中国科学院心理研究所机构知识库

尽管越来越多的证据表明，GABA能系统参与多种形式的学习记忆行为，但目前对该系统在记忆形成过程中的作用仍有待于进一步探讨。研究者们普遍认为，记忆形成过程由数个相对独立的阶段组成，但如何测量这些阶段的时间参数长期以来未得到很好地解决。Gibbs和Ng(1977)通过系统地观察一些工具药物对小鸡一次性被动回避行为的影响，第一次比较精确地测量了这些阶段的起止时间。Rosenzweig(1992)认为，通过以上这类研究方法不仅可以直接探讨GABA能系统与记忆形成过程各阶段之间的关系，而且可能有助于解决目前关于该系统究竟参与记忆的获得还是巩固的争端。虽然已有一些研究从组织学和分子生物学角度为GABA能系统参与小鸡学习记忆的可能性提供了一定的依据，但迄今在行为水平上的研究仍是一项空白.在此背景下，本研究首次通过系统地考察GABA-A受体激动剂和拮抗剂对小鸡一次性被动回避行为的影响对上述问题进行了探讨。另外，由于近年来发现，GABA能系统和胆碱能系统的相互作用在哺乳动物的学习记忆中具有十分重要的意义，故本研究还对小鸡的这种作用进行了研究。

第一部分的系列实验主要观察了在不同强化水平训练时小鸡一次性被动回避试验的行为特点。当动物采用100%MeA训练时，记忆至少可保持至训练后180min，而用20%MeA训练时动物的记忆仅能保持至训练后20min。但不管采用何种强度训练，动物均能在经一次短暂的训练后学会辨别不同颜色的刺激物。以上结果充分肯定了小鸡一次性被动回避行为在研究记忆形成过程中的优越性。

第二部分的系列实验观察了MUS和BIC对小鸡学习记忆的一般作用。结果表明，MUS和BIC可以分别破坏和促进记忆。在一定的时间范围内给药，两种药物对学习记忆的影响均具有明显的剂量依赖性。另一部分实验则观察了M US和BIC联合用药对小鸡学习记忆的影响。结果表明，MUS对记忆的破坏作用可被BIC所逆转，反之亦然。以上结果证明，GABA能系统对小鸡学习记忆的作用特点与哺乳动物相似.由于M US在训练前后给药均可影响学习记忆，因此支持GA BA能系统参与记忆巩固过程的观点.

第三部分的系列实验系统地研究了以Bn能系统在记忆形成过程中的作用。结果表明，MUS和BIC破坏或促进记忆的时间均落在Gibbs和1}'g(1977)提出的记忆形成过程三阶段模型的ITM阶段，而且其作用至少可持续至训练后120min。这些结果不仅提示GABA能系统主要参与ITM的形成过程，而且有力地支持了记忆形成过程各阶段次序依赖(Sequential-dependent)的观点。

第四部分工作的第一部分实验观察了SCOP对小鸡记忆形成过程的影响，结果证实和扩展了Patterson等(1990)关于胆碱能系统参与ITM的观点.第二部分实验则进一步发现，SCOP可以加强MUS对记忆的破坏作用和部分阻断BIC对记忆的促进作用，从而为GABA能系统通过影响胆碱能系统的活动而作用于学习记忆的观点提供了新的依据.但由于SCOP并不能完全阻断BIC对记忆的促进作用，因此不能排除GABA能系统通过其它机制作用于学习记忆的可能性。

第五部分实验分别观察了MUS, BIC和SCOP的镇静、致惊厥和对运动功能的影响。结果表明，上述三种药物确有以上作用，但其剂量远远大于前面一系列实验中影响学习记忆所需的剂量。由此认为，这三种药物对学习记忆的作用是特异性的。

本研究结果提示，在ITM阶段可能存在着一种依赖于GABA能系统的触发形成LTM的机制。大量研究表明，许多被认为是形成LTM的重要基础的生化变化发生于ITM阶段。因此，这种触发机制可能相当复杂，除GABA能系统外，还有其它许多因素参与其中。对遗忘症病人的研究提示，新近获得的记忆不仅容易受到电击等多种因素的影响，而且可能是向LTM转化的最重要的环节.由此可见，新近获得的记忆与ITM可能具有相似的机制。因此进一步探讨ITM及发生于ITM的触发形成LTM的机制将有助于阐明学习记忆的神经机制，并可能为理解遗忘症的病理机制提供新的思路。

总之，从本研究的一系列实验结果可以得到以下一些结论:1)小鸡的GABA能系统与哺乳动物相似也参与学习记忆，其作用环节是记忆的巩固而不是获得;2) GABA能系统较特异地参与ITM的形成过程，但对LTM的形成也具有重要作用，并认为在ITM阶段可能存在着依赖于GABA能系统的触发LTM形成的复杂机制;3) GAQA能系统对学习记忆的作用可能部分是通过影响胆碱能系统的活动而实现的;4)小鸡一次性被动回避行为是研究记忆形成过程的理想的动物模型。

A considerable literatwre demonstrates that the GABAergic system may be involved in several kinds of learning and memory. However, the precise role of this system in memory formation remains to be determined. The hypothesis that memory formation occurs in behaviorally distinct stages has been widely accepted. A strategy emphasizes the time of onset of amnesia induced by pharmacological interventions, has proven to be possible to allow the determination of the time courses of several different memory stages . This strategy employs a one-trial passive avoidance task in chicks. It has been suggested that investigations into the temporal aspects of the memory effects of GABAergic agents may be of use both in determining the precise role of the GABAergic system in memory formation and in settling the debates on whether this system mediates acquisition of or retention. There is extensive GABA distribution in IMHV, which has been implicated both in imprinting and in passive avoidance learning in chicks. In the present study, we therefore focused on the effects of GABA-A receptor ligands, muscimol and bicuculline, on the time courses of retention in a one-trial passive avoidance task in day-old chicks. Since several lines of evidence indicate that there is an important interaction between the cholinergic and the GABAergic systems in modulation of memory processing, we also studied the effect of scopolamine on the memory effects of chicks injected with muscimol and bicuculline.

