“Forcing” new interpretations of molecular tension sensor studies

Molecular tension sensors are central tools for mechanobiology studies but have limitations in interpretation. Reporting in Cell Reports Methods, Shoyer et al. discover that fluorescent protein photoswitching in concert with sensor extension may expand the use and interpretation of common force-sensing tools. Molecular tension sensors are central tools for mechanobiology studies but have limitations in interpretation. Reporting in Cell Reports Methods, Shoyer et al. discover that fluorescent protein photoswitching in concert with sensor extension may expand the use and interpretation of common force-sensing tools. Mechanical forces have emerged as critical stimuli in a surprising diversity of key cellular processes including cell migration, T cell activation, and stem cell differentiation.1Zuela-Sopilniak N. Lammerding J. Can't handle the stress? Mechanobiology and disease.Trends Mol. Med. 2022; 28: 710-725Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar As such, there has been significant interest in decoding the pathways responsible for converting mechanical stimuli into cellular responses (Figure 1). It is increasingly appreciated that this mechanical correspondence is a stepwise process initiated by the application of force to a specific protein (mechanotransmission) that, in response, undergoes conformational changes that result in modifications to its structure, dynamics, and function as a whole (mechanosensing). These changes in conformation are then biochemically interpreted (mechanotransduction), often through the engagement or assembly of transducer proteins, that in concert drive downstream alterations to gene expression (mechanoresponse). Recently, a wide variety of technologies—collectively known as molecular tension sensors (MTSs)2Huang Y. Chen T. Chen X. Chen X. Zhang J. Liu S. Lu M. Chen C. Ding X. Yang C. et al.Decoding Biomechanical Cues Based on DNA Sensors.Small. 2024; 20e2310330Crossref Scopus (0) Google Scholar,3Fischer L.S. Rangarajan S. Sadhanasatish T. Grashoff C. Molecular Force Measurement with Tension Sensors.Annu. Rev. Biophys. 2021; 50: 595-616Crossref PubMed Scopus (33) Google Scholar—have been developed to better characterize and quantify these forces in cultured cells. These technologies range from surface-immobilized sensors capable of digitally quantifying overall force output to genetically encoded sensors integrated into force-related proteins of interest. One of the first genetically encoded MTSs was developed in 2010 by Grashoff et al.4Grashoff C. Hoffman B.D. Brenner M.D. Zhou R. Parsons M. Yang M.T. McLean M.A. Sligar S.G. Chen C.S. Ha T. Schwartz M.A. Measuring mechanical tension across vinculin reveals regulation of focal adhesion dynamics.Nature. 2010; 466: 263-266Crossref PubMed Scopus (1151) Google Scholar and enabled the detection of piconewton forces communicated by individual proteins within cultured cells. This groundbreaking technology has essentially acted as the blueprint for its successors, which are most often composed of a well-characterized force-responsive spring-like element and some kind of reporter element capable of communicating spring extension and thus force transduction. The most common readouts of MTSs are changes in fluorescence that occur as force is applied and the spring element is extended, though other readouts such as bioluminescence resonance energy transfer (BRET)5Aird E.J. Tompkins K.J. Ramirez M.P. Gordon W.R. Enhanced Molecular Tension Sensor Based on Bioluminescence Resonance Energy Transfer (BRET).ACS Sens. 2020; 5: 34-39https://doi.org/10.1021/acssensors.9b00796Crossref PubMed Scopus (24) Google Scholar have also been reported. This can be achieved through the observation of changes in Förster resonance energy transfer (FRET) between donor and acceptor fluorescent proteins (FPs) that flank the force-responsive element. When the spring is stretched, the FPs are physically separated from one another, which yields a change in their FRET ratio proportional to the distance between them that can be used to inform the force requirements of a conformational change. Importantly, a majority of the studies that employ these types of genetically encoded tension sensors are only observing the very first step of force sensing—mechanotransmission—because the spring element extends instantaneously in response to force, leading to an analog signal that lacks information concerning the duration of the force or its loading rate, limiting our understanding of the conformational responses that the sensor-embedded protein is undergoing. There thus remains an untapped source of information that would be critical for understanding how an initial force sensing event across a ligand-receptor pair is sustained and how this allows for the clustering and cascading of conformational changes that mediate the formation of biologically important focal adhesions or immune synapses. In the current issue of Cell Reports Methods, Shoyer et al.6Shoyer T.C. Collins K.L. Ham T.R. Blanchard A.T. Malavade J.N. Johns B.A. West J.L. Hoffman B.D. Detection of fluorescent protein mechanical switching in cellulo.Cell Reports Methods. 2024; 4100815Google Scholar present an elegant approach that aims to elucidate the rich information that escapes traditional MTSs while also establishing a framework for the standardization of this information to enhance reproducibility and enable accurate comparisons. As described above, the majority of genetically encoded tension sensors report changes in force application through the FRET signal of FPs separated by a spring-like force-responsive element. However, it was recently discovered that the application of force to FPs can alter its conformation,7Ganim Z. Rief M. Mechanically switching single-molecule fluorescence of GFP by unfolding and refolding.Proc. Natl. Acad. Sci. USA. 2017; 114: 11052-11056Crossref PubMed Scopus (42) Google Scholar leading to a decrease in fluorescent signal and potentially confounding interpretation of the results. With that in mind, Shoyer et al.6Shoyer T.C. Collins K.L. Ham T.R. Blanchard A.T. Malavade J.N. Johns B.A. West J.L. Hoffman B.D. Detection of fluorescent protein mechanical switching in cellulo.Cell Reports Methods. 