Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Feb 8;12(1):849.
doi: 10.1038/s41467-021-21033-6.

Single-molecule force spectroscopy reveals the dynamic strength of the hair-cell tip-link connection

Eric M Mulhall  1   2 , Andrew Ward  1   3 , Darren Yang  3 , Mounir A Koussa  1   2 , David P Corey  4 , Wesley P Wong  5   6   7
Affiliations

Single-molecule force spectroscopy reveals the dynamic strength of the hair-cell tip-link connection

Eric M Mulhall et al. Nat Commun. .

Abstract

The conversion of auditory and vestibular stimuli into electrical signals is initiated by force transmitted to a mechanotransduction channel through the tip link, a double stranded protein filament held together by two adhesion bonds in the middle. Although thought to form a relatively static structure, the dynamics of the tip-link connection has not been measured. Here, we biophysically characterize the strength of the tip-link connection at single-molecule resolution. We show that a single tip-link bond is more mechanically stable relative to classic cadherins, and our data indicate that the double stranded tip-link connection is stabilized by single strand rebinding facilitated by strong cis-dimerization domains. The measured lifetime of seconds suggests the tip-link is far more dynamic than previously thought. We also show how Ca2+ alters tip-link lifetime through elastic modulation and reveal the mechanical phenotype of a hereditary deafness mutation. Together, these data show how the tip link is likely to function during mechanical stimuli.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. The mechanical strength of a single tip-link bond.
a The hair-cell stereocilia bundle and mechanotransduction complex. Bundle deflection to the right increases tension on tip links, composed of CDH23 and PCDH15. Here and throughout, PCDH15 is shown in purple and CDH23 in blue. b Single-bond fusion proteins containing one EC1-2 binding domain. Antibody Fc domains (CH2, CH3) fused to the EC domains bind to create a dimer, masking non-specific binding of EC3-5. The SNAPtag binds to the DNA tether. c Single-bond fusion proteins in the dual-beam optical trap. Force was calculated as a linear function of displacement from a stiffness-calibrated optical trap. The extension was measured as the distance between the surfaces of the two microspheres. d Representative force-time and force-extension profiles for a single-bond unbinding event. The force-time profile was used to extract the force-loading rate in pN s−1 in the linear regime preceding bond rupture. The force-extension profile was fit with an extensible worm-like chain (WLC) model to extract contour length, persistence length (related to bending stiffness), and unbinding force for each unbinding trace. e Histograms of unbinding forces at different force-loading rates in 50 µM (cyan) and 2 mM (pink) Ca2+. f Most-probable unbinding forces plotted as a function of loading rate. Systematic kernel density estimation was used to determine the most likely unbinding force for each condition. Error is shown as the optimal bandwidth from a kernel density estimation. A weighted linear fit of the data with the Bell–Evans model was used to extract force-dependent unbinding kinetics, where fβ is the slope and koff0∙fβ is the x-intercept. For 50 µM Ca2+, n = 174 unbinding events, and for 2 mM Ca2+ n = 411 unbinding events. g Mean single-bond lifetime as a function of force, calculated using the intrinsic zero-force off rate koff0 and the force scale fβ from f, and compared to the average lifetime of classic cadherin bonds (dashed black line)–. The propagated errors in lifetime due to errors in fit parameters from f are shown as light bands.
Fig. 2
Fig. 2. The dynamics of the double-stranded tip-link connection.
a Schematic of dimeric, full-ectodomain proteins, which were used for both biolayer interferometry and force-spectroscopy experiments, attached to the biolayer-interferometer sensor surface. Fc domains, SNAPtag, and tandem His6 tag as in Fig. 1b. b State diagram for tip-link avidity during dissociation. B2 is the doubly bound state, B1 the singly-bound, and U is unbound. koff is the single-bond off-rate. The rebinding rate B1→B2 depends on the intrinsic on-rate kon and the effective concentration Ceff, which is determined by the volume through which the unbound ends can move and remain in proximity. The N-terminal protein structure shown in the diagram is adapted from the Protein Data Bank (PDB) structure 4AXW. c Sample biolayer interferometry traces for dimeric full-ectodomain PCDH15 and CDH23 fusion proteins. An incident white light transmitted through the fiber-optic sensor was differentially reflected from the glass sensor surface and from the bound protein layer (here CDH23); changes in the protein layer upon binding of a soluble protein (here PCDH15) produced the binding signal. (Left) Association timecourse enables calculation of Ceff kon. (Right) The dissociation timecourse of the dimer (fit in red) indicates a long lifetime relative to the expected lifetime in the absence of rebinding (calculated in blue). Here, 320 nM of both PCDH15 and EC1-2 truncated PCDH15 were used for the association phase (Supplementary Fig. 2c). d Scatter plot of tip-link connection lifetimes measured with biolayer interferometry at 2 mM Ca2+ (63 ± 16 s, mean ± SD, n = 18 independent experiments).
