TractGeoNet: A geometric deep learning framework for pointwise analysis of tract microstructure to predict language assessment performance.

We propose a geometric deep-learning-based framework, TractGeoNet, for performing regression using diffusion magnetic resonance imaging (dMRI) tractography and associated pointwise tissue microstructure measurements. By employing a point cloud representation, TractGeoNet can directly utilize tissue microstructure and positional information from all points within a fiber tract without the need to average or bin data along the streamline as traditionally required by dMRI tractometry methods. To improve regression performance, we propose a novel loss function, the Paired-Siamese Regression loss, which encourages the model to focus on accurately predicting the relative differences between regression label scores rather than just their absolute values. In addition, to gain insight into the brain regions that contribute most strongly to the prediction results, we propose a Critical Region Localization algorithm. This algorithm identifies highly predictive anatomical regions within the white matter fiber tracts for the regression task. We evaluate the effectiveness of the proposed method by predicting individual performance on two neuropsychological assessments of language using a dataset of 20 association white matter fiber tracts from 806 subjects from the Human Connectome Project Young Adult dataset. The results demonstrate superior prediction performance of TractGeoNet compared to several popular regression models that have been applied to predict individual cognitive performance based on neuroimaging features. Of the twenty tracts studied, we find that the left arcuate fasciculus tract is the most highly predictive of the two studied language performance assessments. Within each tract, we localize critical regions whose microstructure and point information are highly and consistently predictive of language performance across different subjects and across multiple independently trained models. These critical regions are widespread and distributed across both hemispheres and all cerebral lobes, including areas of the brain considered important for language function such as superior and anterior temporal regions, pars opercularis, and precentral gyrus. Overall, TractGeoNet demonstrates the potential of geometric deep learning to enhance the study of the brain's white matter fiber tracts and to relate their structure to human traits such as language performance.

Tractography-based automated identification of the retinogeniculate visual pathway with novel microstructure-informed supervised contrastive learning.

The retinogeniculate visual pathway (RGVP) is responsible for carrying visual information from the retina to the lateral geniculate nucleus. Identification and visualization of the RGVP are important in studying the anatomy of the visual system and can inform the treatment of related brain diseases. Diffusion MRI (dMRI) tractography is an advanced imaging method that uniquely enables in vivo mapping of the 3D trajectory of the RGVP. Currently, identification of the RGVP from tractography data relies on expert (manual) selection of tractography streamlines, which is time-consuming, has high clinical and expert labor costs, and is affected by inter-observer variability. In this paper, we present a novel deep learning framework, DeepRGVP , to enable fast and accurate identification of the RGVP from dMRI tractography data. We design a novel microstructure-informed supervised contrastive learning method that leverages both streamline label and tissue microstructure information to determine positive and negative pairs. We propose a simple and successful streamline-level data augmentation method to address highly imbalanced training data, where the number of RGVP streamlines is much lower than that of non-RGVP streamlines. We perform comparisons with several state-of-the-art deep learning methods that were designed for tractography parcellation, and we show superior RGVP identification results using DeepRGVP. In addition, we demonstrate a good generalizability of DeepRGVP to dMRI tractography data from neurosurgical patients with pituitary tumors and we show DeepRGVP can successfully identify RGVPs despite the effect of lesions affecting the RGVPs. Overall, our study shows the high potential of using deep learning to automatically identify the RGVP.

Deep fiber clustering: Anatomically informed fiber clustering with self-supervised deep learning for fast and effective tractography parcellation.

White matter fiber clustering is an important strategy for white matter parcellation, which enables quantitative analysis of brain connections in health and disease. In combination with expert neuroanatomical labeling, data-driven white matter fiber clustering is a powerful tool for creating atlases that can model white matter anatomy across individuals. While widely used fiber clustering approaches have shown good performance using classical unsupervised machine learning techniques, recent advances in deep learning reveal a promising direction toward fast and effective fiber clustering. In this work, we propose a novel deep learning framework for white matter fiber clustering, Deep Fiber Clustering (DFC), which solves the unsupervised clustering problem as a self-supervised learning task with a domain-specific pretext task to predict pairwise fiber distances. This process learns a high-dimensional embedding feature representation for each fiber, regardless of the order of fiber points reconstructed during tractography. We design a novel network architecture that represents input fibers as point clouds and allows the incorporation of additional sources of input information from gray matter parcellation. Thus, DFC makes use of combined information about white matter fiber geometry and gray matter anatomy to improve the anatomical coherence of fiber clusters. In addition, DFC conducts outlier removal naturally by rejecting fibers with low cluster assignment probability. We evaluate DFC on three independently acquired cohorts, including data from 220 individuals across genders, ages (young and elderly adults), and different health conditions (healthy control and multiple neuropsychiatric disorders). We compare DFC to several state-of-the-art white matter fiber clustering algorithms. Experimental results demonstrate superior performance of DFC in terms of cluster compactness, generalization ability, anatomical coherence, and computational efficiency.

