The roles of distinct Ca2+ signaling mediated by Piezo and inositol triphosphate receptor (IP3R) in the remodeling of E-cadherin during cell dissemination

Introduction

Intracellular Ca²⁺ signals regulate a vast array of cellular behaviours, including those associated with cell motility and cancer^1,2. Cytosolic Ca²⁺ concentration is controlled by a number of transport proteins which facilitate the movement of Ca²⁺ across membranes in response to extracellular stimuli. These transport proteins include intracellular Ca²⁺-release channels such as the ubiquitously-expressed inositol 1,4,5-trisphosphate receptor (IP₃R) present on the membrane of the endoplasmic reticulum (ER, the major intracellular Ca²⁺ store), and plasma membrane Ca²⁺-permeable channels such as the mechanosensitive Piezo. Importantly, tight control over the spatial and temporal dynamics of Ca²⁺ signals contributes to the selective activation of Ca²⁺-dependent proteins, and so to the specificity of downstream cellular processes³.

The formation of cell-cell contacts is one process thought to be regulated by intracellular Ca²⁺ signals, as well as being dependent on extracellular Ca²⁺. In epithelial tissue, cell-cell contacts are maintained by the transmembrane adhesion protein E-cadherin, which forms adherens junctions. Extracellularly, E-cadherin binds the E-cadherin of neighbouring cells, and intracellularly, binds a complex of proteins of the catenin family which are anchored to the actin cytoskeleton⁴. Loss of E-cadherin and disintegration of adherens junctions are associated with epithelial-to-mesenchymal transition (EMT), which is a hallmark of metastatic cancer⁵. However, there is a growing body of evidence to suggest that E-cadherin expression is actually required for invasiveness and metastasis⁶, possibly by promoting formation of invasive membrane protrusions⁷.

Recent work from the Kwon lab describes a neat model for cell invasion using Drosophila intestinal epithelial cells which are made to express an oncogenic mutant version of Ras, Ras^V12. These Ras^V12-expressing cells are tumorigenic and, most notably, disseminate from the epithelium by breaching the basal extracellular matrix (ECM) and visceral muscle of the Drosophila midgut. During dissemination, cells form actin- and cortactin-rich protrusions which enable invasion and subsequent migration into the circulatory system⁸. In this pre-print, Cabrera et al. examine the roles of E-cadherin and Ca²⁺ signals mediated by IP₃R and Piezo in epithelial cell dissemination.

 

Key findings

E-cadherin assembly at invasive protrusions is required for cells to invade
Firstly, the authors show that E-cadherin expression is necessary for Ras^V12-expressing cells to disseminate, as cells fail to disseminate when E-cadherin expression is knocked down by RNA interference (RNAi). The authors propose that the subcellular localisation of E-cadherin is remodelled during dissemination. They show, using immunostaining of β-catenin as a proxy, that E-cadherin loses its lateral localisation at adherens junctions and is instead found forming puncta in the basal region of the cell. They go on to show that basal β-catenin/E-cadherin puncta co-localise with actin puncta at invasive protrusions, and although E-cadherin expression is not required for the formation of these protrusions, it is required for the degradation of laminin in the ECM and penetration of the visceral muscle layer during invasion.

Ca²⁺ signalling by IP₃R is needed to disassemble E-cadherin from adherens junctions
The authors propose that Ca²⁺ signals are responsible for E-cadherin remodelling, and attempt to disentangle the Ca²⁺ signalling pathways responsible for disassembly of adherens junctions and re-assembly at invasive protrusions. Ca²⁺ release from the ER via IP₃Rs is triggered by the binding of IP₃, a second messenger that is generated by phospholipase C (PLC) following stimulation of cell surface receptors. These Ca²⁺ signals often activate Ca²⁺/calmodulin-dependent protein kinase (CaMK), which phosphorylates downstream proteins. The authors find that knock-down of PLC, IP₃R and CaMK reduces the number of cells escaping from the epithelium and prevents loss of β-catenin/E-cadherin at lateral adherens junctions, implicating this pathway in E-cadherin disassembly.

Ca²⁺ signalling by Piezo is needed for E-cadherin assembly
Piezo has previously been shown to be required for the function of invasive protrusions and migration by membrane blebbing in Ras^V12-expressing cells, possibly in response to changing mechanical cues in the local microenvironment. This may activate the Ca²⁺-dependent protease calpain, which is thought to upregulate matrix-metalloprotease 1 and thereby degrade the ECM⁸.  Here, the authors find that knock-down of both Piezo and calpain results in a cytosolic localisation of β-catenin/E-cadherin, suggesting that E-cadherin has been lost from adherens junctions but fails to re-assemble basally.

Finally, the authors use pharmacological means to manipulate intracellular Ca²⁺. They use ex vivo preparations of Drosophila midguts at an earlier stage of Ras^V12 expression before cells undergo dissemination. Generating a global cytosolic Ca²⁺ signal by treatment with thapsigargin appears to induce loss of β-catenin/E-cadherin from the lateral region of the Ras^V12-expressing cells, but blocking Ca²⁺ entry via Piezo with GdCl₃ does not alter the localisation of already-assembled E-cadherin. When Piezo is blocked after thapsigargin-induced disassembly, there is less β-catenin/E-cadherin present basally, suggesting that Piezo is needed for E-cadherin assembly at basal regions of the cell. Interestingly, the authors note that knock-down or GdCl₃-inhibition of Piezo also prevents the formation of adherens junctions at an earlier stage of Ras^V12 expression, suggesting a wider role for Piezo-mediated Ca²⁺ signals in E-cadherin assembly.

