 25 seconds at 5 to 10 pN, and a two-state catch-bond model fit the force-lifetime data better than the alternatives tested. A catch bond lives longer as force rises, over a limited range; beyond it, more force breaks it faster, like any bond. And 25 s is the long tail, not the typical lifetime.

Force is not always needed. In 2020 Paul Atherton, Christoph Ballestrem and colleagues anchored talin or vinculin to the outer surface of mitochondria in mouse NIH3T3 fibroblasts. Full-length wild-type proteins did not recruit each other. Releasing autoinhibition, the folded-shut state that hides each protein’s binding sites, let them associate, even with myosin II blocked by blebbistatin.

These results fit together if you keep their systems apart. Purified fragments show that force can expose sites. A cell assay shows that activation can do it without force. Neither tells you which route dominates in a given adhesion in a tissue. And a few piconewtons on one molecule is not the traction a whole cell exerts.

Substrate mechanics also bias cell behaviour in culture. Engler, Sen, Sweeney and Discher found in 2006 that naive mesenchymal stem cells on soft, brain-like gels drifted toward neuronal-like cells and on rigid, bone-like gels toward bone, and that blocking myosin II removed the effect. In 2024 Qiao and colleagues varied how fast gels relax under strain while holding initial stiffness fixed, and greater relaxation favoured astrocytes from rat and mouse neural stem cells. Both are culture systems.

Reading adhesion assays

The classic evidence for these receptors came from three kinds of experiment. A dissociation and reaggregation assay separates cells with a chosen treatment and measures re-clumping. An attachment assay counts cells that stay on a coated well after washing. An antibody-blocking version adds an antibody against a candidate receptor. Here is how to read one.

Worked problem (made-up data). Wells are coated with fibronectin or with bovine serum albumin (BSA) only, a protein cells do not grip. Each well gets 20,000 fibroblasts. After 30 minutes and a standard wash, the retained cells are:

  • fibronectin, no antibody: 14,000
  • fibronectin plus isotype-control antibody (same antibody class, irrelevant target): 13,600
  • fibronectin plus integrin-blocking antibody: 3,200
  • BSA only: 1,000

Viability is above 95% in every well, so lost cells were not dead cells.

Adhesion assay: read the wells⚙ original · interactive
Choose a well

Subtract the BSA background from both wells, then compare the blocking well with the isotype-control well, not with the well that has no antibody. Lower attachment shows the integrin is needed, not that it touches fibronectin. The counts are made-up data.

Step 1: percent attached. 14,000/20,000 = 70%; 13,600/20,000 = 68%; 3,200/20,000 = 16%; 1,000/20,000 = 5%.

Step 2: subtract the background. The BSA well shows cells that stick to plastic or protein nonspecifically. Specific attachment is 13,600 − 1,000 = 12,600 with the isotype control, and 3,200 − 1,000 = 2,200 with the blocking antibody.

Step 3: inhibition.

inhibition = 1 − (blocking well − background) / (isotype well − background)

1 − 2,200/12,600 = 1 − 0.175 = 0.825, or about 83%.

Step 4: why the isotype well. Adding any antibody protein lowered attachment from 70% to 68% here. Comparing with the no-antibody well would credit that 2-point drop to the integrin.

Step 5: what this does not show. An antibody can block by sitting near the receptor, or by bending it. Lower attachment shows the integrin is needed, not that it touches fibronectin. Horwitz and colleagues made that next step in 1985 with purified antigen in equilibrium gel filtration, and the antibody blocked the purified laminin association.

Design problem

Do cadherins sort cells into tissues? Design an experiment in the spirit of Nose, Nagafuchi and Takeichi (1988), using mouse L cells, which have little cadherin of their own.

One way to do it

Make two L-cell lines, one expressing E-cadherin and one expressing P-cadherin. Label them with different dyes, mix them, and let them aggregate in calcium.

Predictions: if any adhesion will do, aggregates contain both colours intermixed. If binding is type-specific and homophilic, each aggregate shows patches of one colour.

