Connectomics at the Single-Cell Level
Total Page:16
File Type:pdf, Size:1020Kb
RESEARCH HIGHLIGHTS NEUROSCIENCE Connectomics at the single-cell level Trans-synaptic tracing that starts from a virus) encoding Cre–GFP into area V1 in single neuron provides a detailed view of the visual cortex. The virus is taken up by that neuron’s inputs. projection neurons and transported back Trans-synaptic retrograde tracing in to their soma, producing a fluorescent bulk has been highly successful in reveal- signal in the nuclei of sparsely distributed ing the connectome in the mouse brain. neurons in the LGN. The researchers then However, bulk labeling of neurons does electroporate various plasmids into a single not provide a detailed view of the connec- GFP-labeled cell. This is the most difficult tions individual neurons make. For exam- part of the approach, as the LGN is “in ple, with this technology, it is difficult to the middle of the brain,” Roska says. The distinguish whether individual neurons in electroporated plasmids prepare the tar- a given brain region receive inputs from geted cell for infection with an mCherry- multiple cell types or even multiple brain encoding rabies virus tracer that then regions, or whether different neurons in trans-synaptically infects the retinal gan- A cluster of ganglion cells in the retina that the region of interest are targeted by dif- glion cells that connect to the LGN neuron. connect to a single LGN neuron. Reprinted from Rompani et al. (2017) with permission from ferent cell types or input from more than Spread of the rabies virus is restricted by Elsevier. one brain region. the monosynaptic nature of the engineered Botond Roska and his team at the virus, which limits infection and mCherry In this mode, several ganglion cells of the Friedrich Miescher Institute for Biomedical expression to the ganglion cells directly same class and an outlier ganglion cell type Research in Basel, Switzerland, are work- connected to the targeted LGN cell. (although with shared features) connected ing to solve this dilemma in the visual To determine the different classes of to a single LGN neuron. Second, they system, where ganglion cells in the retina ganglion cells that connect to individual found LGN cells that received inputs from convey sensory information to the lateral LGN neurons, the researchers imaged the clusters of different ganglion cell types. geniculate nucleus (LGN) and then the labeled ganglion cells in retinal explants. And finally, they observed LGN neurons cortex. The circuitry in the retina is fairly However, the cells had to be accurately that received binocular input. “It just blew well understood, but this is not the case for traced for determination of the ganglion our mind that…we started to see binocular the LGN, which was thought to be a simple cell class. To do this, Roska and his team cells,” says Roska. relay structure. In light of sparse, yet con- tried automated tracing approaches. “But At this stage, the approach is fairly tradictory, electrophysiological evidence, it’s not bulletproof, and in many of these tedious, but Roska and his team are work- Roska and his team set out to address this tracings the problem is that you get one ing to modify the technology in order to conundrum. “It is very clear that you can mistake and the whole thing is totally achieve higher throughput. They are also only ask [this question] at the single-cell wrong,” says Roska. He therefore used the planning to use their technology in other level,” says Roska. “Once you don’t know services of a company that does tracing for organisms. “The next one will be the mar- a brain region, the only thing that makes electron microscopy, and “we just taught moset,” says Roska. He is interested in sense is either to record all of them or one them how to do tracing…for confocal,” understanding the binocular LGN cells by on e .” says Roska. His co-worker Fiona Müllner better, and thinks that correlating their The researchers developed a method spearheaded the subsequent statistical presence with features of the visual system to visualize the retinal ganglion cells that analysis of the data. in various animals could be informative. projected to single LGN neurons, following The team found three modes that Nina Vogt the one-by-one strategy Roska proposed. described how retinal ganglion cells It took more than two years for Roska’s co- connect with LGN neurons. First, RESEARCH PAPERS Rompani, S.B. et al. Different modes of visual worker Santiago Rompani to work out the they observed a simple relay mode, as integration in the lateral geniculate nucleus revealed technical details. The researchers inject low previously postulated, although “with a by single-cell-initiated transsynaptic tracing. Neuron concentrations of AAV (adeno-associated little bit of a twist,” according to Roska. 93, 767–776 (2017). NATURE METHODS | VOL.14 NO.4 | APRIL 2017 | 335.