Critical evaluation of the expression of gastrin-releasing peptide in dorsal root ganglia and spinal cord

There are substantial disagreements about the expression of gastrin-releasing peptide (GRP) in sensory neurons and whether GRP antibody cross-reacts with substance P (SP). These concerns necessitate a critical revaluation of GRP expression using additional approaches. Here, we show that a widely used GRP antibody specifically recognizes GRP but not SP. In the spinal cord of mice lacking SP (Tac1 KO), the expression of not only GRP but also other peptides, notably neuropeptide Y (NPY), is significantly diminished. We detected Grp mRNA in dorsal root ganglias using reverse transcription polymerase chain reaction, in situ hybridization and RNA-seq. We demonstrated that Grp mRNA and protein are upregulated in dorsal root ganglias, but not in the spinal cord, of mice with chronic itch. Few GRP(+) immunostaining signals were detected in spinal sections following dorsal rhizotomy and GRP(+) cell bodies were not detected in dissociated dorsal horn neurons. Ultrastructural analysis further shows that substantially more GRPergic fibers form synaptic contacts with gastrin releasing peptide receptor-positive (GRPR(+)) neurons than SPergic fibers. Our comprehensive study demonstrates that a majority of GRPergic fibers are of primary afferent origin. A number of factors such as low copy number of Grp transcripts, small percentage of cells expressing Grp, and the use of an eGFP GENSAT transgenic as a surrogate for GRP protein have contributed to the controversy. Optimization of experimental procedures facilitates the specific detection of GRP expression in dorsal root ganglia neurons.