Polyamine: A metabolic compass for T helper cell fate direction.

Interplay between metabolic and epigenetic remodeling may be key to cell fate control. In this issue of Cell, Puleston et al. and Wagner et al. use metabolomic, computational, and genetic approaches to uncover that polyamine metabolism directs T helper cell lineage choices, epigenetic state, and pathogenic potential in inflammation.

In vivo CRISPR screening reveals nutrient signaling processes underpinning CD8+ T cell fate decisions.

How early events in effector T cell (TEFF) subsets tune memory T cell (TMEM) responses remains incompletely understood. Here, we systematically investigated metabolic factors in fate determination of TEFF and TMEM cells using in vivo pooled CRISPR screening, focusing on negative regulators of TMEM responses. We found that amino acid transporters Slc7a1 and Slc38a2 dampened the magnitude of TMEM differentiation, in part through modulating mTORC1 signaling. By integrating genetic and systems approaches, we identified cellular and metabolic heterogeneity among TEFF cells, with terminal effector differentiation associated with establishment of metabolic quiescence and exit from the cell cycle. Importantly, Pofut1 (protein-O-fucosyltransferase-1) linked GDP-fucose availability to downstream Notch-Rbpj signaling, and perturbation of this nutrient signaling axis blocked terminal effector differentiation but drove context-dependent TEFF proliferation and TMEM development. Our study establishes that nutrient uptake and signaling are key determinants of T cell fate and shape the quantity and quality of TMEM responses.

CXCR6 orchestrates brain CD8+ T cell residency and limits mouse Alzheimer's disease pathology.

Neurodegenerative diseases, including Alzheimer's disease (AD), are characterized by innate immune-mediated inflammation, but functional and mechanistic effects of the adaptive immune system remain unclear. Here we identify brain-resident CD8+ T cells that coexpress CXCR6 and PD-1 and are in proximity to plaque-associated microglia in human and mouse AD brains. We also establish that CD8+ T cells restrict AD pathologies, including ß-amyloid deposition and cognitive decline. Ligand-receptor interaction analysis identifies CXCL16-CXCR6 intercellular communication between microglia and CD8+ T cells. Further, Cxcr6 deficiency impairs accumulation, tissue residency programming and clonal expansion of brain PD-1+CD8+ T cells. Ablation of Cxcr6 or CD8+ T cells ultimately increases proinflammatory cytokine production from microglia, with CXCR6 orchestrating brain CD8+ T cell-microglia colocalization. Collectively, our study reveals protective roles for brain CD8+ T cells and CXCR6 in mouse AD pathogenesis and highlights that microenvironment-specific, intercellular communication orchestrates tissue homeostasis and protection from neuroinflammation.

Global net climate effects of anthropogenic reactive nitrogen.

Anthropogenic activities have substantially enhanced the loadings of reactive nitrogen (Nr) in the Earth system since pre-industrial times1,2, contributing to widespread eutrophication and air pollution3-6. Increased Nr can also influence global climate through a variety of effects on atmospheric and land processes but the cumulative net climate effect is yet to be unravelled. Here we show that anthropogenic Nr causes a net negative direct radiative forcing of -0.34 [-0.20, -0.50] W m-2 in the year 2019 relative to the year 1850. This net cooling effect is the result of increased aerosol loading, reduced methane lifetime and increased terrestrial carbon sequestration associated with increases in anthropogenic Nr, which are not offset by the warming effects of enhanced atmospheric nitrous oxide and ozone. Future predictions using three representative scenarios show that this cooling effect may be weakened primarily as a result of reduced aerosol loading and increased lifetime of methane, whereas in particular N2O-induced warming will probably continue to increase under all scenarios. Our results indicate that future reductions in anthropogenic Nr to achieve environmental protection goals need to be accompanied by enhanced efforts to reduce anthropogenic greenhouse gas emissions to achieve climate change mitigation in line with the Paris Agreement.

Single-cell CRISPR screens in vivo map T cell fate regulomes in cancer.