Overall, the work points to two distinct Ca²⁺ signalling pathways in E-cadherin remodelling during epithelial cell invasion: IP₃R-mediated signals in E-cadherin disassembly from lateral adherens junctions, and Piezo-mediated signals in assembly at basal invasive protrusions (Figure 1).

 

Figure 1. The model proposed by Cabrera et al., where the PLC>IP₃R>CaMK pathway mediates E-cadherin disassembly from adherens junctions, and the Piezo>calpain pathway mediates assembly of E-cadherin at invasive protrusions in Ras^V12-expressing epithelial cells. Reproduced from Figure S5 from Cabrera et al.

 

 

Why I liked this pre-print

This pre-print caught my attention as it highlights the interplay between Ca²⁺ signals from different sources influencing distinct but related cellular events, especially in circumstances where Ca²⁺ signalling may be dysregulated, such as in cancer. On that note, I liked that the in vivo model for cell invasion takes into account the extracellular context, especially mechanical cues that the cells may be exposed to, such as changes to the ECM. This model seems great for investigating less well-understood aspects of EMT and metastasis.

 

Questions for the authors

1. Thapsigargin treatment is used to elicit a global elevation in cytosolic Ca²⁺ concentration by blocking the sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) pump, preventing re-filling of the ER and causing a leak of Ca²⁺ into the cytosol. However, depletion of the ER triggers the activity of STIM which interacts with Orai channels in the plasma membrane to induce Ca²⁺ influx from the extracellular medium by store-operated Ca²⁺ entry (SOCE). SOCE often follows IP₃R-mediated Ca²⁺ signals after store depletion, and so the two are intimately linked. In terms of what is mediating E-cadherin disassembly, how could Ca²⁺ release from the ER be distinguished from Ca²⁺ influx across the plasma membrane via SOCE?
2. In human cells, inhibition of SERCA by thapsigargin is close to irreversible. Is this also the case for SERCA in Drosophila? If so, one might expect that the 30-minute “recovery” from thapsigargin after washout does not lead to ER re-filling, but Ca²⁺ may still be extruded from the cytosol by various mechanisms such as the plasma membrane Ca²⁺ ATPase (PMCA). How might this impact on the conclusions? Could these experiments be performed with a more easily-reversible SERCA inhibitor, such as cyclopiazonic acid?
3. What could be the mechanisms by which CaMK controls disassembly of E-cadherin?
4. What changes to the extracellular environment might cells experience that trigger activity of Piezo?
5. Ca²⁺ signals are often separated in time and space – in subcellular Ca²⁺ microdomains, for example. This is often the way in which different Ca²⁺-dependent processes can be regulated. Could this be the case for E-cadherin remodelling? For example, is it possible that IP₃Rs located near adherens junctions might preferentially deliver local Ca²⁺ signals there?

 

 

References

1            Monteith, G. R., Prevarskaya, N. & Roberts-Thomson, S. J. The calcium–cancer signalling nexus. Nature Reviews Cancer 17, 373-380, doi:10.1038/nrc.2017.18 (2017).

2            Tsai, F.-C., Kuo, G.-H., Chang, S.-W. & Tsai, P.-J. Ca²⁺ Signaling in Cytoskeletal Reorganization, Cell Migration, and Cancer Metastasis. BioMed Research International 2015, 409245, doi:10.1155/2015/409245 (2015).

3            Berridge, M. J., Lipp, P. & Bootman, M. D. The versatility and universality of calcium signalling. Nature reviews. Molecular cell biology 1, 11-21, doi:10.1038/35036035 (2000).

4            Hartsock, A. & Nelson, W. J. Adherens and tight junctions: structure, function and connections to the actin cytoskeleton. Biochimica et biophysica acta 1778, 660-669, doi:10.1016/j.bbamem.2007.07.012 (2008).

5            Iamshanova, O., Fiorio Pla, A. & Prevarskaya, N. Molecular mechanisms of tumour invasion: regulation by calcium signals. The Journal of physiology 595, 3063-3075, doi:10.1113/jp272844 (2017).

6            Padmanaban, V. et al. E-cadherin is required for metastasis in multiple models of breast cancer. Nature 573, 439-444, doi:10.1038/s41586-019-1526-3 (2019).

7            Dobric, A. et al. E-cadherin: Unexpected actor of invadopodia formation in pancreatic cancer. bioRxiv, 2020.2010.2009.332783, doi:10.1101/2020.10.09.332783 (2020).

8            Lee, J., Cabrera, A. J. H., Nguyen, C. M. T. & Kwon, Y. V. Dissemination of Ras^V12-transformed cells requires the mechanosensitive channel Piezo. Nature Communications 11, 3568, doi:10.1038/s41467-020-17341-y (2020).

Tags: adherens junctions, calcium signalling, cancer, disease model, dissemination, drosophila, e-cadherin, epithelial, fly, invadopodia, invasion, ip3r, metastasis, piezo

doi: https://doi.org/10.1242/prelights.31421

Read preprint (No Ratings Yet)