2024; 4100815Google Scholar developed models to predict the precise characteristics—such as the duration or loading rate of the applied force—required for a force to mediate this mechanical switching in FPs. These models were then applied to simulate what FRET-based MTS data may look like when the FPs responsible for FRET signals are undergoing mechanical switching. These models simulated the FRET efficiency (E, the distance between the FPs on the termini of the molecular spring) and the stochasticity (S, the ratio of each FP detected relative to the theoretical amount) for MTSs and distinct trends emerged depending on if an, or which, FP was experiencing mechanical switching when E versus S was plotted. To test these simulations, Shoyer et al.6Shoyer T.C. Collins K.L. Ham T.R. Blanchard A.T. Malavade J.N. Johns B.A. West J.L. Hoffman B.D. Detection of fluorescent protein mechanical switching in cellulo.Cell Reports Methods. 2024; 4100815Google Scholar developed a genetically encoded MTS composed of a force-responsive element (a "spring" composed of spider silk protein) flanked by two FRET-capable FPs, mTFP1 and mVenus, designed to bind actin filaments to evaluate their forces. From this, both E and S were measured and the data matched (albeit with slight variability attributed to cellular complexity) the simulated results for mechanical switching of the acceptor (mVenus) at lower forces and the donor (mTFP-1) at higher forces. This difference in mechanical stability between mTFP1 and mVenus (mTFP-1 > m-Venus) was also seen in molecular dynamic simulations. This switching was most abundant within the lamellar actin network near the periphery of the cell, suggesting that this region is experiencing the greatest force (in either magnitude, duration, or other metric) between actin filaments. To further ensure this finding was reflecting a force measurement, the cellular actin network was disrupted with Latrunculin A, which resulted in a loss of mechanical switching signatures. This overall approach presents a significant feat, as it comprises the first readout of in cellulo mechanosensing that does not require secondary protein binding8Zhong B.L. Vachharajani V.T. Dunn A.R. Facile detection of mechanical forces across proteins in cells with STReTCh.Cell Rep. Methods. 2022; 2100278Google Scholar and demonstrates that both mVenus and mTFP1 are compatible with in cellulo mechanical switching. The approach by Shoyer et al.6Shoyer T.C. Collins K.L. Ham T.R. Blanchard A.T. Malavade J.N. Johns B.A. West J.L. Hoffman B.D. Detection of fluorescent protein mechanical switching in cellulo.Cell Reports Methods. 2024; 4100815Google Scholar suggests an added benefit as well, which is that existing FRET MTS data can be reanalyzed to reveal mechanical switching. To highlight this, previous data in which an MTS inserted into vinculin were re-assessed. The results indicated that mechanical switching was robust in regions previously identified as transmitting higher forces and previously identified focal adhesion gradients9LaCroix A.S. Lynch A.D. Berginski M.E. Hoffman B.D. Tunable molecular tension sensors reveal extension-based control of vinculin loading.Elife. 2018; 7e33927https://doi.org/10.7554/eLife.33927Crossref PubMed Scopus (63) Google Scholar corresponded to gradients of mechanical switching. Intriguingly, when vinculin point mutants that are known to prevent catch bond formation (bonds that strengthen when force is applied) were analyzed, mechanical switching was either decreased or completely lost depending on the mutant. This suggests that catch bonds within vinculin are necessary for mechanosensing as they regulate force duration and magnitude. It should be noted that in the currently proposed framework, the magnitude, duration, and loading rate of the force cannot be disentangled from each other. However, future work cataloging photoswitching propensity of a broader range of FPs and luciferases may identify pairs that decouple these parameters. Nevertheless, much like the initial demonstration of MTSs, a major milestone toward understanding the steps in cellular force sensing downstream of mechanotransmission has been achieved. A second yet equally important take away from this work is a cautionary tale of how mechanical switching can skew FRET-based MTS results. Factors such as which FPs are used and their orientation can alter the reported FRET efficiency and may be a reason for incongruency of different measurements for the same protein. In summary, the work from Shoyer et al.6Shoyer T.C. Collins K.L. Ham T.R. Blanchard A.T. Malavade J.N. Johns B.A. West J.L. Hoffman B.D. Detection of fluorescent protein mechanical switching in cellulo.Cell Reports Methods. 2024; 4100815Google Scholar demonstrates a significant step toward a deeper understanding of protein mechanosensitivity within cells through the exploitation of FP mechanical switching. The role of forces underlying biology and disease is expanding and tools to elevate understanding of mechanical signals to the level of biochemical signals will "force" mechanobiology as a whole forward. The authors declare no competing interests. Detection of fluorescent protein mechanical switching in celluloShoyer et al.Cell Reports MethodsJuly 9, 2024In BriefShoyer et al. develop an approach to probe how mechanical forces affect the function of fluorescent proteins inside living cells and leverage it to study mechanosensitive molecular processes. Fluorescent protein mechanical switching is manipulated by perturbations to force-activate bonds and substrate stiffness, consistent with its sensitivity to protein loading dynamics. Full-Text PDF Open Access