Dissemination of RasV12-transformed cells requires the mechanosensitive channel Piezo

Lee, Jiae; Cabrera, Alejandra J. H.; Nguyen, Cecilia M. T.; Kwon, Young V.

doi:10.1038/s41467-020-17341-y

Download PDF

Article
Open access
Published: 16 July 2020

Dissemination of Ras^V12-transformed cells requires the mechanosensitive channel Piezo

Nature Communications volume 11, Article number: 3568 (2020) Cite this article

5446 Accesses
12 Citations
7 Altmetric
Metrics details

Subjects

Abstract

Dissemination of transformed cells is a key process in metastasis. Despite its importance, how transformed cells disseminate from an intact tissue and enter the circulation is poorly understood. Here, we use a fully developed tissue, Drosophila midgut, and describe the morphologically distinct steps and the cellular events occurring over the course of Ras^V12-transformed cell dissemination. Notably, Ras^V12-transformed cells formed the Actin- and Cortactin-rich invasive protrusions that were important for breaching the extracellular matrix (ECM) and visceral muscle. Furthermore, we uncovered the essential roles of the mechanosensory channel Piezo in orchestrating dissemination of Ras^V12-transformed cells. Collectively, our study establishes an in vivo model for studying how transformed cells migrate out from a complex tissue and provides unique insights into the roles of Piezo in invasive cell behavior.

Evolution of tissue-specific expression of ancestral genes across vertebrates and insects

Article 15 April 2024

Spatially organized cellular communities form the developing human heart

Article Open access 13 March 2024

Collective intelligence: A unifying concept for integrating biology across scales and substrates

Article Open access 28 March 2024

Introduction

A hallmark of malignancy, metastasis is the major contributor of mortality in cancer patients^1,2. To metastasize, cancer cells undergo a series of processes, including dissemination from the original tumor site, circulation in blood or lymph, extravasation, and then colonization at a secondary site^2,3. Given that metastasis is initiated by dissemination from the original tumor site into the circulation, targeting this initial step could be an ideal strategy for intervening metastasis. Most of our knowledge of the molecular mechanisms of cell dissemination has been acquired from studies utilizing cancer cells in culture. Nevertheless, recapitulating the intricacy of the native microenvironment using in vitro culture systems remains difficult at best. Therefore, a simple in vivo system allowing us to observe the dissemination process in a native context will be useful to achieve a better understanding on the mechanisms that transformed cells utilize for dissemination.

Simple model organisms have provided tools to investigate the mechanisms underlying cell migration and invasion in native contexts⁴. For instance, migration of the border cells during Drosophila oocyte development has illustrated how cells can migrate collectively^4,5. In addition, invasion of Caenorhabditis elegans (C. elegans) anchor cells into the vulval epithelia has provided a powerful tool to study invasive cell behavior in an in vivo microenvironment^6,7. In Drosophila, genetic manipulation of epithelial tissues can induce hyperplasia or tissue overgrowth—a so called ‘tumor’ in Drosophila^8,9,10,11. Notably, these Drosophila tumors also provide a tool to study metastatic behavior^12,13. In particular, a recent study demonstrated that adult hindgut epithelial cells expressing mutant Ras (Ras^V12) could disseminate from the hindgut and metastasize to distant tissues^14,15. Interestingly, the same study elucidated that sustained intestinal infection with pathogenic bacteria could enhance dissemination of Ras^V12-expressing hindgut cells via activation of Imd innate immune signaling¹⁴. Although these tumor models have allowed the identification of several genetic and environmental factors underlying metastatic phenotypes^{12,13,14,16,17}, it still remains obscure how these transformed cells disseminate from the primary tumor site into the hemocoel to initiate metastasis.

Here, we show that expression of Ras^V12 in intestinal stem cells (ISCs) and enteroblasts (EBs) in the adult Drosophila midgut causes them to disseminate from the posterior midgut and transmigrate into the circulation. Our cellular and molecular characterization reveals how some of the molecular mechanisms underlying the migratory and invasive phenotypes of cancer cells are assembled in vivo to form a mode of cell dissemination. Observing the cell dissemination process in a native context allows us to describe actin- and cortactin-rich invasive protrusions that are associated with degradation of the ECM and the visceral muscle (VM) layer in Drosophila and discover the mechanosensitive channel Piezo as a key player of cell dissemination in vivo.

Results

Ras ^V12 cells basally disseminate from the posterior midgut

Ras genes encode small GTPases that are frequently mutated in multiple types of cancers¹⁸. Oncogenic Ras isoforms affect multiple aspects of cancers, including the metastatic transformation of breast cancers^19,20,21. In Drosophila, ectopic expression of Ras^V12 in developing disks increases cell division; however, it is not sufficient to induce malignant transformation. Disruption of polarity in addition to Ras^V12 is required to induce malignant disc tumors with metastatic properties¹². Similarly, ectopic expression of Ras^V12 in midgut ISCs and EBs using a clonal strategy was not sufficient to induce tumors. Instead, Ras^V12-expressing cells gradually disappeared from midguts^22,23. When we expressed Ras^V12 in adult midgut ISCs and EBs using the conditional GAL4 driver esg^ts (escargot-GAL4, tubulin-GAL80^ts, UAS-GFP/+; see Methods), Ras^V12 cells propagated initially and then, progressively disappeared from the midgut. At day 6 of Ras^V12 expression, most of the Ras^V12 cells had been eliminated from the midgut (Fig. 1a). In contrast, cells expressing a gain-of-function Raf allele (Raf^gof) were retained in the midgut (Fig. 1a), indicating that general alteration in cell proliferation or cell crowding is not the cause of the phenotype.

Previously, it has been reported that the hindgut epithelial cells expressing Ras^V12 could disseminate and metastasize to distant tissues¹⁴. Similarly, we noticed that a significant number of GFP-labeled Ras^V12 cells were detected outside of the VM at day 2 of Ras^V12 expression (Fig. 1b, c). Moreover, we detected GFP-labeled cells in hemolymph prepared from flies expressing Ras^V12 with esg^ts, but not from control flies (Fig. 1d, e). To be detected outside the VM or in hemolymph, these cells must have passed through the basement membrane (BM), which resides outside of the midgut epithelium.

Previous studies have shown that metastatic transformation in Drosophila can upregulate Matrix-metalloprotease 1 (Mmp1)^14,16,24, which plays a crucial role in the degradation of the extracellular matrix (ECM). Similarly, we found that Mmp1 levels were increased by expression of Ras^V12 but not Raf^gof (Fig. 1f, g). In addition, expression of Ras^V12 with esg^ts also caused a cell non-autonomous increase in Mmp1 signals in surrounding cells. Although expression of MMPs was thought to be critical for invasive cell behavior, anchor cells in C. elegans could physically breach the basement membrane even in the absence of MMPs by employing an extensive F-actin-rich protrusion²⁵. We found that expression of Tissue inhibitor of metalloproteases (Timp) with esg^ts completely suppressed dissemination of Ras^V12 cells (Fig. 1h and Supplementary Fig. 1a), suggesting the importance of MMPs in dissemination of Ras^V12 cells.