Superficial white matter analysis: An efficient point-cloud-based deep learning framework with supervised contrastive learning for consistent tractography parcellation across populations and dMRI acquisitions.

Diffusion MRI tractography is an advanced imaging technique that enables in vivo mapping of the brain's white matter connections. White matter parcellation classifies tractography streamlines into clusters or anatomically meaningful tracts. It enables quantification and visualization of whole-brain tractography. Currently, most parcellation methods focus on the deep white matter (DWM), whereas fewer methods address the superficial white matter (SWM) due to its complexity. We propose a novel two-stage deep-learning-based framework, Superficial White Matter Analysis (SupWMA), that performs an efficient and consistent parcellation of 198 SWM clusters from whole-brain tractography. A point-cloud-based network is adapted to our SWM parcellation task, and supervised contrastive learning enables more discriminative representations between plausible streamlines and outliers for SWM. We train our model on a large-scale tractography dataset including streamline samples from labeled long- and medium-range (over 40 mm) SWM clusters and anatomically implausible streamline samples, and we perform testing on six independently acquired datasets of different ages and health conditions (including neonates and patients with space-occupying brain tumors). Compared to several state-of-the-art methods, SupWMA obtains highly consistent and accurate SWM parcellation results on all datasets, showing good generalization across the lifespan in health and disease. In addition, the computational speed of SupWMA is much faster than other methods.

Sphingosine-1-phosphate, a novel TREM2 ligand, promotes microglial phagocytosis to protect against ischemic brain injury.

The mechanism of sphingosine-1-phosphate (S1P)-mediated phagocytosis remains unknown. Here, we found that S1P or FTY720 (an analog of S1P) promoted microglial phagocytosis in stroke independent of S1PRs. First, we used computer simulation of molecular docking to predict that S1P might be a ligand for triggering receptor expressed on myeloid cells 2 (TREM2). Next, microscale thermophoresis (MST), surface plasmon resonance (SPR) and liquid chromatography-tandem mass spectrometry (LC-MS/MS) were performed to reveal that S1P was a novel TREM2 ligand. Then, we confirmed the pro-phagocytosis of S1P targeting in Trem2-Dap12 transfected CHO cells and TREM2 knockdown microglia. Point mutation analysis showed that D104 was the critical binding residue. Trem2 -/- mice were used to demonstrate the role of S1P-induced phagocytosis targeting on TREM2 in protecting against ischemic brain injury. Finally, further studies revealed that apolipoprotein E (APOE) loaded with S1P was released by microglia and bound to apoptotic neurons via LDL receptor related protein 1B (LRP1B) and thereby induced microglia to phagocytose apoptotic neurons. Overall, the present work reveals for the first time that S1P acts as a novel endogenous ligand of TREM2 to effectively promote microglial phagocytosis. Our findings provide a new lead compound for developing immunomodulator targeting on TREM2.

One-Shot Learning with Pseudo-Labeling for Cattle Video Segmentation in Smart Livestock Farming.

Computer vision-based technologies play a key role in precision livestock farming, and video-based analysis approaches have been advocated as useful tools for automatic animal monitoring, behavior analysis, and efficient welfare measurement management. Accurately and efficiently segmenting animals' contours from their backgrounds is a prerequisite for vision-based technologies. Deep learning-based segmentation methods have shown good performance through training models on a large amount of pixel-labeled images. However, it is challenging and time-consuming to label animal images due to their irregular contours and changing postures. In order to reduce the reliance on the number of labeled images, one-shot learning with a pseudo-labeling approach is proposed using only one labeled image frame to segment animals in videos. The proposed approach is mainly comprised of an Xception-based Fully Convolutional Neural Network (Xception-FCN) module and a pseudo-labeling (PL) module. Xception-FCN utilizes depth-wise separable convolutions to learn different-level visual features and localize dense prediction based on the one single labeled frame. Then, PL leverages the segmentation results of the Xception-FCN model to fine-tune the model, leading to performance boosts in cattle video segmentation. Systematic experiments were conducted on a challenging feedlot cattle video dataset acquired by the authors, and the proposed approach achieved a mean intersection-over-union score of 88.7% and a contour accuracy of 80.8%, outperforming state-of-the-art methods (OSVOS and OSMN). Our proposed one-shot learning approach could serve as an enabling component for livestock farming-related segmentation and detection applications.

