Tension causes structural unfolding of intracellular intermediate filaments

Background

Cell shape and structure result from a well-orchestrated balance of forces acting between the cytoskeleton, extracellular adhesions, and membrane tension. In mammalian cells, the cytoskeleton is the primary load-bearing unit, and consists of three filamentous protein networks: actin filaments, microtubules, and intermediate filaments (IF). The contributions of IFs to cell mechanics remains relatively understudied. Although molecular dynamic simulations and in vitro studies in hydrogels have shown that IF proteins undergo secondary structural changes to compensate mechanical loads, this phenomenon remains to be investigated in cells. Yet, the structural polymorphism observed in models and hydrogels led to the hypothesis that IFs could play an important role in intracellular mechanotransduction. Few experimental methods allow analyzing protein structure within cells. Methods that allow probing protein secondary structure in a non-invasive way include non-vibrational spectroscopies, such as Raman spectroscopy. In their work, Fleissner et al use in situ nonlinear Raman imaging and multivariate data analysis to quantify the intracellular secondary structure of the IF cytoskeletal protein vimentin under different states of cellular tension. Vimentin is one IF protein that forms cytoplasmic IF networks and plays a role in cell adhesion, mechanical stability, cell migration, and signaling.

Key findings

Overall findings

The authors began by testing how mechanical deformation affected the secondary structure of vimentin IFs when polymerized in vitro. Using a micromanipulator on a non-linear Raman (BCARS) microscope, they acquired BCARS spectra from relaxed and physically tensed vimentin bundles in vitro, and observed clear spectral changes between the relaxed and tensed vimentin IF bundles. The relaxed vimentin IF bundles showed a predominantly alpha-helical structure after decomposition, while tensed samples showed an increased beta-sheet structure and reduced helicity.

Specific findings

Generating D-Vim

The authors discuss that the BCARS spectra from any location inside a cell will consist of the sum of all proteins within the excitation volume, making it nearly impossible to produce vibrational fingerprints from one specific protein in the cell. For that, it is necessary to separate the contribution of the target protein (i.e. vimentin) from that of the remaining cellular background. One way to create spectral contrast is stable isotope substitution, such as hydrogen replacement with deuterium. The authors produced isotopically-substituted vimentin containing deuterated carbohydrates as the only carbon source. This allowed for uniquely identifying signals coming from D-Vim compared to the intracellular cytosolic protein pool.

Drug-treatments and substrate modifications resulting in relaxation of cell-generated tension

To investigate the effect of cell-generated tension on vimentin IFs, cells were treated with blebbistatin- a myosin II inhibitor. Blebbistatin blocks actomyosin contraction, resulting in relaxation of cell-generated tension. In treated cells, the vimentin IFs appeared less wavy and less sprawling than before treatment, but still spanned large distances within the cytosol. Substrate stiffness also regulates cellular tension. Therefore, cells were cultured on collagen-coated soft substrates to reduce cell-generated tensile forces. In this condition, HeLa cells showed an altered morphology with more perinuclear vimentin network and compact shape.

The influence of cellular tension on vimentin secondary structure was analyzed by comparing the shape of the Amide I vibration in D-Vlm RL spectra from cells grown on collagen-coated glasses, or cells in which tension was relaxed (physically or chemically). Comparison of intracellular and in vitro polymerized vimentin showed that intracellular vimentin contained more beta-sheets while the in vitro polymerized vimentin was more helical. Cells treated with blebbistatin had a more alpha-helical (native) structure, and had a narrowed Amide I region, similar to the spectrum of purified vimentin in vitro. This was also the case of cells grown in soft substrates, and of cells treated with latrunculin-A, a marine toxin that promotes depolymerization of the actin cytoskeleton. Meanwhile, the amount of unstructured vimentin was similar in all cell groups, indicating that relaxation of cell tension did not increase random coil amounts in intracellular vimentin IFs.

Correlation of phosphorylation and cell tension

The authors went on to study vimentin organization and level of vimentin phosphorylation at serine 55 (pSer55) in response to the mechanical properties of substrate. HeLa cells expressing GFP-vimentin were investigated in collagen hydrogels (soft surfaces), and stiff-collagen coated glass surfaces. Increased pSer55 was found in vimentin in cells cultured on soft subratres.

Together, authors conclude that cellular tension is sufficient to alter the secondary structure of intracellular vimentin IFs, resulting in more beta-sheet content with increasing tension.

What I like about this preprint

I find the topic interesting, and I think the methods developed, and the findings achieved here are important contributions to our understanding of cell biology and mechanobiology. I think this work opens an important window of study for various areas of research beyond cell biology alone, namely pathology, parasitology, tumour biology, and others.

Open questions 

1. In your work, you used mostly HeLa cells. To what extent do your findings respective to tension translate to other cell lines, and primary cells?
2. In your work you focused on IFs – specifically vimentin. You mention that cell mechanics depend to a large extent on IFs, actin filaments, and microtubules. Using the methods you developed here, or other combination, what is the interplay between these 3 components in cell mechanics?
3. In your work you used mostly physical and chemical alterations to study effects of tension on IFs. Is it possible to generate reporter and genetically altered cells lines to screen a large scope of specific mutants, to gain further insight on the role of IFs in cell mechanics?
4. While you defined very clearly IF alterations under specific conditions, is it possible to visualize or measure changes in a dynamic manner- resembling specific physiological processes, such as cell migration, extravasation, etc.?

References

1. Fleissner F, et al. Tension causes structural unfolding of intracellular intermediate filaments, bioRxiv, 2020.

 

doi: https://doi.org/10.1242/prelights.22736

Read preprint (No Ratings Yet)