Adult midgut is surrounded by thick ECM, which can be visualized by staining laminin: one continuous laminin layer is detected at the basal side of the epithelium (Fig. 1i and Supplementary Fig. 1c, white arrowhead) and another prominent laminin layer is located at the outside of the VM (Supplementary Fig. 1c, yellow arrowhead). This complexity in the midgut ECM structure might account for the requirement of MMPs for dissemination of Ras^V12 cells. In accordance with Mmp1 elevation, we detected deterioration of laminin structure. At day 1 of Ras^V12 expression, we frequently observed a partial degradation of the ECM, which is manifested by loss of the laminin layer adjacent to the intestinal epithelium (Supplementary Fig. 1c, white arrow). At day 2 of Ras^V12 expression, more extensive degradation affecting the entire ECM structure was observed. The inner laminin layer juxtaposed with the intestinal epithelium was almost completely absent (Fig. 1j, white arrow). Moreover, the laminin layer outside visceral muscle was also significantly compromised, leading to patchy pattern of laminin signals outside of the visceral muscle at day 2 of Ras^V12 expression (Fig. 1j, yellow arrow and Supplementary Fig. 1d). Of importance, laminin signals were almost invisible at the basal side of disseminated Ras^V12 cells (Supplementary Fig. 1c, white arrow). This global degradation of the laminin layer could be explained by the fact that Ras^V12 cells covered the entire basal side of the midgut epithelium at day 1 of Ras^V12 expression (Supplementary Fig. 2b). Therefore, a tissue-wide upregulation of Mmp1 signals was detected at day 1 of Ras^V12 expression (Fig. 1f). In addition, we speculate that cell non-autonomous MMP expression may contribute to the profound ECM degradation phenotype although the molecular mechanism underlying this cell non-autonomous Mmp1 expression requires further investigations.

We noticed that Ras^V12 cells also frequently delaminated toward the midgut lumen after 2 days of Ras^V12 expression (Supplementary Fig. 2b’). Altogether, our observations suggest that Ras^V12 cells migrate out from the posterior midgut via at least two different processes: dissemination into the hemocoel and delamination into lumen. Of note, expression of anti-apoptotic protein p35 did not impair the dissemination of Ras^V12 cells (Supplementary Fig. 2c), suggesting that cell death was not responsible for the phenotype.

Disseminating Ras ^V12 cells undergo morphological remodeling

Since the mechanism by which cells disseminate from a complex tissue is poorly described, we further investigated the phenomenon by scrutinizing the morphology of Ras^V12 cells. We found that Ras^V12 cells basally moved out from midgut epithelia at day 1 of Ras^V12 expression (Fig. 2a). They resided in-between the epithelium and the VM layer and were significantly spread out (Fig. 2a). Interestingly, a significant portion of Ras^V12 cells appeared to be bigger than normal ISCs or EBs (Supplementary Fig. 3a, b), an indication of possible misdifferentiation²⁶. Nevertheless, a large portion of Ras^V12 cells residing outside the epithelium could still divide (Fig. 2b) and express the Notch receptor ligand, Delta (Dl) (Fig. 2c and Supplementary Fig. 3c), which are the characteristics of active ISCs. At days 2 and 3 of Ras^V12 expression, Ras^V12 cells produced large protrusions across the VM while these large protrusions were not detected in control or Raf^gof cells (Fig. 2d). Furthermore, we were able to observe cells adopting various morphologies that appeared to be traversing the muscle layer (Fig. 2e).

**Fig. 2: Disseminating *Ras*^V12-transformed cells undergo extensive remodeling of cell morphology.**

Disseminating Ras ^V12 cells assemble invasive protrusions

Actin-based cellular structures play crucial roles in invasive cell behaviors. In particular, invadopodia, which are actin- and cortactin-rich protrusions associated with degradation of the ECM^27,28, are known to be essential for cancer-cell invasive phenotypes^28,29. To address whether Ras^V12 cells are forming similar invasive structures, we decided to elucidate actin cytoskeleton organization. Staining actin with phalloidin did not yield a discernable signal in ISCs and EBs because the adjacent VM expressed a large quantity of actin. Thus, we assessed Actin cytoskeleton organization by expressing actin-mRFP (UAS-Actin-mRFP) with esg^ts. Intriguingly, actin-mRFP was detected as puncta at the basal side of Ras^V12 cells that protruded toward the VM while similar puncta were undetectable in control cells (Fig. 3a, b). Of importance, these actin-rich protrusions frequently reached the outer surface by passing through the BM and VM layers (Fig. 3b). Furthermore, these protrusions were also detected as large blebs at day 2 of Ras^V12 expression (Fig. 3c).

**Fig. 3: Basal actin-rich protrusions disrupt visceral muscle integrity.**

In cancer cells, invadopodia are detected as protrusions enriched for both actin and cortactin—a regulator of actin polymerization²⁸. Since the shape of these protrusions observed in Ras^V12 cells resembles that of invadopodia²⁸, we examined cortactin localization by expressing HA-tagged cortactin (UAS-Cortactin-HA) with esg^ts. We found that cortactin-HA was also detected as protrusions and large blebs at the basal side of Ras^V12 cells (Supplementary Fig. 4b, c). In control cells, cortactin-HA was broadly detected at the cortical region and did not form prominent puncta or protrusions at the basal side (Supplementary Fig. 4a). To further confirm that cortactin-rich protrusions observed in Ras^V12 cells were not formed due to an overexpression artifact, we expressed cortactin-HA in Raf^gof cells. Nevertheless, we found that cortactin-HA overexpression alone was not sufficient to induce similar protrusions and large blebs at the basal side of Raf^gof cells (Supplementary Fig. 4e, f). Notably, these cortactin-rich protrusions found in Ras^V12 cells were also stained with phalloidin (Fig. 3d, arrowhead), indicating that these protrusions were enriched for both cortactin and actin.

Remarkably, these protrusions could even vertically penetrate the VM (Fig. 3d), resulting in damage to the tissue. In particular, we observed that the circular muscles around Ras^V12 cells became torn and thinned (Fig. 3a–c; phalloidin staining). Interestingly, we detected large ruptures in the VM layer where multiple projections are concentrated (Fig. 3e and Supplementary Fig. 4d). When the midguts were viewed from the outside, a severe damage in VM was detected at day 2 of Ras^V12 expression, which was manifested by breakage of the longitudinal muscles that normally span the whole posterior part of the midgut (Fig. 3f, g). We also noticed that a significant number of large blebs were visible outside of the posterior esg^ts > Ras^V12 midguts (Fig. 3f, arrowhead). Altogether, these observations indicate that actin- and cortactin-rich protrusions are associated with the breach of the VM layer.

