Skip to main content Skip to secondary navigation

Assembly of neural circuits in the mouse brain

Main content start
mouse brain

Assembly of neural circuits in the mouse brain

We have studied a broad range of developmental processes in rodent brains using genetic tools we have developed. Some of these studies extend what we are learning in the fly (Section 1), whereas others explore processes more prevalent in vertebrates. As an example of the latter, activity-dependent processes play a prominent role in wiring the mammalian brain. Single-cell knockout (see Section 4) revealed a cell-autonomous function of the NMDA receptor in aligning dendrites of barrel cortex stellate neurons with their presynaptic inputs following Hebb’s rule (Fig. 2-1). As another example, cerebellar Purkinje cells have highly elaborate and planar dendritic trees, each of which receives presynaptic inputs from tens of thousands of granule cells. Our investigations of Purkinje cell dendrite morphogenesis have highlighted the importance of competitive interactions in dendritic growth and branching (Fig. 2-2).

We are also investigating how a limited number of cell-surface proteins can determine the wiring specificity of a much larger number of neurons. Our studies of hippocampal network assembly have revealed that the same cell-surface proteins, teneurin-3 and latrophilin-2, can serve both as ligands and receptors to mediate attraction and repulsion, and these molecules are likely reused in the assembly of multiple nodes of the hippocampal networks (Fig. 2-3). We are investigating the function of these molecules in the assembly of additional circuits as well as how they work both as ligands and receptors.


Selected publications:

Nakayama AY*, Harms, MB* & Luo L (2000) Small GTPases Rac and Rho in the maintenance of dendritic spines and branches in hippocampal pyramidal neurons.  J Neurosci 20: 5329-5338.

Espinosa JS & Luo L (2008) Timing neurogenesis and differentiation: insights from quantitative clonal analyses of cerebellar granule cells. J Neurosci 28: 2301-2312.

Espinosa JS, Wheeler DG, Tsien RW & Luo L (2009) Uncoupling dendrite growth and patterning: Single cell knockout analysis of NMDA receptor 2B. Neuron 62:205-217.

Joo W, Hippenmeyer S & Luo L (2014) Dendrite morphogenesis depends on relative levels of NT-3/TrkC signaling. Science 346:626-629.

DeNardo LA*, Berns DS*, DeLoach K & Luo L (2015) Connectivity of mouse somatosensory and prefrontal cortex examined with trans-synaptic tracing. Nat Neurosci 18:1687-1697.

Berns DS, DeNardo LA, Pederick DT & Luo L (2018) Teneurin-3 controls topographic circuit assembly in the hippocampus. Nature 554:328-333.

Takeo YH*, Shuster SA*, Jiang L, Hu MC, Luginbuhl DJ, Rülicke T, Contreras X, Hippenmeyer S, Wagner MJ, Ganguli S & Luo L (2021) GluD2- and Cbln1-mediated competitive interactions shape the dendritic arbors of cerebellar Purkinje cells. Neuron 109:489-506.

Pederick DT, Lui JH, Gingrich EC, Xu C, Wagner MJ, Liu Y, He Z, Quake SR & Luo L (2021) Reciprocal repulsions instruct the precise assembly of parallel hippocampal networks. Science 372:1068-1073.

* co-first authors

Fig. 2-1 – Cell-autonomous function of NMDA receptor in aligning dendrites with presynaptic input patterns. (A) Top, a control stellate cell in layer 4 of the barrel cortex confines its dendrites to a single barrel (dashed ovals), and thus receives inputs from thalamocortical axons the represent a single whisker. Bottom, a stellate cell with deleted GluN2B (encoding a subunit of the NMDA receptor highly expressed in developing neurons) in an otherwise GluN2B-expressing background extends its dendrites across multiple barrels. Cells were genetically manipulated and visualized using the MADM method (see Section 4). (B) Interpretation of results in Panel A according to Hebb’s rule. In early development, dendrites of a stellate cell (green) are contacted by thalamocortical axons (TCAs) representing multiple whiskers. If TCAs representing one whisker (blue) provide more input to the stellate cell than TCAs representing another (orange), the stellate cell is more likely to fire action potentials (bars perpendicular to the axon) that match the blue TCA firing pattern. Over time, correlated firing leads to strengthening of the synapses and growth of dendritic branches receiving input from the blue whisker; uncorrelated firing in other dendritic branches leads to destabilization of synapses and pruning of the dendrites representing other whiskers. Adapted from Espinosa et al. (2009) Neuron 62:205-217.


Fig. 2-2 – Competitive interactions in dendrite morphogenesis. (A) Sparse TrkC–/– Purkinje cell (green) has reduced dendritic height as well as total branch number and length, compared to a neighboring TrkC+/– cell (yellow). Cells were genetically manipulated and visualized using the MADM method (see Section 4). Dashed line, pial surface, where normal Purkinje cell dendrite arbors terminate. (B) Compared to normal Purkinje cells (left), sparse GluD2–/– Purkinje cells (right) have fewer branches in the deep molecular layer closer to the cell body, but more branches in the superficial molecular layer. Scale bar, 20 µm. (C) Schematic dendritic trees of the TrkC and GluD2 mutants. Global knockouts of TrkC and GluD2 in all Purkinje cells do not exhibit phenotypes as sparse knockouts, highlighting the importance of competitive interactions. Sparse GluD2 overexpression (OE) causes opposite phenotypes as sparse knockout. Adapted from Joo et al. (2014) Science 346:626-629; Takeo et al. (2021) Neuron 109:489-506.


Fig. 2-3 –  Repulsion and attraction in the assembly of the parallel hippocampal networks. (A) Distribution of Ten3 and Lphn2 proteins in the hippocampal network. Ten3 is highly enriched in the interconnected medial entorhinal cortex (MEC), proximal CA1 (pCA1), and distal subiculum (dSub), constituting the medial network, whereas Lphn2 is highly enriched in the interconnected lateral entorhinal cortex (LEC), distal CA1 (dCA1), and proximal subiculum (pSub), constituting the lateral network. (B) Ten3+ pCA1 axons are guided by repulsion from Lphn2 in pSub and attraction from Ten3 in dSub to target precisely to dSub. Lphn2+ dCA1 axons are repelled by Ten3 in dSub, and therefore are confined to pSub. Adapted from Pederick et al. (2021) Science 372:1068-1073; see also Berns et al. (2018) Nature 549:328-333.