Controls: untransfected L cells (little aggregation); aggregation in EGTA, which removes calcium (should fail); a function-blocking antibody for each cadherin (should break up only its own population’s contacts); and matched surface expression, measured by flow cytometry. The last matters because a difference in amount alone can drive sorting, as differential-adhesion arguments predict, so unequal levels would confound the type-specific reading. Score the images with a mixing index, such as the fraction of each cell’s neighbours that share its colour, across many aggregates and independent experiments.

Limit: these are fibroblasts in suspension, not an embryo. The result shows cadherins can sort cells; whether they do in a tissue needs work in the embryo itself, such as the cadherin changes that accompany neural crest cells leaving the neural tube.

How we measure it

Dissociation and reaggregation

Cells are separated with a chosen treatment (EDTA, trypsin with calcium, trypsin with EDTA), then shaken together and counted as they form clumps. Which treatments spare aggregation, and whether calcium is needed during re-aggregation, separates one adhesion mechanism from another.

Attachment assay with a function-blocking antibody

A known number of cells is put into wells coated with a matrix protein, unbound cells are washed off, and the rest are counted. An antibody that lowers the count, compared with a control antibody, implicates its target in attachment but does not show direct binding.

Equilibrium gel filtration

A purified receptor runs through a gel filtration column pre-equilibrated with its ligand, so a complex that falls apart and re-forms quickly stays occupied throughout. A complex that runs larger than the receptor alone is evidence of direct association.

Single-molecule force spectroscopy

One protein or one complex is held between a surface and a bead in magnetic tweezers or an optical trap and pulled with a known force of a few piconewtons. It shows whether domains unfold, binding sites open, or a bond lasts longer under load.

Finish line

Test yourself

8 questions

Practice and MCAT-style, with explained answers.

Next · Lesson 23

Epithelia and Tissue Organization

The gut lining is one cell thick. How does it move glucose one way and keep the gut contents out?

Continue →

← Lesson 21 · Cell Death and Clearance

C-cadherin ectodomain — the outer arm of a classical cadherin
C-cadherin ectodomain — the outer arm of a classical cadherin ● LIVE · RCSB PDB 1L3W
open on RCSB PDB ↗
Loading live structure 1L3W from RCSB PDB…

Five repeated domains form a curved rod, with calcium ions bound at the joints between them. Without calcium the joints go floppy, which is one reason cadherin adhesion needs calcium.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
β-catenin bound to the E-cadherin tail — the first link inside the cell
β-catenin bound to the E-cadherin tail — the first link inside the cell ● LIVE · RCSB PDB 1I7W
open on RCSB PDB ↗
Loading live structure 1I7W from RCSB PDB…

The cadherin tail, normally floppy, lies stretched along the groove of β-catenin's armadillo repeats. This is the contact that ties the adhesion molecule to the catenins and through them to actin.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Integrin αVβ3 with an RGD ligand — a matrix receptor holding its code
Integrin αVβ3 with an RGD ligand — a matrix receptor holding its code ● LIVE · RCSB PDB 1L5G
open on RCSB PDB ↗
Loading live structure 1L5G from RCSB PDB…

Find the small peptide where the α and β heads meet. Its arginine contacts the α subunit and its aspartate reaches a metal ion held by the β subunit, so one ligand bridges both subunits.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Fibronectin type III repeats 7 to 10 — the cell-binding region of fibronectin
Fibronectin type III repeats 7 to 10 — the cell-binding region of fibronectin ● LIVE · RCSB PDB 1FNF
open on RCSB PDB ↗
Loading live structure 1FNF from RCSB PDB…

Four beads on a string. The RGD sequence sits on an exposed loop of the tenth repeat, where an integrin can reach it.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
A collagen-like peptide — the triple helix
A collagen-like peptide — the triple helix ● LIVE · RCSB PDB 1CAG
open on RCSB PDB ↗
Loading live structure 1CAG from RCSB PDB…

Three strands wind around one another. Glycine, the smallest amino acid, falls at every third position, where the strands pack closest at the centre; this peptide carries one deliberate glycine-to-alanine substitution, which loosens the helix where it sits.

Drag to rotate · scroll or pinch to zoom · right-drag to pan
Masatoshi Takeichi

Masatoshi Takeichi

Separated calcium-dependent from calcium-independent adhesion in Chinese hamster V79 cells and tied the calcium-dependent kind to a surface protein of about 150 kDa; his laboratory later showed that cells made to express one cadherin prefer partners with the same one.

