Perforant path long-term potentiation (LTP) in intact mouse hippocampal dentate gyrus increased the neuron-specific, growth-associated protein GAP-43 mRNA in hilar cells 3 days after tetanus, but surprisingly not in granule cells While it is now well established that long-term potentiation (LTP), a physiological model of the memory storage process, activates immediate early genes encoding transcription factors (1, 2), there is no evidence to our knowledge indicating what target genes are activated by these transcription factors. GAP-43, ¶ the presynaptically localized, growth-associated protein whose expression is up-regulated during axonal outgrowth in development and axonal regeneration (3), and whose phosphorylation is increased after LTP (4), is one likely candidate target gene (see also ref. 5). A previous study of chronic recording in intact freely moving rats 3 days after LTP induction revealed that perforant path LTP altered GAP-43 mRNA expression (6). To explore the hypothesis that the alteration in GAP-43 mRNA was related to altered transcriptional activity of the GAP-43 gene in hippocampal neurons after LTP, we first studied the effect of LTP in the intact mouse on the endogenous pattern of GAP-43 mRNA transcript levels with in situ hybridization. We then studied the effect of LTP on promoter activation in transgenic mice bearing a GAP-43 promoter-lacZ reporter construct.
MATERIALS AND METHODSAnimals. Mouse strain C57BL͞6 was obtained from The Jackson Laboratory. Heterozygous GAP-43͞lacZ transgenic mouse line 252 bearing 6-kb 5Ј-flanking and 11-kb first intron sequence driving lacZ reporter gene was bred with C57BL͞6 mice, and homozygous transgenic mice were initially screened by genomic DNA Southern or dot hybridization. Constitutive expression of the transgene in hippocampus in GAP-43͞lacZ transgenic mouse line 252 has been described (7, 8). Thirty-five male homozygous transgenic mice (2 months old) were used in the electrophysiological and histochemical analysis.Electrophysiology. In vivo LTP in anesthetized mice was performed essentially as described previously (9). Potentiated responses were measured by percent baseline population spike amplitude. Mean potentiation of combined 1-, 2-, and 3-day LTP group animals was 255% Ϯ 40.1% (n ϭ 20; mean Ϯ SEM) over the 2-hr recording session and did not show any significant difference from that of wild-type C57BL͞6 mice (229.4% Ϯ 36.3%, P Ͼ 0.10). Mean response of the low-frequency control (LFC) animals during the 2-hr recording period was 101.1 Ϯ 5.5 (n ϭ 15). In the LTP experiment with DL-aminophosphonovalerate (APV; Sigma) ejection, 80 nl (1.6 nmol) of APV dissolved in 20 mM Tris⅐HCl (pH 8.0) was pressure-ejected from the recording micropipette into the dentate gyrus molecular layer, using procedures described earlier (10). APV was delivered 15 min prior to high-frequency tetanic stimulation. Equivalent volumes of 20 mM Tris vehicle were ejected into the same region of the hippocampus in control group animals. Potentiated responses in the molecul...