Our survival depends upon our ability to observe the world around us accurately and reliably. Disturbances in visual perception occur in a range of psychiatric and neurodevelopmental diseases, including autism, schizophrenia and stroke. To perceive the world, we rely on the cerebral cortex to assimilate sensory information that is relayed through the thalamus.
“How the thalamus communicates with the cortex in a fundamental feature of how the brain interprets the world,” said Elly Nedivi, professor in the Picower Institute for Learning and Memory at MIT.
However, there are few connections between the thalamus and cerebral cortex, confounding neuroscientists as to how we can perceive the world.
Simon Schultz, PhD, professor of neurotechnology at the Imperial College London said, “We’ve known for a long while that there are a surprisingly small number of synaptic inputs that come into the cortex from our senses via the thalamus, compared to the number that originate internally from the cortex itself. But there was always the possibility that those inputs were very strong, so that they could nevertheless have a big effect.”
To resolve this issue, Nedivi and her collaborators within and beyond MIT used several innovative methods and reported their findings in the journal Nature Neuroscience last week “Mapping thalamic innervation to individual L2/3 pyramidal neurons and modeling their ‘readout’ of visual input.” They found thalamic inputs into superficial layers of the cortex are not only rare, but also weak and diverse in their distribution. Yet, they help form a reliable and efficient representation of information in the aggregate.
The authors meticulously mapped every thalamic synapse on 15 neurons in layer 2/3 of the visual cortex in mice and modeled how the input affected each neuron’s ability to process visual information. This is the first study to precisely map all thalamic innervations onto entire cortical neurons in live mice.
The researchers found that wide variations in the number and arrangement of thalamic synapses made them differentially sensitive to visual stimulus features. Therefore, individual neurons could not reliably interpret all aspects of the stimulus, but a small group of neurons could effectively assemble the overall picture, working together.
“It seems this heterogeneity is not a bug, it’s a feature that provides not only a cost benefit, but also confers flexibility and robustness to perturbation,” said Nedivi.
Lead author of the study, Aygul Balcioglu, PhD, a scientist in Nedivi’s lab, said that the research has created a way for neuroscientists to track all individual inputs a cell receives as the inputs arrive.
“Thousands of information inputs pour into a single brain cell. The brain cell then interprets all that information before it communicates its own response to the next brain cell,” said Balcioglu. “What is new and exciting is, we can now reliably describe the identity and the characteristics of those inputs, as different inputs and characteristics convey different information to a given brain cell. Our techniques give us the ability to describe in living animals where in the structure of the single cell what kind of information gets incorporated. This was not possible until now.”
Nedivi’s team focused on layer 2/3 of the cortex because of its relatively high flexibility or plasticity, even in the adult brain. The team used a multicolor two-photon microscopy technique established in Nedivi’s lab, to observe whole cortical neurons using three color tags in the same cell simultaneously, to label thalamic inputs contacting the labeled cortical neurons. Overlap of the labels for thalamic inputs and excitatory synapses on cortical neurons revealed putative connections.
To confirm the presence of thalamic inputs, the team used a technique called MAP (Magnified Analysis of Proteome), invented in lab of Kwanghun Chung, PhD, an associate professor of chemical engineering at MIT. MAP physically enlarges tissue, increasing the resolution of standard microscopes. Rebecca Gillani, PhD, a postdoctoral fellow in the Nedivi lab, with help from Taeyun Ku, PhD, a postdoctoral fellow in Chung’s Lab, combined the new labeling technique and MAP to resolve, count, map, and measure all thalamo-cortical synapses on entire cortical neurons.
The analysis showed thalamic inputs were rather small and accounted for two to 10 percent of the excitatory synapses on individual neurons in the visual cortex. The number of thalamic synapses vary not just at a cellular level, but also across different dendritic branches of individual neurons.
The conundrum of the finding lays in how these weak, sparse and highly variable thalamic inputs form the basis of reliable transfer of visual information.
To address this issue, Nedivi collaborated with Idan Segev, PhD, a professor of computational neuroscience at the Hebrew University in Jerusalem. Segev and his student, Michael Doron, created a biophysical model of the cortical neurons based on the findings from Nedivi’s lab and the Allen Brain Atlas.
This showed, when the cortical neurons were fed visual information, their electrical responses varied based on the variation of their thalamic inputs. Some cells responded to a greater degree to contrast or shape, but no single cell revealed much about the overall picture. However, from the combined activity of a cluster of about 20 cells, the entire visual input could be decoded.
“Heterogeneity imparts a cost reduction in terms of the number of synapses required for accurate readout of visual features,” the authors noted.
Schultz said, “This is an extremely intriguing result. This paper shows (using some really neat new tools for studying brain circuit anatomy) that the thalamocortical synapses to layer 2/3 of the cortex are small, sparse and not even particularly reliable! They are still just enough to get the sensory information across.” (Schultz was not involved in the current study).
Given the small size of thalamic synapses, they are likely to exhibit significant plasticity, Nedivi said. She also said it is possible that benefits of diverse neuronal inputs may be a general feature.
“This work is in my view really in accordance with a picture that has been building up in recent years of our ‘perceived world’ as being very much generated internally in our cortex —and only sparsely updated, when needed, by information coming in from the outside. It’s a lovely paper, on an important question,” said Schultz.
Financial support for the study was provided by the National Eye Institute of the National Institutes of Health, the Office of Naval Research, and the JPB Foundation.