Researchers have long understood that the brain’s visual system is not entirely pre-programmed from birth — it is shaped by the experiences of infants — yet the team behind a recent MIT investigation was still taken aback by the extent of reorganization they witnessed when they observed this phenomenon in mice for the first time in real-time.
As the scientists at The Picower Institute for Learning and Memory monitored hundreds of “spine” structures that accommodate individual network connections, or “synapses,” on the dendrite branches of neurons within the visual cortex over a span of 10 days, they discovered that merely 40 percent of those that began the journey persisted. Perfecting binocular vision (merging inputs from both eyes) necessitated numerous additions and deletions of spines along the dendrites to finalize a set of connections.
Former graduate researcher Katya Tsimring spearheaded the study, published this month in Nature Communications, which the team claims is the first to track identical connections throughout the “critical period” when binocular vision becomes fine-tuned.
“What Katya accomplished was imaging the same dendrites on the same neurons repeatedly over 10 days in the same living mouse during a crucial developmental phase, to investigate, what occurs with the synapses or spines on them?,” states senior author Mriganka Sur, the Paul and Lilah Newton Professor in the Picower Institute and MIT’s Department of Brain and Cognitive Sciences. “We were astonished by the amount of change observed.”
Extensive turnover
During the experiments, young mice observed black-and-white patterns with lines of distinct orientations and movement directions passing through their visual field. Concurrently, the scientists examined both the structure and activity of the neurons’ main body (or “soma”) as well as the spines along their dendrites. By monitoring the structure of 793 dendritic spines on 14 neurons at approximately Day 1, Day 5, and Day 10 of the critical period, they were able to measure both the addition and loss of spines, and correspondingly, the synaptic connections they contained. Furthermore, by tracking their activity simultaneously, they quantified the visual input the neurons received at every synaptic connection. For instance, a spine might react to a particular orientation or direction of grating, multiple orientations, or might not respond at all. Ultimately, by relating a spine’s structural modifications throughout the critical period to its activity, they aimed to unveil the mechanism by which synaptic turnover refined binocular vision.
Structurally, the researchers noted that 32 percent of the spines visible on Day 1 were absent by Day 5, and that 24 percent of the spines apparent on Day 5 had been added since Day 1. The interval between Day 5 and Day 10 exhibited a similar turnover: 27 percent were removed, while 24 percent were added. All in all, only 40 percent of the spines identified on Day 1 remained on Day 10.
Meanwhile, only four of the 13 neurons they were monitoring that reacted to visual stimuli continued to respond on Day 10. The scientists remain uncertain as to why the other nine ceased responding, at least to the stimuli they previously reacted to, but it is probable they now serve a different purpose.
What are the rules?
Having witnessed this extensive wiring and rewiring, the researchers then inquired what criteria allowed certain spines to endure throughout the 10-day critical period.
Earlier research has indicated that the initial inputs to reach binocular visual cortex neurons derive from the “contralateral” eye on the opposite side of the head (thus in the left hemisphere, the right eye’s inputs arrive first), Sur remarks. These inputs stimulate a neuron’s soma to respond to specific visual properties such as line orientation — for example, a 45-degree diagonal. By the onset of the critical period, inputs from the “ipsilateral” eye on the same side of the head begin to join the fray for visual cortex neurons, enabling some to evolve into binocular.
It is no coincidence that many neurons in the visual cortex are attuned to lines of various directions in the visual field, Sur comments.
“The world is composed of oriented line segments,” Sur observes. “They may be lengthy; they may be short. But the world does not merely consist of formless blobs with indistinct edges. Objects in our environment — trees, the ground, horizons, blades of grass, tables, chairs — are defined by small line segments.”
Since the researchers were monitoring activity at the spines, they could discern how frequently they were active and which orientation triggered that response. As the data accumulated, they found that spines were more likely to persist if (a) they exhibited greater activity, and (b) they responded to the same orientation preferred by the soma. Significantly, spines that responded to both eyes demonstrated higher activity than spines that only reacted to one, implying that binocular spines had a higher chance of surviving than non-binocular ones.
“This observation offers compelling support for the ‘use it or lose it’ hypothesis,” states Tsimring. “The more active a spine was, the more likely it was to be preserved during development.”
The researchers also observed another pattern. Throughout the 10 days, clusters formed along the dendrites where neighboring spines were increasingly likely to be active simultaneously. Other investigations have revealed that by clustering together, spines can combine their activity to produce a greater impact than if they were acting alone.
According to these principles, during the critical period, neurons seemingly refined their role in binocular vision by selectively keeping inputs that reinforced their incipient orientation preferences, through both their activity volume (a synaptic attribute known as “Hebbian plasticity”) and their correlation with nearby spines (an attribute referred to as “heterosynaptic plasticity”). To verify that these principles were sufficient to explain the outcomes they observed under the microscope, they created a computer model of a neuron, which indeed replicated the same patterns as those observed in the mice.
“Both mechanisms are essential during the critical period to drive the turnover of spines that are misaligned to the soma and to adjacent spine pairs,” the researchers wrote, “ultimately leading to the refinement of [binocular] responses such as orientation matching between the two eyes.”
In addition to Tsimring and Sur, the paper’s other contributors are Kyle Jenks, Claudia Cusseddu, Greggory Heller, Jacque Pak Kan Ip, and Julijana Gjorgjieva. The research was supported by funding from the National Institutes of Health, The Picower Institute for Learning and Memory, and the Freedom Together Foundation.