DNA/RNA heteroduplex technology with cationic oligopeptide reduces class-related adverse effects of nucleic acid drugs.

Antisense oligonucleotides (ASOs) are a therapeutic modality for incurable diseases. However, systemic injection of gapmer-type ASOs causes class-related toxicities, including prolongation of activated partial thromboplastin time (aPTT) and thrombocytopenia. We previously reported that cholesterol-conjugated DNA/RNA heteroduplex oligonucleotides (Chol-HDOs) exhibit significantly enhanced gene-silencing effects compared to ASOs, even in the central nervous system, by crossing the blood-brain barrier. In the present study, we initially evaluated the effect of the HDO structure on class-related toxicities. The HDO structure ameliorated the class-related toxicities associated with ASOs, but they remained to some extent. As a further antidote, we have developed artificial cationic oligopeptides, L-2,4-diaminobutanoic acid oligomers (DabOs), which bind to the phosphates in the major groove of the A-type double-helical structure of HDOs. The DabO/Chol-HDO complex showed significantly improved aPTT prolongation and thrombocytopenia in mice while maintaining gene-silencing efficacy. Moreover, the conjugation with DabOs effectively prevented cerebral infarction, a condition frequently observed in mice intravenously injected with high-dose Chol-HDO. These approaches, combining HDO technology with DabOs, offer distinct advantages over conventional strategies in reducing toxicities. Consequently, the DabO/HDO complex represents a promising platform for overcoming the class-related toxicities associated with therapeutic ASOs.

Usefulness of an Artificial Intelligence Model in Recognizing Recurrent Laryngeal Nerves During Robot-Assisted Minimally Invasive Esophagectomy.

BACKGROUND: Recurrent laryngeal nerve (RLN) palsy is a common complication in esophagectomy and its main risk factor is reportedly intraoperative procedure associated with surgeons' experience. We aimed to improve surgeons' recognition of the RLN during robot-assisted minimally invasive esophagectomy (RAMIE) by developing an artificial intelligence (AI) model. METHODS: We used 120 RAMIE videos from four institutions to develop an AI model and eight other surgical videos from another institution for AI model evaluation. AI performance was measured using the Intersection over Union (IoU). Furthermore, to verify the AI's clinical validity, we conducted the two experiments on the early identification of RLN and recognition of its location by eight trainee surgeons with or without AI. RESULTS: The IoUs for AI recognition of the right and left RLNs were 0.40 ± 0.26 and 0.34 ± 0.27, respectively. The recognition of the right RLN presence in the beginning of right RLN lymph node dissection (LND) by surgeons with AI (81.3%) was significantly more accurate (p = 0.004) than that by surgeons without AI (46.9%). The IoU of right RLN during right RLN LND recognized by surgeons with AI (0.59 ± 0.18) was significantly higher (p = 0.010) than that by surgeons without AI (0.40 ± 0.29). CONCLUSIONS: Surgeons' recognition of anatomical structures in RAMIE was improved by our AI system with high accuracy. Especially in right RLN LND, surgeons could recognize the RLN more quickly and accurately by using the AI model.

Golgi pH homeostasis stabilizes the lysosomal membrane through N-glycosylation of membrane proteins.

Protein glycosylation plays a vital role in various cellular functions, many of which occur within the Golgi apparatus. The Golgi pH regulator (GPHR) is essential for the proper functioning of the Golgi apparatus. The lysosomal membrane contains highly glycosylated membrane proteins in abundance. This study investigated the role of the Golgi luminal pH in N-glycosylation of lysosomal membrane proteins and the effect of this protein modification on membrane stability using Gphr-deficient MEFs. We showed that Gphr deficiency causes an imbalance in the Golgi luminal pH, resulting in abnormal protein N-glycosylation, indicated by a reduction in sialylated glycans and markedly reduced molecular weight of glycoproteins. Further experiments using FRAP and PLA revealed that Gphr deficiency prevented the trafficking dynamics and proximity condition of glycosyltransferases in the Golgi apparatus. In addition, incomplete N-glycosylation of lysosomal membrane proteins affected lysosomal membrane stability, as demonstrated by the increased susceptibility to lysosomal damage. Thus, this study highlights the critical role of Golgi pH regulation in controlling protein glycosylation and the impact of Golgi dysfunction on lysosomal membrane stability.