Fig. 3
Fig. 3. The strength of the tip-link connection under force.
a Unbinding force as a function of loading rate for full-length dimers in 2 mM Ca2+. Displayed are most-probable unbinding forces for double-bond full-length dimer proteins (black circles, n = 247 unbinding events) and single-bond unbinding forces from Fig. 1f (magenta squares, n = 411 unbinding events). Error is shown as the optimal bandwidth from a kernel density estimation. Dimer unbinding forces show enhanced strength relative to single bonds, particularly at slow loading rates. A kinetic model that incorporates rebinding was fitted to the double-bond interaction (solid black line). A no-rebinding model is shown as the blue dashed line; a rebinding model with constant Ceff is shown as the green dashed line. b Calculated mean lifetime from the model fits in a. At forces below about 20 pN, the tip link connection has a pronounced increase in lifetime relative to the single bond due to avidity. At low forces, this results from a combination of force-sharing and rebinding. At high force, unbinding becomes faster and rebinding becomes slower, so load sharing dominates, resulting in half the sensitivity to force relative to a single bond. c Histograms of unbinding forces plotted in a. Unbinding forces at each loading rate were systematically binned using the Freedman–Diaconis rule to yield N total bins of width ∆F. Bin centers were systematically chosen to yield the most likely unbinding force with the maximum number of counts. A kernel smoothing function density estimate is overlaid on the binned unbinding force data.
Fig. 4
Fig. 4. Cis-dimerization and extracellular Ca2+ alter the lifetime of the tip-link connection through elastic modulation.
a Avidity under force is enhanced by cis-dimerization. Dimeric tip-link proteins were truncated to just the first five EC domains, to remove cis-dimerization interfaces further from the N termini. Unbinding forces (blue circles, n = 159 unbinding events) were fit with the force-dependent avidity model, with fixed parameters koff = 0.5 s−1 and fβ = 13.5 pN (EC1-5/1-5). Error is shown as the optimal bandwidth from a kernel density estimation. At the rupture forces tested, the model fit was indistinguishable from a model in which there is no rebinding. a’ Truncated proteins show an enhancement in lifetime above 6 pN primarily attributable to force-sharing. a” Ceff calculated from the fit decreased rapidly with the applied force for EC1-5 dimers. b Low [Ca2+] destabilizes the tip link. Unbinding forces in 30 µM Ca2+ (blue, n = 339 unbinding events) were nearly identical to those in 2 mM Ca2+ (black), but were much lower in 10 µM Ca2+ (magenta, n = 344 unbinding events). Error is shown as the optimal bandwidth from a kernel density estimation. Model fits suggested that 10 µM Ca2+ weakened the connection both by accelerating the single-bond off-rate (koff = 2.6 ± 0.8 s−1) and by altering the mechanical properties of the protein complex (fc = 2.9 ± 1.1 pN). b’ Calculated lifetimes of the tip link in 30 µM Ca2+ at forces between 0 and 30 pN were indistinguishable from lifetimes at 2 mM Ca2+ but were severely decreased at 10 µM Ca2+. b” Calculated Ceff decreased rapidly with applied force at 10 µM Ca2+ through increased protein elasticity. c Scatter plot of full-length dimer lifetimes measured from biolayer interferometry experiments at various concentrations of Ca2+, and of the truncated EC1-5/EC1-5 dimer in 2 mM Ca2+ (mean ± SD). 2 mM: 63 ± 16 s (n = 18 independent experiments), 50 µM: 52 ± 4 s (n = 9 independent experiments), 30 µM: 64 ± 9 s (n = 6 independent experiments), 10 µM: 43 ± 13 s (n = 9 independent experiments), Truncated: 27 ± 12 s. The Student’s two-tailed unpaired t-test was used to determine statistical significance (**p = 0.0028, ***p < 0.001). d Schematic of singly-bound dimeric tip links, illustrating how an increase in elasticity produced by low extracellular Ca2+ can separate unbound ends and lower Ceff.
Fig. 5
Fig. 5. The mechanical phenotype of a hereditary deafness mutation.
