Transsynaptic Tracing of Respiratory Circuits

The CEES studies were conducted 64–66 h after PRV injections in each animal to maximize polysynaptic transport and minimize immune cell infiltration. It is possible that some of the cells expressing PRV-GFP were glia, as these cells phagocytize the debris from infected and lysed cells. However, glia are unlikely to fluoresce with GFP since we used a less virulent PRV and relatively short incubation times (Rinaman et al., 1993 (link)). Labeling glia through synaptic transfer is unlikely as this has not been observed in glial cells (Rinaman et al., 1993 (link)). The interval between PRV injection and CEES allowed sufficient time for polysynaptic labeling of premotor neurons and spinal respiratory-related interneurons (Dobbins and Feldman, 1994 (link); Lane et al., 2008 (link)). During each CEES experiment, the rat was kept on a water-circulating heating pad to prevent hypothermia. Animals were anesthetized with urethane (1200 mg/kg) and α-chloralose (30 mg/kg). A vertical incision was made ventrally on the neck; the sternohyoid muscles were separated to expose the trachea; a small incision was made between the cartilage rings of the trachea; a short segment of PE 200 tubing was inserted to tracheostomize each animal; and the tracheostomy tube was connected to a pneumotach (Validyne) to record respiratory airflow. Two wires (St. Steel 7 Strand, AM-Systems), with the insulation stripped at the end (2 mm), were inserted bilaterally through abdominal incisions into the lateral costal portion of the diaphragm muscle to record electromyographic (EMG) activity. Each animal was placed prone and a laminectomy was performed to expose C2–C7 spinal cord levels. Dorsal CEES was administered using a stimulating electrode (Tungsten Parylene 0.01, AM-Systems) placed ∼2 mm lateral to the midline on the dorsal surface of the cervical spinal cord, and a ground electrode was placed on the dorsal surface of the spinal cord ∼2–3 mm away from the stimulating electrode. The ends of the stimulating and ground electrodes were stripped, leaving ∼2 mm of each electrode tip uninsulated. CEES was delivered as a continuous 30-Hz monophasic (500-μs pulse width) train of impulses for 30 s (Master 9 AMPI). EMG signals were amplified 1000× and bandpass filtered at 300–1000 Hz before digitization. Diaphragmatic activity was sampled throughout each study at a rate of 2 kHz.
Animals were randomized to receive six trials of 30-s active CEES (n = 9) and sham stimulation or sham (n = 3) stimulation only at the intersection of cervical levels 2 and 3 (C2/3). Sham stimulation trials in the CEES group were performed to control for any effects that the electrode pressure on the dura may have had on respiratory behavior in the absence of current. Experiments in which animals received only sham stimulation were performed as a control for c-Fos expression in the unstimulated condition. To execute the sham trials, the stimulation and ground electrodes were placed on the dura with similar pressure as stimulation trials, but no stimulation was delivered. During sham stimulation (Sham) and active stimulation trials (Stim), data were recorded for 1 min of baseline recording (Pre), 30 s. Stim/Sham, and 8–10 min post-Stim/Sham. Data presented are the 30 s before Stim/Sham (Pre), 30 s of Stim/Sham (Intra), and 30 s of after Stim/Sham (Post). Each animal was allowed to survive for at least 1 h after the mid-way point of stimulation to allow c-Fos expression to develop.