In the first group of experiments, the time course of the retention function was investigated in chicks without any pharmacological intervention. Chicks trained with 100% MeA showed high levels of retention at various time points from 5 to 180 min after training. For chicks trained with 20% MeA, retention levels were high up to 20 min after training. These animals did not avoid pecking at the blue beads at any time points, indicating that the chicks were capable of learning to discriminate aversive from neutral stimuli. These results demonstrate that chicks can learn fast and retain memory for quite a long time in a one-trial passive avoidance task, thus providing an excellent model system for researching the time course of retention functions.

In the second group of experiments, dose-response functions and the time windows of action of both muscimol and bicuculline were examined under different reinforcement conditions. When chicks were trained using 100% MeA, muscimol produced amnesia in a dose-response manner, and the time-course of effectiveness of muscimol (25ng) ranged from 10 min before training to immediately after training. Bicuculline injected 10 min before training enhanced retention in animals trained with 20% MeA, and the dose-response curve obtained was expressed as an inverted U shaped function. Bicuculline (100ng) enhanced retention only when injected 5 to 20 min before training. Furthermore, the memory-impairing effect of muscimol could be reversed by bicuculline, and the memory-enhancing effect of bicuculline could be blocked by' muscimol. These results demonstrate that the GABAergic system in the chick brain, like that in mammals, plays an important role in learning and memory processes.

In the third group of experiments, the time courses for the memory-impairing and memory-enhancing effects of muscimol and bicuculline, respectively, were explored. The results showed that the amnesia induced by muscimol (injected 10 min before or immediately after training) developed 30 min after training. Bicuculline appeared to enhance memory in chicks trained with 20%MeA, and the time course of the retention function was very similar to that seen in normal，nonoperated animals trained using 100% MeA. According to the model of memory formation proposed by Gibbs and Ng (1977), there are three sequential-dependent memory stages, of which ITM occurs between 20 and 50 min post -training. The present results thus appear to suggest that the GABAergic system is involved in the formation of ITM. These results also support the hypothesis that the GABAergic system mediates retention rather than acquisition.

In the fourth group of experiments, the interaction of the GABAergic and cholinergic systems were investigated in this task. When injected 30 min before training with scopolamine alone, amnesia was produced in a dose-dependent manner. The time windows of effectiveness of scopolamine was from 30 min to 5 min before training. The, amnesia produced by scopolamine developed 20, min after training. A low dose of muscimol (12.5 ng) alone did not produce significant amnesia, but amnesia did appear when combined with a low, otherwise ineffective dose of scopolamine (6.25 mM). The memory-enhancing effect of bicuculline was partially blocked by scopolamine in a dose-dependent manner. These results were generally consistent with the hypothesis that the memory effects of GABAergic agents resulted from an indirect influence on the activity of the central cholinergic system But GABAergic agents may also directly affect memory processing via other mechanisms since scopolamine could not fully block the memory-enhancing effect of bicuculline.

In the fifth group of experiments, the effects of the above three drugs on sedation，convulsant activity, and locomotor activity were observed. Muscimol at 250 ng, but not at 100 ng; induced sedation at 0.5-15 min after injection. Bicuculliue at 1,000 ng induced convulsant activity in all ten chicks; while. bicuculline at 500ng induced convulsant activity in two of ten chicks observed. The effects of scopolamine on locomotor activity were observed m an open field task at 15 and 30 min after injection, which showed that scopolawine at SOmM (but neither at 25mM nor at 12. SmM) produced significant hyeractivity. These results, combined with those of the previous series of experiments, indicated that the doses of all three. drugs used for the one-trial passive avoidance task were much lower than those inducing the above three non-specific effects, thus supporting the hypothesis that both the GABAergic and the cholinergic systems play specific roles in learning and memory processing.

The main findings in this study indicate that GABAergic system appears to be critical for the formation of LTM,. although it pays a specific role in ITM formation. Thus, it seems reasonable to propose that a GABAergic system-dependent triggering mechanism for LTM formation may occur during the period of ITM. This mechanism may also be related to other events of the biochemical cascade that occurs in ITM but is critical for LTM formation. It is of interest not only for basic researchers who study animal memory, but also for neuropsychologists who work with human memory disorders, because amnesic patients usually exhibit time-limited retrograde amnesia arid show intact STM, but fail to form LTM.

Taken together, the following conclusions can be reached from the present study:1) the GABAergic system in chicks, like that in mammals, is involved in learning and memory processing, mediating retention rather than acquisition; 2) the GABAergic system plays a specific role in the formation of ITM, and there may be a GABAergic system-related triggering mechanism for LTM formation during ITM stage; 3) the memory effect of the GABAergic system is partially related to an indirect influence on the cholinergic system; 4) the described one-trial passive avoidance task does provide an excellent model system for research on memory formation.