Iptakalim improves cerebral microcirculation in mice after ischemic stroke by inhibiting pericyte contraction.

Pericytes are present tight around the intervals of capillaries, play an essential role in stabilizing the blood-brain barrier, regulating blood flow and immunomodulation, and persistent contraction of pericytes eventually leads to impaired blood flow and poor clinical outcomes in ischemic stroke. We previously show that iptakalim, an ATP-sensitive potassium (K-ATP) channel opener, exerts protective effects in neurons, and glia against ischemia-induced injury. In this study we investigated the impacts of iptakalim on pericytes contraction in stroke. Mice were subjected to cerebral artery occlusion (MCAO), then administered iptakalim (10 mg/kg, ip). We showed that iptakalim administration significantly promoted recovery of cerebral blood flow after cerebral ischemia and reperfusion. Furthermore, we found that iptakalim significantly inhibited pericytes contraction, decreased the number of obstructed capillaries, and improved cerebral microcirculation. Using a collagen gel contraction assay, we demonstrated that cultured pericytes subjected to oxygen-glucose deprivation (OGD) consistently contracted from 3 h till 24 h during reoxygenation, whereas iptakalim treatment (10 µM) notably restrained pericyte contraction from 6 h during reoxygenation. We further showed that iptakalim treatment promoted K-ATP channel opening via suppressing SUR2/EPAC1 complex formation. Consequently, it reduced calcium influx and ET-1 release. Taken together, our results demonstrate that iptakalim, targeted K-ATP channels, can improve microvascular disturbance by inhibiting pericyte contraction after ischemic stroke. Our work reveals that iptakalim might be developed as a promising pericyte regulator for treatment of stroke.

White matter association tracts underlying language and theory of mind: An investigation of 809 brains from the Human Connectome Project.

Language and theory of mind (ToM) are the cognitive capacities that allow for the successful interpretation and expression of meaning. While functional MRI investigations are able to consistently localize language and ToM to specific cortical regions, diffusion MRI investigations point to an inconsistent and sometimes overlapping set of white matter tracts associated with these two cognitive domains. To further examine the white matter tracts that may underlie these domains, we use a two-tensor tractography method to investigate the white matter microstructure of 809 participants from the Human Connectome Project. 20 association white matter tracts (10 in each hemisphere) are uniquely identified by leveraging a neuroanatomist-curated automated white matter tract atlas. The fractional anisotropy (FA), mean diffusivity (MD), and number of streamlines (NoS) are measured for each white matter tract. Performance on neuropsychological assessments of semantic memory (NIH Toolbox Picture Vocabulary Test, TPVT) and emotion perception (Penn Emotion Recognition Test, PERT) are used to measure critical subcomponents of the language and ToM networks, respectively. Regression models are constructed to examine how structural measurements of left and right white matter tracts influence performance across these two assessments. We find that semantic memory performance is influenced by the number of streamlines of the left superior longitudinal fasciculus III (SLF-III), and emotion perception performance is influenced by the number of streamlines of the right SLF-III. Additionally, we find that performance on both semantic memory & emotion perception is influenced by the FA of the left arcuate fasciculus (AF). The results point to multiple, overlapping white matter tracts that underlie the cognitive domains of language and ToM. Results are discussed in terms of hemispheric dominance and concordance with prior investigations.

Extracellular vesicles derived from LPS-preconditioned human synovial mesenchymal stem cells inhibit extracellular matrix degradation and prevent osteoarthritis of the knee in a mouse model.