Takeichi's 1977 paper turned on a practical point: how you take cells apart decides what they can do afterwards. He dissociated Chinese hamster V79 cells three ways, with EDTA, with trypsin in the presence of calcium, and with trypsin plus EDTA, and compared how well each preparation re-aggregated. The comparison separated two mechanisms. One needed no calcium and looked like simple clumping. The other needed calcium and physiological temperature, and the cells deformed against each other, widening their contact. Both were sensitive to trypsin, but calcium protected the calcium-dependent one. Labelling surface proteins with iodine picked out a protein of about 150,000 daltons found only in cells with calcium-dependent adhesion, and calcium protected it from trypsin as well. That was a correlation, not proof. In 1988, with A. Nose and A. Nagafuchi, his laboratory put E- or P-cadherin into mouse L cells, which have little cadherin activity, and the transfected cells preferred partners carrying the same cadherin. A calcium requirement had become a family of molecules that could sort cells.

✓ verbatim Takeichi, J. Cell Biol. 75:464-474 (1977); Nose, Nagafuchi & Takeichi, Cell 54:993-1001 (1988).
Alan Horwitz

Alan Horwitz

With Karen Duggan, Richard Greggs, Cynthia Decker and Clayton Buck, showed that the CSAT antibody blocks attachment to laminin and fibronectin and that the purified CSAT antigen binds both, which strongly implicated it as a matrix receptor.

The CSAT monoclonal antibody recognised a complex of membrane glycoproteins and stopped cells from attaching to matrix. In 1985 Alan Horwitz, Karen Duggan, Richard Greggs, Cynthia Decker and Clayton Buck asked whether that complex was the receptor itself. First the phenotype: the antibody inhibited adhesion of cardiac fibroblasts, tendon fibroblasts and skeletal myoblasts to both laminin and fibronectin. A blocked attachment could mean the antibody sat on the receptor, or only near it. So they purified the antigen and ran it through a gel filtration column pre-equilibrated with laminin or fibronectin, which keeps a fast-exchanging complex occupied all the way down the column. Both matrix proteins formed complexes with the antigen. The antibody inhibited the laminin association, and synthetic peptides containing fibronectin's cell-binding sequence inhibited both. All three glycoproteins of the antigen travelled together as one oligomer. Phenotype plus purified binding is the pairing the paper used to argue for a receptor, and the CSAT antigen was soon placed in the family now called integrins.

✓ verbatim Horwitz, Duggan, Greggs, Decker & Buck, J. Cell Biol. 101:2134-2144 (1985).
Erkki Ruoslahti

Erkki Ruoslahti

With Michael Pierschbacher, showed that small synthetic fragments of fibronectin can duplicate its cell-attachment activity, and that the tetrapeptide Arg-Gly-Asp-Ser accounts for it.

Fibronectin is a large matrix protein built from many repeated modules, and cells stick to it. By the early 1980s the question was which small part of it the cell actually recognises. Michael Pierschbacher and Erkki Ruoslahti answered by making short synthetic pieces of the protein's cell-attachment region and testing whether they could do the job of the whole. In 1984 they reported in Nature that small synthetic fragments could duplicate fibronectin's cell-attachment activity, and that the activity could be accounted for by four amino acids: arginine, glycine, aspartate and serine, RGDS. The same sequence appeared in at least five other proteins, so they proposed that it might be a recognition signal shared by several of them. The short code made matrix binding easy to test. A soluble RGD peptide can compete with a matrix protein for its receptor, and the Horwitz group used such peptides the following year to block purified receptor from binding fibronectin and laminin. Later crystal structures caught an RGD ligand sitting in the head of an integrin.

✓ verbatim Pierschbacher & Ruoslahti, Nature 309:30-33 (1984).
Dennis Discher

Dennis Discher

With Adam Engler, Shamik Sen and H. Lee Sweeney, showed that the elasticity of a culture substrate biases the lineage that naive mesenchymal stem cells adopt, and that blocking nonmuscle myosin II removes the effect.

