Cells constantly interact with each other and with the surrounding extracellular matrix through physical forces such as tension, pressure, torque, and shear stress. Over the past 50 years, biologists have increasingly come to recognize the important role biomechanics plays in the function of cellular activities such as gene expression and signaling. (See “May the Force Be with You,” The Scientist, February 2017.)
One key tool for studying how physical forces affect cells is the micropipette aspirator, a tiny glass pipette that applies pressure on a section of the cell membrane; testing gene expression, protein levels, or other factors can point to the effect of such forces on the cell. Researchers also use atomic force microscopy, which senses how cells respond to tiny mechanical pokes; apply fluid shear stresses to perturb membrane mechanosensors; and stick cells onto flexible polymers to investigate how their cytoskeletons are disrupted and repaired when the material is stretched and flexed.
However, these methods can only yield insights by looking at the whole cell at once, and not at smaller entities such as organelles and structural specializations. As the field delves into the mechanobiology of the nucleus, and investigates interactions between cells and between proteins and cell membranes, a new set of techniques has emerged. One widely used approach, three-dimensional microfluidics, segregates subcellular structures such as axons and dendrites into different microfluidic compartments to determine exactly where and how external forces affect cellular biology. Another emerging method deploys magnetic nanoparticles onto the cell to exert forces with better spatial and temporal control than conventional tools such as the micropipette aspirator.
These newer approaches have yielded surprising insights into intracellular processes, from how the cell deforms to how external force affects cell signaling and induce cell migration. For example, scientists have found that the stiffness of extracellular matrix can influence stem cell differentiation, that stretching chromatin can upregulate transcription, and that cytotoxic T cells use mechanical forces to recognize pathogens in order to eliminate them from the body.
Here, The Scientist reports on recently developed methods—from upgraded versions of conventional tools to newer micro- and nanotechnologies—in the proliferating tool chest of cellular mechanobiology research.
Problem: Micropipette aspiration of a section of an intact cell is a useful technique for measuring mechanical properties such as the cell’s stiffness, but setting up the experiment is time-consuming, and throughput is extremely low (~10 minutes/cell).
Solution: Liu and colleagues fabricated a microfluidic device consisting of a group of micropipettes that can perturb up to 128 cells simultaneously. They attached the micropipettes to a simple, cheap, calibrated pump that generates fluid pressure to exert forces on the cells through the micropipettes. The team then used two computational models to describe how the deformations produced by the device reflect mechanical properties of the cells. The researchers showed that, consistent with previous findings, the stiffness of cultured metastatic breast cancer cells was lower than that of normal breast epithelial cells (Lab on a Chip, 15:264-73, 2015).
- The device is much cheaper (~$3,000) than a conventional micropipette aspirator machine (>$10,000) as it does not require expensive electronics such as a piezo motor, an electrical motor that senses deflections and is used for force calibration.
- It offers higher throughput than a conventional micropipette aspirator.
- Without the electronics of the traditional micropipette, the tool cannot exert small-magnitude forces of under a few hundred nano-Newtons and has limited sensitivity.
- The micropipettes are grouped together and cannot be controlled individually. Channels can get clogged by debris, which can affect the measurements and system throughput.
- The device cannot be applied to mechanoreceptors on cellular extensions such as neurites of neurons, because the cells, which are detached during mechanical phenotyping, lose their extensions.
Future plans: Liu plans to develop a device to monitor mechanical cues in addition to pressure, to investigate how “multiple physical cues can orchestrate convergent mechanotransduction,” he says.
INDUCING NUCLEAR DEFORMATION
Researcher: Jan Lammerding, Associate Professor of Biomedical Engineering, Cornell University
Problem: Changes to the biophysical properties of the nucleus, such as loss of nuclear envelope integrity and nuclear pore selectivity, are linked to aging and diseases such as cancer. However, it is difficult to reproduce these changes in the lab in order to understand how they regulate DNA stability.
Solution: To investigate nuclear deformation in cancer cells, Lammerding’s group fabricated a microchannel device with a series of posts that narrow the cross-section of the channel as the cells travel along. By applying a chemotactic gradient, the researchers could induce cancer cells to migrate through pores with cross-sections that ranged from 5µm2 to more than 20 µm2 in size. They tagged nuclear envelope proteins with green and red fluorescent markers and imaged the rupture of the nuclear envelope to determine how constriction affected the nucleus’s contents. Cancer cells pass through tight spaces in the extracellular matrix and between endothelial cells lining capillaries during metastasis, so Lammerding’s group used the tool to show that the DNA of cancer cells breaks as cells migrate through constrictions; the smaller the constrictions, the higher the probability of damage (Science, 352:353-58, 2016).