CD8+ cytotoxic T cells (CTLs) orchestrate antitumour immunity and exhibit inherent heterogeneity1,2, with precursor exhausted T (Tpex) cells but not terminally exhausted T (Tex) cells capable of responding to existing immunotherapies3-7. The gene regulatory network that underlies CTL differentiation and whether Tex cell responses can be functionally reinvigorated are incompletely understood. Here we systematically mapped causal gene regulatory networks using single-cell CRISPR screens in vivo and discovered checkpoints for CTL differentiation. First, the exit from quiescence of Tpex cells initiated successive differentiation into intermediate Tex cells. This process is differentially regulated by IKAROS and ETS1, the deficiencies of which dampened and increased mTORC1-associated metabolic activities, respectively. IKAROS-deficient cells accumulated as a metabolically quiescent Tpex cell population with limited differentiation potential following immune checkpoint blockade (ICB). Conversely, targeting ETS1 improved antitumour immunity and ICB efficacy by boosting differentiation of Tpex to intermediate Tex cells and metabolic rewiring. Mechanistically, TCF-1 and BATF are the targets for IKAROS and ETS1, respectively. Second, the RBPJ-IRF1 axis promoted differentiation of intermediate Tex to terminal Tex cells. Accordingly, targeting RBPJ enhanced functional and epigenetic reprogramming of Tex cells towards the proliferative state and improved therapeutic effects and ICB efficacy. Collectively, our study reveals that promoting the exit from quiescence of Tpex cells and enriching the proliferative Tex cell state act as key modalities for antitumour effects and provides a systemic framework to integrate cell fate regulomes and reprogrammable functional determinants for cancer immunity.

SLC38A2 and glutamine signalling in cDC1s dictate anti-tumour immunity.

Cancer cells evade T cell-mediated killing through tumour-immune interactions whose mechanisms are not well understood1,2. Dendritic cells (DCs), especially type-1 conventional DCs (cDC1s), mediate T cell priming and therapeutic efficacy against tumours3. DC functions are orchestrated by pattern recognition receptors3-5, although other signals involved remain incompletely defined. Nutrients are emerging mediators of adaptive immunity6-8, but whether nutrients affect DC function or communication between innate and adaptive immune cells is largely unresolved. Here we establish glutamine as an intercellular metabolic checkpoint that dictates tumour-cDC1 crosstalk and licenses cDC1 function in activating cytotoxic T cells. Intratumoral glutamine supplementation inhibits tumour growth by augmenting cDC1-mediated CD8+ T cell immunity, and overcomes therapeutic resistance to checkpoint blockade and T cell-mediated immunotherapies. Mechanistically, tumour cells and cDC1s compete for glutamine uptake via the transporter SLC38A2 to tune anti-tumour immunity. Nutrient screening and integrative analyses show that glutamine is the dominant amino acid in promoting cDC1 function. Further, glutamine signalling via FLCN impinges on TFEB function. Loss of FLCN in DCs selectively impairs cDC1 function in vivo in a TFEB-dependent manner and phenocopies SLC38A2 deficiency by eliminating the anti-tumour therapeutic effect of glutamine supplementation. Our findings establish glutamine-mediated intercellular metabolic crosstalk between tumour cells and cDC1s that underpins tumour immune evasion, and reveal glutamine acquisition and signalling in cDC1s as limiting events for DC activation and putative targets for cancer treatment.

Amino Acids License Kinase mTORC1 Activity and Treg Cell Function via Small G Proteins Rag and Rheb.

