Learning alters neurotransmitter release across axons in a compartmentalized manner
March 14, 2022
Neuronal dendrites carry out computations through compartmentalized signaling, while axons have long been thought to carry signals relatively uniformly to their terminal fields. How axonal compartmentalization influences information flow in the nervous system is not understood. A new study reveals that compartmentalized changes in neurotransmitter release following learning alter the flow of information across the brain circuits that mediate learned behaviors, such as olfactory approach and avoidance.
Neurons consist of basic building blocks, including dendrites, a cell body, and axons. The dendrites are the input side of the neuron, the cell body functions as the "command center" and carries out protein synthesis, and the axon relays output to downstream neurons. Recent studies have shown that axons are anatomically complex and can be subdivided into compartments. Neurons that are involved in learning and memory have axons that exhibit such compartmentalization. These axons have multiple compartments, each of which receives its own inputs and projects to different downstream circuits. The compartments could, in theory, allow learning to modulate the output of the neuron in different ways across each compartment. For instance, reward learning could enhance the activation of downstream circuits that drive reward behaviors, while simultaneously suppressing downstream circuits that drive the opposite aversive behaviors.
How does learning alter release of neurotransmitter from different axonal "compartments"?
To test how learning alters compartmentalized responses of neurons, a new study from the Tomchik lab examined how learning alters release of neurotransmitters from different axonal compartments in neurons that are involved in olfactory learning, Drosophila Kenyon cells. A state-of-the art fluorescent reporter was genetically expressed in Kenyon cells, allowing the researchers to image the release of the neurotransmitter across each of the axonal compartments. The researchers then paired an odor with either a reward (sucrose) or an aversive stimulus (mild shock) - such that the animal learns that the odor is associated with the outcome. Following this learning event, neurotransmitter release was increased from some compartments and decreased in others. The pattern of these changes was not random, however. If the animal underwent reward learning, neurotransmitter release increased in compartments that project to brain regions driving approach behaviors (and decreased in those projecting to brain regions mediating behavioral avoidance). This suggested that the learning events were creating memories that could drive appropriate behavioral responses when the animal encounters the odor again later - a major reason that brains are so apt to form memories.
What are the molecular mechanisms of compartmentalized, learning-mediated plasticity?
To determine how such a process could occur at the molecular level, the researchers tested how manipulating certain signaling molecules affected the neurotransmitter release. In particular, movement of calcium in and out of axon terminals is a major mechanism controlling the neurotransmitter release. When they reduced the level of one channel that calcium flows through into the axon terminal, the cacophony Cav2 channel, increases in neurotransmitter release were no longer observed following learning. Further, the animals were unable to form memories, as revealed in behavioral tests. Additional experiments revealed that a second mechanism of calcium regulation - release of calcium from intracellular stores via IP3 receptors - helps to maintain responsivity of the olfactory pathway over time and therefore plays a different role in learning.
For more information, see the manuscript at eLife.
The study was led by Aaron Stahl, in collaboration with Nathaniel Noyes in the lab of Ronald L. Davis, with additional contributions from Tamara Boto, Valentina Botero, Connor Broyles, and Lanikea King (Tomchik lab), as well as Miao Jing and Jianzhi Zhang in the lab of Yulong Li.