In cancer cells, cortactin plays a crucial role in formation of invadopodia as well as secretion of MMPs³⁰. Knockdown of cortactin in cancer cells decreased MMP secretion and impaired ability to degrade the ECM by invadopodia³¹. Given the strong enrichment of cortactin in the protrusions detected in Ras^V12 cells, we decided to deplete cortactin to gain more insight into the role of these protrusions. Cortactin knockdown in Ras^V12 cells suppressed ECM degradation (Supplementary Fig. 5a, b) and rescued the muscle damage phenotype caused by Ras^V12 cells (Supplementary Fig. 5c). Notably, cortactin depletion caused a slight, but significant reduction in Mmp1 levels (Supplementary Fig. 5d, e). As a consequence, cell dissemination is almost completely suppressed by cortactin depletion in Ras^V12 cells (Supplementary Fig. 5f). Altogether, these results demonstrate that these actin- and cortactin-rich protrusions observed in Ras^V12 cells are linked to the breach of the ECM as well as the VM layer.

Note that we detected Lifeact-mRFP also forming puncta at the basal side of Ras^V12 cells (Supplementary Fig. 6). In general, these puncta marked with Lifeact-mRFP remained underneath the VM layer and did not grow into large protrusions, which might be explained by defect in actin caused by the overexpression of Lifeact³².

Our study describes the actin- and cortactin-rich protrusions that are associated with degradation of the ECM and the VM layer in Drosophila. The characteristics and the invasive nature of these protrusions resemble those of invadopodia observed in cancer cells²⁸. Altogether, our observations demonstrate that dissemination of Ras^V12 cells is an active process requiring cell-autonomous remodeling of cell morphology and formation of invasive protrusions at the basal side.

Ras ^V12 cells produce large blebs and extracellular vesicles

To gain further insights into how midgut epithelial cells disseminate by passing through the VM layers, we performed ex vivo live imaging of midguts. We noticed that Ras^V12 cells produced large blebs, which were also formed across the VM and eventually released as extracellular vesicles (Fig. 4a, b and Supplementary Movies 1–3). In contrast, Raf^gof cells did not produce a significant number of blebs or extracellular vesicles even though expression of Raf^gof induced comparable cell proliferation (Fig. 4b and Supplementary Movie 4). The average size of these vesicles was ~3.1 ± 1.5 µm (mean ± SD) (Fig. 4c). Thus, they are bigger than exosomes (<100 nm) and microvesicles (100–1000 nm). Since these vesicles have not been previously described in Drosophila, we refer them as extremely-large extracellular vesicles (ELEVs). In addition, GFP-positive particles bigger than 1 µm were detected in hemolymph prepared from flies expressing Ras^V12 but not from controls or flies expressing Raf^gof, indicating that ELEVs were also produced in vivo (Fig. 4d, e). Interestingly, these ELEVs are reminiscent of large extracellular vesicles, including large oncosomes and cytoplasts, which are also implicated in the invasive phenotypes of cancer cells^33,34,35.

**Fig. 4: *Ras*^V12 cells generate blebs and ELEVs and disseminate by extensive blebbing.**

Ameboid movement, which is characterized by extensive blebbing, is used by cancer cells to migrate through holes in the matrix and move without support of adhesion^36,37,38. Production of ELEVs implies that Ras^V12 cells may utilize ameboid movement. Indeed, we observed that Ras^V12 cells in the posterior midgut often extensively remodeled their cell shape. Tracing the outline of representative Ras^V12 cells in ex vivo time-lapse live-imaging videos further visualized these dynamic remodeling events (Fig. 4f–h). Notably, we observed that Ras^V12 cells could migrate out from the midgut involving extensive blebbing (Fig. 4i and Supplementary Movie 5). In addition, Ras^V12 cells could also slip out of the tissue without significant blebbing through a region where a large quantity of ELEVs had been generated (Supplementary Fig. 7 and Supplementary Movie 6). These observations suggest that Ras^V12 cells mainly use blebbing to migrate out from the midgut.

Disruption of Piezo impairs dissemination of Ras ^V12 cells

Given the breach of the BM and VM layers, Ras^V12 cells need to sense the cues associated with it and elicit appropriate cellular responses to successfully transmigrate into the hemocoel. Since degradation of the ECM and ruptures in the VM layers could substantially alter the biomechanical microenvironment, we looked into the role of the mechanosensitive cation channel Piezo in the cell dissemination process.

Disruption of Piezo in Ras^V12 cells using RNA inference (RNAi) lines (NIG#8486R-3 and VDRC#v2796)^39,40 led to a significant suppression in cell dissemination (Fig. 5a, b). It has been shown that a blocker of mechanically-activated cationic currents, gadolinium also efficiently inhibits mouse Piezo1-induced current⁴¹. We found that feeding gadolinium was also sufficient to suppress dissemination of Ras^V12 cells (Fig. 5c). This suppression could not be simply explained by a reduction in cell number since piezo knockdown did not decrease the number of ISCs nor the division of Ras^V12 cells (Supplementary Fig. 8a–c). Similarly, the division of Ras^V12 cells was not decreased by gadolinium feeding (Supplementary Fig. 8d).

**Fig. 5: The mechanosensory channel Piezo is required for dissemination of *Ras*^V12 cells.**

A recent study has demonstrated that Piezo regulates differentiation of enteroendocrine (EE) cells in Drosophila midgut⁴². However, we found that the overall EE cell number remained unchanged in esg^ts > Ras^V12 midgut, indicating that expression of Ras^V12 did not facilitate production of EEs (Supplementary Fig. 8e–g). In addition, the effect of piezo knockdown in Ras^V12 cells on EE cell population was negligible (Supplementary Fig. 8e–g). Thus, a possible change in EE cell population due to Piezo inhibition cannot account for the suppression of Ras^V12 cell dissemination.

To determine which process requires Piezo, we scrutinized the cellular phenotypes caused by piezo knockdown. Interestingly, formation of invasive protrusions was not inhibited by piezo knockdown as Lifeact puncta were still detected at the basal side of Ras^V12, piezo RNAi cells (Supplementary Fig. 9). In contrast, the muscle damage phenotype was rescued by piezo knockdown in Ras^V12 cells, and the large blebs/protrusions detected outside the posterior midguts were almost invisible in the esg^ts > Ras^V12, piezo RNAi posterior midgut (Fig. 5a, d). Furthermore, we found that piezo knockdown significantly suppressed ELEV production (Fig. 5e and Supplementary Movies 7, 8). Similarly, blebbing cells were no longer detected in the posterior esg^ts > Ras^V12, piezo RNAi midguts; the boundary of Ras^V12, piezo RNAi cells was not altered dramatically over an hour of tracing (Fig. 4f). Altogether, our characterization of the piezo knockdown phenotypes suggests that during dissemination of Ras^V12 cells, Piezo impinges on two different processes: the breach of the VM layer and the induction of blebbing.