Evaluating surgical expertise with AI-based automated instrument recognition for robotic distal gastrectomy.

Introduction: Complexities of robotic distal gastrectomy (RDG) give reason to assess physician's surgical skill. Varying levels in surgical skill affect patient outcomes. We aim to investigate how a novel artificial intelligence (AI) model can be used to evaluate surgical skill in RDG by recognizing surgical instruments. Methods: Fifty-five consecutive robotic surgical videos of RDG for gastric cancer were analyzed. We used Deeplab, a multi-stage temporal convolutional network, and it trained on 1234 manually annotated images. The model was then tested on 149 annotated images for accuracy. Deep learning metrics such as Intersection over Union (IoU) and accuracy were assessed, and the comparison between experienced and non-experienced surgeons based on usage of instruments during infrapyloric lymph node dissection was performed. Results: We annotated 540 Cadiere forceps, 898 Fenestrated bipolars, 359 Suction tubes, 307 Maryland bipolars, 688 Harmonic scalpels, 400 Staplers, and 59 Large clips. The average IoU and accuracy were 0.82 ± 0.12 and 87.2 ± 11.9% respectively. Moreover, the percentage of each instrument's usage to overall infrapyloric lymphadenectomy duration predicted by AI were compared. The use of Stapler and Large clip were significantly shorter in the experienced group compared to the non-experienced group. Conclusions: This study is the first to report that surgical skill can be successfully and accurately determined by an AI model for RDG. Our AI gives us a way to recognize and automatically generate instance segmentation of the surgical instruments present in this procedure. Use of this technology allows unbiased, more accessible RDG surgical skill.

Improvement in Epigenetic Aging Clock Induced by BioBran Containing Rice Kefiran in Relation to Various Biomarkers: A Pilot Study.

Many lifestyle-related diseases such as cancer, dementia, myocardial infarction, and stroke are known to be caused by aging, and the WHO's ICD-11 (International Classification of Diseases, 11th edition) created the code "aging-related" in 2022. In other words, aging is irreversible but aging-related diseases are reversible, so taking measures to treat them is important for health longevity and preventing other diseases. Therefore, in this study, we used BioBran containing rice kefiran as an approach to improve aging. Rice kefiran has been reported to improve the intestinal microflora, regulate the intestines, and have anti-aging effects. BioBran has also been reported to have antioxidant effects and improve liver function, and human studies have shown that it affects the diversity of the intestinal microbiota. Quantitative measures of aging that correlate with disease risk are now available through the epigenetic clock test, which examines the entire gene sequence and determines biological age based on the methylation level. Horvath's Clock is the best known of many epigenetic clock tests and was published by Steve Horvath in 2013. In this study, we examine the effect of using Horvath's Clock to improve aging and report on the results, which show a certain effect.

Vero cell-adapted SARS-CoV-2 strain shows increased viral growth through furin-mediated efficient spike cleavage.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) utilizes several host proteases to cleave the spike (S) protein to enter host cells. SARS-CoV-2 S protein is cleaved into S1 and S2 subunits by furin, which is closely involved in the pathogenicity of SARS-CoV-2. However, the effects of the modulated protease cleavage activity due to S protein mutations on viral replication and pathogenesis remain unclear. Herein, we serially passaged two SARS-CoV-2 strains in Vero cells and characterized the cell-adapted SARS-CoV-2 strains in vitro and in vivo. The adapted strains showed high viral growth, effective S1/S2 cleavage of the S protein, and low pathogenicity compared with the wild-type strain. Furthermore, the viral growth and S1/S2 cleavage were enhanced by the combination of the Δ68-76 and H655Y mutations using recombinant SARS-CoV-2 strains generated by the circular polymerase extension reaction. The recombinant SARS-CoV-2 strain, which contained the mutation of the adapted strain, showed increased susceptibility to the furin inhibitor, suggesting that the adapted SARS-CoV-2 strain utilized furin more effectively than the wild-type strain. Pathogenicity was attenuated by infection with effectively cleaved recombinant SARS-CoV-2 strains, suggesting that the excessive cleavage of the S proteins decreases virulence. Finally, the high-growth-adapted SARS-CoV-2 strain could be used as the seed for a low-cost inactivated vaccine; immunization with this vaccine can effectively protect the host from SARS-CoV-2 variants. Our findings provide novel insights into the growth and pathogenicity of SARS-CoV-2 in the evolution of cell-cell transmission. IMPORTANCE: The efficacy of the S protein cleavage generally differs among the SARS-CoV-2 variants, resulting in distinct viral characteristics. The relationship between a mutation and the entry of SARS-CoV-2 into host cells remains unclear. In this study, we analyzed the sequence of high-growth Vero cell-adapted SARS-CoV-2 and factors determining the enhancement of the growth of the adapted virus and confirmed the characteristics of the adapted strain by analyzing the recombinant SARS-CoV-2 strain. We successfully identified mutations Δ68-76 and H655Y, which enhance viral growth and the S protein cleavage by furin. Using recombinant viruses enabled us to conduct a virus challenge experiment in vivo. The pathogenicity of SARS-CoV-2 introduced with the mutations Δ68-76, H655Y, P812L, and Q853L was attenuated in hamsters, indicating the possibility of the attenuation of excessive cleaved SARS-CoV-2. These findings provide novel insights into the infectivity and pathogenesis of SARS-CoV-2 strains, thereby significantly contributing to the field of virology.

