Project Summary
Behavior is movement, and the effective and efficient execution of movement has served as a fundamental
evolutionary force shaping the form and function of the nervous system. Control of the forelimbs to interact with
the world is one of the most essential achievements of the mammalian motor system, yet unfortunately these
behaviors are particularly vulnerable to disease and injury. The execution of skilled limb movements requires the
continuous refinement of motor output across dozens of muscles, suggesting the existence of feedback
pathways that enable rapid adjustments. The temporal delays of peripheral pathways, however, suggest that
sensory feedback alone cannot explain the sophistication of online motor control. In principle, a more rapid
source of feedback would be to convey copies of motor commands internally to the cerebellum to generate
predictions of motor outcome, reducing dependence on delayed sensory information. Yet putative copy circuits
have been difficult to isolate experimentally, leaving their contributions to movement unclear. Mouse genetic
tools offer a means to explore a diverse class of spinal interneurons as a neural substrate for internal copies.
Cervical propriospinal neurons (PNs) receive descending motor command input and extend bifurcating axons;
one branch projects to forelimb motor neurons and the other projects to the lateral reticular nucleus (LRN), a
major cerebellar input, providing an anatomically straightforward means to convey motor copies internally. Yet
how diverse classes of PN-LRN circuits are organized and how they each contribute to distinct elements of limb
behavior remain unclear. Complicating the problem, the field lacks robust ways for deconstructing complex limb
movements into component elements (e.g. reaching, grasping, postural control), and objective means for
quantifying these behaviors in mice. Hypothesis: Discrete classes of PN circuits convey distinct types of spinal
motor copy information to the LRN, each necessary for separate aspects of forelimb control; this functional logic
can be resolved with more quantitative, high-resolution and standardized behavioral assays. To test this
overarching hypothesis, Aim 1 uses molecular-genetic circuit mapping approaches and single-cell RNA-
sequencing to define the anatomical and molecular organization of four classes of PN-LRN circuits. Identifying
the fine-grained structure of these diverse pathways will be essential for establishing how internal copies are
conveyed to the cerebellum to control forelimb behavior. Aim 2 addresses the need for more sensitive and
unbiased behavioral tools by developing novel assays of discrete elements of forelimb behavior and machine
learning approaches for automated quantification of forelimb kinematics. Finally, Aim 3 merges these novel
behavioral approaches with intersectional genetic tools, electrophysiological recording and circuit-specific
perturbation to functionally dissect PN-LRN circuits and define their modular contributions to dexterous limb
control. Ultimately, these studies will yield insight into the function of internal copy circuits throughout the nervous
system, and help to lay the foundation for better diagnosis and treatment of motor deficits.