The human brain is a dynamic and active organ, constantly processing information in an adaptive way, which allows it to navigate the complexities of the environment. This adaptive ability, so called psychological resilience, ensures that the brain is always striving for maintaining normative brain function in the face of unpredictable and challenging circumstances. In this study, we explore a biomarker of resilience using resting-state functional magnetic resonance imaging (rs-fMRI). Importantly, resilience is an abstract concept presently defined based on self-report questionnaires. We suggest a general definition of resilience as a dynamic state of mind that fluctuates due to external circumstances but maintains a stable level of psychological functioning. In this light, we considered that the stability of mental states should be reflected in neural activity, especially in the default mode network (DMN), which is implicated in mental representations of self-states. Based on this, higher levels of mental stability should be represented by smaller variation of DMN activity. To quantify DMN stability, the multivariate signals in all DMN voxels at each time point were projected in two-dimensional space using multidimensional scaling (MDS). The average distances between state coordinates at consecutive time points were extracted as individual indices of DMN stability. 279 healthy (163 females) completed a rs-fMRI scan, the Connor-Davidson Resilience Scale (CD-RISC), Brief Resilience Scale (BRS), and Resilience Scale for Adults (RSA). Regression models showed that DMN stability indices and age could successfully predict the scores of CD-RISC (r2 = 0.08***) and BRS (r2 = 0.17***). These results are in line with our hypothesis that more stable of internal states represented by DMN activity. This neural-based objective definition of mental state dynamics might be a more accurate and reliable physiological assessment of mental health, which can be applied in the development of targeted interventions that promotes mental health.

Habituation to threatening stimuli is adaptive for animals to pursue valuable goals under threat, yet the underlying neuro-cognitive mechanisms remain elusive. Here, three experiments were performed to elucidate the cerebral mechanism of human pain habituation from an emotional-motivational perspective. We showed, using these experiments, that habituation to painful stimuli at the repetitively stimulated site entailed reduced fear, covaried with individual fear-relevant personality traits, involved amygdala-prefrontal fear inhibitory mechanisms, engaged reward circuitry. These phenomena appeared to emanate from the context in which repetitive stimulation were not associated with harmful outcomes, and the neural networks underlying context-associated fear inhibition via safety (i.e., ventromedial prefrontal cortex and hippocampus) and pain modulation (i.e., periaqueductal grey) before stimulus encounter modulated fear inhibitory circuits during the experience of pain. Of note, pain habituation at the repetitively stimulated site was enhanced when participants were asked to perform a cognitively demanding task, which echoed with the motivational purpose of threat habituation. Moreover, the acquired fear inhibition was transferrable to non-stimulated sites, which involved brain structures implicated in applying encoded safety under threat to relevant contexts (i.e., anterior insula). These results provide an emotional-motivational account for habituation under threat, indicating that this adaptation is for other valuable purposes and relies on context-relevant fear regulation.

RBFOX3 is a important gene which has been used as a post-mitotic neuron maker for a long time. Here, in our lab, we found that Rbfox3 knockout mice show hyperactivity as Attention-Deficit/Hyperactivity Disorder (ADHD)-like behavior. Besides, we also found variations of RBFOX family genes, (RBFOX1, RBFOX2, RBFOX3), in Taiwan ADHD cohort. Hence, we’re going to investigate how RBFOX3 deficit will cause ADHD both in mice and human. In my study, I’d used human pluripotent stem cell (hiPSC)-derived neuron as our model to overcome the unreachable limitation of getting human neurons directly. We’d differentiated hiPSC into cortical neurons and cortical organoids, trying to approach the unknow of ADHD by morphology, molecular, and electrophysiology methods.

Dietary restriction (DR) is widely recognized as a beneficial food regimen to improve physiological conditions and alleviate age-related dysfunctions. Recent studies demonstrated DR exerts several benefits through the microbiota-gut-brain axis. Our current research further established the role of the gut microbiota in DR-induced memory enhancement in middle age mice and aged mice, and lifespan extension. In middle age mice, we adopted two loss-of-function models to examine the role of the gut microbiota: antibiotics-treated mice and germ-free mice. Both approaches demonstrated that DR-induced memory enhancement requires the present of gut microbiota. Moreover, the enhanced memory performance could be accomplished in mice under ad libitum (AL) condition receiving fecal microbiota transplantation from DR mice (FMT-DR). 16s rRNA sequencing identified common bacterial change in DR and FMT-DR. We further confirm the memory enhancing effect of two commonly identified species. Furthermore, serum metabolomic experiments on revealed several overlapped metabolites among DR/AL, FMT-DR/FMT-AL, and probiotics/saline, providing a potential interface between microbial alteration and memory enhancement. In aged mice, DR or FMT treatment starting from 14-month-old mice improves recognition memory and spatial memory. In addition to memory-enhancing effect, both treatments significantly promote lifespan extension, demonstrating that the gut microbiota also mediates lifespan extension effect of DR. Further microbiome and metabolome studies are pending to further decipher the underlying mechanism. Our study would provide a comprehensive understanding of the mechanism underlying DR-induced benefits. The ultimate goal is to identify a suitable intervention to improve memory function as well as mitigate disease-related deficits.

