Imaging spinal cord activity in behaving animals

Abstract

The spinal cord is the primary neurological link between the brain and peripheral organs. How important it is in everyday life is apparent in patients with spinal cord injury or motoneuron disease, who have dramatically reduced musculoskeletal control or capacity to sense their environment. Despite its crucial role in sensory and motor processing little is known about the cellular and molecular signaling events that underlie spinal cord function under naturalistic conditions. While genetic, electrophysiological, pharmacological, and circuit tracing studies have revealed important roles for different molecularly defined neurons, these approaches insufficiently describe the moment-to-moment neuronal and non-neuronal activity patterns that underlie sensory-guided motor behaviors in health and disease. The recent development of imaging methods for real-time interrogation of cellular activity in the spinal cord of behaving mice has removed longstanding technical obstacles to spinal cord research and allowed new insight into how different cell types encode sensory information from mechanoreceptors and nociceptors in the skin. Here, we review the current state-of-the-art in interrogating cellular and microcircuit function in the spinal cord of behaving mammals and discuss current opportunities and technological challenges.

Keywords: Astrocyte; Behavior; Calcium imaging; Miniature microscope; Motor system; Multi-photon microscope; Neuron; Nociception; Sensory system; Spinal cord.

Figure 1.

Current approaches and opportunities. (A) Imaging in focally restrained animals with conventional microscopes offers tight control over sensory input and precise readout of motor responses. The image shows a head- and vertebra-restrained mouse on an exercise ball. Two-photon imaging during cutaneous (e.g., tail pinch) stimulation allows real-time readout of sensory- and motor-evoked cellular activity. (B) Imaging in unrestrained animals with wearable miniature microscopes allows use of well- validated sensory and motor tests that require a broad range of animal movement. The image shows a mouse with a 2.5-gram miniature microscope on a linear track. (C-D) Use of transgenic mouse lines (and viral vectors) allows measurement of sensory-evoked calcium activity in genetically defined neurons. (C) Coronal spinal cord section showing GCaMP6-positive excitatory neurons in a Vglut2-Cre x Ai95D transgenic mouse (top). Bottom, two-photon images showing pinch-evoked calcium activity. (D) Coronal spinal cord section showing GCaMP6-positive inhibitory neurons in a Viaat-Cre x Ai95D transgenic mouse (top). Bottom, two-photon images showing pinch-evoked calcium activity. (E) Sensory-evoked calcium activity in AAV9-CaMKII-GCaMP6-transduced excitatory neurons in the spinal dorsal horn of an awake unrestrained mouse imaged with miniaturized one-photon microscopy. Air puff (blue) and pinch (yellow) applied to the animal’s proximal tail activates partially overlapping cell ensembles (filled arrowhead). (F) Wearable miniature microscopes allow repeated measurement of sensory-evoked calcium spiking from the same neurons across days. Top, dorsal view of lumbar spinal cord blood vessels as seen through an implanted optical window 0 (left) and 6 days (right) after laminectomy. Bottom, responsive neurons in the black boxed region (two somata indicated). (G-H) Combined two- and miniaturized one-photon microscopy allows simultaneous readout of sensory-evoked activity in brain and spinal cord of behaving mice. (G) Top left, image of a head-restrained mouse on an exercise ball. Top right, two-photon image showing responsive GCaMP6-expressing neurons in primary somatosensory cortex. Bottom, one-photon image showing concomitantly active GCaMP6-expressing neurons in the spinal dorsal horn. (H) Top, calcium spiking in 24 regions of interest (ROIs) from cortical neurons in response to two cutaneous stimuli (pinch and air puff). Bottom, calcium spiking in 17 ROIs from spinal dorsal horn neurons in response to the same stimuli. ROIs are indicated in G by yellow outlines and numbers. Arrows and dashed vertical lines indicate type and onset of stimuli. (A-B) and (E-F) adapted from (Sekiguchi et al., 2016) with permission from Nature Publishing Group.

Figure 2.

Select challenges. (A) Multiplex imaging in behaving animals promises to allow direct measurement of the cellular and molecular interactions that underlie tissue physiology and pathology. While multi-color imaging over a wide wavelength range (450–650 nm) remains challenging in unrestrained animals, multi-photon imaging is readily adaptable to different indicator combinations. Top, example two-photon fluorescence images from a time-lapse recording in the mouse lumbar dorsal horn, showing jRGECO1a-expressing neurons and GCaMP6-positive astrocytes. Bottom, calcium transients in 9 neuronal and 26 astrocyte regions of interest (ROIs) in response to two cutaneous pinch stimuli of different amplitude. Select ROIs are indicated. Arrows and dashed vertical lines indicate type and onset of stimuli. (B) The ability to perform all-optical interrogation in unrestrained animals promises to shed light onto how activity patterns in different cell types relate to one another and are causally linked to behavior. New miniature microscopes have the potential to perform simultaneous imaging and optogenetic manipulation. Top, coronal spinal cord section showing AAV-mediated co-expression of GCaMP6 (left) and the excitatory opsin C1V1 (center). The overlay is shown on the right. Bottom, pinch- (left) and optically evoked (right) calcium excitation in dorsal horn neurons. (C) Current imaging approaches provide optical access up to lamina V, as illustrated in this schematic. Implantation of micro-optics has been successfully used in the brain to extend imaging depth but remains unexplored in the spinal cord. (D) Standard two- and miniaturized one-photon microscopes permit high-speed, cellular-resolution activity measurements across modest FOVs in behaving animals. New widefield multi-photon microscopes now enable cellular-resolution recordings across millimeter-sized FOVs, potentially facilitating more comprehensive study of how sensory or motor information is normally encoded in the spinal cord at cellular and population levels. Additionally, they will facilitate study of disease conditions and treatment strategies, such as neural stem cell transplantation for improved functional recovery after spinal cord injury (top image). However, frame rates of these microscopes are currently low, complicating motion correction and signal extraction. (C) and (D) adapted from (Sengul et al., 2012) and (Lu et al., 2014) with permission from Academic Press and Elsevier, respectively.