Regulatory T (Treg) cells are critical mediators of immune tolerance whose activity depends upon T cell receptor (TCR) and mTORC1 kinase signaling, but the mechanisms that dictate functional activation of these pathways are incompletely understood. Here, we showed that amino acids license Treg cell function by priming and sustaining TCR-induced mTORC1 activity. mTORC1 activation was induced by amino acids, especially arginine and leucine, accompanied by the dynamic lysosomal localization of the mTOR and Tsc complexes. Rag and Rheb GTPases were central regulators of amino acid-dependent mTORC1 activation in effector Treg (eTreg) cells. Mice bearing RagA-RagB- or Rheb1-Rheb2-deficient Treg cells developed a fatal autoimmune disease and had reduced eTreg cell accumulation and function. RagA-RagB regulated mitochondrial and lysosomal fitness, while Rheb1-Rheb2 enforced eTreg cell suppressive gene signature. Together, these findings reveal a crucial requirement of amino acid signaling for licensing and sustaining mTORC1 activation and functional programming of Treg cells.

cBAF complex components and MYC cooperate early in CD8+ T cell fate.

The identification of mechanisms to promote memory T (Tmem) cells has important implications for vaccination and anti-cancer immunotherapy1-4. Using a CRISPR-based screen for negative regulators of Tmem cell generation in vivo5, here we identify multiple components of the mammalian canonical BRG1/BRM-associated factor (cBAF)6,7. Several components of the cBAF complex are essential for the differentiation of activated CD8+ T cells into T effector (Teff) cells, and their loss promotes Tmem cell formation in vivo. During the first division of activated CD8+ T cells, cBAF and MYC8 frequently co-assort asymmetrically to the two daughter cells. Daughter cells with high MYC and high cBAF display a cell fate trajectory towards Teff cells, whereas those with low MYC and low cBAF preferentially differentiate towards Tmem cells. The cBAF complex and MYC physically interact to establish the chromatin landscape in activated CD8+ T cells. Treatment of naive CD8+ T cells with a putative cBAF inhibitor during the first 48 h of activation, before the generation of chimeric antigen receptor T (CAR-T) cells, markedly improves efficacy in a mouse solid tumour model. Our results establish cBAF as a negative determinant of Tmem cell fate and suggest that manipulation of cBAF early in T cell differentiation can improve cancer immunotherapy.

Hippo Kinases Mst1 and Mst2 Sense and Amplify IL-2R-STAT5 Signaling in Regulatory T Cells to Establish Stable Regulatory Activity.

Interleukin-2 (IL-2) and downstream transcription factor STAT5 are important for maintaining regulatory T (Treg) cell homeostasis and function. Treg cells can respond to low IL-2 levels, but the mechanisms of STAT5 activation during partial IL-2 deficiency remain uncertain. We identified the serine-threonine kinase Mst1 as a signal-dependent amplifier of IL-2-STAT5 activity in Treg cells. High Mst1 and Mst2 (Mst1-Mst2) activity in Treg cells was crucial to prevent tumor resistance and autoimmunity. Mechanistically, Mst1-Mst2 sensed IL-2 signals to promote the STAT5 activation necessary for Treg cell homeostasis and lineage stability and to maintain the highly suppressive phosphorylated-STAT5+ Treg cell subpopulation. Unbiased quantitative proteomics revealed association of Mst1 with the cytoskeletal DOCK8-LRCHs module. Mst1 deficiency limited Treg cell migration and access to IL-2 and activity of the small GTPase Rac, which mediated downstream STAT5 activation. Collectively, IL-2-STAT5 signaling depends upon Mst1-Mst2 functions to maintain a stable Treg cell pool and immune tolerance.

Lipid signalling enforces functional specialization of Treg cells in tumours.

