Hisashi Haga

Why study collective cell migration in a test tube?

Tissue rearrangements and deformations lie at the heart of shaping an organism. Whether gastrulation in Xenopus, neural tube formation in a mouse embryo or dorsal closure during Drosophila development, they all share an aspect that is fundamental to morphogenesis: the dynamic and coordinated movement of cells. Which forces control such collective behavior? Which mechanisms are employed? In search for clues, Prof. Hisashi Haga and his team are walking along a new and unconventional research path: they are trying to unravel the puzzle by studying cell movements and 3D tissue rearrangements in vitro. As such, their aim is to isolate developmental events from the complexity of the embryo in which they occur, enabling them to observe the behavior of individual cells and making tissue rearrangements amenable to experimental intervention.

A soft substrate is required for collective cell migration

Fig 1. Collective migration of MDCK epithelial cells, cultured on a collagen gel, is directed by a leader cell (1 s in movie: 150 min).

When Madin-Darby Canine Kidney (MDCK) cells were cultured on a glass slide (using conventional methods) no signs of collective migration could be observed - quite unlike their behavior in vivo - as these cells proliferate and move randomly. How come? Why do epithelial cells demonstrate no coordinated migration in vitro? Pondering over this question, Prof. Haga could eventually connect this aberrant cell behavior to the physical properties of the extracellular environment: the stiffness of the substrate. Indeed, the substrate dramatically differs between both settings: it is very soft (~0.1 kPa) in vivo, as compared to the conventional coating of a glass slide, which is very hard and approaches the mega Pascal range. Experimenting with soft substrates, such as collagen type I (Nitta gelatin) - thereby mimicking in vivo conditions - he could eventually demonstrate that the stiffness of the extracellular environment can indeed control cell behavior. As a matter of fact, Prof. Haga and his co-workers were the first to demonstrate that cells can move in a collective, directional way on a glass slide.

Having thus developed the appropriate conditions for coordinated cell behavior in vitro, the road was paved for unraveling the mechanisms involved, through manipulation of biological and physical parameters. And indeed, this research work led to several important findings:

* A single leader cell determines the direction of migration. Removing this leader cell by means of micromanipulation, results in random and uncontrolled cell movements.
* The leader cell is highly enriched for Integrin beta-1, a transmembrane receptor involved in cell-cell and cell-extracellular matrix interactions. Using inhibitory antibodies, it could be demonstrated that Integrin beta-1is essential for leader cell function.
* Collective cell migration is under control of the Rac1 (Rho GTPase) signaling cascade.

Lumen formation in a Petri dish

Fig 2. Epithelial sheet folding and lumen formation of epithelial cells on a collagen gel after an overlay with another collagen gel (1 s in movie: 150 min).

Developmental events in vivo occur in a 2D or a 3D field, whereas under conventional in vitro conditions, they invariantly take place in a two-dimensional space. Therefore, exploring (literally) new dimensions, Prof. Haga and his team set out for a strategy aimed at observing 3D morphogenetic movements in vitro. In a first step, they tried creating a lumen in isolation, as lumen formation - a well known key aspect of 3D morphogenesis - is vital to innumerable embryological processes and organogenesis. To that end, an MDCK epithelial sheet was cultured on a collagen gel and overlaid with another collagen gel, as to mimic a 3D extracellular environment. Apparently not without success, as immediately after this overlay the MDCK cells began to fold from the periphery and migrated collectively towards the center of the field: a coordinated behavior that ultimately resulted in (and turned out to be indispensable for) the formation of a luminal structure.

Using atomic force microscopy, Prof. Haga and his co-workers succeeded in measuring the contractile force that provokes these coordinated cell movements during lumen formation. They could demonstrate that this force is linked to Myosin Regulatory Light Chain (MRLC): a key player in the process. Moreover, in an elegant set of experiments it was conclusively shown that the strength of this contractile force is tightly regulated in a spatiotemporal manner through the phosphorylation state of MRLC.

Shaping an organism in a test tube?

3D morphogenesis in vitro: MDCK cells collectively form a tulip hat structure on Matrigel, as observed by immunofluorescence (Green: F-actin, Magenta: Laminin 111).

Fig 3. 3D morphogenesis in vitro: MDCK cells collectively form a tulip hat structure on Matrigel, as observed by immunofluorescence (Green: F-actin, Magenta: Laminin 111).

Digging deeper and deeper into the unexplored territories of in vitro 3D morphogenesis, Prof. Haga pursued the idea of shaping a biological body under controlled conditions. Culturing MDCK cells on matrigel - a complex extracellular matrix-like material, mainly composed of laminin, Type IV collagen, entactin and soluble growth factors - they could steer epithelial cells into a hitherto unobserved organized behavior. Under the given biomimetic settings, these cells collectively adopt a peculiar morphology: a 3D shape reminiscent of a tulip hat. Intriguingly, similar 3D structures are observed during notochord tubulogenesis, brain morphogenesis and optic cup formation.

