Matthias Kaschube 

		Lewis-Sigler Theory Fellow
Lewis-Sigler Institute for Integrative Genomics
Department of Physics
Princeton University

261 Carl-Icahn Laboratory
Princeton NJ, 08544

Email: kaschube[a]princeton edu
Phone: +1 (609) 258-1937




Biological physics of collective behavior of cells


Many interesting phenomena arise when passing from the single-cell to the multi-cell level. Striking examples are networks of neurons in sensory cortical areas that are capable of 'processing' sensory input, or the precise morphgenetic changes that take place in a developing embryo. These complex tasks require the orchestrated interplay of a large collection of individual cells. What mechanisms organize cells to such a highly elaborate level?


Whereas processes in individual cells are relatively well studied, the mechanisms by which these processes are coordinated among multiple cells are much less understood. In my lab we explore the principles governing the formation of collective processes that involve many cells and underlie such remarkable tasks as vision or morphogenesis. For instance, in the primary visual cortex we study the role of cortical interactions in learning to perceive motion in viusal stimuli. In fruit fly embryos we ask how individual cells coordinate their shape changes to robustlyly realize the formation of the ventral furrow during early gastrulation.


Some specific projects:

Apical constriction and its coordination across cells in ventral furrow formation in fruit flies

With Adam Martin and Eric Wieschaus (Department of Molecular Biology)

How do individual cells dynamically coordinate their behavior to realize morphogenetic changes on the tissue level? By combining quantitative analysis of live-cell imaging data with physical models we seek to reveal the nature of cell-cell interactions that play a role in ventral furrow formation during early gastrulation in fruit flies. In a recent study (A. Martin, M. Kaschube, E. Wieschaus, Nature, in press)  we revealed a novel ‘ratchet’ mechanism for apical constriction in ventral furrow cells.  We showed that apical constriction is pulsed, with phases of constriction interrupted by pauses in which the cell size is maintained.  Actin-myosin network contractions on the apical cortex drive constriction.  Our results contrast with the conventional model in which an actin-myosin purse-string causes continuous constriction.  We found that the transcription factors twist and snail differentially regulate the separate phases of the contraction cycle, demonstrating how the activities of these genes are coordinated to produce shape changes in individual cells. Currently, we study how these processes are coordinated among different cells.


The role of self-organization in development of direction selectivity in the visual cortex.

In collaboration with Len White, Steve van Hooser and David Fitzpatrick (Duke University)

Experience with moving visual stimuli drives the early development of direction selectivity in the visual cortex  (Ye et al., Nature, 2008). However, alsothe state of the cortical network appears to have a strong affect on whether a given neuron becomes selective for one stimulus direction or the other. If most of its surrounding neurons prefer a given direction, a cell is more likely to become selective for the same direction, even if exposed predominantly to stimuli of the opposite direction (Van Hooser et al., SfN, 2008). We combine mathematical modelling and quantitative analysis of two-photon calcium imaging data of a collection of visual cortical cells to dissect the interplay between the external training stimulus and neuronal interactions that shapes a neural circuitry capable of robustly processing motion components in natural stimuli. In one recent study we focused on the response properties of individual direction selective simple cells and revealed the important role of inhibitory inputs in rendering these neurons selective (Liu et al. , SfN, 2008). In particular, when inhibitory inputs are tuned in the same direction as excitatory inputs (as observed by Priebe and Ferster, Neuron, 2005) cells respond highly selective, even for stimuli presented only over a short period (~100ms).


Reorganization of neuronal circuits in growing visual cortex

Together with Wolfgang Keil, a visiting grad student at Princeton University.

The dynamics of reorganization of large cortical circuits is rooted in plasticity of individual synapses, but rules governing the collective behavior of large networks of neurons are only poorly understood. The postnatal brain growth partly evoked by extensive formation of new synaptic connections may expose cortical areas to a 'natural perturbation' sufficiently strong to observe signatures of large scale reorganization. In a recent study (Keil et al, SfN, 2008) we observed by quantifying large sets of imaging data from juvenile cat visual cortex  a novel mode of reorganization of domains that prefer inputs from one eye or the other.  Our theoretical analysis shows that this mode can be explained quantitatively by the so called Zigzag instability, a dynamical reorganization, well-known in the field of pattern formation in physics, by which 2D isotropic Turing patterns respond to an increase in their typical spatial scale with a zigzag-like bending of domains. We point out that this instability has in fact been predicted, albeit implicitly, by most models of visual cortical development that have been proposed so far. These results indicate that during normal postnatal development cortical networks can undergo large scale reorganizations during which cortical neurons shift their response properties in a spatially coordinated and activity dependent fashion.


Universality in the evolution of orientation columns in the visual cortex 

With Fred Wolf (MPIDS Goettingen), Len White (Duke), Siegrid Loewel (Jena) and David Coppola (Randolph Macon)

Since the basal radiation of placental mammals 65 million years ago, genetic changes have accumulated that underlie marked phenotypic differences expressed across mammalian clades.  In this project, we analyzed systems of orientation columns in the visual cortex of long-separated species that occupy widely different ecological niches and demonstrate that there are quantitative rules that govern the design of neuronal networks from which evolutionary modification can hardly deviate. Thus, in primate (galago), carnivora (ferret) and scandentia (tree shrew) the layout of orientation columns precisely follows a single universal design. We characterize the universal design by a series of invariant quantitative laws that govern the layout of orientation columns in the visual cortex. Furthermore, we show that the robust formation and evolutionary preservation of the universal design can be explained as a consequence of cortical network self-organization. This is achieved through a mathematical analysis of models for the dynamical self-organization of orientation columns. We demonstrate that all features of the universal design robustly emerge, when neuronal self-organization is dominated by long-ranging interactions. Thus the universal design will emerge in any species that ontogenetically builds its contour processing networks by such a process. This fact can explain the evolutionary emergence and preservation of the universal design without reference to selective pressures on the visual system and without the existence of a common ancestor exhibiting the universal design.



Selected publications

Pulsed contractions of an actin-myosin network drive apical constriction. A.C. Martin, M. Kaschube, E.F. Wieschaus, Nature, 2008, doi: 10.1038/nature07522 PubMed

Self-organization and the selection of pinwheel density in visual cortical development. M. Kaschube, M. Schnabel, and F. Wolf, New Journal of Physics, 10, 015009, 2008. Journal

Random Waves in the Brain: Symmetries and Defect Generation in the Visual Cortex. M. Schnabel, M. Kaschube, S. Loewel and F. Wolf,  Europhysics Journal Special Topics 145, 137-157, 2007. Journal

The pattern of ocular dominance columns in cat primary visual cortex: Intra- and interindividual variability of column spacing and its dependence on genetic background. M. Kaschube, F. Wolf, M. Puhlmann, S. Rathjen, K.-F. Schmidt, T. Geisel, and S. Loewel. European Journal of Neuroscience, 18:3251-3266, 2003. PubMed

Genetic influence on quantitative features of neocortical architecture. M. Kaschube, F. Wolf, T. Geisel, and S. Loewel.  Journal of Neuroscience, 22(16):7206-7217, 2002. PubMed

On the archive


Pinwheel stability, pattern selection and the geometry of visual space. M. Schnabel, M. Kaschube, F. Wolf, arXiv:0801.3832 arXiv

Inter-areal coordination of columnar architectures during visual cortical development. M. Kaschube, M. Schnabel, S. Löwel, F. Wolf, arXiv:0801.4164 arXiv