Piezo plays two discrete roles in Ras ^V12 cell dissemination

Our observations indicate that actin- and cortactin-rich invasive protrusions were associated with the breach of the ECM and the VM layer. Given the rescue of muscle damage by piezo depletion in Ras^V12 cells, Piezo might regulate the function of these invasive protrusions. Since we found that Cortactin knockdown in Ras^V12 cells suppressed the degradation of the ECM (Supplementary Fig. 5a,b), we tested whether piezo knockdown also rescued ECM degradation. Stainings of laminin B1 showed that Piezo was required for degradation of the ECM by Ras^V12 cells (Fig. 6a). Furthermore, we found that piezo knockdown rescued the circular muscle thinning and segregation phenotype caused by Ras^V12 expression (Fig. 6b). Of importance, piezo knockdown in Ras^V12 cells greatly reduced Mmp1 levels (Fig. 6c), indicating that Mmp1 expression in Ras^V12 cells was dependent upon Piezo. Multiple studies have shown that the calcium-dependent proteases calpains are critical mediators of calcium signaling induced by Piezo channels^43,44. To elucidate the role of calpain in Mmp1 induction, we depleted Drosophila Calpain A (CalpA) and Calpain-B (CalpB) in Ras^V12 cells, respectively, by RNAi approach. We found that knockdown of either CalpA or CalpB in Ras^V12 cells significantly reduced Mmp1 levels (Supplementary Fig. 10a, b). In addition, Calpain knockdown in Ras^V12 cells rescued the muscle damage phenotype (Supplementary Fig. 10a). Of importance, dissemination of Ras^V12 cells was almost completely inhibited by depletion of either CalpA or CalpB (Supplementary Fig. 10c). Altogether, these results suggest that Piezo is a critical determinant of Mmp1 levels, and calpains might be a mediator of Piezo-induced calcium signaling in Ras^V12 cells.

**Fig. 6: Piezo is required for induction of Mmp1 expression and blebbing.**

Piezo disruption induces additional distinct phenotypes: reduction in ELEVs and impairment in blebbing. If the molecular function of Piezo during cell dissemination can be attributed solely to the control of Mmp1 expression, the cell dissemination defect caused by piezo knockdown could be rescued by promoting Mmp1 expression. Cortactin is a key regulator of invadopodia assembly and function in cancer cells, and amplification of cortactin is linked to an increase in metastasis of carcinoma^31,45. Furthermore, it has been shown that cortactin overexpression promotes invadopodia function and MMP activity in many types of cancer cells³¹. Given the conservation of the molecular function of cortactin in the control of actin assembly, we hypothesized that cortactin overexpression might rescue the cell dissemination defect caused by piezo knockdown by enhancing the invasive protrusion function. We found that cortactin overexpression in Ras^V12, piezo RNAi cells restored Mmp1 levels (Fig. 6d), the formation of large blebs/protrusions across VM (Fig. 6e), and the VM damage (Fig. 6f) to the extent caused by Ras^V12 cells. These results suggest that augmentation of cortactin levels in Ras^V12 cells was sufficient to promote Mmp1 expression and compromise the integrity of the tissue. Nevertheless, it could not rescue the defect in dissemination of Ras^V12, piezo RNAi cells (Fig. 6g), indicating that creating ruptures in the BM and VM layers was not sufficient to induce dissemination of Ras^V12, piezo RNAi cells. Of significance, cortactin overexpression failed to restore ELEV production to the levels of Ras^V12 cells even though it could slightly increase ELEV formation from Ras^V12, piezo RNAi cells (Fig. 6h). Furthermore, Ras^V12, piezo RNAi cells failed to produce blebs despite of ectopic cortactin expression (Fig. 6i and Supplementary Movie 9). Time-lapse outline tracing of representative cells in the tissue demonstrated that Ras^V12, piezo RNAi cells with ectopic cortactin expression still remained steady and did not exert extensive blebbing (Figs. 4h and 6i). Altogether, these results suggest that Piezo plays a discrete role in eliciting blebbing, which is a way to transmigrate into the hemocoel.

Discussion

Although previous studies have demonstrated that Drosophila transformed cells can metastasize to distant tissues^{8,12,14,16,17}, it has been largely unknown how these transformed cells migrate from their primary site into the hemocoel. Here, we define a series of cellular processes and molecular mechanisms required for cell dissemination in Drosophila (Fig. 7). Initially, expression of Ras^V12 in ISCs and EBs makes them propagate. These Ras^V12 cells produce actin- and cortactin-rich invasive protrusions at the basal side, which grow into large blebs/protrusions penetrating the BM and VM layers (Fig. 7, stage 2). Ras^V12 cells release ELEVs across the VM layer presumably from where these large blebs/protrusions are formed. Finally, Ras^V12 cells transverse the VM layer by involving extensive blebbing or simply slipping through where tissue integrity is compromised (Fig. 7, stage 3) and complete the transmigration process (Fig. 7, stage 4). Note that we did not detect metastasis of Ras^V12 cells although they were found in hemolymph. To metastasize, these cells might require additional genetic alterations to evade anoikis in hemolymph and reactivate cell division in distant tissues, avenues for further investigation. Altogether, our observations demonstrate how multiple invasive and migratory mechanisms are incorporated in vivo to form a mode of cell dissemination.

**Fig. 7: Schematic stages of cell dissemination.**

In this study, we describe the actin- and cortactin-rich invasive protrusions in Drosophila. Our observations show the striking resemblance between these invasive protrusions and invadopodia observed in cancer cells. In particular, our findings indicate that these invasive protrusions in Ras^V12 cells are associated with the breach of the ECM. In cancer cells, disruption of cortactin, which is a major component in invadopodia, impaired invadopodia-mediated ECM degradation^31,30. Similarly, cortactin was enriched at the invasive protrusions, and cortactin depletion in Ras^V12 cells significantly suppressed the ECM degradation (Supplementary Fig. 5a,b) and dissemination of Ras^V12 cells (Supplementary Fig. 5f). These results suggest that the invasive protrusions in Drosophila are functionally orthologous to invadopodia observed in cancer cells. Therefore, testing localization of additional invadopodium markers, such as tyrosine kinase substrate with five SH3 domains (Tks5), and measuring the release of MMPs at the invasive protrusions will help us to further highlight the resemblance between two invasive protrusions. Interestingly, a robust ECM degradation is a phenotype commonly observed in metastatic tumors in Drosophila^{12,13,14,16,24}. However, it is unknown whether other transformed Drosophila cells also utilize similar invasive protrusions for cell dissemination. Therefore, it would be interesting to address other metastatic cells in Drosophila, such as Ras^V12/scrib^−/− eye disc tumor cells¹² and Ras^V12-expressing hindgut epithelial cells¹⁴, utilize a similar invasive structure for cell dissemination.

