br Materials and methods Rat forebrains were obtained

Materials and methods Rat forebrains were obtained from Sprague–Dawley rats at 2, 7, 14, 21, and 30–35 postnatal days and were homogenized in 1:5 volumes (w/v) of 0.32 M sucrose containing 1 mM EGTA, 1 mM EDTA, 1 μg/ml TPCK, 50 μM leupeptin, and 3 μM antipain with 10 strokes in a glass-teflon homogenizer at 1200 rpm. Subcellular fractionation of tissue was performed as described by Whitaker et al. [34]. Briefly, homogenates were first centrifuged at 1000×g for 10 min to yield the nuclear pellet (P1); the supernatant (S1) was then centrifuged at 10,000×g for 20 min to yield the mitochondrial fraction (P2). S2 was then centrifuged at 100,000×g for 1 h to yield the microsomal fraction (P3). Pellets were resuspended in homogenization medium for SDS-PAGE.
Results Western blots of crude mitochondrial and synaptosomal fractions from forebrains of rats sacrificed at different postnatal days were stained with anti-GluR1 and anti-GluR2/3 Concanamycin A (Fig. 1). Western blots labeled with antibodies to GluR1 exhibited a sharp band with an apparent Mr of 108 kDa in crude mitochondrial fraction (P2), with the lowest levels at PND 2 and a steady rise until PND 21 (Fig. 1A, top, Fig. 2A). Postnatal changes in GluR1 levels in microsomal fractions (P3) were significantly different, with a rapid increase between PND 2 and PND 7 and a small overexpression at PND 7 and PND 14 as compared to adult levels (Fig. 1A, Fig. 2A). In addition, we observed qualitative differences in GluR1 migration pattern between fractions P2 and P3, similar to the changes we previously reported [27]. While P2 showed a sharp, single band corresponding to a protein with an apparent Mr of 108 kDa, P3 fractions exhibited a doublet band corresponding to proteins with apparent Mr of 108 kDa and 105 kDa, respectively. Anti-GluR2/3 antibody immunoreactivity in Western blots from P2 fractions exhibited an ontogenic pattern similar to that obtained with anti-GluR1 antibodies, with a steady increase between PND 2 and PND 21 (Fig. 1B, Fig. 2B). In addition, anti-GluR2/3 antibodies did label a second band with a lower molecular weight in P2 fractions (in particular at PND 2 and PND 7), with an apparent Mr of 103.5 kDa, similar to the one observed in all P3 fractions labeled with anti-GluR2/3. We previously showed that the species of GluR1 and GluR2/3 migrating at a slightly lower Mr than the 108 kDa represented partially glycosylated forms of GluR subunits [27]. Developmental changes in GluR2/3 immunoreactivity in P3 fractions were also similar to those observed with GluR1, demonstrating a more rapid increase between PND 2 and PND 14 and an overexpression above adult levels at PND 14 (Fig. 1B, Fig. 2B). In a number of blots, it was possible to separately analyze the upper and lower bands of the doublet labeled with anti-GluR2/3 antibodies in P2 and P3 fractions. We then plotted the changes in the upper band in P2 fractions, as well as those in the lower band in P3 fractions as a function of age (Fig. 2C). GluR2/3 immunoreactivity in the 108 kDa band of P2 fractions increased about 4-fold between PND 2 and PND 21. In contrast, immunoreactivity in the 103.5-kDa band of P3 fractions increased sharply (almost 3-fold) between PND 2 and PND 7, and was present above adult levels until PND 21, and reached adult levels by PND 30. To better illustrate the differential changes in GluR subunits between subcellular fractions with postnatal development, we plotted the ratio between the values in P3 and P2 expressed as percent of adult values for both GluR1 and GluR2/3 (Fig. 2D). This clearly indicated a large developmental shift in subcellular localization of both GluR1 and GluR2/3. We and others have employed quantitative autoradiography with -AMPA to examine the ontogeny of AMPA/GluR receptors in various regions of the rat brain 8, 13, 26. To determine whether the developmental changes observed in subcellular fractions with Western blotting paralleled those reported in ligand binding studies, we compared data obtained by Insel et al. [13], and Standley et al. [26], with those obtained in the present study. Developmental changes in ligand binding in hippocampus were highly significantly correlated with those in the whole forebrain (Fig. 3A). Moreover, overall binding in hippocampus (derived from Insel et al. [13]), and hippocampal fields alone (field measurements separated from ligand binding in cell body layers of the hippocampus; Standley et al. [26]) exhibited highly significant correlations with data from Western blots of P2 fractions obtained in the present study (Fig. 3B). However, ligand binding measured in the cell body layers of hippocampus did not correlate with data from Western blots of P2 fractions, but was highly significantly correlated with data from Western blots of P3 fractions (Fig. 3C).