Regulatory T cells (Treg cells) are essential for immune tolerance1, but also drive immunosuppression in the tumour microenvironment2. Therapeutic targeting of Treg cells in cancer will therefore require the identification of context-specific mechanisms that affect their function. Here we show that inhibiting lipid synthesis and metabolic signalling that are dependent on sterol-regulatory-element-binding proteins (SREBPs) in Treg cells unleashes effective antitumour immune responses without autoimmune toxicity. We find that the activity of SREBPs is upregulated in intratumoral Treg cells. Moreover, deletion of SREBP-cleavage-activating protein (SCAP)-a factor required for SREBP activity-in these cells inhibits tumour growth and boosts immunotherapy that is triggered by targeting the immune-checkpoint protein PD-1. These effects of SCAP deletion are associated with uncontrolled production of interferon-γ and impaired function of intratumoral Treg cells. Mechanistically, signalling through SCAP and SREBPs coordinates cellular programs for lipid synthesis and inhibitory receptor signalling in these cells. First, de novo fatty-acid synthesis mediated by fatty-acid synthase (FASN) contributes to functional maturation of Treg cells, and loss of FASN from Treg cells inhibits tumour growth. Second, Treg cells in tumours show enhanced expression of the PD-1 gene, through a process that depends on SREBP activity and signals via mevalonate metabolism to protein geranylgeranylation. Blocking PD-1 or SREBP signalling results in dysregulated activation of phosphatidylinositol-3-kinase in intratumoral Treg cells. Our findings show that metabolic reprogramming enforces the functional specialization of Treg cells in tumours, pointing to new ways of targeting these cells for cancer therapy.

Metabolic control of TFH cells and humoral immunity by phosphatidylethanolamine.

T follicular helper (TFH) cells are crucial for B cell-mediated humoral immunity1. Although transcription factors such as BCL6 drive the differentiation of TFH cells2,3, it is unclear whether and how post-transcriptional and metabolic programs enforce TFH cell programming. Here we show that the cytidine diphosphate (CDP)-ethanolamine pathway co-ordinates the expression and localization of CXCR5 with the responses of TFH cells and humoral immunity. Using in vivo CRISPR-Cas9 screening and functional validation in mice, we identify ETNK1, PCYT2, and SELENOI-enzymes in the CDP-ethanolamine pathway for de novo synthesis of phosphatidylethanolamine (PE)-as selective post-transcriptional regulators of TFH cell differentiation that act by promoting the surface expression and functional effects of CXCR5. TFH cells exhibit unique lipid metabolic programs and PE is distributed to the outer layer of the plasma membrane, where it colocalizes with CXCR5. De novo synthesis of PE through the CDP-ethanolamine pathway co-ordinates these events to prevent the internalization and degradation of CXCR5. Genetic deletion of Pcyt2, but not of Pcyt1a (which mediates the CDP-choline pathway), in activated T cells impairs the differentiation of TFH cells, and this is associated with reduced humoral immune responses. Surface levels of PE and CXCR5 expression on B cells also depend on Pcyt2. Our results reveal that phospholipid metabolism orchestrates post-transcriptional mechanisms for TFH cell differentiation and humoral immunity, highlighting the metabolic control of context-dependent immune signalling and effector programs.

CRISPR screens unveil signal hubs for nutrient licensing of T cell immunity.

Nutrients are emerging regulators of adaptive immunity1. Selective nutrients interplay with immunological signals to activate mechanistic target of rapamycin complex 1 (mTORC1), a key driver of cell metabolism2-4, but how these environmental signals are integrated for immune regulation remains unclear. Here we use genome-wide CRISPR screening combined with protein-protein interaction networks to identify regulatory modules that mediate immune receptor- and nutrient-dependent signalling to mTORC1 in mouse regulatory T (Treg) cells. SEC31A is identified to promote mTORC1 activation by interacting with the GATOR2 component SEC13 to protect it from SKP1-dependent proteasomal degradation. Accordingly, loss of SEC31A impairs T cell priming and Treg suppressive function in mice. In addition, the SWI/SNF complex restricts expression of the amino acid sensor CASTOR1, thereby enhancing mTORC1 activation. Moreover, we reveal that the CCDC101-associated SAGA complex is a potent inhibitor of mTORC1, which limits the expression of glucose and amino acid transporters and maintains T cell quiescence in vivo. Specific deletion of Ccdc101 in mouse Treg cells results in uncontrolled inflammation but improved antitumour immunity. Collectively, our results establish epigenetic and post-translational mechanisms that underpin how nutrient transporters, sensors and transducers interplay with immune signals for three-tiered regulation of mTORC1 activity and identify their pivotal roles in licensing T cell immunity and immune tolerance.