The ventromedial hypothalamic nucleus (VMH) is vital for various innate behaviors. Among multiple neural subsets within the VMH, the estrogen-receptor-1-expressing neurons in the ventrolateral division (VMHvl) are well-known for their involvements in consummatory social behaviors. Another non-overlapping neural subset expresses steroidogenic factor-1, and occupies the dorsomedial VMH (VMHdm). Studies showed that VMHdmSF-1 neurons encode a predator-orientated defensive state. Nevertheless, neuroanatomical evidences suggested that VMHdmSF-1 neurons axonal innervation from other VMH neural subpopulations, and they may receive and process social-related sensory cues. To address the functional involvement of VMHdmSF-1 neurons in social behaviors, we performed cell-type-specific in vivo calcium imaging in freely roaming animals. We revealed that the VMHdmSF-1 neurons were robustly activated by social- but not predator-associated stimuli, with a male-biased conspecific sex representation. In addition, conspecifics with different sexes selectively recruited distinct VMHdmSF-1 neural subsets. Through selective ablation of particular olfactory transmitting pathways, we found that male-biased populational responses of the VMHdmSF-1 neurons depend on VNO-relayed pheromonal signals, which are majorly transmitted to hypothalamus through the bed nucleus of stria terminalis (BNST). By optogenetically silencing the BNST-VMHdm pathway, we could diminish the male-preference among the VMHdm neural population. Moreover, VMHdmSF-1 neuronal activities are highly correlated with social investigative behaviors. Altogether, we suggest that apart from defensive behavioral state, a large portion of f VMHdmSF-1 neurons are capable of encoding conspecific social cues. The conspecific sex representation of VMHdm/cSF1 neurons may prompt animals’ investigative behaviors upon encountering other mice and facilitate proper behavioral decision making by accessing more information. 

Depression is a complex psychiatric disorder caused by chronic stress, affecting more than 300 million people worldwide and resulting in negative emotions and cognitive impairments. Novel strategies are urgently needed, as current treatments like selective serotonin reuptake inhibitors and ketamine-based antidepressants only address emotional responses and are less effective in managing cognitive impairments. In this study, we used a repeated social defeat stress mouse model to evaluate the cognitive effects of dietary restriction (DR), defined as 60% of daily food intake compared to ad libitum (AL) feeding. Our results showed that DR induced anti-depressant-like behaviors and improved memory performance in mice, accompanied by altered neuronal structure and increased neurogenesis in the dentate gyrus of the hippocampus. Importantly, the neurobehavioral changes induced by DR were abolished in mice treated with antibiotics and could be mimicked in AL-fed mice receiving fecal microbiota transplantation (FMT) from DR donors, suggesting the critical role of gut microbiota in mediating the beneficial effects of DR. We further used 16s rRNA sequencing and LC/MS-based serum metabolomics to identify commonly changed bacterial composition and metabolites in both DR and FMT-DR groups, revealing certain probiotics and postbiotics that had positive impacts on memory performance. Our ongoing work aims to unravel the underlying mechanism behind the pro-cognitive effects of DR. We believe that our study will have a significant impact on the future treatment of mental dysfunction associated with depression.

The gut microbiota is known to play a critical role in regulating multiple behaviors in Drosophila, including courtship, aggression, locomotion, sleep, and memory. To investigate the specific mechanism by which gut microbiota influences long-term memory (LTM) formation, we developed a model of chronic ovipositional depression induced by parasitoid wasp exposure. Our results demonstrate that intact LTM genes, such as amnesiac, rutabaga, and dunce, are essential for the sustained decrease in oviposition observed for several days following acute wasp exposure. Both antibiotic-treated and germ-free flies showed acute ovipositional depression on the day of exposure but failed to form LTM in the absence of gut microbiota. Interestingly, gut microbiota transplantation was able to rescue the impaired LTM, suggesting that gut microbiota mediates LTM formation without affecting learning in Drosophila. Metabolomics analysis of fly hemolymph revealed a significant alteration in amino acid biosynthesis, and we found that wasp-induced LTM was abolished in flies with a defective amino acid sensing molecule, target of rapamycin. Our ongoing experiments aim to determine the critical metabolites and underlying molecular mechanisms responsible for LTM formation.

Dietary restriction (DR) is a dietary regimen that reduces food intake by 20-40% compared to normal ad libitum (AL) feeding. Recent studies have shown that DR can extend lifespan and reduce aging-related diseases in different model organisms. In addition, DR also enhances memory function by modulating the mTOR signaling pathway in mice brains. Our recent work found that DR improves mice memory by changing the composition of gut microbiota and is independent of the vagus nerve that connects the gut to the brain. Accordingly, we hypothesize that modulated gut microbiota may generate beneficial metabolites that confer pro-cognitive activity through the circulatory system. We began with serum metabolomic analysis by an untargeted metabolomics strategy using liquid chromatography-mass spectrometry (LC-MS/MS). A few metabolites that highly correlated with improved memory performance were identified and validated through water supplementation to mice. In summary, we discovered beneficial metabolites from DR-associated gut microbiota, and we identified a group of metabolites that are involved in memory enhancement.

