The ratio of mRNA not selected for polyadenylation (non-poly(A)+ selected) to mRNA selected for polyadenylation (poly(A)+) for the beta-1, alpha-1 and gamma-2 subunits of the GABA(A) receptor complex was examined in rats as a function of age. RNA was extracted from whole brain of rats that were either 0, 1, 3, 5 or over 60 days of postnatal age. Poly(A)+ mRNA was purified by oligo(dT)-cellulose chromatography. Non-poly(A)+ selected mRNA and poly(A)+ mRNA for the GABA(A) receptor beta-1, alpha-1 and gamma-2 subunits were examined by Northern blot analysis using cDNA probes specific for these subunits. Levels of GABA(A) receptor beta-1 subunit mRNA were also examined by solution hybridization analysis with beta-1 riboprobe. Analysis of Northern blots revealed that levels of poly(A)+ beta-1 subunit mRNA were highest at 0 days of age, but decreased and reached adult levels by 5 days of postnatal age. However, levels of the beta-1 subunit message extracted from non-poly(A)+ selected mRNA were not significantly different at any of the ages examined, suggesting the existence of a population of beta-1 subunit mRNA that is not polyadenylated. The age-related discrepancy between beta-1 subunit levels measured in non-poly(A)+ selected mRNA and poly(A)+ mRNA was also observed using solution hybridization analysis. In contrast, levels of both non-poly(A)+ selected mRNA and poly(A)+ mRNA for the alpha-1 subunit of the GABA(A) complex increased from 0 days of age to adulthood. Similarly, levels of both non-poly(A)+ selected mRNA and poly(A)+ mRNA for the GABA(A) receptor gamma-2 subunit increased with age. The findings demonstrate that: (1) the polyadenlyated fraction of mRNA for the GABA(A) receptor beta-1, alpha-1, and gamma-2 subunits is developmentally regulated, and (2) polyadenylated mRNA for the beta-1 subunit is not representative of the entire mRNA population for this subunit, as reflected by the lack of developmental change in levels of beta-1 subunit message extracted from non-poly(A)+ selected mRNA. While the significance of a non-polyadenylated population of beta-1 subunit mRNA remains unknown, it is clear that both non-poly(A)+ selected mRNA and poly(A)+ mRNA should be examined when evaluating developmental changes in receptor subunit expression.