Final Report Summary - MESENCHYMAL COLL MOT (Physical forces involved in collective migration of mesenchymal versus epithelial cells)
Coordinated migration of cells plays an important role during normal development of multicellular organisms, as well as during events under abnormal conditions such as cancer metastasis or wound healing. Understanding the mechanisms that coordinate this collective migration of cells is therefore of high importance. The majority of studies investigate the collective migration of tightly connected cells, epithelia, which typically form the skin or the boundary of organs. These studies show that physical interaction, shear, pushing and pulling, between cells plays a major role in coordinating the migration of the cell sheet. While the importance of epithelial behaviour remains unquestionable, a recent discovery showed that loosely connected mesenchymal cells also undergo collective migration. These cells do not form a stable contact and therefore the role of physical forces in coordinating the collective migration of mesenchymal cells is questionable, and the mechanism for coordinating their migration is lesser known. To investigate this question, the project focused on the collective migration of the mesenchymal neural crest cells (NCCs). During embryo development of vertebrates, the NCCs differentiate at the side of the neural tube and undergo an epithelial-to-mesenchymal transition (EMT) and begin to migrate as a collective of individually migrating cells. Migration of these cells is exceptionally efficient, whereby a large population of cells migrates long distances in a short time, and has been likened to the migration of metastatic cancer cells.
The project was carried out in a developmental biology group specialised in studies of NCC migration, using the Xenopus laevis animal model system. Cells were observed both in the embryo and in culture using time-lapse microscopy. Experimental data was processed using both established and novel analytical methods, making sure that the newly developed software is compatible with the previous methods and is accessible through both a user-friendly interface and high-throughput processing. To measure the forces exerted by the cells, traction force microscopy was established in the host lab by improving experimental protocols and introducing a post-processing pipeline partly from re-using published algorithms and custom software to implement traction force calculations. The establishment of the technique was facilitated by national and international collaborations with labs specialising in mechanobiology. Cell-cell adhesion forces were measured in collaboration with a specialist lab. A new data storage facility was installed in the host lab to ensure the re-use of data and prevent unnecessary replication of experiments. An online temperature logging system was introduced to ensure stability of experimental conditions.
To characterise the difference between epithelial and mesenchymal cells, the movement of NCCs was compared before they undergo EMT (early NCCs) and after EMT (late NCCs), and with an epithelial tissue adjacent to the NCCs, the placodes. NCC and placode movements were examined both separately and in interaction with each other, showing that early NCCs behave as epithelial cells while late NCCs behave as mesenchymal. The advantage of comparing early and late NCCs is that we use the exact same cell type and therefore minimise the effect of unrelated differences between the epithelial and the mesenchymal types. We analysed the behaviour of cells with altered adhesion properties and found that the difference in the adhesion molecule profile of cells could be sufficient to explain the difference in their collective behaviour. We discovered that when mesenchymal cells come into contact, their protrusions are actively inhibited at the site of contact. New protrusions are established away from the contact, which generate sufficient traction forces to tear the adhesion complex at the contact apart, allowing the cells to move away. In epithelial cells, this inhibition does not occur and the cells do not separate. The phenomenon whereby cells repolarise and move away from one another after contact is termed contact inhibition of locomotion (CIL), and has been described in a wide variety of cell types, including some cancer cells. Our results uncover the mechanism of CIL, and pinpoint the molecular components of epithelial and mesenchymal cells that differentiate between the respective absence and presence of CIL, with potential relevance to understanding and preventing cancer metastasis.
To investigate the effect of CIL on collective migration, we revisited previous computational studies conducted in the host lab. These show that CIL is necessary but insufficient to explain mesenchymal collective migration, and that cells also require co-attraction (CoA), whereby cells secrete a diffusible substance towards which the surrounding cells migrate. CoA together with CIL was sufficient to give rise to collective migration in previous models, and this explains why cells lacking CIL are less prone to migrate. However, a closer inspection of the models revealed that contact-driven alignment was also required in these models. Based on the observation that persistence of cells is increased when they come in contact with a group of cells, we constructed a new model of mesenchymal collective migration. In addition, we observed that confining neural crest clusters onto a lane using the inhibitory molecule Versican enhances their collective directional migration. These results are consistent with experiments using Versican deficient embryos and explain the puzzling fact that all neural crest migrate in streams.
