Defining the spinal cord circuitry that controls muscles is key to our understanding of movement and movement-related disorders. However, elucidation of this spinal circuitry has been problematic due to the lack of track tracing methods capable of labeling muscle-specific motoneurons and synaptically-connected interneurons in spinal microcircuits. Recently, modified rabies virus (RV) has become the vector of choice for tracing neural circuits. The attenuated RV was modified by deletion of the glycoprotein (G) gene necessary for virus propagation and neuronal uptake and by the addition of a fluorescent protein gene in its place. When combined with approaches that supply G to the initially infected neurons (starter neurons) through trans- complementation, the RV can then move one synaptic step and repeat the replication/labelling process in synaptically connected neurons. Since the RV always lacks the G-gene and cannot acquire it, virus spread stops at this point in the circuit revealing just monosynaptic connections. We inoculate specific hind limb muscles with RV, retrogradely label the muscle-specific motor neurons and then transynaptically label monosynaptically connected interneurons. This approach holds the promise of revealing premotor networks that modulate function of specific motor pools and even single motoneurons. This information is essential to understand how these connections change after nerve or spinal cord injury, neurodegenerative diseases, or aging. However, experiments in our lab and others soon demonstrated several limitations of the technique when applied to the tracing of muscle-specific premotor interneuronal networks. The virus was, in fact, lethal to motoneurons and spinal interneurons reducing the temporal window for labeling. Second, it is not taken up by adult motor axons. Third, even when we were able to infect large numbers of motoneurons in mature animals the transynaptic transfer inside the spinal cord did not occur. We hypothesized that the virus lethality, a G protein with low neurotropism and the robust microglia reaction around infected motoneurons all contribute to the lack of consistency in neonates and its complete failure in mature animals. In this proposal we aim to test significant modifications to increase the reliability and replicability of the method for revealing premotor spinal networks in neonates and adults. We will specifically test the feasibility of a new strain of virus with a cre- dependent self-inactivation cassette that limits the time of viral amplification in infected neurons and that we show extends motoneuron viability for at least a month. We will combine this with an optimized G-protein that increases neurotoropism and virus packaging and we will interfere with the microglia reaction to overcome the limited transynaptic spread in mature animals. We believe these improvements could be transformative in the field by making the technique robust enough to be used by many labs, augmenting the scope towards analyses of different diseases and allowing validation in many different labs and experimental conditions.
This proposal aims to optimize existing methods using modified rabies viruses for labeling the spinal cord neuronal networks that are synaptically connected to motoneuron pools controlling specific muscles, such that studies can be carried out in neonates with higher reliability and the method can be extended to the adult. A more robust method that can be implemented in adults will permit analyses of how these connections change after nerve or spinal cord injury, neurodegenerative disease or aging. These studies are at present not possible.