BACKGROUND: Previous studies report that lipopolysaccharide (LPS)-preconditioned mesenchymal stem cells have enhanced trophic support and improved regenerative and repair properties. Extracellular vesicles secreted by synovial mesenchymal stem cells (EVs) can reduce cartilage damage caused by osteoarthritis (OA). Previous studies show that extracellular vesicles secreted by LPS-preconditioned synovial mesenchymal stem cells (LPS-pre EVs) can improve the response to treatment of osteoarthritis (OA). This study sought to explore effects of LPS-pre EVs on chondrocyte proliferation, migration, and chondrocyte apoptosis, as well as the protective effect of LPS-pre EVs on mouse articular cartilage. METHODS: Chondrocytes were extracted to explore the effect of LPS-pre EVs on proliferation, migration, and apoptosis of chondrocytes. In addition, the effect of LPS-pre EVs on expression level of important proteins of chondrocytes was explored suing in vitro experiments. Further, intraarticular injection of LPS-pre EVs was performed on the destabilization of the medial meniscus (DMM)-induced mouse models of OA to explore the therapeutic effect of LPS-pre EVs on osteoarthritis in vivo. RESULTS: Analysis showed that LPS-pre EVs significantly promoted proliferation and migration of chondrocytes and inhibited the apoptosis of chondrocytes compared with PBS and EVs. Moreover, LPS-pre EVs inhibited decrease of aggrecan and COL2A1 and increase of ADAMTS5 caused by IL-1ß through let-7b. Furthermore, LPS-pre EVs significantly prevented development of OA in DMM-induced mouse models of OA. CONCLUSIONS: LPS pretreatment is an effective and promising method to improve therapeutic effect of extracellular vesicles secreted from SMSCs on OA.

Neuronal extracellular vesicle derived miR-98 prevents salvageable neurons from microglial phagocytosis in acute ischemic stroke.

Extracellular vesicles (EVs), as a novel intercellular communication carrier transferring cargo microRNAs (miRNAs), could play important roles in the brain remodeling process after ischemic stroke. However, the detailed mechanisms involved in EVs derived miRNAs-mediated cellular interactions in the brain remain unclear. Several studies indicated that microRNA-98 (miR-98) might participate in the pathogenesis of ischemic stroke. Here, we showed that expression of miR-98 in penumbra field kept up on the first day but dropped sharply on the 3rd day after ischemic stroke in rats, indicating that miR-98 could function as an endogenous protective factor post-ischemia. Overexpression of miR-98 targeted inhibiting platelet activating factor receptor-mediated microglial phagocytosis to attenuate neuronal death. Furthermore, we showed that neurons transferred miR-98 to microglia via EVs secretion after ischemic stroke, to prevent the stress-but-viable neurons from microglial phagocytosis. Therefore, we reveal that EVs derived miR-98 act as an intercellular signal mediating neurons and microglia communication during the brain remodeling after ischemic stroke. The present work provides a novel insight into the roles of EVs in the stroke pathogenesis and a new EVs-miRNAs-based therapeutic strategy for stroke.

Rabies Virus-Inspired Metal-Organic Frameworks (MOFs) for Targeted Imaging and Chemotherapy of Glioma.

The blood-brain barrier (BBB) restricts access to the brain of more than 98 % of therapeutic agents and is largely responsible for treatment failure of glioblastoma multiforme (GBM). Therefore, it is of great importance to develop a safe and efficient strategy for more effective drug delivery across the BBB into the brain. Inspired by the extraordinary capability of rabies virus (RABV) to enter the central nervous system, we report the development and evaluation of the metal-organic framework-based nanocarrier MILB@LR, which closely mimicked both the bullet-shape structure and surface functions of natural RABV. MILB@LR benefited from a more comprehensive RABV-mimic strategy than mimicking individual features of RABV and exhibited significantly enhanced BBB penetration and brain tumor targeting. MILB@LR also displayed superior inhibition of tumor growth when loaded with oxaliplatin. The results demonstrated that MILB@LR may be valuable for GBM targeting and treatment.

FTY720 Exerts Anti-Glioma Effects by Regulating the Glioma Microenvironment Through Increased CXCR4 Internalization by Glioma-Associated Microglia.

