Martinez-Arias Lab
Department of Genetics, University of Cambridge
Research: Stochastic events in cell fate decisions
Analysis of gene expression at the level of single cells has revealed the existence of 'molecular noise' which can be detected as variability in the rate and patterns of expression of single genes within a population (for definitions see e.g. Kaern et al.2005 Nature Rev Genet. 6, 451-464 and Raser and O'Sheah 2005 Science 309, 2010-2013). During development, transitions between cells states lead to the activation of cohorts of genes whose expression levels have to be modulated and established in a coordinated manner, sometimes over large cell populations and in limited amounts of time. In the context of the inevitability of noise in gene expression, this situation creates some problems. In particular, how is it that a process that is noisy, and probably stochastic, at the microscopic level (transcription and the web of molecular interactions) has deterministic and reproducible effects at the macroscopic level (the behaviour of cells and tissues)? Reasoning through this question has led us to suggest that in multicellular organisms there might be molecular devices dedicated to the control of transcriptional noise and, specifically, that Wnt and Notch signalling configure a module that executes this task (Martinez Arias and Hayward 2006 Nature Rev. Genet. 7, 34-44).
Examples of how Wnt signalling influences the probability with
which a cell will adopt a fate. During the development and patterning
of the PNS in Drosophila (left), SOPs arise from so called proneural
clusters (see above). Most significantly, the SOPs arise from defined
and reproducible positions within the clusters. Genetic analysis
suggests that there is a cluster of cells that have similar
probability to become an SOP, but only some cells (green) adopt this
fate. These cells emerge, in different clusters, from the edge of the
cells expressing Wingless (Drosophila Wnt1) and this, in the context
of genetic analusis (see refs below) indicates that Wingless biases
the probability that cells have to adopt the fate. In this case there
is evidence that Notch is setting up a threshold for Wnt signalling
and that cells measure the relative levels of both signalling
activities. On the right hand side you can see a second example of
Wingless influencing the probability with which a cell adopts a
fate. The pictures show the ectoderm of 6 hours old embryos of
Drosophila and the image shows the expression of Engrailed. In blue
one can see the cells that have expressed engrailed at some time
during development as reflected from a lineage marker
(β-galactosidase) under the control of engrailed. In brown, the cells
that actually express engrailed, show that it is only a subset of the
cells that have expressed engrailed. The subset is patterned and has a
sharp boundary in the anterior side (left). Adjacent to the engrailed
expressing cells there is a group of cells expressing Wingless which
will diffuse thus creating a gradient that will be transformed into a
gradient of probabilities. In both, the SOPs and engrailed expression,
Wingless does not influence the initial expression but rather that
probability with which a cell will maintain (stably express) the
expression.
In the last few years we have initiated a project centered on the study of transcriptional noise and its regulation on cell fate decision. We have chosen mouse embryonic stem cells (ES cells) as a model for these studies because we deem them a simple model system with limited informational content and few Gene Regulatory Networks active, which will translate into defined experimental settings. Our focus is on the effect that Wnt signalling has on the Nanog/Oct4 network.
In order to study the regulation of transcriptional
noise in stem cells, we have begun by assemblying an experimental
system in P19 embryocarcinoma cells (EC cells) in which we have
transfected GFP under the control of the Nanog promoter and, in the
same cells, RFP under the control of a Wnt reporter. In this manner we
can monitor simultaneously in the same cell a transcriptional output,
Nanog expression, and Wnt signalling. We use FACs profiles from single
cells as the observable. We are also modelling the system from a
theoretical point of view.
Whereas in telecommunications and electronics noise is a nuisance and there is an investment to create devices that filter and remove noise, in biology noise might be something not just desirable but necessary. So much so that not only there migh exist molecular devices dedicated to the filtering of noise to generate specific patterns, but there might be devices dedicated to the generation of noise i.e. noise could be a controlled output of the system. Thus, in the proneural clusters (see our work on interactions between signalling pathways) the 'noisy' expression of Achaete might be carefully tuned for the pattern that will emerge. The analysis of these processes will require a quantitative and analytical study of defined observables.
The group working on stochastic events in cell fate decisions are:
Alfonso Martinez Arias - Lab Leader
Silvia Munoz Descalzo - Postdoc
Penny Hayward - Postdoc
Tina Balayo - Postdoc
Jamie Trott - PhD Student
Joaquin de Navascues - Postdoc
Chea Lim - PhD Student
Ana Mateus - PhD Student
Fernando Faunes - Postdoc
In collaboration with
Jennifer Nichols (Cambridge Stem Cell Initiative, University of Cambridge)
Kat Hadjanntonakis (Sloan Kettering Institute, New York, USA)
Jeremy Gunawardena (Department of Systems Biology, Harvard, USA)
Ben Simons (The Cavendish Laboratory, University of Cambridge)
Jordi Garcia Ojalvo (Departament de Fisica I Enginyeria Nuclear Universitat Politecnica de Catalunya, Terrassa, Spain)
Selected references on this subject
Hayward P, Kalmar T, and Martinez Arias A. (2008)
Wnt/Notch signalling and information processing during development. Development 135, 411-424
Maheshri N, O'Shea EK. (2007)
Living with noisy genes: how cells function reliably with inherent variability in gene expression. Annu Rev Biophys Biomol Struct.36:413-434
Martinez Arias AM, Hayward P. (2006)
Filtering transcriptional noise during development: concepts and mechanisms. Nat Rev Genet. 7, 34-44