We have isolated a brain-specific cDNA that encodes a Na+-dependent inorganic phosphate (Po) Subtractive Cloning. Subtractive hybridization was used to identify genes differentially expressed after NMDA exposure (12 hr) of cerebellar granule cells. Rat cerebellar granule cells (P8) were cultured and exposed to NMDA as described (6). Poly(A)+ RNA was isolated by using oligo(dT)-cellulose columns (9) from NMDA-treated (12 hr) (tester) and control (untreated) cerebellar granule cells (driver). The tester mRNA (1 I.g) was used to generate[32P]dCTP-labeled first-strand cDNA (6000 Ci/mmol; NEN; 1 Ci = 37 GBq). Driver mRNA (10 pg) was photobiotinylated with a 300 W Sylvania clear bulb and was used to hybridize to tester cDNAs. Subtractive hybridization was done by using subtractor 1, as described by the manufacturer (Invitrogen). After hybridization, common sequences were removed by adding streptavidin and extracting with phenol/chloroform. Resulting subtracted cDNA was used to screen a Zap II cDNA library constructed from NMDA-treated cerebellar granule cells. Positive clones were rescued, replated individually, and rescreened with labeled, single-stranded cDNAs derived from driver and tester mRNAs. Those cDNAs that were detected only by tester probes, and were not detected by driver probes, were isolated for further analysis. Plasmids were excised from A phage, as described by the manufacturer (Stratagene), digested with EcoRI, and electrophoresed in a 1% agarose gel before DNA blotting. cDNA inserts that were specifically detected by tester probes were sequenced.DNA Sequencing and Sequence Analysis. The nucleotide sequence of the BNPI cDNA clone was determined for both strands. Sequence reactions were done by using doublestranded DNA templates, sequence-specific oligonucleotide primers, fluorescently labeled dideoxynucleotide terminators (Applied Biosystems) and Ampli-Taq polymerase in cyclesequencing reactions modified as described (10). Individual sequences were assembled with an Applied Biosystems Abbreviations: NMDA, N-methyl-D-aspartate; BNPI, brain Na+-dependent inorganic phosphate cotransporter I. tPresent address:

No abstract

A series of neuroanatomical, biochemical, and histochemical studies have been conducted to determine the sources of cholinergic afferents to the main olfactory bulb (MOB) in the hamster. Following horseradish peroxidase (HRP) injections that are restricted to the MOB, retrograde neuronal labeling is observed bilaterally in the anterior olfactory nucleus, locus coeruleus, and raphe nuclei, and ipsilaterally in the ventral hippocampal rudiment, dorsal peduncular cortex, piriform cortex, nucleus of the lateral olfactory tract, anterior pole of the medial septal area and vertical limb of the diagonal band, nucleus of the horizontal limb of the diagonal band (HDB), and hypothalamus. Spread of HRP into the accessory olfactory bulb results in additional neuronal labeling ipsilaterally in the bed nucleus of the accessory olfactory tract, medial amygdaloid nucleus, and bed nucleus of the stria terminalis, and bilaterally in the posteromedial cortical amygdaloid nucleus. Retrograde tracing studies also have been conducted in cases with lesions in the basal forebrain or hypothalamus to assess the extent to which such lesions interrupt fibers of passage from other sources of centrifugal afferents, and the effects of such lesions on choline acetyltransferase (CAT) activity and catecholamine content in the MOB and on acetylcholinesterase (AChE) activity in the forebrain have been evaluated. Lesions in the basal forebrain reduce or eliminate CAT and AChE activity in the MOB in direct relationship to the extent of damage to the HDB. Norepinephrine (NE) content in the MOB also is reduced by basal forebrain lesions, but in relationship to damage of the medial forebrain bundle (MFB). The hypothalamic lesions have no effect on AChE activity in the forebrain or on CAT activity in the MOB, but they eliminate retrograde labeling in the locus coeruleus and raphe nuclei and reduce the NE content of the MOB to undetectable levels. The dopamine content of the MOB is not reduced by any of the lesions. Anterograde tracing studies have been conducted to compare the rostral projection patterns of the HDB with the distribution of AChE activity. Most of the rostrally directed axons travel in association with the MFB. A small component of axons travels in association with the lateral olfactory tract. Within the MOB, the axons terminate predominantly in the glomerular layer and in the vicinity of the internal plexiform layer. The projection and termination patterns of the HDB correspond well with the distribution of AChE activity. These various results indicate that the HDB is the major source of cholinergic afferents to the MOB.

Galanin coexists with acetylcholine in medial septal neurons projecting to the ventral hippocampus, a projection thought to modulate memory functions. Neurochemical lesions of the nucleus basalis-medial septal area in rats impaired choice accuracy on a delayed alternation t-maze task. Acetylcholine (7.5 or 10 ,sg intraventricularly or 1 ,.g microinjected into the ventral hippocampus) significantly improved performance in the lesioned rats. Atropine (5 mg/kg intraperitoneally or 10 ,.g intraventricularly), but not mecamylamine (3 mg/kg intraperitoneally or 20 ,ug intraventricularly), blocked this action of acetylcholine, suggesting involvement of a muscarinic receptor. Galanin (100-500 ng intraventricularly or 200 ng into the ventral hippocampus) attenuated the ability of acetylcholine to reverse the deficit in working memory in the lesioned rats. The antagonistic interaction between galanin and acetylcholine suggests that endogenous galanin may inhibit cholinergic function in memory processes, particularly in pathologies such as Alzheimer disease that involve degeneration of basal forebrain neurons.Galanin (GAL) is a 29-amino acid peptide (1) localized in many regions of the mammalian central nervous system (2-5). Immunohistochemical studies revealed the coexistence of GAL and choline acetyltransferase (ChAT), the synthetic enzyme for acetylcholine (ACh), in the medial septal nucleus and nucleus basalis/diagonal band complex of primates (6, 7).In the rat, GAL coexists with ACh in the medial septal neurons projecting to the ventral hippocampus but not in the nucleus basalis neurons projecting to the cerebral cortex (8, 9). The localization of this peptide in the septo-hippocampal pathway, and the high density of GAL binding sites labeled with 1251I-labeled GAL in the ventral hippocampus (10), raise the possibility of a functional role for GAL in memory processes. To investigate the functional significance of the GAL-ACh coexistence, rats were microinjected with GAL, ACh, GAL plus ACh, or saline before behavioral testing with a t-maze delayed alternation task, in which performance was disrupted by lesions of the nucleus basalis-medial septal area (11). MATERIALS AND METHODSMale Sprague-Dawley rats, 200 g (starting weight), were housed in a temperature and humidity controlled vivarium, with lights on from 0700 to 1900. Lesions of the nucleus basalis magnocellularis (NBM) and medial septal area (MSA) were performed according to the methods of Wenk and coworkers (12). Briefly, rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and intracerebrally microinjected through 31-gauge hypodermic tubing with ibotenic acid (in isotonic phosphate-buffered saline, pH 7.7, shielded from light; Sigma), at five individual stereotaxic sites: 1.3 and 1.7 mm posterior to bregma, bilaterally 2.6 mm lateral to the midline, and 7.5 mm ventral to the surface of the skull with 4 ,ug in 0.4 ,ul over 2.5 min for NBM, and 0.8 mm anterior to bregma, at the midline, and 6.5 mm ventral to the surface of the skull with 6 ,u...