Characterization of Dentate Mossy Cell-restricted Genetic Manipulation Mice

早些时候，我们建立了一个条件转基因小鼠系，其中白喉毒素受体在苔藓细胞中使用Cre/loxP系统选择性表达。白喉毒素注射后一周内，我们观察到80%的苔藓细胞在纵轴上消失。在活体局部场电位记录中，我们在海马区没有发现明显或持续的癫痫样放电。有趣的是，TIMM染色未检测到苔藓纤维发芽。这些结果表明，与以前的报道相反，整个肺门区的损伤会导致大量苔藓纤维发芽和癫痫，体内选择性消除苔藓细胞不会引发行为性癫痫或苔藓纤维发芽。今年，我们发现，在体外切片制备中，DT处理的突变体中的齿状颗粒细胞对传入刺激变得过度兴奋，并在这种过度兴奋状态下检测到上下文模式分离的缺陷。我们还评估了海人酸(KA)注射后即刻早期基因(IEG)的表达，假设与对照相比，兴奋性刺激会导致突变体中更多的颗粒细胞释放并激活IEG表达。KA注射诱导突变组Zif268在颗粒细胞中的表达高于对照组。我们还检测了KA诱导的癫痫发作强度。突变组在注射KA后1小时内的累积癫痫评分显著高于对照组。总之，这些结果都表明苔藓细胞消融后颗粒细胞的兴奋性增加。综上所述，我们得出结论，体内苔藓细胞的丢失使颗粒细胞过度兴奋。然而，与休眠篮子细胞假说中隐含的癫痫发生预测相反，它不足以触发苔藓纤维发芽和癫痫放电。也许，除了苔藓细胞的丧失外，其他边缘区域的神经退化，如内嗅觉皮质，也是诱发内侧颞叶癫痫所必需的。这些发现为研究边缘皮质癫痫的发生机制提供了新的见解。 今年，我们为苔藓细胞消融项目进行了几个额外的实验，这是审稿人在手稿修改过程中提出的建议。首先，我们需要有决定性的证据，证明从苔藓细胞到齿状颗粒细胞的局部中间神经元存在主要的前馈抑制投射。其次，我们需要揭示肝门部苔藓细胞消融的时间进程。为了确定苔藓细胞投射到颗粒细胞的性质，我们测量了谷氨酸受体阻滞剂(AP5和NBQX)存在下的自发性IPSCs(SIPSCs)。我们在今年之前已经证明，在没有这些阻滞剂的情况下，在DT处理后5-7天，突变体小鼠来自齿状颗粒细胞的sIPSCs频率比对照小鼠减少。这表明苔藓细胞消融后颗粒细胞的抑制性输入减少。如果在突变脑片中存在谷氨酸能兴奋受体阻滞剂，这种减少就消失了，这可能表明苔藓细胞通过局部中间神经元向颗粒细胞投射前馈抑制。换句话说，谷氨酸能苔藓细胞投射刺激局部抑制性中间神经元，进而抑制苔藓细胞。事实上，我们证实，在存在阻滞剂的情况下，sIPSC的频率在不同的基因型之间没有差异，这表明从苔藓细胞到颗粒细胞的前馈抑制。 为了揭示肝门部苔藓细胞消融的时间进程，我们使用了荧光玉B(FJB)染料，它对细胞消融过程中的神经变性进行了染色。我们发现，治疗后第3-4天，苔藓细胞胞体被FJB染色，到治疗后第7天，这些细胞胞体消失，这一点被腹面苔藓细胞的标志物-抗钙视黄素染色所证实。目前，手稿仍在修改中，同时我们将此数据与其他仍在进行中的免疫细胞化学相结合。

Earlier, we generated a conditional transgenic mouse line in which diphtheria toxin receptor was selectively expressed in mossy cells using the Cre/loxP system. Within one week after diphtheria toxin injection, we observed 80% loss of mossy cells throughout the longitudinal axis. We found no obvious or sustained epilepsy-like discharges in the hippocampus as measured by in vivo local field potential recordings. Interestingly, no mossy fiber sprouting was detected by Timm staining. These results suggested that, in contrast to previous reports showing that lesions of the entire hilar region induce massive mossy fiber sprouting and epilepsy, selective in vivo elimination of mossy cells does not trigger behavioral epilepsy or mossy fiber sprouting. This year, we found that dentate granule cells in the DT-treated mutants became hyperexcitable to afferent stimulation in in vitro slice preparation, and during this hyperexcitable state deficits in contextual pattern separation were detected. We also evaluated the immediate-early gene (IEG) expression in response to kainic acid (KA) injection under the assumption that an excitatory stimulus would cause more granule cells to discharge and activate IEG expression in mutants compared to controls. KA injection evoked Zif268 expression in more granule cells in mutants than in controls. We also examined the KA-induced seizure intensity. The cumulative seizure score of mutants for the hour following KA injection was significantly higher than controls. Together, these results all suggested an increase in granule cell excitability following mossy cell ablation. In summary, we concluded that mossy cell loss in vivo renders the granule cells hyperexcitable. Contrary to the predicted epileptogenesis implicit in the dormant basket cell hypothesis, however, it was insufficient to trigger the mossy fiber sprouting and epileptic discharges. Perhaps, in addition to the loss of mossy cells, neurodegeneration of other limbic areas, such as entorhinal cortex, is necessary to induce medial temporal lobe epilepsy. These findings provide new insights into the mechanisms of epileptogenesis in the limbic cortex. This year we conducted several additional experiments for the mossy cell ablation project, which were suggested by reviewers during the manuscript revision process. First, we needed to have decisive evidence for a major feed-forward inhibitory projection via local interneurons from mossy cells to dentate granule cells. Second, we needed to uncover the time course of mossy cell ablation in the hilus. To determine the nature of the mossy cell projections to the granule cells, we measured spontaneous IPSCs (sIPSCs) in the presence of glutamate receptor blockers (AP5 and NBQX) in hippocampal slice preparation. We had already demonstrated prior to this year that, in the absence of those blockers, the frequency of sIPSCs from dentate granule cells were reduced in the mutants compared to the control mice 5-7 days after DT treatment. This suggested that the inhibitory inputs to granule cells are reduced upon mossy cell ablation. If this reduction disappears in the presence of glutamatergic excitatory receptor blockers in the mutant slices, it would indicate the presence of a feed-forward inhibitory projection from mossy cells to granule cells via local interneurons. In other words, glutamatergic mossy cell projections excite local inhibitory interneurons, which in turn inhibit mossy cells. Indeed, we confirmed no difference in the sIPSC frequency in the presence of the blockers between genotypes, suggesting a feed-forward inhibition from mossy cells to granule cells. To uncover the time course of mossy cell ablation in the hilus we used the dye Fluoro-Jade B (FJB), which stains the neurodegeneration during the process of the cell ablation. We found that by post-treatment day 3-4, mossy cell somata were FJB-stained, and these were gone by post-treatment day 7, which was confirmed by anti-calretinin staining, a marker of ventral mossy cells. Presently the manuscript is still under revision while we combine this data with other immunocytochemistry still under way.