skip to primary navigationskip to content

Ongoing projects

Specification of network components


NB 1-2 lineage visualised using the MultiColorFLPout method [Eva Higginbotham].

Neuroblasts and the lineages they generate are the fundamental developmental modules of the nervous system. How do lineages relate the eventual network? To date exceedingly little is known in this system about the assignment of neurotransmitter identity within and between lineages. Using newly developed tools for reporting on transmitter identity, lineage-specific Gal4 expression lines and a clonal approach we have begun to uncover the pattern of transmitter phenotypes in the developing nerve cord (Diao et al., 2015) (in collaboration with Ben White, NIH, Bethesda; and Haluk Lacin & Jim Truman, HHMI Janelia). 


Organised growth of axons and dendrites

We study the mechanisms that regulate the growth, branching and targeting of postsynaptic dendritic arbors (Ou 2008). We have shown that the terminal phases of growth in both axons and dendrites depend on the responses of individual neurons to positioning cues in the mediolateral (Zlatic 2003; Mauss 2009) and dorsoventral axes of the developing nervous system (Zlatic 2009). We have identified these cues and shown that axons and dendrites respond to them by growing to specific locations within the network in an autonomous, target independent fashion. The implication of these findings is that the terminals of connecting pre- and postsynaptic partners are delivered autonomously to common 'meeting regions'.

Image: Snapshot of live imaging the emergence of connections between presynaptic (red) and postsynaptic (green) partner neurons [Javier Aleman].


Patterns of synaptic connections

At the output level, connectivity in the motor system aligns with a straightforward organizational logic: the dendrites of motoneurons form ‘myotopic’ map of the musculature. In collaboration with the Albert Cardona lab, HHMI Janelia Research Campus, we established the neuronal connectivity network upstream of this myotopic map of motoneuron dendrites. This revealed new connectivity motifs for generating sequences of motor unit activation that are characteristic for larval crawling behaviours (Zwart 2016).


Reconstructed pre-motor network upstream of one of the principal set of motor units [Maarten Zwart]. 

To understand how such patterns of connections form, we developed genetic tools for imaging synapse formation between identified nerve cells in the central nervous system. In collaboration with Felix Evers, Centre for Organismal Studies, University of Heidelberg, we demonstrated the importance dendritic arbors in determining connectivity (Zwart 2013) and of axon-dendritic geography for connectivity (Couton 2015). In collaboration with Marta Zlatic’ lab at HHMI Janelia Research Campus, we are now investigating cellular and molecular mechanisms that underlie the observed specificity of connections.




RP2 motoneuron (blue), making connections with one of its partner pre-motor interneurons (yellow), with synaptic sites visualised using bimolecular fluorescence complementation (GRASP; magenta) [Louise Couton].


Activity dependent structural homeostasis & maturation of network properties

Nervous systems manifest considerable levels of variability. How can networks reliably generate specific outputs in the face or naturally occurring variability? We discovered that nerve cells adjust the size of their dendritic arbors so as to regulate the number of input synapses they receive (Tripodi 2008). In collaboration with the lab of Richard Baines, University of Manchester, we are studying mechanisms, including structural homeostasis of dendrites, which networks use to adjust as they assemble and mature.

By imaging function in the motor system we have precisely identified the point at which the motor network first becomes active. During this early phase network activity is episodic, recurring at regular intervals separated by extended periods of quiescence. Early episodes are unpatterned, but as the episodes continue, elements of coordinated output begin to appear, culminating finally in episodes of well-patterned crawling-like activity (Crisp 2008). We find that early activity is instructive and required for the normal maturation of motor circuitry (Crisp 2011), suggesting that the requirement for tuning and adjustment in developing circuitry is far more widespread than previously supposed.


Oxidative stress - synaptic oxygen species regulate synaptic growth.

High levels of reactive oxygen species (ROS) are a well studied hallmark of ageing and neurodegenerative conditions, leading to cellular damage, and, eventually cell death. Most ROS are a generated as metabolic by-products of mitochondrial respiration. Given that the nervous system is the most energetically demanding organ, we wondered whether neurons might utilise ROS as a readout of their activity. Indeed, we discovered that ROS are both necessary and sufficient for adaptive structural changes of synaptic terminal size. Studying identified neurons in their entirety, we find that adjustments of presynaptic (NMJ) and postsynaptic (dendrites) terminals are coordinated. We further discovered redox sensitive DJ-1b acting as a ROS sensor, and PI3Kinase signalling as the critical downstream pathway (Oswald 2016).


Top: 3D reconstructions of aCC motoneurons from control and after progressive over-activation, leading to progressively smaller dendritic arbors. Below: the opposite growth phenotype (more but smaller boutons) occurs in the periphery at the neuromuscular junction (NMJ) [Matt Oswald].

We are now investigating other sources of activity-induced ROS generation and the molecular pathways by which these bring about distinct structural changes to presynaptic NMJs and postsynaptic dendrites.

This is a longstanding BBSRC funded collaboration with Sean Sweeney and Sangeeta Chawla, University of York, currently translated to a vertebrate system.


Connectomics of memory circuits

Together with Gregory Jefferis (MRC-LMB Cambridge), Scott Waddell (Oxford), Gerry Rubin and Davi Bock (HHMI Janelia Research Campus), funded by the Wellcome Trust, we recently established ‘Drosophila Connectomics’, a consortium to reconstruct the circuitry for learning, memory and memory retrieval.


Left: correlated threshold images from XRM and FIBSEM tomograms. Middle: FIBSEM volume of Mushroom body calyx (in μm); thresholding shows  miniSOG label. Right: Single section and higher magnification inset below; the miniSOG label appears as high contrast/black [Julian Ng].

In parallel, Dr Julian Ng in the lab has been developing new reagents and methods for targeted volume ssTEM acquisition (Ng 2016). New genetically encoded miniSOG-based EM probes, when targeted to specific cells and sub-cellular compartments, allow targeted investigation of pre- and postsynaptic partner neurons at synaptic resolution. X-Ray Microscopy (XRM) offers both quality control and the definition of sub-volumes of interest for subsequent high resolution block face-SEM or FIB-SEM.




Our research is supported by a variety of funding agencies, including The BBSRC, The Wellcome Trust, The Sir Isaac Newton Trust with previous support from The Royal Society, The Gates Cambridge Trust, The Katsia Foundation, Fondation Fyssen, and a Henslow Research Fellowship. 

If you are interested in joining our group, please e-mail Matthias Landgraf to discuss research projects and positions/lab space. We have been successful in obtaining support for promising scientists and are happy to help.


Richard Baines (Faculty of Life Sciences, University of Manchester)

Raymond Budjoso (Department of Veterinary Medicine, University of Cambridge)

Albert Cardona (HHMI Janelia Farm Research Campus)

Jan Felix Evers (Centre for Organismal Studies Heidelberg, Ruprecht-Karls- Universität, 69120 Heidelberg, Germany)

Lee Fradkin (Department of Molecular Cell Biology, Leiden University Medical Center)

Don van Meyel (Centre for Research in Neuroscience, McGill University)

Steve Stowers (Montana State University)

Sean Sweeney (Department of Zoology, University of York)

Gerd Technau (Institut für Genetik, Johannes Gutenberg Universität Mainz)

Davie van Vactor (Department of Cell Biology, Harvard Medical School)

Benjamin White (National Institute of Mental Health, NIH)

Marta Zlatic (HHMI Janelia Farm Research Campus)