Enhanced flexibility of place discrimination learning by targeting striatal cholinergic interneurons

Kana Okada, Kayo Nishizawa, Ryoji Fukabori, Nobuyuki Kai, Akira Shiota, Masatsugu Ueda, Yuji Tsutsui,Shogo Sakata, Natsuki Matsushita & Kazuto Kobayashi

Nature communications (2014)

Doi: 10.1038/ncomms4778

Brief summary: previous studies reported contradictory results about the role of cholinergic interneurons (ChIs) for flexible behaviors. In this study, Okada et al, shown that elimination of ChIs in dorsolateral striatum improved spatial reversal learning.

Previous pharmacologic studies implied the involvement of ChIs in reversal learning. Because of the shortcomings of drug applications (such as dose effects, non-specificity, etc.), contradictory results were reported about the functional role of ChIs for reversal learning. To resolve the conflicts, Okada and colleagues developed cell-type dependent transgenic rats to specifically eliminate ChIs in striatum. Firstly, they trained rats to perform a spatial (with cross maze) reversal learning task, during which ChIs were eliminated either at distinct parts of striatum or at different phases of learning. They found that the ablation of ChIs in the dorsomedial, but not dorsolateral parts of striatum significantly enhanced behavioral performance after reversal, while the initial acquisition learning remained intact. The effects could be consistently observed no matter ChIs were eliminated at the initial or reversal learning phases.  Furthermore, the improvement effect was also dramatic for the extinction learning following the initial learning.

Taken together, the results convincingly demonstrated that ChIs inhibited behavioral flexibility under physiological conditions. On the contrary, it might be critical for maintaining the reward value contingency stably across a long time period.

In the paper shared in 2021-07-05 (Ventral tegmental area GABA projections pause accumbal cholinergic interneurons to enhance associative learning, Nature, 2012), the authors shown that activating GABAergic projections in nucleus accumbens (NAc) could suppress ChIs, and then improve the stimulus-dependent fear conditioning. They interpreted the results in terms of the improved discrimination of CS+ and CS-. Enlightened by the current study, the better learning for the fear conditioning in that study could also be explained by enhanced maintenance of stimulus-outcome associations.

It is an interesting attempt to compare the findings from different studies, and to cross validate the hypotheses/theories they used to explain the findings. The more consistency we can reach, the larger the change we can construct a grand theory. So let’s do more comparisons.