Our observations suggest that the breach of the ECM and the VM layer associated with the action of the invasive protrusions could be exploited for cell dissemination. Ras^V12 cells transmigrate into the hemocoel through the ruptures in the VM layer by employing extensive blebbing. In addition, Ras^V12 cells can even slide out from the midgut through the regions that produce ELEVs. Notably, our observations show that some of these disseminated cells are not originally located near the dissemination site (Supplementary Movie 8). Altogether, these observations suggest that the breach of the ECM and VM layers provides an opportunity that nearby cells can utilize for dissemination. Nevertheless, our observations also demonstrate that this opportunity cannot be utilized when Piezo is disrupted (Fig. 6d–i). Given the extensive degradation of the ECM, mesenchymal mode of migration is not a proper migratory strategy that Ras^V12 cells can adopt for dissemination since it requires formation of strong adhesions at the leading edge^37,46. Multiple mechanisms that cells use to migrate without support of focal adhesions have been proposed³⁸. In particular, bleb-driven ameboid movement allows cells to pass through holes in the three-dimensional substrate. Interestingly, we found that Ras^V12 cells generate a large amount of ELEVs and blebs, suggesting that Ras^V12 cells adopt bleb-driven ameboid movement. Notably, Piezo depletion in Ras^V12 cells impaired production of ELEVs and blebs. The breach of the BM and VM layers would cause a significant remodeling of the biomechanical microenvironment. Therefore, we propose that Piezo plays a key role in transducing the opportunistic cues associated with the compromise in the tissue integrity to allow Ras^V12 cells to adapt appropriate migratory modes for dissemination. A recent study demonstrated that confinement and low adhesion could induce ameboid movement in various mammalian cells⁴⁷, which highlighted how the mechanical cues in the substrates influences a cell’s migratory strategy. Interestingly, Srivastava et al. showed that pressure sensing via Piezo makes Dictyostelium to generate blebs instead of pseudopods for migration⁴⁸. Thus, it would be interesting to address what environmental cues drive Piezo activation in the midguts for coordination of the cell’s dissemination process.

We elucidated a mode of cell dissemination and uncovered Piezo as a key player of cell dissemination in vivo. Piezo disruption interferes with specific cellular phenotypes observed over the course of cell dissemination, indicating that some of the morphologically distinct stages and cellular processes observed during cell dissemination are also genetically discriminable. Given the description of the molecular and cellular mechanisms during an actual cell dissemination process, our study underscores the usefulness of Drosophila in deciphering the genetic basis of these invasive mechanisms in a native context.

Methods

Fly genetics and husbandry

Fly crosses were raised in vials containing standard cornmeal-agar medium and incubated at 18 °C throughout development and adulthood. Three- to ten-day-old nonvirgin female flies were used for all experiments. Flies were shifted to 29 °C for 1- to 6-days prior to dissection. During incubation at 29 °C, flies were transferred onto fresh food every 2 days.

To manipulate intestinal stem cells (ISCs) and enteroblasts (EBs), we used esg-GAL4, tub-GAL80^ts, UAS-GFP (referred as esg^ts). The experimental procedures are described previously⁴⁹. Strains obtained from the Bloomington Drosophila Stock Center (BDSC) are the followings: UAS-piezo-GFP/TM6B (#58773), UAS-Ras^V12 (III) (#4847), UAS-Raf^gof (#2033), UAS-p35 (#5073), UAS-Lifeact-mRFP (#58362), UAS-Actin-mRFP (#24778), UAS-cortactin-HA (#9368), ey-GAL4 (#5534), and UAS-EGFP (#5430). We also used UAS-Ras^V12 (II) (laboratory stock). Cell dissemination phenotypes induced by our laboratory UAS-Ras^V12 (II) and BDSC UAS-Ras^V12 (III) (#4847) alleles were comparable. For the experiments to check subcellular distribution of Piezo-GFP and cortactin-HA, we omitted UAS-GFP from esg^ts.

We used two piezo RNAi lines: 8486R-3³⁹ (shown as UAS-piezo-i⁸⁴⁸⁶) from the National Institute of Genetics, Japan (NIG-Fly; https://shigen.nig.ac.jp/fly/nigfly) and v2796⁴⁰ (shown as UAS-piezo-i^v2796) from the Vienna Drosophila Resource Center (VDRC; www.vdrc.at). Knockdown efficiency of the piezo RNAi lines was assessed by expressing each line with a piezo promoter trap line, piezo-GAL4 (BDSC #78335)⁴⁰. We extracted RNA samples from five adult female flies to measure piezo mRNA levels by qPCR using the primer sequences: 5′-TTGCTCGTTCAGTGAGCGTC-3′ and 5′-AGGCACTAGCCATTCGATGAT-3′. piezo transcript levels were reduced to ~43% by UAS-piezo-i⁸⁴⁸⁶ and ~61% by UAS-piezo-i^v2796 compared with the levels in piezo-GAL4 only.

For other knockdown experiments, we obtained from BDSC: UAS-cortactin RNAi (BDSC #32871), UAS-Calpain-A RNAi (BDSC #29455), and UAS-Calpain-B RNAi (BDSC #25963). We also used UAS-Calpain A RNAi (NIG #7563R-3) from NIG-Fly.

Antibodies and immunofluorescence imaging

We used the following primary antibodies: anti-GFP antibody, Alexa Fluor® 488 (1:1000; Thermo Fisher Scientific, A-21311; rabbit), anti-Dl antibody (1:1000; Developmental Studies Hybridoma Bank, C594.9B; mouse), anti-Mmp1 antibody (1:1000; Developmental Studies Hybridoma Bank, 3B8D12; mouse), anti-phospho-histone H3 antibody (1:1000; Millipore, 06-570; rabbit), anti-phospho-histone H3 antibody (1:1000; Abcam, ab14955; mouse), anti-laminin B1 antibody (1:1000; Abcam, ab47650; rabbit), and anti-HA antibody (1:2000; Santa Cruz, SC7392; mouse). Secondary antibodies used in this study were anti-rabbit and anti-mouse IgGs conjugated to Alexa Flour® 594 or Alexa Fluor® 647 (1:1000; Thermo Fisher Scientific, A-11012, A11005, A-21244, A-21235; goat). Filamentous actin was stained with phalloidin conjugated to Alexa Fluor® 594 or 647 (1:1000; Thermo Fisher Scientific, A-12381, A-22287). Nuclei were stained with DAPI (1:2000; Sigma, D9542).