Spatiotemporal formation of the large vacuole regulated by the BIN2-VLG module is required for female gametophyte development in Arabidopsis.

In Arabidopsis thaliana, female gametophyte (FG) development is accompanied by the formation and expansion of the large vacuole in the FG; this is essential for FG expansion, nuclear polar localization, and cell fate determination. Arabidopsis VACUOLELESS GAMETOPHYTES (VLG) facilitates vesicular fusion to form large vacuole in the FG, but the regulation of VLG remains largely unknown. Here, we found that gain-of-function mutation of BRASSINOSTEROID INSENSITIVE2 (BIN2) (bin2-1) increases VLG abundance to induce the vacuole formation at stage FG1, and leads to abortion of FG. Loss-of-function mutation of BIN2 and its homologs (bin2-3 bil1 bil2) reduced VLG abundance and mimicked vlg/VLG phenotypes. Knocking down VLG in bin2-1 decreased the ratio of aberrant vacuole formation at stage FG1, whereas FG1-specific overexpression of VLG mimicked the bin2-1 phenotype. VLG partially rescued the bin2-3 bil1 bil2 phenotype, demonstrating that VLG acts downstream of BIN2. Mutation of VLG residues that are phosphorylated by BIN2 altered VLG stability and a phosphorylation mimic of VLG causes similar defects as did bin2-1. Therefore, BIN2 may function by interacting with and phosphorylating VLG in the FG to enhance its stability and abundance, thus facilitating vacuole formation. Our findings provide mechanistic insight into how the BIN2-VLG module regulates the spatiotemporal formation of the large vacuole in FG development.

Highly multiplexed spatial mapping of microbial communities.

Mapping the complex biogeography of microbial communities in situ with high taxonomic and spatial resolution poses a major challenge because of the high density1 and rich diversity2 of species in environmental microbiomes and the limitations of optical imaging technology3-6. Here we introduce high-phylogenetic-resolution microbiome mapping by fluorescence in situ hybridization (HiPR-FISH), a versatile technology that uses binary encoding, spectral imaging and decoding based on machine learning to create micrometre-scale maps of the locations and identities of hundreds of microbial species in complex communities. We show that 10-bit HiPR-FISH can distinguish between 1,023 isolates of Escherichia coli, each fluorescently labelled with a unique binary barcode. HiPR-FISH, in conjunction with custom algorithms for automated probe design and analysis of single-cell images, reveals the disruption of spatial networks in the mouse gut microbiome in response to treatment with antibiotics, and the longitudinal stability of spatial architectures in the human oral plaque microbiome. Combined with super-resolution imaging, HiPR-FISH shows the diverse strategies of ribosome organization that are exhibited by taxa in the human oral microbiome. HiPR-FISH provides a framework for analysing the spatial ecology of environmental microbial communities at single-cell resolution.

A comprehensive quantification of global nitrous oxide sources and sinks.

Nitrous oxide (N2O), like carbon dioxide, is a long-lived greenhouse gas that accumulates in the atmosphere. Over the past 150 years, increasing atmospheric N2O concentrations have contributed to stratospheric ozone depletion1 and climate change2, with the current rate of increase estimated at 2 per cent per decade. Existing national inventories do not provide a full picture of N2O emissions, owing to their omission of natural sources and limitations in methodology for attributing anthropogenic sources. Here we present a global N2O inventory that incorporates both natural and anthropogenic sources and accounts for the interaction between nitrogen additions and the biochemical processes that control N2O emissions. We use bottom-up (inventory, statistical extrapolation of flux measurements, process-based land and ocean modelling) and top-down (atmospheric inversion) approaches to provide a comprehensive quantification of global N2O sources and sinks resulting from 21 natural and human sectors between 1980 and 2016. Global N2O emissions were 17.0 (minimum-maximum estimates: 12.2-23.5) teragrams of nitrogen per year (bottom-up) and 16.9 (15.9-17.7) teragrams of nitrogen per year (top-down) between 2007 and 2016. Global human-induced emissions, which are dominated by nitrogen additions to croplands, increased by 30% over the past four decades to 7.3 (4.2-11.4) teragrams of nitrogen per year. This increase was mainly responsible for the growth in the atmospheric burden. Our findings point to growing N2O emissions in emerging economies-particularly Brazil, China and India. Analysis of process-based model estimates reveals an emerging N2O-climate feedback resulting from interactions between nitrogen additions and climate change. The recent growth in N2O emissions exceeds some of the highest projected emission scenarios3,4, underscoring the urgency to mitigate N2O emissions.

