We move by contracting and relaxing our muscles; the resulting changes in muscle length and tension are detected by proprioceptive sensory neurons. Therefore, unlike all other senses, proprioception is inexorably bound to movement. The central nervous system uses proprioceptive sensory information to construct a representation of body position and muscle activity which is critical in initiating and refining centrally-driven motor commands. Although the impact of muscle and joint afferent input on spinal reflexes is well characterized, we know considerably less about how proprioceptive information influences voluntary commands in supraspinal motor centers.
The primary transmission pathway for proprioceptive information from the periphery to the motor control centers of the brain engages the spino- and cuneo-cerebellar relay neurons. These two classes of neurons receive proprioceptive information from different regions of the body and transmit this information to the cerebellum. Defining how sensory information is transformed at this first central relay station is critical to an overall understanding of the role of proprioceptive afference in the motor control system.
In the lab, we are approaching these questions at the cellular and network levels using an approach that integrates genetics, physiology, optical-based circuit tracing, and behavioral assays.
Spino-cerebellar neurons integrate proprioceptive sensory, spinal inhibitory, and descending corticospinal inputs. Proprioceptive synapses are formed along the dendrites, inhibitory inputs are found along the somato-dendritic surface, and corticospinal inputs target spino-cerebellar somata. The spatial integration of these inputs is likely to influence the output function of spino-cerebellar neurons. To investigate somato-dendritic integration in proprioceptive relay neurons, we are using a combination of dendritic calcium imaging, in vitro intracellular electrophysiology, and optogenetic stimulation techniques.
Anatomy and in vitro electrophysiology have taught us a great deal about the synaptic partners of the proprioceptive relay neurons, but our grasp of the neuronal activity of the spino- and cuneo-cerebellar system in vivo is quite rudimentary. To gain access to the cuneo-cerebellar neurons in vivo we are developing a whole-animal hindbrain preparation compatible with calcium imaging and electrophysiology. This preparation opens many avenues of research. By simultaneously monitoring presynaptic sensory and postsynaptic cuneo-cerebellar in vivo activity, we will watch the transformation of proprioceptive information across the first central synapse. The use of optogenetic approaches will permit us to examine how selective recruitment or suppression of circuit elements affects the transmission of incoming natural proprioceptive stimuli. Eventually we hope to monitor and test the effects of our optogenetic manipulations in animals during the course of purposeful movement. This approach will be used to characterize the effects of (1) inhibition (2) modulation and (3) descending cortical control of proprioceptive processing in precerebellar neurons.
Taking descending control as an example, spinocerebellar neurons receive cortical input. Our recent findings indicate that these inputs may contribute to a corollary discharge circuit capable of predicting or suppressing self-generated sensory feedback. Many aspects of this descending control remain unknown; we do not know the origins or the in vivo nature of the cortical inputs to the spinocerebellar neurons. Anatomical evidence suggests a similar circuit may exist for external cuneo-spinocerebellar neurons. Our in vivo cuneate preparation allows us to probe the functional role of cortical inputs and test the corollary discharge hypothesis. First, we will map the cortical areas providing input to the external cuneate nucleus. Then we will assess the nature of the descending signals by expressing genetically-encoded calcium indicators in these cortical areas and monitoring calcium changes of the labeled terminals in the external cuneate nucleus. Using optogenetics we can next selectively activate or suppress these inputs while the animal is receiving peripheral proprioceptive input. Eventually, we will use this strategy in moving animals to examine the role of descending control of cuneo-cerebellar pathways in behavior.
In addition to using optogenetics, we will also use genetic methods to probe cortical influence of proprioception. Cortical activity can suppress proprioceptive transmission in spino-cerebellar neurons. Since cortical output to spinocerebellar and motor neurons may be linked, cortical suppression of spinocerebellar neurons may provide an effective means of ignoring self-generated, expected sensory information while enhancing attention to unexpected sensory information. By specifically manipulating cortically-driven GABA and glycine mediated inhibition of spino-cerebellar neurons, we will explore the behavioral consequences of this mode of inhibition and test if the nervous system uses it as a filter for self-generated information.
Spino- and cuneo-cerebellar neurons end as mossy fiber terminations in the cerebellum. In addition to the multiple classes of spino- and cuneo-cerebellar neurons, mossy fiber inputs can originate from vestibular nuclei, the lateral reticular nucleus, and the pons. Granule cells are the primary target of mossy fiber terminals in the cerebellum; consequently granule cells serve as the principal processor of this diverse input to the cerebellum. Granule neurons have ~4 dendrites with each dendrite receiving one excitatory mossy fiber input. The simplicity in number and the diversity of origin of mossy fiber inputs make granule neurons an interesting and tractable system to explore how input convergence creates meaningful neuronal codes. We are characterizing the global and cellular patterns of mossy fiber convergence in the cerebellum as well as how these arrangements affect granule cell output. To this end, we will first determine unique genetic identities of each precerebellar neuronal class; knowledge of the genetic identities of these neurons will facilitate a number of crucial experiments including circuit tracing, optogenetic experiments, and ablation studies.
Currently, we have genetic control over one class of spinocerebellar neurons providing an example of our plan for each precerebellar type. Glial derived neurotrophic factor (GDNF) is selectively expressed in Clarke’s column of the dorsal spinocerebellar system. We have produced several lines of GDNF BAC transgenic mice with the aim of dissecting the role Clarke’s column in motor control. To this end, we have generated a GDNF::diphtheria toxin receptor mouse line that renders only GDNF-expressing neurons susceptible to toxin-induced cell death. The specificity of the genetics and the control of toxin administration permit us to test the behavioral role of Clarke’s column.
Despite much effort, a black box has remained in between sensory input and motor output. To reveal aspects of this black box we will chase sensory inputs into the brain in order to understand how the nervous system changes natural stimuli into a code it can use for motor control.
Since arriving at Janelia in October of 2010, I have been gearing up the lab with microscopes, electrophysiology equipment, and genetic tools. Collaborating with Dan Flickinger and Jason Osbourne in Instrument, Design, and Fabrication, we have developed a multi-photon microscope that enables us to image and perform optogenetic experiments in vivo in the brainstem. Also during this time we have mined the brainstem for specific and selective gene expression patterns. The transgenic core at Janelia has used these results to design and produce mice for the lab which will be instrumental in obtaining genetic control over select populations of precerebellar neurons.
Cerebellar granule cells constitute the majority of neurons in the brain and are the primary conveyors of sensory and motor-related mossy fiber information to Purkinje cells. The functional capability of the cerebellum hinges on whether individual granule cells receive mossy fiber inputs from multiple precerebellar nuclei or are instead unimodal; this distinction is unresolved. Using cell-type-specific projection mapping with synaptic resolution, we observed the convergence of separate sensory (upper body proprioceptive) and basilar pontine pathways onto individual granule cells and mapped this convergence across cerebellar cortex. These findings inform the long-standing debate about the multimodality of mammalian granule cells and substantiate their associative capacity predicted in the Marr-Albus theory of cerebellar function. We also provide evidence that the convergent basilar pontine pathways carry corollary discharges from upper body motor cortical areas. Such merging of related corollary and sensory streams is a critical component of circuit models of predictive motor control. DOI:http://dx.doi.org/10.7554/eLife.00400.001.