Automated artificial intelligence-based phase-recognition system for esophageal endoscopic submucosal dissection (with video).

BACKGROUND AND AIMS: Endoscopic submucosal dissection (ESD) for superficial esophageal cancer is a multistep treatment involving several endoscopic processes. Although analyzing each phase separately is worthwhile, it is not realistic in practice owing to the need for considerable manpower. To solve this problem, we aimed to establish a state-of-the-art artificial intelligence (AI)-based system, specifically, an automated phase-recognition system that can automatically identify each endoscopic phase based on video images. METHODS: Ninety-four videos of ESD procedures for superficial esophageal cancer were evaluated in this single-center study. A deep neural network-based phase-recognition system was developed in an automated manner to recognize each of the endoscopic phases. The system was trained with the use of videos that were annotated and verified by 2 GI endoscopists. RESULTS: The overall accuracy of the AI model for automated phase recognition was 90%, and the average precision, recall, and F value rates were 91%, 90%, and 90%, respectively. Two representative ESD videos predicted by the model indicated the usability of AI in clinical practice. CONCLUSIONS: We demonstrated that an AI-based automated phase-recognition system for esophageal ESD can be established with high accuracy. To the best of our knowledge, this is the first report on automated recognition of ESD treatment phases. Because this system enabled a detailed analysis of phases, collecting large volumes of data in the future may help to identify quality indicators for treatment techniques and uncover unmet medical needs that necessitate the creation of new treatment methods and devices.

Unveiling the physics underlying symmetry breaking of the actin cytoskeleton: An artificial cell-based approach.

Single-cell behaviors cover many biological functions, such as cell division during morphogenesis and tissue metastasis, and cell migration during cancer cell invasion and immune cell responses. Symmetry breaking of the positioning of organelles and the cell shape are often associated with these biological functions. One of the main players in symmetry breaking at the cellular scale is the actin cytoskeleton, comprising actin filaments and myosin motors that generate contractile forces. However, because the self-organization of the actomyosin network is regulated by the biochemical signaling in cells, how the mechanical contraction of the actin cytoskeleton induces diverse self-organized behaviors and drives the cell-scale symmetry breaking remains unclear. In recent times, to understand the physical underpinnings of the symmetry breaking exhibited in the actin cytoskeleton, artificial cell models encapsulating the cytoplasmic actomyosin networks covered with lipid monolayers have been developed. By decoupling the actomyosin mechanics from the complex biochemical signaling within living cells, this system allows one to study the self-organization of actomyosin networks confined in cell-sized spaces. We review the recent developments in the physics of confined actomyosin networks and provide future perspectives on the artificial cell-based approach. This review article is an extended version of the Japanese article, The Physical Principle of Cell Migration Under Confinement: Artificial Cell-based Bottom-up Approach, published in SEIBUTSU BUTSURI Vol. 63, p. 163-164 (2023).

Geometric confinement guides topological defect pairings and emergent flow in nematic cell populations.

