Characterization of Dentate Mossy Cell-restricted Geneti

众所周知，海马区与情景记忆的获得和提取密切相关，而齿状回是海马区感觉输入的中央信息处理器。齿状门区的谷氨酸能苔藓细胞沿海马体纵轴向齿状颗粒细胞和门区GABA能中间神经元提供重要的经常性兴奋性联系。颗粒细胞和肺门中间神经元的轴突投射仅限于板层平面内，苔藓细胞则沿纵轴介导远端颗粒细胞的前馈抑制和反馈兴奋，并可能在决定齿状体内的整体兴奋性方面发挥重要作用。有人认为，在正常情况下，苔藓细胞驱动的颗粒细胞跨板层抑制占主导地位，从而维持颗粒细胞的稀疏编码，但在脆弱的苔藓细胞丢失后，跨层去抑制发生，这可能是损伤后海马功能障碍的原因。事实上，众所周知，肺门苔藓细胞是最容易受到损伤的海马神经元。因此，据推测，在损伤性刺激，如癫痫发作期间，苔藓细胞丢失后，将发生跨板层去抑制。然而，体内苔藓细胞在记忆形成或暴露于强烈刺激期间对齿状回兴奋性的贡献尚未得到解决。 为了了解这种苔藓细胞介导的齿状回网络在体内的作用，我们首先用携带小鼠降钙素受体样受体(Crlr)基因转录调控区的BAC DNA片段和Cre基因构建了Cre重组酶转基因小鼠。其中一个品系Cre#4688仅在1~2月龄的齿状细胞中显示限制性Cre免疫反应(IR)，它不与GAD65/67-IR共定位，而与Calretinin-IR共定位，提示Cre阳性细胞为苔藓细胞。与rosa26LacZ报告小鼠杂交的Cre系X-Gal染色显示，Cre蛋白在齿状苔藓细胞和CA3c细胞中有功能性重组，在小脑核和视前核团中有稀疏染色，表明Cre蛋白在青春期前在CA3c细胞和其他区域瞬时表达。基于NMDA受体在海马区依赖记忆中的关键作用，我们建立了海马苔藓细胞限制性NMDA受体(NR)基因敲除小鼠。用原位杂交和免疫组织化学方法检测到12周龄突变体与NMDA受体R1(NR1)品系杂交后，其齿状体内NR1mRNA和NR1-IR的表达均显著降低。令人惊讶的是，NR1mRNA在CA3区、小脑核团和视前核团中的减少并不明显，很可能是由于CRE介导的NR1基因座重组的相对低效。因此，我们建立的海马区苔藓细胞限制性NR基因敲除系可为苔藓细胞功能的研究提供理想的动物模型。利用这些小鼠，Seiichiro jin de和其他人现在正在开发行为研究策略，以在体内阐明苔藓细胞在获得和检索联想记忆中的上下文专一性方面的NR功能，例如区分上下文中两个相似输入模式的能力(模式分离)。 我们还在构建另一个突变系，在αCaMKII启动子的控制下，转基因表达Cre重组酶介导的可诱导白喉毒素受体(DTR)，该启动子限制了前脑神经元的表达。与表达Cre的小鼠品系Cre#4688杂交后，得到的双转基因突变体将仅在表达Cre的细胞表面表达DTR，仅在前脑表达。由于DTR的表达允许结合DT的B亚单位和随后的受体介导的内吞作用，白喉毒素(DT)的应用可以诱导选择性地消融表达Cre的细胞。我们计划利用这些可诱导苔藓细胞消融小鼠来评估苔藓细胞在齿状突起兴奋性和癫痫发生中的作用。

It is widely known that the hippocampus is critically involved in acquisition and retrieval of episodic memory and the dentate gyrus is the central information processor for sensory inputs in the hippocampus. The glutamatergic mossy cells of the dentate hilus provide an important recurrent excitatory connection to dentate granule cells and hilar GABAergic interneurons along the longitudinal axis of the hippocampus. While the axonal projections of granule cells and hilar interneurons are restricted within a lamellar plane, mossy cells mediate feedforward inhibition and feedback excitation of distal granule cells along the longitudinal axis and may play a significant role in determining the overall excitability of the dentate. It has been suggested that while mossy cell-driven translamellar inhibition of granule cells is dominant in the normal condition, thereby maintaining sparse coding of granule cells, translamellar disinhibition occurs following the loss of vulnerable mossy cells, which may underlie postinjury hippocampal dysfunction. In fact, hilar mossy cells are known to be the most injury-prone hippocampal neurons. Accordingly, it is presumed that translamellar disinhibition would occur following the loss of mossy cells during injurious stimuli, such as epileptic seizure. However, in vivo contribution of mossy cells to dentate gyrus excitability during either memory formation or exposure to intense stimuli has not been addressed. To understand the role of this mossy cell-mediated dentate gyrus network in vivo, we first generated Cre recombinase transgenic mice by using a BAC DNA fragment carrying 5?-transcriptional regulatory region of murine calcitonin receptor-like receptor (Crlr) gene and Cre cDNA. One of the lines, Cre#4688, showed restricted Cre immunoreactivity (IR) only in the dentate hilar cells at 1-2 months of age, which was co-localized not with GAD65/67-IR but with calretinin-IR, suggesting that the Cre positive cells are mossy cells. X-Gal staining of Cre lines crossed with Rosa26 LacZ reporter mice revealed a functional Cre recombination in dentate mossy cells and CA3c cells, with sparse staining in cerebellar nuclei and preoptic nuclei, suggesting that Cre protein is transiently expressed in CA3c cells and other regions before adolescence. We created hippocampal mossy cell-restricted NMDA receptor (NR) knockout mice based on the critical role of NRs on hippocampus-dependent memory. After crossing with a floxed-NMDA receptor R1 (NR1) strain, significant reduction of NR1 mRNA and NR1-IR was found in the dentate hilus of the mutants at 12 weeks of age, using in situ hybridization and immunohistochemistry. Surprisingly, reduction of NR1 mRNA was not evident in area CA3, cerebellar nuclei, and preoptic nuclei, most likely due to the relative inefficiency of Cre-mediated recombination of the floxed NR1 loci. Therefore, our hippocampal mossy cell-restricted NR knockout line could provide an ideal animal model for the study of mossy cell function. Using these mice, Seiichiro Jinde and others are now developing behavioral research strategies to elucidate in vivo NR function of mossy cells in the acquisition and retrieval of context specificity in an associative memory, such as the ability to distinguish two similar input patterns in a context (pattern separation). We are also generating another mutant line with transgenic expression of Cre recombinase-mediated inducible diphtheria toxin receptor (DTR) under control of the alphaCaMKII promoter, which limits expression to neurons of the forebrain. After crossed with the Cre-expressing mouse strain, Cre#4688, the resulting double-transgenic mutant will express DTR in the surface of Cre-expressing cells only in the forebrain. Since expression of DTR allows binding of the B subunit of DT and subsequent receptor-mediated endocytosis, application of diphtheria toxin (DT) induces the selective ablation of Cre-expressing cells. We plan to evaluate the role of mossy cells in dentate excitability and epileptogenesis using these inducible-mossy cell ablation mice.