Background: Glioblastoma (GBM) is one of the most malignant and aggressive primary brain tumors. The incurability of glioblastoma is heavily influenced by the glioma microenvironment. FTY720, a potent immunosuppressant, has been reported to exert anti-tumor effects in glioblastoma. However, the impact of FTY720 on the glioma microenvironment remains unclear. Methods: We examined the effects of FTY720 on the distribution and polarization of glioma-associated microglia and macrophages (GAMs) in glioma-bearing rats using immunofluorescence staining. qRT-PCR and Western blotting were used to detect the expressions of CXCR4 and MAPK pathway-related signal molecules on microglia in the coculture system. The levels of inflammatory factors were tested via ELISA. Wound healing assay and Matrigel invasion assay were used to determine the migration and invasion of C6 glioma cells. Results: We discovered that FTY720 could inhibit the growth, migration, and invasion of glioma by targeting GAMs to impede their effect on glioma cells. Simultaneously, FTY720 could block the chemoattraction of GAMs by inhibiting MAPK-mediated secretion of IL-6 through increased internalization of CXCR4. Moreover, microglia and macrophages are polarized from pro-glioma to an anti-tumor phenotype. Conclusion: These results provide novel insights into the inhibitory effects of FTY720 on glioma by targeting GAMs-glioma interaction in the tumor microenvironment.

The Intra-nuclear SphK2-S1P Axis Facilitates M1-to-M2 Shift of Microglia via Suppressing HDAC1-Mediated KLF4 Deacetylation.

Sphingosine 1-phosphate (S1P) is involved in a variety of cellular responses including microglial activation and polarization. However, the impacts of S1P on ischemia-induced microglial activation and polarization remain unclear. In the present study, Sprague-Dawley rats were selected for middle cerebral artery occlusion (MCAO) establishment and treated with S1P analog FTY720 (0.5, 1, 2 mg/kg) for 24 h. The impacts of FTY720 on oxygen-glucose deprivation (OGD)-induced microglial polarization were examined in the primary cultured microglia. FTY720 treatment could prevent ischemia-induced brain injury and neurological dysfunction, also decrease the levels of IL-1ß and TNF-α and promote M2 microglial polarization in rats. Further, we found that FTY720 inhibited the expressions of M1 markers, but increased the expressions of M2 markers in the OGD-insulted microglia. And FTY720 could enhance the phagocytic function of microglia. The sphingosine kinase 1/2 (SphK1/2) or the Sphk2 inhibitor could prevent the M1 to M2 phenotype shift improved by FTY720, but the Sphk1 inhibitor failed to affect the roles of FTY720. Furthermore, the Sphk1/2 or Sphk2 inhibitor promoted the activities of histone deacetylase (HDAC1) and inhibited the histone acetylation of the Krüppel-like factor 4 (KLF4) promoter regions, indicating that intra-nuclear pFTY720 inhibited HDAC1 activations and prevented KLF4 to interact with HDAC1, and thereby suppresses KLF4 deacetylation. Therefore, our data reveals that intra-nuclear SphK2-S1P axis might facilitate the transformation of microglial polarization from M1 to M2 phenotype, which might be intra-nuclear regulatory mechanisms of FTY720-prevented neuroinflammation.

Establishment and evaluation of a novel mouse model of peri/postmenopausal depression.

Women are believed to be more vulnerable to develop depressive symptoms during the perimenopause compared to postmenopause. The traditional bilateral ovariectomy and chronic mild stress (CMS) stimulation animal model produces a postmenopausal depressive-like state but the transition from perimenopausal period to postmenopausal period was ignored. Thus we establish a novel animal model in which the mice were stimulated by CMS for three months and removed the ovaries by two-step operation, and then evaluate whether this novel model could be much better for preclinical study used as a peri/postmenopause depressive model. The present study systemically evaluated the changes induced by two-step ovariectomy plus CMS in the mice. The depression-like behaviors, the levels of corticosterone, estrogen, pro-inflammatory factors, neurotransmitters, as well as brain-derived neurotrophic factor were determined; the changes of estrogen receptors, serotonin receptors, uterine weight and bone microarchitecture were also observed. The results show that the behaviors and biochemical indexes of mice changed gradually over time. Our study suggests that this two-step ovariectomy operation plus CMS successfully establishes a more reasonable peri/postmenopausal depression animal model which effectively simulates the clinical symptoms of peri/postmenopausal depressive women.

Antagonizing peroxisome proliferator-activated receptor γ facilitates M1-to-M2 shift of microglia by enhancing autophagy via the LKB1-AMPK signaling pathway.