As a by-product of the project, we discovered a surprising and an unexpected molecular mechanism that is essential for embryo gastrulation. Gastrulation is the first major morphogenetic event in embryogenesis during which the flat embryo, consisting of two cell layers, involutes to give rise to its middle layer that is essential for its three-dimensional organisation. We found that epiboly, the expansion of the region opposite the involuting side of the embryo, is an active and tissue-autonomous process. Our results show that the previously reported radial intercalation of cells occurring in this region is driven by chemotaxis towards the complement component C3a, a molecule well known for its role in the immune system.
The project has provided several important discoveries in the field of developmental biology with implications to cancer biology. This project would not have been possible without the cooperation of the excellent research groups within Europe, and together it raises the impact of the European research community in the world.
The project was carried out in a developmental biology group specialised in studies of NCC migration, using the Xenopus laevis animal model system. Cells were observed both in the embryo and in culture using time-lapse microscopy. Experimental data was processed using both established and novel analytical methods, making sure that the newly developed software is compatible with the previous methods and is accessible through both a user-friendly interface and high-throughput processing. To measure the forces exerted by the cells, traction force microscopy was established in the host lab by improving experimental protocols and introducing a post-processing pipeline partly from re-using published algorithms and custom software to implement traction force calculations. The establishment of the technique was facilitated by national and international collaborations with labs specialising in mechanobiology. Cell-cell adhesion forces were measured in collaboration with a specialist lab. A new data storage facility was installed in the host lab to ensure the re-use of data and prevent unnecessary replication of experiments. An online temperature logging system was introduced to ensure stability of experimental conditions.
To characterise the difference between epithelial and mesenchymal cells, the movement of NCCs was compared before they undergo EMT (early NCCs) and after EMT (late NCCs), and with an epithelial tissue adjacent to the NCCs, the placodes. NCC and placode movements were examined both separately and in interaction with each other, showing that early NCCs behave as epithelial cells while late NCCs behave as mesenchymal. The advantage of comparing early and late NCCs is that we use the exact same cell type and therefore minimise the effect of unrelated differences between the epithelial and the mesenchymal types. We analysed the behaviour of cells with altered adhesion properties and found that the difference in the adhesion molecule profile of cells could be sufficient to explain the difference in their collective behaviour. We discovered that when mesenchymal cells come into contact, their protrusions are actively inhibited at the site of contact. New protrusions are established away from the contact, which generate sufficient traction forces to tear the adhesion complex at the contact apart, allowing the cells to move away. In epithelial cells, this inhibition does not occur and the cells do not separate. The phenomenon whereby cells repolarise and move away from one another after contact is termed contact inhibition of locomotion (CIL), and has been described in a wide variety of cell types, including some cancer cells. Our results uncover the mechanism of CIL, and pinpoint the molecular components of epithelial and mesenchymal cells that differentiate between the respective absence and presence of CIL, with potential relevance to understanding and preventing cancer metastasis.
To investigate the effect of CIL on collective migration, we revisited previous computational studies conducted in the host lab. These show that CIL is necessary but insufficient to explain mesenchymal collective migration, and that cells also require co-attraction (CoA), whereby cells secrete a diffusible substance towards which the surrounding cells migrate. CoA together with CIL was sufficient to give rise to collective migration in previous models, and this explains why cells lacking CIL are less prone to migrate. However, a closer inspection of the models revealed that contact-driven alignment was also required in these models. Based on the observation that persistence of cells is increased when they come in contact with a group of cells, we constructed a new model of mesenchymal collective migration. In addition, we observed that confining neural crest clusters onto a lane using the inhibitory molecule Versican enhances their collective directional migration. These results are consistent with experiments using Versican deficient embryos and explain the puzzling fact that all neural crest migrate in streams.
As a by-product of the project, we discovered a surprising and an unexpected molecular mechanism that is essential for embryo gastrulation. Gastrulation is the first major morphogenetic event in embryogenesis during which the flat embryo, consisting of two cell layers, involutes to give rise to its middle layer that is essential for its three-dimensional organisation. We found that epiboly, the expansion of the region opposite the involuting side of the embryo, is an active and tissue-autonomous process. Our results show that the previously reported radial intercalation of cells occurring in this region is driven by chemotaxis towards the complement component C3a, a molecule well known for its role in the immune system.
The project has provided several important discoveries in the field of developmental biology with implications to cancer biology. This project would not have been possible without the cooperation of the excellent research groups within Europe, and together it raises the impact of the European research community in the world.