トラウマ記憶と解離の治療（推敲）3

老眼が進んだ。０．５では済まなくなり、１．５とか、ひょっとして２とか(わかる人にはわかる)。ナサケけない。

1968年にこんな画期的な実験があったという。（Misanin JR, Miller RR, Lewis DJ. Retrograde amnesia produced by electroconvulsive shock after reactivation of a consolidated memory trace. Science. 1968;160:554. [PubMed]）　ある研究者がラットを使って記憶の実験をした。まずラットに籠の中のボトルから水を飲むという動作を覚えさせた。その後にラットに条件付けを行った。例の、音を鳴らして足にショックを与えるというものである。そしてそのラットを二群に分けた。この条件付けの際に、ショック後直後に頭に電気ショックを与えたグループと与えないグループである。ちなみに頭に電気ショックを与えると一種のてんかん発作のようになり、その直前のことを忘れてしまう。きのう書いたアニソマイシンなどのタンパク合成剤を与えるよりずっと手早くできる。（ただし脳全体にショックを与えるわけであるから、脳のどこの部分の機能を阻害したかということはわからないという欠点はある。）

さてこのような条件付けを二群のラットにしっかり行ったとしよう。当然ながら、電気ショックを与えなかった群のラットは、音を鳴らすとボトルをなめる動作に影響が出た。なめるのが遅くなるのだ。いつ足にショックが加わるかと思うとそわそわするのだろう。それに比べて電気ショックを与えられたラットは音を聞いても平気でボトルを同じペースでなめ続ける。
　さてこの実験は、次の点が面白かった。電気ショックを与えて条件付けした群のラットに、音を鳴らした後に電気ショックを与えたのである。するとこの条件付けを忘れてしまったということだ。条件付けを施した直後の状態と同じことが起きてしまったのである。

この実験の理解のために、読者用（結局は自分のためだ）にこんな例を考えた。このラットと同じことが人間に起きた場合、たとえばてんかん発作を起こした場合、直前に起きたことは覚えられない。記憶が定着していないからだ。しかし昔の記憶については影響がない。ところがある時てんかんの患者さんが、昔あるところでAさんにばったり出会った時のことを思い出していたとする。その直後にてんかん発作を起こしたとしよう。その結果としてその患者さんは、Aさんとのエピソードを、それが長期記憶としてすでに保存されていたにもかかわらず、「忘れて」しまう可能性があるわけだ。（もちろん実際のてんかん発作を持つ人にこういうことが起きるかを確かめたことはない。しかし理屈上はあり得る話だろう。）

ちなみに原典の情報については以下の通り（誰が読むか！！）。

11.3. RECONSOLIDATION ERA

As noted earlier, consolidation was seen as a process achieved only on newly acquired memories with the intent of long-term storage. However, pioneer studies indicated that consolidated memories may undergo a consolidation-like process more than once under certain conditions. In 1968, Misanin et al.¹⁰ habituated rats to lick from a drinking bottle in a conditioning chamber, after which they were trained in a fear conditioning task in which a tone (conditioned stimulus, CS) was paired to a footshock (unconditioned stimulus, US). As a result, a conditioned response was obtained and used as a measure of memory, in this case, a reduced licking rate from the water bottle after the tone onset. They reported that an ECS applied immediately after conditioning disrupted memory consolidation (Figure 11.1b, Group 2). The interesting point arose from Group 3. Those animals were trained but without delivery of an ECS. A day later, the consolidated fear memory was reactivated by presenting the tone again. Immediately after this memory reactivation, an ECS was applied with the surprising result that memory was impaired when tested 24 hours later (Figure 11.1b, Group 3). Notably, ECS was unable to disrupt memory if the tone cue was not presented (Figure 11.1b, Group 4) and the phenomenon was referred as cue-dependent amnesia.¹⁰

Even though these results were at first not replicated,¹¹ they encouraged further (mainly unnoticed) work on the possibility that consolidated memories enter into an active stage upon retrieval. For example, Gordon showed that as occurs with newly acquired memories, retrieved memories are susceptible to disruption in a time-dependent manner.¹² Cue-dependent amnesia was further studied in the active–inactive memory model proposed by Lewis.¹³ who claimed that memories become active under two conditions: when newly acquired and when reactivated by means of retrieval. Any other memory is in an inactive stable state. Recently, cue-dependent amnesia was taken up again and is now referred as reconsolidation.

Reconsolidation proposes that after a memory trace is activated by means of retrieval, it is susceptible to disruption by the same treatments that disrupt memory during consolidation.^14,15 In 1992, Bucherelli and Tassoni¹⁶ reported that inactivation of the parabrachial nuclei by infusions of tetrodotoxin disrupted previously consolidated memories when reactivated. Similarly, Susan Sara’s group reported that infusions of either NMDA or β-adrenergic antagonists (which disrupted LTM when applied after training) disrupted a clearly established memory trace upon retrieval.^17–19 Since then, memory reconsolidation has actively been studied.

The most acknowledged study is the one carried out by Nader and coworkers in 2000.²⁰ This work brought general attention to the reconsolidation phenomenon because of the clean data reported and because of the use of a translational inhibitor that interfered with protein synthesis, considered to be the main cellular substrate for memory consolidation. The experiments were performed in the widely studied fear conditioning task and showed that the same treatment applied under circumstances that disrupt consolidation also impairs memory after retrieval. Similar to the report by Misanin and coworkers, Nader et al. conditioned rats in a tone-foot-shock association but memory was assessed by the percentage of the time that rats were immobile (except for movements required for breathing) to the total time the tone was presented (freezing). The day after conditioning, the protein synthesis inhibitor anisomycin was injected in the amygdala after the tone presentation.

When the subjects were tested 24 hours later, they performed poorly compared to the rats that were not anisomycin-injected (Figure 11.2b). The same treatment was unable to disrupt memory if a retrieval session was not performed (Figure 11.2b, Group 3). The researchers also showed that the effects of anisomycin were time-dependent. When injected 6 hours after memory reactivation, it is unable to disrupt memory. In the years following the Nader study, a wide variety of reports have shown that reconsolidation is indeed a general process achieved in different species and different kinds of memories.^21–30