a A kinetic state diagram representing a tip-link connection heterozygous for the R113G mutation (red) in PCDH15 (PCDH15 R113G+/−). The doubly bound complex (B2) can transition to a single-bound state containing a mutated PCDH15 (B1*) or to a single-bound state with the wild-type PCDH15 (B1), with rates that differ between mutant and wild-type. b The location of the human deafness mutation R113G in the mouse PCDH15–CDH23 bond interface (PDB 4AXW). c Unbinding forces from dimeric, full-ectodomain proteins measured at 2 mM Ca2+, with one (brown, n = 316 unbinding events) or both (purple, n = 264 unbinding events) PCDH15 strands mutated, simultaneously fit with the three- and four-state models. Error is shown as the optimal bandwidth from a kernel density estimation. d Calculated mean lifetimes of the wild-type and mutant tip-link connections. e Scatter plot of zero-force tip-link lifetimes measured from biolayer interferometry experiments with PCDH15 R113G+/+ proteins (mean ± SD). Wild-type lifetime = 64 ± 16 s (n = 18 independent experiments), R113G+/+ lifetime = 37 ± 18 s (n = 5 independent experiments). The Student’s two-tailed unpaired t-test was used to determine statistical significance (**p = 0.0041). f Mean CDH23–PCDH15 single-bond lifetime as a function of force, calculated from single-bond kinetics of the PCDH15R113G–CDH23 bond with parameters extracted from simultaneous fits to PCDH15 R113G+/+ and PCDH15 R113G+/− force spectroscopy data (c). Shaded bands are the propagated error of the fit parameters fβ and koff0. Relative to the wild-type tip-link bond, the R113G mutation increased both the zero-force single-bond off rate (from 0.5 to 1.4 s−1) and the force sensitivity fβ of the bond (from 13.5 to 1.8 pN). At a resting tension of 10 pN, a single wild-type bond lasts ~1 s, while a single PCDH15R113G–CDH23 bond lasts <10 ms.
Fig. 6
Fig. 6. The lifetime of the tip link is insensitive to a broad range of physiological stimuli.
a A schematic of the modified motor-based adaptation model (see “Methods” section) used to calculate mean tip-link lifetimes. ks is the pivot stiffness of the stereocilia bundle and kg is the gating spring stiffness. If a compliant probe is used to stimulate a hair bundle, xp is the probe deflection magnitude and kp is the probe stiffness (all normalized to a single tip link using a geometric factor γ = 0.12),. The force on the tip link Ft is then calculated. b Force on an individual tip link Ft as a function of time, at a 3 Hz stimulus frequency, with (blue) and without (red) “slow” adaptation (xp = 18 nm). The axis for probe deflection xp is at right. c A single Monte Carlo simulation trajectory of tip-link lifetime performed using force-dependent off-rates and concentration-dependent on-rates obtained from force spectroscopy of full-length dimers (Fig. 3), incorporating the effect of adaptation. Unbinding (red) and rebinding events (green) are marked with squares. An animation of the simulation trajectory is in Supplemental Movie 1. d Mean tip-link lifetimes, based on thousands of such simulations at frequencies between 1 and 10,000 Hz and at hair bundle deflection magnitudes of 0–1000 nm (error bars = SEM, n = 5000 simulations per data point). Without adaptation, tip-link lifetime is insensitive to both frequency and amplitude within the physiological range of hearing in humans (~20–10,000 Hz) for hair bundle deflections up to 133 nm, corresponding to a peak force of ~9 pN above resting tension on the tip link. e Mean tip-link lifetimes, with motor-based adaptation (error bars = SEM, n = 5000 simulations per data point). Slippage of the adaptation motors increases mean tip-link lifetime by reducing the force with each oscillation at low frequencies, and by reducing the resting force at high frequencies. With adaptation, tip-link lifetime is mostly insensitive to both frequency and amplitude for frequencies of 1–10,000 Hz and deflections to 500 nm.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Hudspeth AJ. How the ear’s works work. Nature. 1989;341:397–404. doi: 10.1038/341397a0. - DOI - PubMed
    1. Robles L, Ruggero MA. Mechanics of the mammalian cochlea. Physiol. Rev. 2001;81:1305–1352. doi: 10.1152/physrev.2001.81.3.1305. - DOI - PMC - PubMed
    1. Hudspeth AJ, Konishi M. Auditory neuroscience: development, transduction, and integration. Proc. Natl Acad. Sci. USA. 2000;97:11690–11691. doi: 10.1073/pnas.97.22.11690. - DOI - PMC - PubMed
    1. Pickles JO, Comis SD, Osborne MP. Cross-links between stereocilia in the guinea pig organ of Corti, and their possible relation to sensory transduction. Hear Res. 1984;15:103–112. doi: 10.1016/0378-5955(84)90041-8. - DOI - PubMed
    1. Kachar B, Parakkal M, Kurc M, Zhao Y, Gillespie PG. High-resolution structure of hair-cell tip links. Proc. Natl Acad. Sci. USA. 2000;97:13336–13341. doi: 10.1073/pnas.97.24.13336. - DOI - PMC - PubMed

Publication types

LinkOut - more resources