On March 15, 2018, a research team led by Professor Yi Zhong from the School of Life Sciences at Tsinghua University published a research article entitled Active Protection: Learning-Activated Raf/MAPK Activity Protects Labile Memory from Rac1-Independent Forgetting in Neuron. In this paper, the authors found that learning/training itself activates Raf/MAPK pathway to protect the newly formed labile memory from non-associative disruption. In other words, if such active protection mechanism is activated physiologically or by drug feeding, the English words you just remembered will not be forgotten because of a sudden phone call or a fire.

As early as 1885, the German psychologist Hermann Ebbinghaus proposed the forgetting curve of the human brain for the newly acquired information. From then on, many researchers began the journey of exploration of the mystery of forgetting. In the following hundred years, many researchers used experimental psychology methods to try to find out the mechanism of forgetting, and then proposed a variety of theories to explain forgetting phenomenon. However, up to now, there is not a unified explanation of forgetting. Although forgetting is the main feature of human memory and has been widely studied in psychology, little attention is paid in neurobiology making forgetting become a "forgotten" corner. The possible reason is that forgetting is widely considered as the opposite of memory and is a passive process, thus ignoring the unique molecular mechanism that may exist for forgetting itself.

In 2010, Zhong’s group revealed an active forgetting mechanism: learning itself activates Rac1 pathway to actively forget the newly formed memory (Shuai et al., 2010). Such forgetting mechanism did not affect acquisition of the memory. In contrast, this mechanism specifically affected the rate of forgetting. This research used Drosophila as experimental model and found that Rac1, a small G protein, can regulate the forgetting of labile memory lasting for several hours. This Rac1-dependent forgetting mechanism was reported to be highly conserved in mice in a recent paper of Zhong’s group (Liu et al., 2016). Moreover, Zhong’s group found that consolidated memory also has its specific active forgetting mechanism like labile memory. Such forgetting mechanism is regulated by Cdc42, another small G protein (Zhang et al., 2016). This study suggested that different memory components may have different forgetting mechanisms.

Although inhibiting Racl-dependent active forgetting significantly slowed down labile memory forgetting, the decay still happened, indicating other forgetting mechanisms might coexist. In searching of these forgetting mechanisms, Zhong’s group found an active protection mechanism of labile memory: learning itself can activate Raf/MAPK pathway to protect newly formed memory from forgetting induced by non-associative experience, such as electric shock and temperature changes. Like Rac1-dependent forgetting, Raf/MAPK activity also did not affect memory acquisition but specifically suppressed forgetting of labile memory. Surprisingly, Rac1-depednent forgetting and Raf-suppressed forgetting seemed to be independent and additive. Simultaneously blocking these two kinds of forgetting led to no memory decay more than three hours and reservation of large amount of labile memory more than one day.

Figure 1. Dissection of labile memory forgetting

A resulting question is why animals keep two independent forgetting mechanisms to forget a memory component lasting for only several hours. The authors found that Rac1 pathway is only responsible for forgetting induced by associative experience including interference and reversal learning. The significance of this kind of forgetting can be explained to clear useless memory to keep enough space for memory storage. In contrast, Raf/MAPK pathway is only responsible for forgetting induced by non-associative experience including electric shock and temperature changes. The possible significance of this kind of forgetting may be to stabilize useful memory and prevent them from unexpected disruptions. Together, learning activates Rac1 pathway to actively forget the newly formed useless memory and activates Raf pathway to actively protect useful memory. Such sophisticated design may allow animals to better adapt the rapidly changing environment.

Figure 2. Active forgetting and active protection.

Alzheimer‘s disease, which is a worldwide problem, can cause severe memory impairment. With the failure of many newly developed drugs, whether we can develop tablets to improve memory has become the focus of researchers. In the past, the development of such drugs has focused on improving memory acquisition, and the discovery of forgetting mechanisms independent of acquisition has provided us with new perspectives.

Xuchen Zhang, a Ph.D student, and Qian Li, a postdoctoral fellow, at the School of Life Sciences at Tsinghua University, are the co-first authors of this article. Dr. Qian Li and Dr. Yi Zhong are correspondence authors of this article. Dr. Zhong-jian Liu is a co-director of Qian Li. This study thanked Lianzhang Wang, Dr. Wantong Hu, Yunchuan Zhang, and Bohan Zhao for their help. The authors would like to acknowledge Wenjuan Wang and Yue Sun for their assistance at the Imaging Core Facility, Technology Center for Protein Sciences, Tsinghua University and for assistance in using Imaris software, the Zeiss LSM 710 META microscope, and the Zeiss Airyscan LSM880 super-resolution microscope. This work was supported by grants from the National Science Foundation of China, the National Basic Research Project (973 program) of the Ministry of Science and Technology of China, the Beijing Municipal Science and Technology Commission, the China Postdoctoral Science Foundation, and the Tsinghua-Peking Joint Center for Life Sciences.