Most culture dishes are far stiffer than any soft tissue. In 2006 Adam Engler, Shamik Sen, H. Lee Sweeney and Dennis Discher grew naive mesenchymal stem cells on gels whose elasticity they could set, from soft to comparatively rigid. Soft gels that mimic brain pushed the cells toward a neuronal-like phenotype, stiffer gels that mimic muscle toward muscle, and rigid gels that mimic collagenous bone toward bone. During the first week the cells could still be redirected by adding soluble induction factors; after several weeks they stayed with the lineage the substrate had specified. Inhibiting nonmuscle myosin II, the motor that lets a cell pull on what it holds, blocked all of these elasticity-directed choices. The finding is a culture result, and it does not say that stiffness sets fate in a body. It made the mechanical properties of a substrate into an experimental variable, and later work in other culture systems separated stiffness from other material properties such as how fast a gel relaxes under strain.

✓ verbatim Engler, Sen, Sweeney & Discher, Cell 126:677-689 (2006).
Test yourself
Question 1 of 8 Score 0 / 0

Which chain of proteins connects an epithelial adherens junction to the cytoskeleton?
Sources & Integrity

Every structure, slide, simulation, and clip on this page is listed below with the source it came from. None of the 3D structures is drawn by hand or by AI: each one loads by accession ID from RCSB PDB or PubChem. This panel is generated from the lesson's own data file, so what it lists is what the page loads.

Molecular structures live data

  • C-cadherin ectodomain — the outer arm of a classical cadherin — RCSB PDB 1L3W · coordinates CC0 · Xenopus laevis C-cadherin (EP-cadherin) extracellular region, X-ray crystal structure (Boggon et al., Science, 2002)
  • β-catenin bound to the E-cadherin tail — the first link inside the cell — RCSB PDB 1I7W · coordinates CC0 · Mouse β-catenin armadillo repeats with a phosphorylated E-cadherin cytoplasmic domain, X-ray crystal structure (Huber & Weis, Cell, 2001)
  • Integrin αVβ3 with an RGD ligand — a matrix receptor holding its code — RCSB PDB 1L5G · coordinates CC0 · Human integrin αVβ3 extracellular segment with a cyclic Arg-Gly-Asp peptide, X-ray crystal structure (Xiong et al., Science, 2002)
  • Fibronectin type III repeats 7 to 10 — the cell-binding region of fibronectin — RCSB PDB 1FNF · coordinates CC0 · Human fibronectin type III repeats 7-10, X-ray crystal structure (Leahy, Aukhil & Erickson, Cell, 1996)
  • A collagen-like peptide — the triple helix — RCSB PDB 1CAG · coordinates CC0 · Synthetic collagen-like peptide, X-ray crystal structure at 1.9 Å (Bella et al., Science, 1994)

Simulations

  • Reading an adhesion assay · Original interactive — Biochemistrypedia (Deppmann Lab).

Scientist profiles & quotes

  • Masatoshi Takeichi — Takeichi, J. Cell Biol. 75:464-474 (1977); Nose, Nagafuchi & Takeichi, Cell 54:993-1001 (1988).
  • Alan Horwitz — Horwitz, Duggan, Greggs, Decker & Buck, J. Cell Biol. 101:2134-2144 (1985).
  • Erkki Ruoslahti — Pierschbacher & Ruoslahti, Nature 309:30-33 (1984).
  • Dennis Discher — Engler, Sen, Sweeney & Discher, Cell 126:677-689 (2006).

Quotes are verbatim from the cited source. Profile text is original to this site and was checked against the sources cited.

Animated explainer

Two Grips, Each Wired to a Cable · narration: an AI narrator voice (Gemini, Charon) reading narration written from this lesson's text · animation: AI-generated in code (Manim) by Claude, a schematic with no drawn molecular structures, plots computed from their equations · captions from the narration script.

Images & figures

This lesson carries no raster figures. Every structure is rendered live from RCSB PDB or PubChem, and every plot is drawn from its equation in the browser.

integrity Machines check the mechanics. The build confirms that every local file exists, that every question's answer key points at a real choice, and that any AI-generated scientific figure names the person who checked it; a weekly job re-resolves every accession ID against its database. Questions are original; MCAT-style items are written for this site and are not real AAMC questions. Banned sources (the Tymoczko textbook, publisher decks, copyrighted figures) are not used.