Topological defects in nematically aligned cell populations play a critical role in modulating collective motion, ranging from microbial colonies to epithelial tissues. Despite the potential of manipulating such topological defects to control diverse self-organized structures and collective dynamics, controlling the position of defects in active matter remains a challenging area of research. In this study, we investigated the geometry-guided control of defect positioning and alignment in a nematic cell population by imposing spatial constraints consisting of two or three overlapping circular boundaries. The confined cell population exhibited a paired and ordered distribution of half-integer topological defects that remained stable even when the size of the spatial constraint was altered using geometric parameters. These defects direct the inward flow of cells, induced by the curved boundary shape, towards the geometric center of the confined space. This inward flow contributes to an increase in a local cell density, and furthermore the geometry-induced nematic order provides mechanical stimulation to confined cells, as indicated by the elongated cell nucleus. Our geometry-based approach sets the foundation for controlling defect pairing and provides insights into the interplay among geometry, topology, and collective dynamics.

Effects of Paired Associative Stimulation on Cortical Plasticity in Agonist-Antagonist Muscle Representations.

Paired associative stimulation (PAS) increases and decreases cortical excitability in primary motor cortex (M1) neurons, depending on the spike timing-dependent plasticity, i.e., long-term potentiation (LTP)- and long-term depression (LTD)-like plasticity, respectively. However, how PAS affects the cortical circuits for the agonist and antagonist muscles of M1 is unclear. Here, we investigated the changes in the LTP- and LTD-like plasticity for agonist and antagonist muscles during PAS: 200 pairs of 0.25-Hz peripheral electric stimulation of the right median nerve at the wrist, followed by a transcranial magnetic stimulation of the left M1 with an interstimulus interval of 25 ms (PAS-25 ms) and 10 ms (PAS-10 ms). The unconditioned motor evoked potential amplitudes of the agonist muscles were larger after PAS-25 ms than after PAS-10 ms, while those of the antagonist muscles were smaller after PAS-25 ms than after PAS-10 ms. The γ-aminobutyric acid A (GABAA)- and GABAB-mediated cortical inhibition for the agonist and antagonist muscles were higher after PAS-25 ms than after PAS-10 ms. The cortical excitability for the agonist and antagonist muscles reciprocally and topographically increased and decreased after PAS, respectively; however, GABAA and GABAB-mediated cortical inhibitory functions for the agonist and antagonist muscles were less topographically decreased after PAS-10 ms. Thus, PAS-25 ms and PAS-10 ms differentially affect the LTP- and LTD-like plasticity in agonist and antagonist muscles.

Influence of history of falls and physical function on obstacle-straddling behavior.

[Purpose] This study aimed to clarify the relationship between falls and lower leg motion during obstacle crossing, in which stumbling or tripping is the most common cause of falls in the elderly population. [Participants and Methods] This study included 32 older adults who performed the obstacle crossing motion. The heights of the obstacles were 20, 40, and 60 mm. To analyze the leg motion, a video analysis system was used. The hip, knee, and ankle joint angles during the crossing motion were calculated by the video analysis software, Kinovea. To evaluate the risk of falls, one leg stance time and timed up and go test were measured, and data on fall history were collected using a questionnaire. Participants were divided into two groups: high-risk and low-risk groups, according to the degree of fall risk. [Results] The high-risk group showed greater changes in hip flexion angle in the forelimb. The hip flexion angle in the hindlimb and the angle change of lower extremities among the high-risk group became larger. [Conclusion] Participants in the high-risk group should lift their legs high when performing the crossing motion to ensure foot clearance and avoid stumbling over the obstacle.

Evaluation of surgical complexity by automated surgical process recognition in robotic distal gastrectomy using artificial intelligence.