The ability to pursue good fortune and avoid disaster is vital for organisms to survive. In nature, a high percentage of Drosophila have been attacked by Leptopilina parasitic wasps, which inject their eggs into Drosophila larvae during their larval stage. To combat this predator threat, Drosophila have developed several responses, including innate immune responses and behavioral defenses, such as suppressing oviposition upon seeing wasps in adult female flies. Our laboratory has developed a new paradigm of non-associative learning and memory in Drosophila through an interesting interaction between wasps and flies. We exposed wasps to Drosophila for 24 hours and discovered that the ovipositional suppression can last for several days after the removal of wasps. This long-lasting ovipositional repression appears to be mediated through long-term memory (LTM) formation. To better understand this phenomenon at a circuit level, we began to manipulate neuronal activity through several different genetic approaches. We discovered that NPF neurons are necessary for Drosophila to react to the threat upon seeing Leptopilina wasps, and artificially boosting the neural activity of NPF neurons enhances LTM formation. Moreover, NPF neurons further communicate with dopaminergic neurons to influence mushroom body-mediated LTM formation. Future studies will focus on the downstream targets in order to map the neural circuit induced by wasp exposure.

Spatial navigation is one of the cognitive functions compromised in aging. Older adults require more time to construct a cognitive map and make more mistakes reaching target locations. Previous studies implicate inaccurate distance judgement as a cause. Older adults underestimate actual distances more than younger adults as actual distances increase. Such underestimation in distance judgement imply that older adults may have distortions in mental representations of space. In the present study, we investigated the neural correlates of distance representation in younger and older adults using a distance judgement paradigm within a virtual environment with eight landmarks. Participants freely navigated the virtual environment and encoded the map to criterion. After the cognitive map is acquired, participants were tested on the environment while undergoing functional magnetic resonance imaging. Specifically, for a given start landmark, participants provided their estimates of distances to target landmarks and also navigated there. Compared to younger adults, older adults under- and over-estimated closer and farther distances, respectively. Also, representation similarity analysis showed there were more similarity in visual system on older adults. In conclusion, our findings show how the shift in neural navigational strategy underlies older adult distance representations during spatial navigation.

RTL1/PEG11, which has been associated with anxiety disorders, is a retrotransposon-derived imprinted gene in the placenta. However, imprinting patterns and functions of RTL1 in the brain have not been well-investigated. We investigated Rtl1 imprinting status in mice by immunohistology. According to the result, most of the brain regions that expressed Rtl1 is paternal. Quantification of Rtl1-expressing brain regions showed highly expressed levels in locus coeruleus (LC), ventral tegmental area (VTA), and substantia nigra (SN). Importantly, the results of mice behavior tests show PKO mice have more anxiety-like behaviors. Rtl1 function in mice brain is associated with emotion-related behavior and stress regulation. Based on the previous findings, we try to precisely investigate which brain region does Rtl1 modulates anxious behavior. To achieve the goal, we used CRISPR Cas9 system with AAV to specific knockout Rtl1 in the target brain region. Meanwhile with timescales as parameters to rescue or knockout Rtl1 in mice brain that help we know more about the story that covers under Rtl1 functional deficit. Furthermore, which anxiety-related pathway is Rtl1 involved in the brain and the under mechanism of Rtl1 plays are the topic we are curious about. This study can help us knowing the mechanism which modulates Rtl1 and the role of RTL1 in the brain in more details.

Rbfox3 encoded a member of the RNA-binding FOX protein family which regulated alternative splicing of pre-mRNA. Some studies suggested that RBFOX3 mutation was related to neurodevelopmental delay, cognitive impairments, and epilepsy. Previous studies from our lab reported that Rbfox3 conventional knock-out mice had some deficits in seizure susceptibility, spatial reversal learning, long-term depression (LTD) of the dentate gyrus (DG), adult neurogenesis of the DG, and Attention Deficit/Hyperactivity Disorder (ADHD)-related phenotypes like hyperactivity and impulsivity. We found that seizure was associated with Rbfox3 deletion in GABAergic neurons. Therefore, we aimed to investigate whether the deletion of Rbfox3 in excitatory neurons affected those deficits except for epilepsy.

Human sensory perception stems from a proactive interplay between expectations based on past experiences and sensory inputs. Taking into account that anticipating imminent pain is a salient emotional event that receives priority in processing, it is plausible that the emotional responses elicited by pain expectancy may modulate its integration with nociceptive inputs to underlie pain perception. To confirm this putative modulatory role of emotion, we asked participants to down-regulate expectation-elicited emotions when they anticipated painful stimulation, and adopted a Bayesian approach to analyze behavioral and brain fMRI responses. We found that pain modulation by expectations of increased pain was predicted by anticipatory anxiety, which involved functional coupling between the amygdala and anterior cingulate – two key structures of anxiety processing – to integrate pain expectations and nociceptive inputs. This was accompanied by no updating of pain predictions via suppressed expression of prediction errors in the anterior cingulate, making negative expectation effects resistant to extinction over time. By contrast, for pain modulation by expectations of decreased pain, this integration was mainly linked to anticipatory pleasantness, which engaged functional coupling between the medial orbitofrontal cortex and anterior hippocampus – two regions implicated in positive emotion processing and inhibition of threat responding via safety. These findings suggest that stimulus expectancy effects on pain involve emotion-modulated Bayesian integration, explaining why chronic pain is frequently accompanied by both dysregulation of emotion and aberrant expectations and perception to aversive stimuli.