Like I'm five

Cells in your body hold hands with their neighbours, and they also hold on to a kind of net and jelly that fills the spaces between them. Inside each cell, ropes run from those handholds so a pull on the outside is felt inside. Some handholds even grip harder when they are pulled a little.
High school

Tissues stay together because cells stick to each other and to the extracellular matrix, the material outside cells. Cell-to-cell sticking mostly uses proteins called cadherins, which need calcium and bind the same kind of cadherin on the next cell. Cell-to-matrix sticking uses integrins, which bind matrix proteins such as fibronectin and laminin. Both connect inside the cell to the cytoskeleton. The matrix itself has parts with jobs: collagen cables resist pulling, elastic fibres stretch and recoil, and water-holding gels resist squeezing.
College

Classical cadherins mediate calcium-dependent, homophilic adhesion; inside the cell β-catenin binds the cadherin tail and recruits αE-catenin to couple the junction to actin, while p120-catenin binds the juxtamembrane region. Integrins are α-β heterodimers (24 in vertebrates, from 18 α and 8 β subunits) that bind matrix ligands such as fibronectin's RGD sequence and connect through talin and vinculin to actin at focal adhesions; the hemidesmosome integrin α6β4 instead binds laminin-332 and links through plectin to keratin. The matrix combines fibrillar collagens (tensile strength), elastic fibres, fibronectin networks, basement membranes of laminin and type IV collagen, and proteoglycans and hyaluronan that form hydrated gels resisting compression and that store growth factors. Antibody blocking shows a molecule is needed for attachment; purified binding is needed to show it is the receptor.
Graduate

Mechanical regulation is supported at single-molecule scale but its necessity in cells is contested. Stretching a purified talin rod fragment (residues 482-889) at 12 pN increased observed vinculin-head binding events, and magnetic-tweezer work on the mouse talin1 rod showed its 13 helical bundles unfold and refold at forces of a few piconewtons, which a parameterised model reads as force buffering. Reconstituted zebrafish αE-catenin and β-catenin bound to F-actin showed force-lifetime data better fit by a two-state catch-bond model, with a long-lived subpopulation lasting up to 25 s at 5-10 pN. Yet in NIH3T3 cells, relieving talin or vinculin autoinhibition was enough for stable association at mitochondria with myosin II inhibited by 50 µM blebbistatin. The open question is how much of adhesion maturation in a tissue depends on force exposing cryptic sites, and how much on biochemical activation that force merely accompanies.
Four amino acids for a whole protein

Fibronectin is a large protein, yet Michael Pierschbacher and Erkki Ruoslahti reported that its ability to bind cells could be accounted for by a tetrapeptide, arginine-glycine- aspartate-serine. They also noted that the sequence was present in at least five other proteins and suggested it might be a recognition signal shared among them. Within a year the Horwitz group was using synthetic peptides carrying fibronectin's cell-binding sequence to block a purified receptor from binding both fibronectin and laminin.

— Pierschbacher & Ruoslahti, Nature 309:30-33 (1984).
Meeting at the mitochondrion

To ask whether talin and vinculin need force to bind, Paul Atherton, Christoph Ballestrem and colleagues moved the question away from focal adhesions. In mouse NIH3T3 fibroblasts they anchored one protein to the outer surface of mitochondria and watched whether the other was recruited there. Wild-type full-length proteins ignored each other. An activated vinculin recruited talin, even with myosin II blocked by 50 µM blebbistatin, and photoactivation showed that less than 20% of the bound talin was mobile.

— Atherton et al., J. Cell Biol. 219:e201903134 (2020).
A bond that lasts longer when pulled

Craig Buckley, Jiongyi Tan, William Weis, Alexander Dunn and colleagues rebuilt the cadherin-catenin complex from purified zebrafish αE-catenin and β-catenin and held it against an actin filament in an optical trap. Some bonds lasted up to 25 seconds at forces of 5 to 10 piconewtons, and a two-state catch-bond model fit the force-lifetime data better than the alternatives they tested. The longest events were a subpopulation, not the typical bond.

— Buckley et al., Science 346:1254211 (2014).

 ↑ Top