The genetic complexity of cell fate specification...

The Chambers laboratory uses state-of-the-art genomic technologies and associated bioinformatic analyses in combination with well defined neural patterning systems to establish causal links between regionally-restricted gene expression and the acquisition of neural identity.

The findings from these studies will provide important insights into the genetic complexity of cell fate specification and should help to reveal the underlying molecular logic of pattern formation throughout the developing CNS. Together, these data provide a foundation for developing acutely patterned ESC lines for use in restorative medicine.

Laboratory interests

The vertebrate central nervous system (CNS) consists of thousands of distinct neuronal cells, each with unique molecular genetic profiles, organised into complex patterns of circuitry that ensure appropriate functional output. A key goal of modern developmental neuroscience is to unravel the molecular logic that first establishes the diversity in neuronal cell type and subsequently dictates their development into integrated higher order networks.

Recent years have seen considerable progress in establishing the initial framework that underpins the acquisition of neuronal identity. Several lines of evidence have shown that cells are specified in discrete steps within the neuroepithelium influenced by the anteroposterior (AP) and dorsoventral (DV) position of the precursor cell. In response to diffusible molecular cues released by discrete cell groups known as local organisers a series of high level 'selector' genes establish the first molecular co-ordinates that imprint progenitor cells with an outline fate. However, despite this being a well-recognised and investigated paradigm, we still remain relatively ignorant of the repertoire of subsequent molecular 'effectors' that then impart specific facets of neuronal phenotype, behaviour and connectivity. The developing vertebrate hindbrain and the adjacent mid-hindbrain boundary (MHB) signalling centre provide the ideal accessible systems in which to examine the molecular controls underpinning the emergence of neuronal diversity. Exploiting the relationship between early anatomical distinctions and key selector gene activities of these territories we have generated highly representative and validated genome-wide expression libraries for each division of the hindbrain and the caudal portion of the midbrain as well as some associated specific neuronal populations.

These data can be exploited to investigate how neuronal specificity is achieved by the recruitment of effector genes and will afford insights into 3 key areas;

• What and how are the developmental genetic factors used in response the patterning influences of the mid-hindbrain signalling centre?

• For the understanding of how Hox genes and signalling molecules drive downstream networks of effector gene activity

• Driving the differentiation of embryonic-derived stem cells towards specific fates

A more complete understanding of the molecular logic of neuronal specification and how patterning instructions are relayed to a defined set of cellular cues is a prerequisite to expanding our knowledge of nervous system function in health and disease and for developing rational approaches to progressing stem cells towards a defined fate for use in restorative medicine.

Typical Experimental Setup: neuronal patterning

The set of genetic factors that regulate the positioning, establishment, maintenance and signalling properties of the MHB organiser (e.g. Otx2, Gbx2, En1/2, Pax2/5/8, Lmx1b, Grg4, Fgf8, Spty1/2/4, Irx2 & Wnt1) are reasonably well understood. However, the exact nature of how MHB organiser is maintained, signals are transduced, regulated at the cellular level, interpreted by responding cells and translated into instructive cues that define neuronal cell types remain unclear. Candidate genes whose expression is significantly enriched in midbrain (m), r1 or m & r1 (>300 total) represent the likely pool from which further insights can be drawn into the patterning of this region. We have performed a preliminary in situ hybridisation screen on this subset of genes that suggests previously unexplored mechanisms. We concentrate on two aspects of molecular activity associated with the MHB organiser; Firstly on the nature of the molecular machinery influencing signals from the organiser and secondly, how those signals are perceived and instruct a transcriptional response in either the midbrain or r1.

We also use the candidate gene set to ask two related questions about the acquisition of segmental identity in the r2 and r4 region: (1) we look for genetic cues that underpin the similarity of these territories (e.g. adhesive properties) and (2) we look at the set of effectors that drive fine distinctions in the identity of the motor nerves that originate from them (i.e. trigeminal [V] and facial [VII]/contralateral vestibuloacoustic efferent neurons [VIII])(see also Figure 2). In the first instance, we use the segmental disposition of the motor nerves and their characterized rhombomere-specific properties to begin to understand how regional identity is transcriptionally encoded at the genome wide level. Our analyses in mouse have been able to demarcate the set of genes associated with each rhombomere as well as those genes shared by related rhombomeres.

Figure 1. Determining the genetic profile of the mid-hindbrain region during early development.

A. Schematic of the E9.5 MHB region and expression of key patterning genes. B. Outline of experimental methodology involving dissection into m, r1, r2, r3, r4 & r5 regions C. Bioinformatic identification of genes significantly differentially expressed where each coloured line represents the expression profile of a single gene. D. Proof of principle: key midbrain and hindbrain patterning genes are identified and found in the correct compartment. E. Using clustering algorithms candidate genes can grouped into 'expression motifs'. Schematic representation of the major expression motifs (grey = genes enriched in that region compared to others [white]) and numbers present in each group.

The data summarised in Figure 1 can therefore be used to address several questions about patterning of the developing midbrain and hindbrain. Firstly, what set of genes are significantly differentially expressed between m-r5 of the developing neural tube, summarised in Figure 1, panel 3. Second, can these sets of genes be classified into genetic motifs that logically correspond to known signalling and patterning events? Third, what molecular insights can be derived from this dataset to enhance our understanding of the observed pattern of neurogenesis in the developing neural tube?

Typical Experimental Setup: Identification and characterisation of the genetic determinants of cranial motor neuron identity

Although the initial framework of the Hox transcriptional code that patterns the cranial MNs has been established, the exact repertoire of molecular effectors that drive neuronal morphology, axon guidance, muscle target specificity, and specificity as a target for upper motor control remain largely unknown. We have previously compiled a genome-wide gene expression database of individual cranial MNs (V, VII, X, & XII) at key stages of their specification and maturation (E9.5, 10.5, 11.5 & 12.5) (See Figure 2), as well as generating expression profiles from specific sub populations of retrograde traced neurons (V & VII).

Figure 2. Determining the genetic profile of cranial and spinal motor neuron populations.

Comprehensive bioinformatics and regulatory pathway analysis (See Figure 3 for example) together with confirmatory techniques has yielded a clearly defined set of new candidate effectors, across the spectrum of cellular machinery (i.e. cell surface adhesion molecules, receptors, signal transduction components and transcription factors) that may contribute to the developmental program of individual MNs. Furthermore, we have shown that many of these components are expressed only in one subset of cranial MN or another and are thus strong candidates for the molecular effectors of specific neuronal identity. Our investigations focus upon the molecular determinants of cranial nerves V (trigeminal) and VII (facial) where we can readily assess the impact of candidate gene function against the background of well defined neuronal cell settling position (r2 and (r4)r6 respectively), target selection (the myogenic core of branchial arches 1 and 2 respectively) and genetic markers.

Figure 3. Example of gene regulatory networks constructed from microarray expression data.

To determine the functional relevance of the candidates we systematically manipulate the expression of candidate genes, alone or in combination with other factors, and assess the consequences on neuronal behaviour and maturation.