Empathy, the capacity to share the feelings of others with self-other distinction, plays an important role in human emotional experiences and social interaction. Since empathy comprises self-other distinction, social comparison, a process that we compare a target (i.e., ourselves) to a certain standard (i.e., another person), may involve in empathy processing. Previous empathy-related studies often used pain-related paradigm and revealed that the empathy for others’ pain activates the anterior insula cortex (aIC), a key brain region implicated in processing first-hand painful experiences. Moreover, some studies indicated that aIC was also involved in upward comparison. These observations raise a possibility that the neural processing associated with our first-hand emotional experiences may interact with that of our empathic responses, and the interaction could be integrated in the social comparison framework. However, few studies so far have been conducted to clarify the possibility yet. By using functional magnetic resonance imaging (fMRI), the current project attempts to elucidate the neural mechanisms underlying the influence of self emotional processing on empathic responses within the social comparison framework through a pain-related social emotion paradigm. We found that when participants received a worse-than-others’ outcome, their positive empathic responses toward confederates were countered or even reversed, which were accompanied by the activations in aIC and other brain regions. It depicts the complex interactions between self emotional processing and empathic responses may be integrated into the social comparison framework. Results obtained from this study will enhance our knowledge about cerebral mechanisms underlying emotional processing in interpersonal contexts.

The human brain is a dynamic and active organ, constantly processing information in an adaptive way, which allows it to navigate the complexities of the environment. This adaptive ability, so called psychological resilience, ensures that the brain is always striving for maintaining normative brain function in the face of unpredictable and challenging circumstances. In this study, we explore a biomarker of resilience using resting-state functional magnetic resonance imaging (rs-fMRI). Importantly, resilience is an abstract concept presently defined based on self-report questionnaires. We suggest a general definition of resilience as a dynamic state of mind that fluctuates due to external circumstances but maintains a stable level of psychological functioning. In this light, we considered that the stability of mental states should be reflected in neural activity, especially in the default mode network (DMN), which is implicated in mental representations of self-states. Based on this, higher levels of mental stability should be represented by smaller variation of DMN activity. To quantify DMN stability, the multivariate signals in all DMN voxels at each time point were projected in two-dimensional space using multidimensional scaling (MDS). The average distances between state coordinates at consecutive time points were extracted as individual indices of DMN stability. 279 healthy (163 females) completed a rs-fMRI scan, the Connor-Davidson Resilience Scale (CD-RISC), Brief Resilience Scale (BRS), and Resilience Scale for Adults (RSA). Regression models showed that DMN stability indices and age could successfully predict the scores of CD-RISC (r2 = 0.08***) and BRS (r2 = 0.17***). These results are in line with our hypothesis that more stable of internal states represented by DMN activity. This neural-based objective definition of mental state dynamics might be a more accurate and reliable physiological assessment of mental health, which can be applied in the development of targeted interventions that promotes mental health.

Habituation to threatening stimuli is adaptive for animals to pursue valuable goals under threat, yet the underlying neuro-cognitive mechanisms remain elusive. Here, three experiments were performed to elucidate the cerebral mechanism of human pain habituation from an emotional-motivational perspective. We showed, using these experiments, that habituation to painful stimuli at the repetitively stimulated site entailed reduced fear, covaried with individual fear-relevant personality traits, involved amygdala-prefrontal fear inhibitory mechanisms, engaged reward circuitry. These phenomena appeared to emanate from the context in which repetitive stimulation were not associated with harmful outcomes, and the neural networks underlying context-associated fear inhibition via safety (i.e., ventromedial prefrontal cortex and hippocampus) and pain modulation (i.e., periaqueductal grey) before stimulus encounter modulated fear inhibitory circuits during the experience of pain. Of note, pain habituation at the repetitively stimulated site was enhanced when participants were asked to perform a cognitively demanding task, which echoed with the motivational purpose of threat habituation. Moreover, the acquired fear inhibition was transferrable to non-stimulated sites, which involved brain structures implicated in applying encoded safety under threat to relevant contexts (i.e., anterior insula). These results provide an emotional-motivational account for habituation under threat, indicating that this adaptation is for other valuable purposes and relies on context-relevant fear regulation.

RBFOX3 is a important gene which has been used as a post-mitotic neuron maker for a long time. Here, in our lab, we found that Rbfox3 knockout mice show hyperactivity as Attention-Deficit/Hyperactivity Disorder (ADHD)-like behavior. Besides, we also found variations of RBFOX family genes, (RBFOX1, RBFOX2, RBFOX3), in Taiwan ADHD cohort. Hence, we’re going to investigate how RBFOX3 deficit will cause ADHD both in mice and human. In my study, I’d used human pluripotent stem cell (hiPSC)-derived neuron as our model to overcome the unreachable limitation of getting human neurons directly. We’d differentiated hiPSC into cortical neurons and cortical organoids, trying to approach the unknow of ADHD by morphology, molecular, and electrophysiology methods.