Following intensive research - going hand in hand with 3D computer simulations - the mechanisms controlling this 3D shape are beginning to be untangled. By experimentally manipulating the viscosity of the matrigel (varying the concentration of an amino acid cross-linking agent) and/or the contractile force of the epithelial cells (varying the phosphorylation state of MRLC with a specific kinase inhibitor) different 3D shapes could be induced in vitro. It was convincingly shown that the viscous matrigel is deformed by cellular contractile force and “molded” into a particular 3D structure, depending on the viscoelastic properties of the substrate and the strength of the physical force applied to it. Eventually a hypothesis was formulated, stating that substrate viscosity, substrate deformations and the cellular contractile force play a crucial role during morphogenesis and are essential to the 3D shaping of a biological structure.

A look into the Future
Science Fiction
Science or Fiction?
Prof. Haga, in which directions are your research projects developing?
Prof. Haga:
At present, we are trying to answer two basic questions. The first is: “How do cells sense the stiffness of the substrate?” And the second question is “How do cells collectively determine their direction of migration?”
Would you be willing to speculate?
Prof. Haga:
Considering the first question, we have identified molecules, possibly related to a cellular apparatus that might register substrate stiffness.
What is the nature of these molecules?
Prof. Haga:
We found that the expression level of some cytoskeletal proteins - constituents of intermediate filaments - changes in response to alterations of substrate stiffness, experimentally induced by modifying the concentration of an appropriate cross-linker. So, this could mean that intermediate filaments are integrated in some kind of sensor mechanism.
And the second question?
Prof. Haga:
Well, maybe we can rephrase that question into “Which mechanism allows cells to follow neighboring cells?” and now, we are entering the field of contact following between individual cells. So, I am inclined to believe that again, a general sensing mechanism might be at work here, where cell adhesion undoubtedly plays a role, but where transmembrane proteins could act as sensors, coupled to some feed-back system. Integrin beta-1is a candidate for such a sensing factor, integrated in a signaling pathway.
Are you following other research trajectories at the moment?
Prof. Haga:
Considering future projects, we want to generate many different types of 3D shapes from epithelial sheets in vitro. We succeeded in creating tulip-like structures, but this is of course a relatively simple 3D shape, compared to the complex structures that can be observed in an organism. One goal of our research program is to build a complex biological structure in vitro, thereby identifying every biomechanical parameter involved, and satisfactorily explaining it in biophysical terms. Recently, we succeeded in creating doughnut-like and blood vessel-like tubular structures, by altering the in vitro conditions. So, we are at the level where we can experimentally determine the shape of the lumen, created in the epithelial field.
What is your scientific dream?
Prof. Haga:
My ultimate dream is to understand the phenomenon “life”, in both a scientific and spiritual way. But somewhat closer to our lab projects (laughs), I would like to understand the mechanisms that determine the 3D shape of a biological body. And what I hope to find is one or a few universal principles, like in physics. For example, Newtonian dynamics is explained by a few laws. Also, in Thermodynamics, there are only 3 laws explaining an immense complexity of phenomena. Likewise, quantum mechanics is simple: the mathematics behind it may be complex, but the underlying ideas are simple. So, I hope that in the natural sciences, not only Physics but also Biology is guided by a few, basic universal principles.
  1. Yamaguchi N, Mizutani T, Kawabata K, *Haga H. Leader cells regulate collective cell migration via Rac activation in the downstream signaling of Integrin beta1 and PI3K. Scientific Reports 5, 7656-, 2015.
  2. Ishida S, Tanaka R, Yamaguchi N, Ogata G, Mizutani T, Kawabata K, *Haga H. Epithelial sheet folding induces lumen formation by Madin-Darby canine kidney cells in a collagen gel. PLoS ONE 9, e99655-, 2014.
  3. Imai M, Furusawa K, Mizutani T, Kawabata K, *Haga H. Three-dimensional morphogenesis of MDCK cells induced by cellular contractile forces on a viscous substrate. Scientific Reports 5, 14208-, 2015.
Who is Who?

Prof. Haga

Prof. Hisashi Haga

Prof. Hisashi Haga was born in Hokkaido and finished his studies in Physics at Hokkaido University. There he also obtained his PhD, at the department of Physics, following his research on phase transitions of ferroelectric crystals. After his doctoral studies, Dr. Haga decided to broaden his horizon and crossed the Atlantic to start a postdoctoral training at the Massachusetts Institute of Technology (MIT) in Boston, studying phase transitions of smectic liquid crystals. Constantly searching for new scientific challenges, Prof. Haga decided that the last field in science where new phenomena could be described was Biology. He therefore returned to Japan to the Department of Physics at Hokkaido University, where at that moment new scientific horizons were being explored in the field of Biological Physics. At present, Prof. Haga is leading a research group that studies collective and cooperative phenomena in biological systems, such as cell migration and the spatiotemporal distribution of cellular contractile force.