Center for Human-Intelligence Research

Cognitive Science, Brain Science, Robotics, Society, Civilization

■ Activities Continuing from FY2019: Background, Rationale, and Goals

Following the previous fiscal year, the center's groups—"Elucidation from Molecular Biology and Developmental Engineering," "Elucidation from Comparative Cognitive Science," "Elucidation from Brain Science," and "Elucidation from Robotics"—will continue to work together in unison. They will explore human intelligence broadly, from the molecular to the behavioral level, through approaches not only from the natural sciences but also from the humanities and social sciences. The project items are as follows.

(01) In research on the induction of human cognitive evolution through the interaction of environment, genes, and neural activity, we will continue to conduct behavioral and imaging analyses (MRI, PET) using genetically modified marmosets, a primate model of disease, to detect behavioral abnormalities related to psychiatric and neurological disorders from an early stage, and prepare to publish the results.

For items (02) through (07) from the FY2019 business report, we will aim to publish them as papers or establish a clear path toward publication as significant achievements related to human intelligence by the final fiscal year.

(07) In the development of marmoset brain function mapping technology using fMRI, we will conduct detailed analyses of the relationship between marmoset gene expression and functionally connected regions.

(08) As part of the development of brain function mapping technology using electron microscopy, we will work on developing a method to visualize neural activity with the world's fastest multi-beam scanning electron microscope.

(09) For cell-type-specific long-term imaging of the marmoset brain using fluorescent calcium imaging technology, we will aim to develop deep-brain function mapping technology in a free-moving environment using an ultra-compact fluorescence microscope weighing only 2 grams.

(10) We will conduct correlative analysis using fluorescence and electron microscopy at the center. We will compile a paper on the analysis results regarding the characteristic axonal degeneration in Alzheimer's disease model mice and the localization of molecules related to early-stage Alzheimer's lesions, using cell and tissue labeling techniques that allow for comparative analysis with electron and light microscopy, and employing the world's fastest multi-beam scanning electron microscope.

(11) For the analysis of post-mortem human brains with Alzheimer's disease, our policy is to acquire actual brains from Alzheimer's patients and conduct analyses of axonal degeneration and the localization of molecules related to early-stage Alzheimer's lesions.

(12) In a project to clarify the characteristics of human intelligence by comparing the neural mechanisms of spatial cognition in eels with those in humans, we will identify global cues for spatial cognition and the selection of these cues, as well as investigate sensory deprivation and the effects of brain damage in fish.

■ New Activity Goals, Content, and Background for FY2020

We will begin the following new items from the next fiscal year.

(13) We will conduct research aimed at elucidating the complex motor control mechanisms in the striatum using deep-brain fluorescent calcium imaging technology.

(14) We will induce neural cells from patient-derived iPS cell lines or genome-edited lines and conduct research using human cells to replicate epilepsy-like symptoms in vitro and to study the mechanisms of Alzheimer's disease onset.

■ Implementation Details, Research Outcomes, and Degree of Achievement for the Fiscal Year's Business Plan

The research outcomes for the fiscal year are described for each research item.

In research on the induction of human cognitive evolution through the interaction of environment, genes, and neural activity, we will continue to conduct behavioral and imaging analyses (MRI, PET) using genetically modified marmosets, a primate model of disease, to detect behavioral abnormalities related to psychiatric and neurological disorders from an early stage, and prepare to publish the results.

>> Using marmoset gene modification technology, we created models for Parkinson's disease and Rett syndrome, which exhibits autism-like symptoms. We conducted behavioral and imaging analyses (MRI, PET), and a paper on the former has been submitted (Kobayashi et al., Submitted). Additionally, we used marmoset gene modification technology to analyze the function of the human-specific gene ARHGAP11B, and the results were published in *Science* (Heide et al., *Science*, 2020). Furthermore, in collaboration with Professor Walsh of Harvard University, we analyzed the promoter function of the GPR56 gene using transgenic marmosets (Murayama et al., *Sci Rep*, 2020).