Dietary restriction (DR) is widely recognized as a beneficial food regimen to improve physiological conditions and alleviate age-related dysfunctions. Recent studies demonstrated DR exerts several benefits through the microbiota-gut-brain axis. Our current research further established the role of the gut microbiota in DR-induced memory enhancement in middle age mice and aged mice, and lifespan extension. In middle age mice, we adopted two loss-of-function models to examine the role of the gut microbiota: antibiotics-treated mice and germ-free mice. Both approaches demonstrated that DR-induced memory enhancement requires the present of gut microbiota. Moreover, the enhanced memory performance could be accomplished in mice under ad libitum (AL) condition receiving fecal microbiota transplantation from DR mice (FMT-DR). 16s rRNA sequencing identified common bacterial change in DR and FMT-DR. We further confirm the memory enhancing effect of two commonly identified species. Furthermore, serum metabolomic experiments on revealed several overlapped metabolites among DR/AL, FMT-DR/FMT-AL, and probiotics/saline, providing a potential interface between microbial alteration and memory enhancement. In aged mice, DR or FMT treatment starting from 14-month-old mice improves recognition memory and spatial memory. In addition to memory-enhancing effect, both treatments significantly promote lifespan extension, demonstrating that the gut microbiota also mediates lifespan extension effect of DR. Further microbiome and metabolome studies are pending to further decipher the underlying mechanism. Our study would provide a comprehensive understanding of the mechanism underlying DR-induced benefits. The ultimate goal is to identify a suitable intervention to improve memory function as well as mitigate disease-related deficits.

The ventromedial hypothalamic nucleus (VMH) is vital for various innate behaviors. Among multiple neural subsets within the VMH, the estrogen-receptor-1-expressing neurons in the ventrolateral division (VMHvl) are well-known for their involvements in consummatory social behaviors. Another non-overlapping neural subset expresses steroidogenic factor-1, and occupies the dorsomedial VMH (VMHdm). Studies showed that VMHdmSF-1 neurons encode a predator-orientated defensive state. Nevertheless, neuroanatomical evidences suggested that VMHdmSF-1 neurons axonal innervation from other VMH neural subpopulations, and they may receive and process social-related sensory cues. To address the functional involvement of VMHdmSF-1 neurons in social behaviors, we performed cell-type-specific in vivo calcium imaging in freely roaming animals. We revealed that the VMHdmSF-1 neurons were robustly activated by social- but not predator-associated stimuli, with a male-biased conspecific sex representation. In addition, conspecifics with different sexes selectively recruited distinct VMHdmSF-1 neural subsets. Through selective ablation of particular olfactory transmitting pathways, we found that male-biased populational responses of the VMHdmSF-1 neurons depend on VNO-relayed pheromonal signals, which are majorly transmitted to hypothalamus through the bed nucleus of stria terminalis (BNST). By optogenetically silencing the BNST-VMHdm pathway, we could diminish the male-preference among the VMHdm neural population. Moreover, VMHdmSF-1 neuronal activities are highly correlated with social investigative behaviors. Altogether, we suggest that apart from defensive behavioral state, a large portion of f VMHdmSF-1 neurons are capable of encoding conspecific social cues. The conspecific sex representation of VMHdm/cSF1 neurons may prompt animals’ investigative behaviors upon encountering other mice and facilitate proper behavioral decision making by accessing more information. 

Depression is a complex psychiatric disorder caused by chronic stress, affecting more than 300 million people worldwide and resulting in negative emotions and cognitive impairments. Novel strategies are urgently needed, as current treatments like selective serotonin reuptake inhibitors and ketamine-based antidepressants only address emotional responses and are less effective in managing cognitive impairments. In this study, we used a repeated social defeat stress mouse model to evaluate the cognitive effects of dietary restriction (DR), defined as 60% of daily food intake compared to ad libitum (AL) feeding. Our results showed that DR induced anti-depressant-like behaviors and improved memory performance in mice, accompanied by altered neuronal structure and increased neurogenesis in the dentate gyrus of the hippocampus. Importantly, the neurobehavioral changes induced by DR were abolished in mice treated with antibiotics and could be mimicked in AL-fed mice receiving fecal microbiota transplantation (FMT) from DR donors, suggesting the critical role of gut microbiota in mediating the beneficial effects of DR. We further used 16s rRNA sequencing and LC/MS-based serum metabolomics to identify commonly changed bacterial composition and metabolites in both DR and FMT-DR groups, revealing certain probiotics and postbiotics that had positive impacts on memory performance. Our ongoing work aims to unravel the underlying mechanism behind the pro-cognitive effects of DR. We believe that our study will have a significant impact on the future treatment of mental dysfunction associated with depression.