- Lammerding’s platform is higher-throughput than existing methods, such as the nuclear patch clamp technique, which can only deform the nuclear membrane of one nucleus at a time.
- Microchannels are easy to fabricate.
- Pore sizes used for the microchannels (a few µm in diameter) do not correspond to actual capillary pore sizes (tens of nm), so the tool’s physiological relevance is unclear.
- The platform probes nuclear deformation, but doesn’t control for the effects of cell membrane deformation—a known biomarker of disease—which also occurs upon constriction.
- The technology does not account for the fact that different cells have different nucleus to cytoplasm volumetric ratios, which can affect nuclear envelope deformability.
Future: Lammerding plans to further improve the tool’s throughput to allow assessment of DNA damage in hundreds of cells per run. His lab is also developing software to automate the analysis of the video data it gathers.
Problem: Cells respond to mechanical signals, but researchers lack effective tools to investigate how force magnitudes regulate the activities of specific mechanosensitive receptors on the cell membrane.
Solution: The team synthesized magnetic nanoparticles coated with antibodies to target specific mechanosensitive membrane proteins. Using magnetic fields, they then applied forces on the mechanosensitive proteins via the nanoparticles bound to them. By varying the magnetic field strength, different forces could be applied to the mechanosensitive targets. With their setup, the team found that cells can sense different magnitudes of forces, influencing actin cytoskeleton assembly (Cell, 165:1507-18, 2016).
- The nanoparticles are small enough (~50 nm) not to cluster on the cell surface or immobilize mechanosensitive receptors in the absence of external magnetic fields, avoiding these potential confounding factors.
- The shell of the ferrite nanoparticles used here allows clearer optical imaging than the microparticles used in most previous techniques.
- The nanoparticles that Gartner and Jun designed cannot sustain as strong a force as can most previously used microparticles, so they don’t engage mechanosensitive channels that are activated by forces beyond their limit of detection.
- This technology targets one cell at a time, limiting throughput and prohibiting its use in studying mechanotransduction in cellular networks, such as neuronal populations.
- Temporal resolution is limited by the use of alternating magnetic fields which operate on the millisecond time scale; probing faster (microsecond- to millisecond-scale) events such as synaptic transmission would require better time resolution.
Future plans: Jun hopes to apply the technology to other mechanosensitive proteins such as ion channels. Duke University neurobiologist Jörg Grandl, who developed a similar tool but was not involved in this study, hopes to see improvements in methods for adding labeled antibodies to the nanoparticles to better target specific mechanoreceptors and to control force amplitude, space, and time. He envisions using this technology to “mechanically probe a protein, domain by domain,” while analyzing the effects on protein properties and functions.
Problem: Mechanical forces in cellular microenvironments and their spatiotemporal variations are known to affect cellular behaviors such as migration, but there is no way to make direct in vivo and in situ measurements of such forces in tissues and organs.
Solution: The team created biocompatible microdroplets of ferrofluid oil—a suspension of magnetic iron nanoparticles—and injected single microdroplets into cells of early zebrafish embryos. To exert local and controlled forces within the cells, they then applied uniform magnetic fields to deform the microdroplets without exerting traction on them. Finally, they used high-resolution microscopy to observe tissue deformations in response to forces exerted by the shape change of the microdroplets, and estimated tissue mechanical properties by comparing the deformations of the tissues to a reference library generated from materials with known physical properties. Using the technique, they found that tissue stiffness in live, developing zebrafish embryos varies along the tailbud of the animals (Nat Methods, 14:181-86, 2017).
- Microdroplets of ferrofluid are biocompatible and can be used in vivo, so this tool is widely applicable.
- Injecting microdroplets into tissues is invasive, and immune responses could affect the readout.
- Magnetic fields decay rapidly across distance, so this method may not work in deep tissues or organs such as pancreas and brain.
Future plans: Campàs is using his platform to study the mechanisms of tumor formation in multicellular spheroids and hopes to understand how abnormal biomechanics can cause or promote cancer and other diseases.
Andy Tay is a bioengineering graduate student in the lab of Dino Di Carlo at UCLA, where he uses and develops tools to probe the role of mechanics in cancer metastasis and neural stimulation.
Correction: The article incorrectly stated that pore sizes in the device designed by the Lammerding lab did not reflect actual capillary pore sizes. That statement has been removed from the article. The Scientist regrets the error.