Post-embedding double-gold labeling immunoelectron microscopic co-localization of neurotransmitters in the rat brain

In this report, we describe a new quantitative electron microscopic protocol based on the use of double colloidal-gold post-embedding immunostaining procedure as markers to analyze the subcellular distribution of enkephalin (ENK) and GABA neurotransmitters simultaneously in the same ultrathin tissue sections in the periaqueductal gray area (PAG) of the rat brain.

Double gold particle signals were significantly improved using a number of technical adjustments for the immunogold electron microscopic co-localization technique of ENK and GABA. The GABA-like neuronal elements were immuno-reacted with 20 nm gold particles and the enkephalin-like immunoreactive neurons were labeled with 10 nm gold particles. This double labeling was more apparent in tissue sections that were deactivated for the gold staining of the first antibody. Excellent double labeling was obtained when we blocked antigenicity of the first antiserum with hot (80 degrees C) paraformaldehyde fumes. To minimize the clumping of the second gold particles around the first gold particles used against the first antibody, we tried different staining order for the neurotransmitters tested in this study. It was necessary to use a detergent (Triton X-100) at very low concentration (0.1%) instead of etching to expose the antigenic determinants of the neurotransmitters and at the same time to reduce the deleterious effects on the morphology of the tissue sections. Furthermore, the high-glutaraldehyde fixation and the decrease in the interval between cutting and labeling of the ultrathin sections significantly improved the results obtained in this study.

Double labeling of sections with ENK and GABA produced co-localization in 23.1% and 1.2% of the immunoreactive axonal terminals and dendrites, respectively. Most of the double-labeled terminals contained more GABA-like than ENK-like immunolabeling. Half of the axon terminals [51%] and dendrites [56%] in the ventrolateral PAG were not labeled with either of GABA or ENK immunoreactivity.

This procedure was found to be completely compatible with good double immunolabeling and ultrastructural preservation.