The gut microbiota is known to play a critical role in regulating multiple behaviors in Drosophila, including courtship, aggression, locomotion, sleep, and memory. To investigate the specific mechanism by which gut microbiota influences long-term memory (LTM) formation, we developed a model of chronic ovipositional depression induced by parasitoid wasp exposure. Our results demonstrate that intact LTM genes, such as amnesiac, rutabaga, and dunce, are essential for the sustained decrease in oviposition observed for several days following acute wasp exposure. Both antibiotic-treated and germ-free flies showed acute ovipositional depression on the day of exposure but failed to form LTM in the absence of gut microbiota. Interestingly, gut microbiota transplantation was able to rescue the impaired LTM, suggesting that gut microbiota mediates LTM formation without affecting learning in Drosophila. Metabolomics analysis of fly hemolymph revealed a significant alteration in amino acid biosynthesis, and we found that wasp-induced LTM was abolished in flies with a defective amino acid sensing molecule, target of rapamycin. Our ongoing experiments aim to determine the critical metabolites and underlying molecular mechanisms responsible for LTM formation.

Dietary restriction (DR) is a dietary regimen that reduces food intake by 20-40% compared to normal ad libitum (AL) feeding. Recent studies have shown that DR can extend lifespan and reduce aging-related diseases in different model organisms. In addition, DR also enhances memory function by modulating the mTOR signaling pathway in mice brains. Our recent work found that DR improves mice memory by changing the composition of gut microbiota and is independent of the vagus nerve that connects the gut to the brain. Accordingly, we hypothesize that modulated gut microbiota may generate beneficial metabolites that confer pro-cognitive activity through the circulatory system. We began with serum metabolomic analysis by an untargeted metabolomics strategy using liquid chromatography-mass spectrometry (LC-MS/MS). A few metabolites that highly correlated with improved memory performance were identified and validated through water supplementation to mice. In summary, we discovered beneficial metabolites from DR-associated gut microbiota, and we identified a group of metabolites that are involved in memory enhancement.

The ability to pursue good fortune and avoid disaster is vital for organisms to survive. In nature, a high percentage of Drosophila have been attacked by Leptopilina parasitic wasps, which inject their eggs into Drosophila larvae during their larval stage. To combat this predator threat, Drosophila have developed several responses, including innate immune responses and behavioral defenses, such as suppressing oviposition upon seeing wasps in adult female flies. Our laboratory has developed a new paradigm of non-associative learning and memory in Drosophila through an interesting interaction between wasps and flies. We exposed wasps to Drosophila for 24 hours and discovered that the ovipositional suppression can last for several days after the removal of wasps. This long-lasting ovipositional repression appears to be mediated through long-term memory (LTM) formation. To better understand this phenomenon at a circuit level, we began to manipulate neuronal activity through several different genetic approaches. We discovered that NPF neurons are necessary for Drosophila to react to the threat upon seeing Leptopilina wasps, and artificially boosting the neural activity of NPF neurons enhances LTM formation. Moreover, NPF neurons further communicate with dopaminergic neurons to influence mushroom body-mediated LTM formation. Future studies will focus on the downstream targets in order to map the neural circuit induced by wasp exposure.

Spatial navigation is one of the cognitive functions compromised in aging. Older adults require more time to construct a cognitive map and make more mistakes reaching target locations. Previous studies implicate inaccurate distance judgement as a cause. Older adults underestimate actual distances more than younger adults as actual distances increase. Such underestimation in distance judgement imply that older adults may have distortions in mental representations of space. In the present study, we investigated the neural correlates of distance representation in younger and older adults using a distance judgement paradigm within a virtual environment with eight landmarks. Participants freely navigated the virtual environment and encoded the map to criterion. After the cognitive map is acquired, participants were tested on the environment while undergoing functional magnetic resonance imaging. Specifically, for a given start landmark, participants provided their estimates of distances to target landmarks and also navigated there. Compared to younger adults, older adults under- and over-estimated closer and farther distances, respectively. Also, representation similarity analysis showed there were more similarity in visual system on older adults. In conclusion, our findings show how the shift in neural navigational strategy underlies older adult distance representations during spatial navigation.

RTL1/PEG11, which has been associated with anxiety disorders, is a retrotransposon-derived imprinted gene in the placenta. However, imprinting patterns and functions of RTL1 in the brain have not been well-investigated. We investigated Rtl1 imprinting status in mice by immunohistology. According to the result, most of the brain regions that expressed Rtl1 is paternal. Quantification of Rtl1-expressing brain regions showed highly expressed levels in locus coeruleus (LC), ventral tegmental area (VTA), and substantia nigra (SN). Importantly, the results of mice behavior tests show PKO mice have more anxiety-like behaviors. Rtl1 function in mice brain is associated with emotion-related behavior and stress regulation. Based on the previous findings, we try to precisely investigate which brain region does Rtl1 modulates anxious behavior. To achieve the goal, we used CRISPR Cas9 system with AAV to specific knockout Rtl1 in the target brain region. Meanwhile with timescales as parameters to rescue or knockout Rtl1 in mice brain that help we know more about the story that covers under Rtl1 functional deficit. Furthermore, which anxiety-related pathway is Rtl1 involved in the brain and the under mechanism of Rtl1 plays are the topic we are curious about. This study can help us knowing the mechanism which modulates Rtl1 and the role of RTL1 in the brain in more details.