Microglia-mediated neuroinflammation plays a dual role in various brain diseases due to distinct microglial phenotypes, including deleterious M1 and neuroprotective M2. There is growing evidence that the peroxisome proliferator-activated receptor γ (PPARγ) agonist rosiglitazone prevents lipopolysaccharide (LPS)-induced microglial activation. Here, we observed that antagonizing PPARγ promoted LPS-stimulated changes in polarization from the M1 to the M2 phenotype in primary microglia. PPARγ antagonist T0070907 increased the expression of M2 markers, including CD206, IL-4, IGF-1, TGF-ß1, TGF-ß2, TGF-ß3, G-CSF, and GM-CSF, and reduced the expression of M1 markers, such as CD86, Cox-2, iNOS, IL-1ß, IL-6, TNF-α, IFN-γ, and CCL2, thereby inhibiting NFκB-IKKß activation. Moreover, antagonizing PPARγ promoted microglial autophagy, as indicated by the downregulation of P62 and the upregulation of Beclin1, Atg5, and LC3-II/LC3-I, thereby enhancing the formation of autophagosomes and their degradation by lysosomes in microglia. Furthermore, we found that an increase in LKB1-STRAD-MO25 complex formation enhances autophagy. The LKB1 inhibitor radicicol or knocking down LKB1 prevented autophagy improvement and the M1-to-M2 phenotype shift by T0070907. Simultaneously, we found that knocking down PPARγ in BV2 microglial cells also activated LKB1-AMPK signaling and inhibited NFκB-IKKß activation, which are similar to the effects of antagonizing PPARγ. Taken together, our findings demonstrate that antagonizing PPARγ promotes the M1-to-M2 phenotypic shift in LPS-induced microglia, which might be due to improved autophagy via the activation of the LKB1-AMPK signaling pathway.

PD149163 induces hypothermia to protect against brain injury in acute cerebral ischemic rats.

Therapeutic hypothermia is a promising strategy for acute cerebral ischemia via physical or pharmacological methods. In this study, we pharmacologically induced hypothermia on Sprague Dawley rats by intraperitoneally injecting PD149163. We found that mild hypothermia was induced by PD149163 treatment without local cerebral blood flow (LCBF) alteration. To evaluate the neuroprotective effects of PD149163, TTC staining, HE staining and Nissl's staining were performed in our study. We found that PD149163 could prevent neuronal damage, and inhibit proliferation and activation of glial cells induced by ischemia. Simultaneously, we observed PD149163 ameliorated apoptosis characterized by down-regulated caspase-3 and Bax, but elevated Bcl-2. Moreover, PD149163 dramatically reduced JNK and AMPK/mTOR signaling pathway activation, and thereby inhibited autophagy by increased P62 expression, decreased the ratio of LC3-Ⅱ to LC3-Ⅰ and the expression of Beclin. Taken together, the present findings reveal the therapeutic effects of PD149163-induced hypothermia in brain ischemia, and provide a new strategy for stroke treatment.

Rosiglitazone Exerts an Anti-depressive Effect in Unpredictable Chronic Mild-Stress-Induced Depressive Mice by Maintaining Essential Neuron Autophagy and Inhibiting Excessive Astrocytic Apoptosis.

There is increasing interest in the association between depression and the development of metabolic diseases. Rosiglitazone, a therapeutic drug used to treat type 2 diabetes mellitus, has shown neuroprotective effects in patients with stroke and Alzheimer's disease. The present study was performed to evaluate the possible roles of rosiglitazone in in vivo (unpredictable chronic mild stress-induced depressive mouse model) and in vitro (corticosterone-induced cellular model) depressive models. The results showed that rosiglitazone reversed depressive behaviors in mice, as indicated by the forced swimming test and open field test. Rosiglitazone was also found to inhibit the inflammatory response, decrease corticosterone levels, and promote astrocyte proliferation and neuronal axon plasticity in the prefrontal cortex of mice. This series of in vivo and in vitro experiments showed that autophagy among neurons was inhibited in depressive models and that rosiglitazone promoted autophagy by upregulating LKB1, which exerted neuroprotective effects. Rosiglitazone was also found to activate the Akt/CREB pathway by increasing IGF-1R expression and IGF-1 protein levels, thereby playing an anti-apoptotic role in astrocytes. Rosiglitazone's autophagy promotion and neuroprotective effects were found to be reversed by the PPARγ antagonist T0070907 in primary neurons and by PPARγ knockdown in an N2a cell line. In conclusion, we found that rosiglitazone protects both neurons and astrocytes in in vivo and in vitro depressive models, thereby playing an anti-depressive role. These findings suggest that PPARγ could be a new target in the development of anti-depressive drugs.