To remove food from the midguts, we fed flies on 4% sucrose for ~4 h prior to dissection. Female flies were dissected in PBS. The dissected midguts were fixed in 4% paraformaldehyde (PFA) (Electron Microscopy Sciences, RT15710) diluted in PBS for 20 min and then washed three times with PBST (PBS supplemented with 0.2% Triton X-100) for 5 min each. For permeabilization and blocking, we incubated the tissue samples in blocking buffer (PBST supplemented with 5% normal goat serum) for 1 h at room temperature. The tissue samples were incubated with primary antibodies in the blocking buffer either overnight at 4 °C or for 2–3 h at room temperature. The samples were washed three times with PBST and then incubated with secondary antibodies for 2–3 h at room temperature. Stained midguts were washed three times with PBST and mounted with Vectashield (Vector Laboratories, H-1000). Fluorescence micrographs were acquired with a Leica SP8 laser scanning confocal microscope with 20×/0.7 dry or 40×/1.25 oil objective lenses. Higher resolution images were acquired using Leica LIGHTNING—a detection package for image information extraction of confocal images. 3D reconstructions were created from z-stack images using a LAS X 3D viewer. NIH ImageJ software was used for further adjustment and assembly of the acquired images.

Quantification of disseminated cells

We defined disseminated cells as GFP- and DAPI-positive cells residing more basally than the visceral muscle layer, which was labeled with phalloidin. To determine the position of a cell, we captured a series of z-stacks using confocal microscopy. Orthogonal view reconstituted from z-stacks was used for further confirmation. The number of disseminated cells was counted from the R5 region of the posterior midgut captured in 388 µm × 388 µm confocal microscope fields.

Quantification of Mmp1 intensity

To measure the fluorescent intensity of Mmp1, a z-projection of 388 µm × 388 µm microscope field was created from the R5 region of a posterior midgut. The projection included stacks of one leaflet of the intestine. We collected mean gray value of the red channel from three random 100 µm × 100 µm fields per intestine using NIH ImageJ software and subtracted the background values measured from the outside area surrounding the intestine.

Quantification of phospho-histone H3 (pHH3)-positive cells

To determine the number of dividing cells, midguts were dissected and stained with DAPI and anti-pHH3 antibody. pHH3-positive nuclei were counted from the entire midgut.

Actin time-lapse imaging

To monitor filamentous actin during ex vivo live imaging, we expressed Lifeact-mRFP with esg^ts. Five- to seven-day-old adult flies were incubated at 29 °C for 2 days before imaging. Time-lapse imaging of the posterior midgut was captured with a Leica SP8 laser scanning confocal microscope using a 40×/1.25 oil objective. Z-stacks of the dual-color images were recorded every 1 min for 2 h. A maximum intensity z-projection was obtained using Leica LAS X imaging software.

Ex vivo live imaging

The experimental procedures described in ref. ⁵⁰ were adapted for ex vivo live imaging of midguts. Our ex vivo live-imaging medium comprises Schneider’s Drosophila medium (Thermo Fisher Scientific, 21720024), 2% FBS (Life Technologies, 16140071), and 0.5% penicillin-streptomycin (Thermo Fisher Scientific, 15140122).

Drosophila were dissected in the imaging media and the entire midgut and hindgut sections were recovered. Then, the dissected samples were mounted on a 35-mm glass bottom dish (MatTek, P35G-1.0-14-C). To prevent squeezing, vacuum grease was applied as two lines on the glass bottom dish, ~2 mm apart. Approximately 100 µl of imaging media was added to the center of the dish, and the dissected guts were placed perpendicular to the vacuum grease lines. Anterior end of the midgut and hindgut segment were embedded in each vacuum grease line to prevent drifting. A cover glass was gently placed on top of the vacuum grease lines to prevent excessive movement of the midgut. Approximately 1 ml of imaging medium was added to the dish to prevent dehydration and facilitate gas exchange. To prevent media evaporation, the samples were covered with the dish lid and were imaged from the bottom side.

Quantification of vesicles released from gut

Ex vivo live imaging of midguts was processed into z-projection videos with the Leica LAS X software. The number and diameter of the GFP-positive vesicles detected outside the midguts at the indicated time points were quantified using NIH ImageJ software.

Collection of circulating ELEVs and disseminated cells

Hemolymph was collected through punctures on the head as described⁵¹. Hemolymph prepared from three flies was mixed with 3 μl of Vectashield with DAPI (1:2000) for immediate imaging. Seven random views of 500 µm × 500 µm confocal microscope fields were obtained with a Leica SP8 laser scanning confocal microscope with 40×/1.25 oil objective. ELEVs larger than 1 µm were counted using the NIH ImageJ analysis particle function. The number of circulating cells was counted manually by finding foci that were positive for both GFP and DAPI.

Quantification of longitudinal muscle breakage

The number of discontinuous longitudinal muscle was counted from one leaflet of the R5 region of the posterior midgut captured in 388 µm × 388 µm confocal microscope fields. F-actin-rich visceral muscle (VM) fibers were revealed by phalloidin staining. Z-stack images were acquired using a Leica SP8 laser scanning confocal microscope with 40×/1.25 oil objective.

Gd³⁺ feeding

Three- to seven-day-old adult flies were transferred onto standard media supplemented with either solvent (distilled water) or 100 µM GdCl₃ (EMD Millipore, G7532-5G), then allowed to feed while the transgenes were induced at 29 °C.

Cell outline tracing

We picked cells with distinguishable boundary and followed them for 1 h in ex vivo live-imaging z-projection videos. Still shots were captured every 10 min. At each timepoint, cell outline was traced using freehand area selection tool and subjected to create mask in NIH ImageJ software. To visualize the changes in cell boundary overtime, the traced cell outlines were overlaid. To quantify the changes in cell boundary, we measured non-overlapping area of two cell silhouettes acquired at consecutive time points (n′ and n + 10′, n = 0′, 10′, 20′, 30′, 40′, 50′).

Statistics and reproducibility

All the images presented and used for quantification are from the posterior R5 region of adult female fly midguts, except for the pHH3, which is counted from the entire midgut. All experiments were independently repeated at least three times. Statistical analyses were performed using Microsoft Excel, GraphPad Prism 8, and ‘R’ software. All P values were determined by two-tailed Student’s t-test with unequal variances. Level of significance are depicted by asterisks in the figures: *P < 0.01. P values are indicated in graph. Sample sizes were chosen empirically based on the observed effects and listed in the figure legends.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

The authors declare that the data supporting the findings of this study are available within the paper and its supplementary information files. Additional datasets are available from the corresponding author upon reasonable request.