Rbfox3 encoded a member of the RNA-binding FOX protein family which regulated alternative splicing of pre-mRNA. Some studies suggested that RBFOX3 mutation was related to neurodevelopmental delay, cognitive impairments, and epilepsy. Previous studies from our lab reported that Rbfox3 conventional knock-out mice had some deficits in seizure susceptibility, spatial reversal learning, long-term depression (LTD) of the dentate gyrus (DG), adult neurogenesis of the DG, and Attention Deficit/Hyperactivity Disorder (ADHD)-related phenotypes like hyperactivity and impulsivity. We found that seizure was associated with Rbfox3 deletion in GABAergic neurons. Therefore, we aimed to investigate whether the deletion of Rbfox3 in excitatory neurons affected those deficits except for epilepsy.

Human sensory perception stems from a proactive interplay between expectations based on past experiences and sensory inputs. Taking into account that anticipating imminent pain is a salient emotional event that receives priority in processing, it is plausible that the emotional responses elicited by pain expectancy may modulate its integration with nociceptive inputs to underlie pain perception. To confirm this putative modulatory role of emotion, we asked participants to down-regulate expectation-elicited emotions when they anticipated painful stimulation, and adopted a Bayesian approach to analyze behavioral and brain fMRI responses. We found that pain modulation by expectations of increased pain was predicted by anticipatory anxiety, which involved functional coupling between the amygdala and anterior cingulate – two key structures of anxiety processing – to integrate pain expectations and nociceptive inputs. This was accompanied by no updating of pain predictions via suppressed expression of prediction errors in the anterior cingulate, making negative expectation effects resistant to extinction over time. By contrast, for pain modulation by expectations of decreased pain, this integration was mainly linked to anticipatory pleasantness, which engaged functional coupling between the medial orbitofrontal cortex and anterior hippocampus – two regions implicated in positive emotion processing and inhibition of threat responding via safety. These findings suggest that stimulus expectancy effects on pain involve emotion-modulated Bayesian integration, explaining why chronic pain is frequently accompanied by both dysregulation of emotion and aberrant expectations and perception to aversive stimuli.

Empathy, the capacity to share the feelings of others with self-other distinction, plays an important role in human emotional experiences and social interaction. Since empathy comprises self-other distinction, social comparison, a process that we compare a target (i.e., ourselves) to a certain standard (i.e., another person), may involve in empathy processing. Previous empathy-related studies often used pain-related paradigm and revealed that the empathy for others’ pain activates the anterior insula cortex (aIC), a key brain region implicated in processing first-hand painful experiences. Moreover, some studies indicated that aIC was also involved in upward comparison. These observations raise a possibility that the neural processing associated with our first-hand emotional experiences may interact with that of our empathic responses, and the interaction could be integrated in the social comparison framework. However, few studies so far have been conducted to clarify the possibility yet. By using functional magnetic resonance imaging (fMRI), the current project attempts to elucidate the neural mechanisms underlying the influence of self emotional processing on empathic responses within the social comparison framework through a pain-related social emotion paradigm. We found that when participants received a worse-than-others’ outcome, their positive empathic responses toward confederates were countered or even reversed, which were accompanied by the activations in aIC and other brain regions. It depicts the complex interactions between self emotional processing and empathic responses may be integrated into the social comparison framework. Results obtained from this study will enhance our knowledge about cerebral mechanisms underlying emotional processing in interpersonal contexts.

Reward and punishment often act as important factors that modulate human behavior through learning to maximize the former, and minimizing the latter. Although much is known about how reward and punishment contribute to guiding our behavior independently, how these two types of learning interact with each other remains largely unclear. At the neural level, whether reward learning and punishment learning involve common or distinct neural substrates is still under debate. The aim of the current study was to investigate (1) the neural mechanisms underlying reward learning and punishment learning by comparing between these two types learning in a single experiment, and (2) the interaction between reward learning and punishment learning at both behavioral and neural levels. By using a probabilistic instrumental learning task in combination with functional magnetic resonance imaging, healthy participants were required to try their best to earn money and to avoid losing money or avoid painful stimulation at the same time. We found that when an option was simultaneously associated with rewarding and punishing outcomes, only the performance of reward learning was interfered by punishment learning, but not vice versa. We discovered that punishment learning seemed to interfere reward learning by altering participants exploration behavior, which might reflect that participants adopted a different learning strategy for the interfered reward learning. Taken together, these findings provide novel insights into the understanding about the difference between reward learning and punishment learning in humans.

Dietary restriction (DR) is often defined as a 20-40% reduction in daily food intake without malnutrition compared to ad libitum (AL) feeding. It has been shown to promote physiological and psychological health and increasing lifespan in multiple species. Previous studies in our laboratory demonstrated that DR can significantly improve memory performance of mice, although the underlying mechanism at the circuit level is largely unknown. For this purpose, we began to examine cFos expression as a neuron activity indicator across the whole brain of mice immediately following the memory phase of the novel object recognition test. So far, we have observed several brain regions that are positively and negatively correlated with the behavioral outcomes, respectively. There is increased cFos signal in the anterior cingulate area, CA1, CA3, dentate gyrus, anterodorsal nucleus, paraventricular thalamus, and prelimbic area of DR brains, while decreased cFos signals are observed in the mediodorsal nucleus of thalamus and thalamic nucleus reuniens. Future studies will employ an optical genetic approach to manipulate neuronal activities during learning and memory phases to help decipher the neural circuit that mediates DR-induced memory enhancement.