Leveraging scaffold information to predict protein-ligand binding affinity with an empirical graph neural network.

Protein-ligand binding affinity prediction is an important task in structural bioinformatics for drug discovery and design. Although various scoring functions (SFs) have been proposed, it remains challenging to accurately evaluate the binding affinity of a protein-ligand complex with the known bound structure because of the potential preference of scoring system. In recent years, deep learning (DL) techniques have been applied to SFs without sophisticated feature engineering. Nevertheless, existing methods cannot model the differential contribution of atoms in various regions of proteins, and the relationship between atom properties and intermolecular distance is also not fully explored. We propose a novel empirical graph neural network for accurate protein-ligand binding affinity prediction (EGNA). Graphs of protein, ligand and their interactions are constructed based on different regions of each bound complex. Proteins and ligands are effectively represented by graph convolutional layers, enabling the EGNA to capture interaction patterns precisely by simulating empirical SFs. The contributions of different factors on binding affinity can thus be transparently investigated. EGNA is compared with the state-of-the-art machine learning-based SFs on two widely used benchmark data sets. The results demonstrate the superiority of EGNA and its good generalization capability.

Electrocatalytic Generation of Singlet Oxygen via ROS-Mediated Redox Chain Reaction for Efficient Disinfection.

The risk of harmful microorganisms to ecosystems and human health has stimulated exploration of singlet oxygen (1O2)-based disinfection. It can be potentially generated via an electrocatalytic process, but is limited by the low production yield and unclear intermediate-mediated mechanism. Herein, we designed a two-site catalyst (Fe/Mo-N/C) for the selective 1O2 generation. The Mo sites enhance the generation of 1O2 precursors (H2O2), accompanied by the generation of intermediate •HO2/•O2-. The Fe site facilitates activation of H2O2 into •OH, which accelerates the •HO2/•O2- into 1O2. A possible mechanism for promoting 1O2 production through the ROS-mediated chain reaction is reported. The as-developed electrochemical disinfection system can kill 1 × 107 CFU mL-1 of E. coli within 8 min, leading to cell membrane damage and DNA degradation. It can be effectively applied for the disinfection of medical wastewater. This work provides a general strategy for promoting the production of 1O2 through electrocatalysis and for efficient electrochemical disinfection.

An Autoclavable and Transparent Thermal Cutter for Reliably Sealing Wet Nanofibrous Membranes.

Sealing wet porous membranes is a major challenge when fabricating cell encapsulation devices. Herein, we report the development of an Autoclavable Transparent Thermal Cutter (ATTC) for reliably sealing wet nanofibrous membranes. Notably, the ATTC is autoclavable and transparent, thus enabling in situ visualization of the sealing process in a sterile environment and ensuring an appropriate seal. In addition, the ATTC could generate smooth, arbitrary-shaped sealing ends with excellent mechanical properties when sealing PA6, PVDF, and TPU nanofibrous tubes and PP microporous membranes. Importantly, the ATTC could reliably seal wet nanofibrous tubes, which can shoulder a burst pressure up to 313.2 ± 19.3 kPa without bursting at the sealing ends. Furthermore, the ATTC sealing process is highly compatible with the fabrication of cell encapsulation devices, as verified by viability, proliferation, cell escape, and cell function tests. We believe that the ATTC could be used to reliably seal cell encapsulation devices with minimal side effects.