References

Gomez-Cuadrado, L., Tracey, N., Ma, R., Qian, B. & Brunton, V. G. Mouse models of metastasis: progress and prospects. Dis. Model Mech. 10, 1061–1074 (2017).
CAS PubMed PubMed Central Google Scholar
Saxena, M. & Christofori, G. Rebuilding cancer metastasis in the mouse. Mol. Oncol. 7, 283–296 (2013).
CAS PubMed PubMed Central Google Scholar
Kang, Y. & Pantel, K. Tumor cell dissemination: emerging biological insights from animal models and cancer patients. Cancer Cell 23, 573–581 (2013).
CAS PubMed PubMed Central Google Scholar
Cai, D. & Montell, D. J. Diverse and dynamic sources and sinks in gradient formation and directed migration. Curr. Opin. Cell Biol. 30, 91–98 (2014).
CAS PubMed Google Scholar
Bai, J., Uehara, Y. & Montell, D. J. Regulation of invasive cell behavior by taiman, a Drosophila protein related to AIB1, a steroid receptor coactivator amplified in breast cancer. Cell 103, 1047–1058 (2000).
CAS PubMed Google Scholar
Sherwood, D. R. & Plastino, J. Invading, leading and navigating cells in Caenorhabditis elegans: insights into cell movement in vivo. Genetics 208, 53–78 (2018).
CAS PubMed Google Scholar
Sherwood, D. R. & Sternberg, P. W. Anchor cell invasion into the vulval epithelium in C. elegans. Dev. Cell 5, 21–31 (2003).
CAS PubMed Google Scholar
Sonoshita, M. & Cagan, R. L. Modeling human cancers in Drosophila. Curr. Top. Dev. Biol. 121, 287–309 (2017).
CAS PubMed Google Scholar
Yu, F. X., Zhao, B. & Guan, K. L. Hippo pathway in organ size control, tissue homeostasis, and cancer. Cell 163, 811–828 (2015).
CAS PubMed PubMed Central Google Scholar
Hariharan, I. K. Organ size control: lessons from Drosophila. Dev. Cell 34, 255–265 (2015).
CAS PubMed PubMed Central Google Scholar
Johnson, R. & Halder, G. The two faces of Hippo: targeting the Hippo pathway for regenerative medicine and cancer treatment. Nat. Rev. Drug Disco. 13, 63–79 (2014).
CAS Google Scholar
Pagliarini, R. A. & Xu, T. A genetic screen in Drosophila for metastatic behavior. Science 302, 1227–1231 (2003).
ADS CAS PubMed Google Scholar
Campbell, K. et al. Collective cell migration and metastases induced by an epithelial-to-mesenchymal transition in Drosophila intestinal tumors. Nat. Commun. 10, 2311 (2019).
ADS PubMed PubMed Central Google Scholar
Bangi, E., Pitsouli, C., Rahme, L. G., Cagan, R. & Apidianakis, Y. Immune response to bacteria induces dissemination of Ras-activated Drosophila hindgut cells. EMBO Rep. 13, 569–576 (2012).
CAS PubMed PubMed Central Google Scholar
Bangi, E., Murgia, C., Teague, A. G., Sansom, O. J. & Cagan, R. L. Functional exploration of colorectal cancer genomes using Drosophila. Nat. Commun. 7, 13615 (2016).
ADS CAS PubMed PubMed Central Google Scholar
Hirabayashi, S., Baranski, T. J. & Cagan, R. L. Transformed Drosophila cells evade diet-mediated insulin resistance through wingless signaling. Cell 154, 664–675 (2013).
CAS PubMed PubMed Central Google Scholar
Miles, W. O., Dyson, N. J. & Walker, J. A. Modeling tumor invasion and metastasis in Drosophila. Dis. Model Mech. 4, 753–761 (2011).
CAS PubMed PubMed Central Google Scholar
Simanshu, D. K., Nissley, D. V. & McCormick, F. RAS proteins and their regulators in human disease. Cell 170, 17–33 (2017).
CAS PubMed PubMed Central Google Scholar
Wright, K. L. et al. Ras signaling is a key determinant for metastatic dissemination and poor survival of luminal breast cancer patients. Cancer Res. 75, 4960–4972 (2015).
CAS PubMed Google Scholar
Egan, S. E. et al. Expression of H-ras correlates with metastatic potential: evidence for direct regulation of the metastatic phenotype in 10T1/2 and NIH 3T3 cells. Mol. Cell Biol. 7, 830–837 (1987).
CAS PubMed PubMed Central Google Scholar
McCaffrey, L. M., Montalbano, J., Mihai, C. & Macara, I. G. Loss of the Par3 polarity protein promotes breast tumorigenesis and metastasis. Cancer Cell 22, 601–614 (2012).
CAS PubMed PubMed Central Google Scholar
Martorell, O. et al. Conserved mechanisms of tumorigenesis in the Drosophila adult midgut. PLoS ONE 9, e88413 (2014).
ADS PubMed PubMed Central Google Scholar
Wang, C. et al. APC loss-induced intestinal tumorigenesis in Drosophila: roles of Ras in Wnt signaling activation and tumor progression. Dev. Biol. 378, 122–140 (2013).
CAS PubMed Google Scholar
Uhlirova, M. & Bohmann, D. JNK- and Fos-regulated Mmp1 expression cooperates with Ras to induce invasive tumors in Drosophila. Embo J. 25, 5294–5304 (2006).
CAS PubMed PubMed Central Google Scholar
Kelley, L. C. et al. Adaptive F-actin polymerization and localized ATP production drive basement membrane invasion in the absence of MMPs. Dev. Cell 48, 313–328 e318 (2019).
CAS PubMed PubMed Central Google Scholar
Biteau, B., Hochmuth, C. E. & Jasper, H. JNK activity in somatic stem cells causes loss of tissue homeostasis in the aging Drosophila gut. Cell Stem Cell 3, 442–455 (2008).
CAS PubMed PubMed Central Google Scholar
Lohmer, L. L., Kelley, L. C., Hagedorn, E. J. & Sherwood, D. R. Invadopodia and basement membrane invasion in vivo. Cell Adh Migr. 8, 246–255 (2014).
PubMed PubMed Central Google Scholar
Eddy, R. J., Weidmann, M. D., Sharma, V. P. & Condeelis, J. S. Tumor cell invadopodia: invasive protrusions that orchestrate metastasis. Trends Cell Biol. 27, 595–607 (2017).
CAS PubMed PubMed Central Google Scholar
Parekh, A. & Weaver, A. M. Regulation of invadopodia by mechanical signaling. Exp. Cell Res. 343, 89–95 (2016).
CAS PubMed Google Scholar
Clark, E. S. & Weaver, A. M. A new role for cortactin in invadopodia: regulation of protease secretion. Eur. J. Cell Biol. 87, 581–590 (2008).
CAS PubMed PubMed Central Google Scholar
Clark, E. S., Whigham, A. S., Yarbrough, W. G. & Weaver, A. M. Cortactin is an essential regulator of matrix metalloproteinase secretion and extracellular matrix degradation in invadopodia. Cancer Res. 67, 4227–4235 (2007).
CAS PubMed Google Scholar
Spracklen, A. J., Fagan, T. N., Lovander, K. E. & Tootle, T. L. The pros and cons of common actin labeling tools for visualizing actin dynamics during Drosophila oogenesis. Dev. Biol. 393, 209–226 (2014).
CAS PubMed PubMed Central Google Scholar
Meehan, B., Rak, J. & Di Vizio, D. Oncosomes—large and small: what are they, where they came from? J. Extracell. Vesicles 5, 33109 (2016).
PubMed Google Scholar
Headley, M. B. et al. Visualization of immediate immune responses to pioneer metastatic cells in the lung. Nature 531, 513–517 (2016).
ADS CAS PubMed PubMed Central Google Scholar
Becker, A. et al. Extracellular vesicles in cancer: cell-to-cell mediators of metastasis. Cancer Cell 30, 836–848 (2016).
CAS PubMed PubMed Central Google Scholar
Wolf, K. et al. Compensation mechanism in tumor cell migration: mesenchymal-amoeboid transition after blocking of pericellular proteolysis. J. Cell Biol. 160, 267–277 (2003).
CAS PubMed PubMed Central Google Scholar
Pandya, P., Orgaz, J. L. & Sanz-Moreno, V. Modes of invasion during tumour dissemination. Mol. Oncol. 11, 5–27 (2017).
PubMed Google Scholar
Paluch, E. K., Aspalter, I. M. & Sixt, M. Focal adhesion-independent cell migration. Annu Rev. Cell Dev. Biol. 32, 469–490 (2016).
CAS PubMed Google Scholar
Kim, S. E., Coste, B., Chadha, A., Cook, B. & Patapoutian, A. The role of Drosophila Piezo in mechanical nociception. Nature 483, 209–212 (2012).
ADS CAS PubMed PubMed Central Google Scholar
He, L., Si, G., Huang, J., Samuel, A. D. T. & Perrimon, N. Mechanical regulation of stem-cell differentiation by the stretch-activated Piezo channel. Nature 555, 103–106 (2018).
Coste, B. et al. Piezo1 and Piezo2 are essential components of distinct mechanically activated cation channels. Science 330, 55–60 (2010).
ADS CAS PubMed PubMed Central Google Scholar
He, L., Si, G., Huang, J., Samuel, A. D. T. & Perrimon, N. Mechanical regulation of stem-cell differentiation by the stretch-activated Piezo channel. Nature 555, 103–106 (2018).
ADS CAS PubMed PubMed Central Google Scholar
Pardo-Pastor, C. et al. Piezo2 channel regulates RhoA and actin cytoskeleton to promote cell mechanobiological responses. Proc. Natl Acad. Sci. USA 115, 1925–1930 (2018).
CAS PubMed PubMed Central Google Scholar
McHugh, B. J. et al. Integrin activation by Fam38A uses a novel mechanism of R-Ras targeting to the endoplasmic reticulum. J. Cell Sci. 123, 51–61 (2010).
PubMed Google Scholar
MacGrath, S. M. & Koleske, A. J. Cortactin in cell migration and cancer at a glance. J. Cell Sci. 125, 1621–1626 (2012).
CAS PubMed PubMed Central Google Scholar
Te Boekhorst, V., Preziosi, L. & Friedl, P. Plasticity of cell migration in vivo and in silico. Annu Rev. Cell Dev. Biol. 32, 491–526 (2016).
Google Scholar
Liu, Y. J. et al. Confinement and low adhesion induce fast amoeboid migration of slow mesenchymal cells. Cell 160, 659–672 (2015).
CAS PubMed Google Scholar
Srivastava, N., Traynor, D., Piel, M., Kabla, A. J. & Kay, R. R. Pressure sensing through Piezo channels controls whether cells migrate with blebs or pseudopods. Proc. Natl Acad. Sci. USA 117, 2506–2512 (2020).
CAS PubMed PubMed Central Google Scholar
Apidianakis, Y., Pitsouli, C., Perrimon, N. & Rahme, L. Synergy between bacterial infection and genetic predisposition in intestinal dysplasia. Proc. Natl Acad. Sci. USA 106, 20883–20888 (2009).
ADS CAS PubMed PubMed Central Google Scholar
Aldaz, S., Escudero, L. M. & Freeman, M. Live imaging of Drosophila imaginal disc development. Proc. Natl Acad. Sci. USA 107, 14217–14222 (2010).
ADS CAS PubMed PubMed Central Google Scholar
Haselton, A. T. & Fridell, Y. W. Insulin injection and hemolymph extraction to measure insulin sensitivity in adult Drosophila melanogaster. J. Vis. Exp. 52, e2722 (2011).
Google Scholar

