Observations of HRP labeling following injection through a chronically implanted cannula--a method to avoid diffusion of HRP into injured fibers.

One of the limitations of the horseradish peroxidase (HRP) tracer method is the diffusion of HRP into injured axons resulting in unintended labeling of neurons not terminating in the injection area. To overcome this limitation, an experiment was designed to inject the HRP through an implanted cannula after degeneration and healing had taken place. It was shown that implantation of a cannula into the internal capsule significantly decreased the number of labeled axons in the injection site, thus limiting the unintended labeling of neurons from that injection. When injections followed implantation of the cannula by 24 h or more, fibers damaged by the cannula had healed or degenerated sufficiently that intraaxonal diffusion of HRP into those injured fibers did not occur. A significant difference between control (without the cannula) and experimental (with the cannula) injections was observed. Extensive axonal and neuronal labeling following the control injections was seen at the injection site and caudate nucleus, and in the thalamus and parietal cortex, respectively. Experimental injections resulted in sparse axonal and neuronal labeling evident mostly with the larger injections of HRP.