G protein-coupled receptor kinases, beta-arrestin-2 and associated regulatory proteins in the human brain: postmortem...

G protein-coupled receptor kinases (GRKs) and beta-arrestin-2 play a crucial role in the regulation of neurotransmitter receptors in brain. In this study, GRK2, GRK6, beta-arrestin-2 and associated regulatory proteins (Gbeta proteins and protein phosphatase (PP)-2A) were quantitated in human brains (immunodensity with specific antibodies) to assess for postmortem changes (pattern of protein degradation) and to investigate the effect of aging on these regulatory proteins as well as their subcellular distribution (cytosol and membrane fractions). In brain (prefrontal cortex, total homogenate) of healthy subjects (n=14) the immunodensities of GRK2 (r=-0.76), GRK6 (r=-0.64), beta-arrestin-2 (r=-0.57), Gbeta proteins (r=-0.59) and neurofilament (NF)-L (r=-0.64), but not PP-2A, declined markedly with the length of postmortem delay (PMD, 3-81 h). With these linear decay models, the average decreases per 12 h of PMD (from 12 to 72 h) were 7-11% for the various proteins. The immunodensities of GRK2 (r=-0.71), GRK6 (r=-0.61), and beta-arrestin-2 (r=-0.54) in human brain (n=12) also declined with aging (16 to 87 years) and the average decreases per decade (from 20 to 80 years) were 3-5%. In contrast, the immunodensities of PP-2A, Gbeta and NF-L in brain did not correlate significantly with the age of the subject at death (16-87 years). The immunodensities of GRK2/6 and beta-arrestin-2 showed marked individual variations and were strongly reduced after several freeze/thaw cycles. In the prefrontal cortex the subcellular distribution (cytosol/membrane) of the two GRKs differed markedly (GRK2: 60%/40%; GRK6: 5%/95%), and that of beta-arrestin-2 was as expected for a soluble protein (60%/40%). In brains of healthy subjects, the immunodensities of cytosolic GRK2 and beta-arrestin-2 correlated, respectively, with those of membrane-associated GRK2 (r=0.67, P=0.049, n=9) and membrane-associated beta-arrestin-2 (r=0.77, P=0.01, n=9). The results of this study emphasize the importance of examining relevant variables (PMD, age) and potential artifacts (individual variation, freeze-thawing effect) when designing signal transduction studies in neuropsychiatric disorders using the postmortem human brain.