2015 has seen a great variety of excellent research papers in the field of neuroscience. For the very first time, the FENS-Kavli Network of Excellence has made its own selection and voted on the most prominent breakthroughs of the year. They range from groundbreaking high-resolution insights into synaptic, dendritic and engram-specific memory traces, over hitherto undiscovered anatomical connections in the mouse brain, to single-cell RNA sequencing of different brain cells, and, finally, to a study urging for caution when interpreting real-time manipulations of brain circuits.
1 – Synaptic memory traces – come and gone
That learning and memory engage reorganization of dendritic spines has been widely described, but the possibility of selectively labeling and manipulating spines previously activated by learning in vivo has thus far not been possible. Kasai and co-workers1 (Nature, 2015) from the University of Tokyo have now generated a toolkit named “synaptic optogenetics”, in which a probe targetable to dendrites of pyramidal neurons in the motor cortex leads to the expression of a modified form of PSD and a photoactivatable version of the Rac1 protein: While on the one hand PSD accumulates in potentiated (enlarged) spines after activity-driven processes, on the other hand photoactivation of Rac1 shrinks previously activated spines. This approach not only reverses learning-induced spine remodeling in behaving mice, but also erases the acquired motor task, and thus highlights causality between dendritic plasticity and specific behaviors.
2 – Storing new while maintaining old information
The overwhelming majority of synapses that a cortical pyramidal neuron receives connect onto the dendritic tree, yet we still know very little about the specific roles of dendritic signal integration and maintenance in the intact, behaving animal. Cichon and Gan2 (Nature, 2015) at NYU used calcium imaging and learning of different motor tasks to show that dendritic spikes are highly compartmentalized between different dendritic branches of the same neuron, and also highly selective for storing newly learned motor patterns. Together with the demonstration that dendritic inhibition is a crucial factor ensuring such compartmentalization of dendritic spikes, this suggests – as has long been hypothesized – that individual dendritic branches can function as independent storage units of new information, while at the same time allowing to retain previously acquired information in the same neuron.
3 – Anatomy rewritten
By 2015, one would assume that the major anatomical connections in the mouse brain have all been described. Well, not quite. It has long been known that hippocampal area CA1 – a major site of memory formation – monosynaptically connects to the prefrontal cortex (PFC) – a major site of memory storage – and that this connection is important for memory consolidation. Reciprocally, however, a direct connection from the PFC, in particular the anterior cingulate cortex (ACC), had never been found. Now, Deisseroth and colleagues3 (Nature, 2015) from Stanford University not only trace that such an ACC-CA1 connection exists, but also optogenetically prove that this pathway is crucial in mediating memory retrieval in both a fear conditioning and a virtual reality paradigm.
4 – The brain’s heterogeneity at single-cell resolution
In stem cell research, pluripotency can be achieved with and identified by changes in a cell’s transcriptomic signature at single-cell resolution. Using an unbiased approach to systematically examine the genome-wide transcriptional profile of every single cell in the mouse somatosensory cortex, Linnarsson and collaborators4 (Science, 2015) at the Karolinska Institute in Stockholm now provide the first such comprehensive molecular description of cell types in the mouse brain. By doing so, the authors are able to establish a classification of cell types (including neurons, glial cells, and mesenchymal cells) and link this classification with the functional anatomy of the neocortex. This study paves the way for future work on how cell-type specific transcriptomic programs emerge during development and how they are affected by plastic processes throughout their lifetime.
5 – Oh memory, where are thou?
For long, memory consolidation was considered to be dependent upon synaptic plasticity and associated protein synthesis within cells activated during learning, e.g., engram cells. According to this theoretical framework, retrograde amnesia induced by inhibiting protein synthesis during consolidation was thought to stem from the loss of the engram. Yet, Tonegawa and coworkers5 (Science, 2015) from MIT demonstrate that despite protein synthesis inhibition during consolidation clearly inducing retrograde amnesia of contextual fear, the memory was still present in the brain, as optogenetic stimulation of the neuronal ensembles formed by engram cells activated during conditioning was capable of retrieving it. Because these engram cells did not undergo synaptic plasticity during consolidation, this suggests that the long-term storage of the engram depends upon neural ensembles rather than structural plasticity changes in single cells, thereby challenging our understanding of the mechanisms subserving long-term storage in the brain.
6 – Side effects of neuronal circuit manipulations
Over the past decade, ever-refined time-, site- and cell-type-specific reversible manipulations of neuronal functions have led to important insights into the circuitry underlying a vast diversity of behavioral outcomes. Nevertheless, few studies have gone so far as to also study off-target effects on independent functions of downstream circuits. Ölveczky and colleagues6 (Nature, 2015) from Harvard find that transient optogenetic perturbations of the rat motor cortex and the songbird nucleus interface causes impairments in behavioral outputs that are normally not controlled by these targets. Strikingly, after chronic lesions of these same brain areas the behavior recovered spontaneously. An underlying mechanism of such off-target effects could be that the transiently inactivated brain circuit acutely pushes its downstream area below threshold for the required computations. In contrast, long-term inactivation by lesions would give the required brain area time for homeostatic up-regulation of excitability, and thereby prevent the off-target effect. These findings suggest that such unwanted off-target effects ought to be taken into consideration when interpreting transient neuronal circuit manipulations.
1. Hayashi-Takagi, A. et al. Labelling and optical erasure of synaptic memory traces in the motor cortex. Nature 525, 333-338, doi:10.1038/nature15257 (2015).
2. Cichon, J. & Gan, W. B. Branch-specific dendritic Ca(2+) spikes cause persistent synaptic plasticity. Nature 520, 180-185, doi:10.1038/nature14251 (2015).
3. Rajasethupathy, P. et al. Projections from neocortex mediate top-down control of memory retrieval. Nature 526, 653-659, doi:10.1038/nature15389 (2015).
4. Zeisel, A. et al. Brain structure. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq. Science 347, 1138-1142, doi:10.1126/science.aaa1934 (2015).
5. Ryan, T. J., Roy, D. S., Pignatelli, M., Arons, A. & Tonegawa, S. Memory. Engram cells retain memory under retrograde amnesia. Science 348, 1007-1013, doi:10.1126/science.aaa5542 (2015).
6. Otchy, T. M. et al. Acute off-target effects of neural circuit manipulations. Nature 528, 358-363, doi:10.1038/nature16442 (2015).