Download references

Acknowledgements

We thank Drs. Andrea Wills, Eric Muller, and Akhila Rajan for comments on the manuscript. This work was supported by R35GM128752 to Y.V.K. from the National Institutes of Health.

Author information

These authors contributed equally: Jiae Lee, Alejandra J. H. Cabrera

Authors and Affiliations

Department of Biochemistry, University of Washington, Seattle, WA, 98195, USA
Jiae Lee, Alejandra J. H. Cabrera, Cecilia M. T. Nguyen & Young V. Kwon

Authors

Jiae Lee
View author publications
You can also search for this author in PubMed Google Scholar
Alejandra J. H. Cabrera
View author publications
You can also search for this author in PubMed Google Scholar
Cecilia M. T. Nguyen
View author publications
You can also search for this author in PubMed Google Scholar
Young V. Kwon
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

J.L., A.J.H.C., and Y.V.K. designed and performed the experiments and analyzed the data. C.M.T.N. contributed to the development of the project. J.L., A.J.H.C., and Y.V.K. wrote the paper.

Corresponding author

Correspondence to Young V. Kwon.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Sebastien Roger and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Description of Additional Supplementary Files

Supplementary Movie 1

Supplementary Movie 2

Supplementary Movie 3

Supplementary Movie 4

Supplementary Movie 5

Supplementary Movie 6

Supplementary Movie 7

Supplementary Movie 8

Supplementary Movie 9

Reporting Summary

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Lee, J., Cabrera, A.J.H., Nguyen, C.M.T. et al. Dissemination of Ras^V12-transformed cells requires the mechanosensitive channel Piezo. Nat Commun 11, 3568 (2020). https://doi.org/10.1038/s41467-020-17341-y

Download citation

Received: 30 May 2019
Accepted: 24 June 2020
Published: 16 July 2020
DOI: https://doi.org/10.1038/s41467-020-17341-y

This article is cited by

Intrinsically disordered intracellular domains control key features of the mechanically-gated ion channel PIEZO2
- Clement Verkest
- Irina Schaefer
- Stefan G. Lechner
Nature Communications (2022)
Non-canonical Wnt signaling promotes directed migration of intestinal stem cells to sites of injury
- Daniel Jun-Kit Hu
- Jina Yun
- Heinrich Jasper
Nature Communications (2021)
Tumour–host interactions through the lens of Drosophila
- David Bilder
- Katy Ong
- Jung Kim
Nature Reviews Cancer (2021)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.