ARID5B as a critical downstream target of the TAL1 complex that activates the oncogenic transcriptional program and promotes T-cell leukemogenesis.

The oncogenic transcription factor TAL1/SCL induces an aberrant transcriptional program in T-cell acute lymphoblastic leukemia (T-ALL) cells. However, the critical factors that are directly activated by TAL1 and contribute to T-ALL pathogenesis are largely unknown. Here, we identified AT-rich interactive domain 5B (ARID5B) as a collaborating oncogenic factor involved in the transcriptional program in T-ALL. ARID5B expression is down-regulated at the double-negative 2-4 stages in normal thymocytes, while it is induced by the TAL1 complex in human T-ALL cells. The enhancer located 135 kb upstream of the ARID5B gene locus is activated under a superenhancer in T-ALL cells but not in normal T cells. Notably, ARID5B-bound regions are associated predominantly with active transcription. ARID5B and TAL1 frequently co-occupy target genes and coordinately control their expression. ARID5B positively regulates the expression of TAL1 and its regulatory partners. ARID5B also activates the expression of the oncogene MYC Importantly, ARID5B is required for the survival and growth of T-ALL cells, and forced expression of ARID5B in immature thymocytes results in thymus retention, differentiation arrest, radioresistance, and tumor formation in zebrafish. Our results indicate that ARID5B reinforces the oncogenic transcriptional program by positively regulating the TAL1-induced regulatory circuit and MYC in T-ALL, thereby contributing to T-cell leukemogenesis.

Intermittent Theta-Burst Stimulation for Stroke: Primary Motor Cortex Versus Cerebellar Stimulation: A Randomized Sham-Controlled Trial.

BACKGROUND: Stroke survivors with impaired balance and motor function tend to have relatively poor functional outcomes. The cerebellum and primary motor cortex (M1) have been suggested as targets for neuromodulation of balance and motor recovery after stroke. This study aimed to compare the efficacy and safety of intermittent theta-burst stimulation (iTBS) to the cerebellum or M1 on balance and motor recovery in patients with stroke. METHODS: In this randomized, double-blind, sham-controlled clinical trial, patients with subacute stroke were randomly divided into 3 groups: M1-, cerebellar-, and sham-iTBS (n=12 per group; 15 sessions, 3 weeks). All outcomes were evaluated before intervention (T0), after 1 week of intervention (T1), after 3 weeks of intervention (T2), and at follow-up (T3). The primary outcome was the Berg balance scale score at T2. Secondary outcomes include the Fugl-Meyer assessment scale for lower extremities, the trunk impairment scale, the Barthel index, the modified Rankin Scale, the functional ambulation categories, and cortical excitability. RESULTS: A total of 167 inpatients were screened, 36 patients (age, 57.50±2.41 years; 10 women, 12 ischemic) were enrolled between December 2020 and January 2023. At T2, M1- or cerebellar-iTBS significantly improved Berg balance scale scores by 10.7 points ([95% CI, 2.7-18.6], P=0.009) and 14.2 points ([95% CI, 1.2-27.2], P=0.032) compared with the sham-iTBS group. Moreover, the cerebellar-iTBS group showed a significantly greater improvement in Fugl-Meyer assessment scale for lower extremities scores by 5.6 points than the M1-iTBS ([95% CI, 0.3-10.9], P=0.037) and by 7.8 points than the sham-iTBS ([95% CI, 1.1-14.5], P=0.021) groups at T2. The motor-evoked potential amplitudes of the M1- and cerebellar-iTBS groups were higher than those of the sham-iTBS group (P<0.001). CONCLUSIONS: Both M1- and cerebellar-iTBS could improve balance function. Moreover, cerebellar-iTBS, but not M1-iTBS, induced significant effects on motor recovery. Thus, cerebellar-iTBS may be a valuable new therapeutic option in stroke rehabilitation programs. REGISTRATION: URL: https://www.chictr.org.cn/; Unique identifier: ChiCTR2100047002.