Chronic kidney disease (CKD) has been associated with metabolic disturbance and cognitive impairment. Currently, nutritional therapy, such as low protein diet, is able to manage kidney dysfunction to a certain degree, but the underlying mechanism has been largely unexplored. Our recent studies showed that dietary restriction [DR, defined as 60% daily food intake of ad libitum (AL)] could significantly improve health span and extend life span in several model organisms. In this study, we examined the effects of DR on adenine-induced CKD mouse model. We found that DR could dramatically attenuate CKD-induced depressive-like behaviors and memory impairments. These beneficial effects are mediated by gut microbiota since antibiotics treatment and fecal microbiota transplantation could abolish or mimic DR-induced responses, respectively. Besides, tryptophan of microbe-derived metabolites abolish DR-induced benefits. Future studies will try to unveil more specific bacterium and microbe-derived metabolites in mediating DR-induced cognitive effects.

Spinocerebellar ataxia (SCA) is a progressive neurodegenerative disease with autosomal dominant inheritance. SCA type 22 was originally discovered in a four- generation Chinese pedigree, and KCND3 was identified as the causative gene that encodes the protein Kv4.3, a voltage-gated potassium channel that functions during repolarization. Each subunit of the voltage-gated potassium channel contains six transmembrane helices, and an in-frame three-nucleotide deletion c.679_681delTTC p.F227del in the S2 helix leads to SCA22. Previously generated humanized KCND3 mutation mice with Kv4.3-F227del (KCND3 knockin mice) displayed significant locomotor dysfunction as well as cognitive impairments. This is consistent with the wide expression pattern of KCND3 across different brain regions, as seen in a KCND3 reporter mouse. Moreover, the locomotor and cognitive phenotypes of SCA22 are likely caused by gained neurotoxicity rather than loss of KCND3 protein function, since these phenotypes are largely normal in KCND3 knockout mice. Finally, we conducted a dietary restriction regime (60% daily food intake compared to ad libitum feeding) on KCND3 knockin mice and observed improved memory function but not locomotor performance. Together, our results reveal molecular insights into SCA22 and provide useful clinical implications for the future treatments of the disease.

Cognitive training is touted as a promising means to ameliorate age-related cognitive decline. However, the results from past studies only focused on repetitive engagement of lower-level cognitive processes, such as in attention, memory, or inhibitory tasks. In this registered clinical trial (no. NCT05341232), we aimed to improve higher-level cognitive processes, specifically active inference, through a 12-week Lego Robot Programming Intervention (LRPI) and investigate its neural correlates. Active inference refers that human brain dynamically integrates multiple information to perform the trial-and-error process of observation, prediction and action. To operate active inference, the brain would be required to involve and coordinate the above lower-level cognitive processes. We supposed that targeting on active inference would further stimulate general neural activation and improvement on untrained cognitive functions. In this research, participants were randomly assigned to the experimental or active control group, and neuropsychological tests were performed pre- and post-intervention to assess improvement on untargeted cognitive functions. The Visual Rule Inference Task (VRIT) was applied during fMRI scanning to investigate the change in brain activity and behaviors during inference processing. So far, 8 participants in the experimental group and 4 participants in the control group have completed the intervention. Preliminary data suggests that our experimental training could relate to changes for inference processing in older adults at both behavioral and neural levels. However, due to the small sample size, further research is required to more adequately evaluate the training effects. 

Food seeking and consumption provide animals energy to execute physiological functions and are essential for survival. Nevertheless, the feeding behavior might be affected by internal or external conditions, such as distress or threats, animals would cope with these circumstances and adjust feeding behaviors accordingly. The parabrachial nucleus (PBN) is known to regulate the feeding system and convey negative information. It also sends dense projections to the ventral tegmental area (VTA), a heterogenous brain region that regulates motivational behaviors. To characterize the function of this connection, I first used different behavior paradigms and found that VTA-projecting PBN cells are significantly engaged in stress or threat-related negative emotions. To further examine the circuitry mechanism of PBN-to-VTA input, I employed an activity-dependent targeting approach with optogenetic manipulation and found that PBN afferent activation mostly recruits VTA non-dopamine (non-DA) cells, which send dense projections to dorsal raphe nucleus (DR), lateral preoptic area (LPO) and peduncular part of the lateral hypothalamus (PLH), but not nucleus accumbens (NAc). I further examined the sufficient role of PBN-recruited VTA neurons and showed that optogenetic activation of these cells results in aversion, dampens freely palatable feeding, and disturbs instrumental food-seeking behavior in food self-administration task. Altogether, these results suggest that PBN recruits VTA non-DA cells to convey negative emotions and disengage food-seeking behavior.