In the development of marmoset brain function mapping technology using fMRI, we will conduct detailed analyses of the relationship between marmoset gene expression and functionally connected regions.

>> We advanced our analysis of the marmoset visual system and submitted a paper (Kaneko et al., Submitted).

We will conduct research aimed at elucidating the complex motor control mechanisms in the striatum using deep-brain fluorescent calcium imaging technology.

>> Following our work on the primary motor cortex (Kondo et al., *Cell Reports*, 2018), we successfully performed deep-brain fluorescent calcium imaging of the marmoset striatum.

We will induce neural cells from patient-derived iPS cell lines or genome-edited lines and conduct research using human cells to replicate epilepsy-like symptoms in vitro and to study the mechanisms of Alzheimer's disease onset.

>> In collaboration with the Keio University Hospital Memory Center, we have successfully established iPS cells derived from patients with sporadic Alzheimer's disease and MCI, induced them into neural cells, and are conducting pathological analysis. Furthermore, our research group focused on the fact that there are two types (PS1 and PS2) of presenilin (PS), the catalytic subunit that constitutes the γ-secretase complex producing Aβ. Using genome-editing technology, we succeeded for the first time in the world in developing a human neural cell model in which each or both of these catalytic subunits were deleted (Watanabe et al., *eNeuro*, 2021). Examination of the Aβ produced from these human neural cells showed that the γ-secretase complex with PS2 produces more toxic types of Aβ. Additionally, specific immunocytochemical staining revealed that the γ-secretase complex with PS2 was mainly localized in late endosomes. These findings revealed that differences in Aβ production capacity can arise from the type of catalytic subunit of the γ-secretase complex and its intracellular localization.

Number of published papers: 61

*Nature*, *Science*, *Nature Communications*, *Annual Review of Neuroscience*, *PNAS*, *J Biol Chem*, etc.

Number of conference presentations: 9 domestic, 7 international

Achievements in social contribution, such as events

(1) LINK-J Symposium / June 3, 2020 / Web

(2) Patient and Public Involvement Event "Regenerative Medicine: Thinking with Patients and Society" / September 5, 2020

(3) 63rd Annual Meeting of the Japanese Society for Neurochemistry, Public Lecture for University Students / September 12, 2020

(4) First Keio-UCSD Webinar / January 15, 2021)

(5) Second Keio-UCSD Webinar / January 29, 2021

(6) First Keio-Stanford Webinar / January 30, 2021)

To date, with researchers from the Center for Human-Intelligence Research—established under the comprehensive agreement between Keio University and RIKEN—playing a central role, Keio University has been recognized for undertaking important research as a member institution in the "Brain Mapping by Integrated Neurotechnologies for Disease Studies (Brain/MINDS)" project. This project was launched in FY2014 by the Japan Agency for Medical Research and Development (AMED). As a result, a multi-beam scanning electron microscope with 61 beams for the world's fastest wide-area imaging, one of only three such units in regular operation worldwide, was installed at Keio University in 2016. A state-of-the-art, super-resolution fluorescence microscope equipped with 64 detectors was also installed at Keio University in FY2017, and various results utilizing them are now being reported. Furthermore, to carry out cell-type-specific long-term imaging of the marmoset brain using fluorescent calcium imaging technology, the newly developed nVista, an ultra-compact fluorescence microscope weighing only 2 grams, and the nVoke microscope, which also allows for optical stimulation, were introduced in 2017 and 2018, respectively. An upgrade including an autofocus function was implemented in February 2019. This has made it possible to visualize neural activity in the living primate brain with fluorescent probes and to artificially induce neural activity through optical stimulation, with the results also being prepared for publication. In FY2020, due to the COVID-19 pandemic, face-to-face meetings and events were not possible, but we were able to deepen exchanges among members both inside and outside the center through numerous webinars. We were also able to publish many papers, including a functional analysis of the human-specific gene ARHGAP11B using marmoset gene modification technology, which was published in *Science* (Heide et al., *Science*, 2020).