BACKGROUND: Although radical gastrectomy with lymph node dissection is the standard treatment for gastric cancer, the complication rate remains high. Thus, estimation of surgical complexity is required for safety. We aim to investigate the association between the surgical process and complexity, such as a risk of complications in robotic distal gastrectomy (RDG), to establish an artificial intelligence (AI)-based automated surgical phase recognition by analyzing robotic surgical videos, and to investigate the predictability of surgical complexity by AI. METHOD: This study assessed clinical data and robotic surgical videos for 56 patients who underwent RDG for gastric cancer. We investigated (1) the relationship between surgical complexity and perioperative factors (patient characteristics, surgical process); (2) AI training for automated phase recognition and model performance was assessed by comparing predictions to the surgeon-annotated reference; (3) AI model predictability for surgical complexity was calculated by the area under the curve. RESULT: Surgical complexity score comprised extended total surgical duration, bleeding, and complications and was strongly associated with the intraoperative surgical process, especially in the beginning phases (area under the curve 0.913). We established an AI model that can recognize surgical phases from video with 87% accuracy; AI can determine intraoperative surgical complexity by calculating the duration of beginning phases from phases 1-3 (area under the curve 0.859). CONCLUSION: Surgical complexity, as a surrogate of short-term outcomes, can be predicted by the surgical process, especially in the extended duration of beginning phases. Surgical complexity can also be evaluated with automation using our artificial intelligence-based model.

SARS-CoV-2 ORF8 is a viral cytokine regulating immune responses.

Many patients with severe COVID-19 suffer from pneumonia and the elucidation of the mechanisms underlying the development of this severe condition is important. The in vivo function of the ORF8 protein secreted by SARS-CoV-2 is not well understood. Here, we analyzed the function of ORF8 protein by generating ORF8-knockout SARS-CoV-2 and found that the lung inflammation observed in wild-type SARS-CoV-2-infected hamsters was decreased in ORF8-knockout SARS-CoV-2-infected hamsters. Administration of recombinant ORF8 protein to hamsters also induced lymphocyte infiltration into the lungs. Similar pro-inflammatory cytokine production was observed in primary human monocytes treated with recombinant ORF8 protein. Furthermore, we demonstrated that the serum ORF8 protein levels are well-correlated with clinical markers of inflammation. These results demonstrated that the ORF8 protein is a SARS-CoV-2 viral cytokine involved in the immune dysregulation observed in COVID-19 patients, and that the ORF8 protein could be a novel therapeutic target in severe COVID-19 patients.

The relationship between the esophageal endoscopic submucosal dissection technical difficulty and its intraoperative process.

BACKGROUND: Estimating the esophageal endoscopic submucosal dissection (ESD) technical difficulty is important to reduce complications. Endoscopic duration is one of the related factors to a technical difficulty. The relationship between the esophageal ESD technical difficulty and its intraoperative process was analyzed as a first step toward automatic technical difficulty recognition using artificial intelligence. METHODS: This study enrolled 75 patients with superficial esophageal cancer who underwent esophageal ESD. The technical difficulty score was established, which consisted of three factors, including total procedure duration, en bloc resection, and complications. Additionally, technical difficulty-related factors, which were perioperative factors that included the intraoperative process, were investigated. RESULTS: Eight (11%) patients were allocated to high difficulty, whereas 67 patients (89%) were allocated to low difficulty. The intraoperative process, which was shown as the extension of each endoscopic phase, was significantly related to a technical difficulty. The area under the curve (AUC) values were higher at all the phase duration than at the clinical characteristics. Submucosal dissection phase (AUC 0.902; 95% confidence intervals (CI) 0.752-1.000), marking phase (AUC 0.827; 95% CI 0.703-0.951), and early phase which was defined as the duration from the start of marking to the end of submucosal injection (AUC 0.847; 95% CI 0.701-0.992) were significantly related to technical difficulty. CONCLUSIONS: The intraoperative process, particularly early phase, was strongly associated with esophageal ESD technical difficulty. This study demonstrated the potential for automatic evaluation of esophageal ESD technical difficulty when combined with an AI-based automatic phase evaluation system.

Membrane Sphingomyelin in Host Cells Is Essential for Nucleocapsid Penetration into the Cytoplasm after Hemifusion during Rubella Virus Entry.

The lipid composition of the host cell membrane is one of the key determinants of the entry of enveloped viruses into cells. To elucidate the detailed mechanisms behind the cell entry of rubella virus (RuV), one of the enveloped viruses, we searched for host factors involved in such entry by using CRISPR/Cas9 genome-wide knockout screening, and we found sphingomyelin synthase 1 (SMS1), encoded by the SGMS1 gene, as a candidate. RuV growth was strictly suppressed in SGMS1-knockout cells and was completely recovered by the overexpression of enzymatically active SMS1 and partially recovered by that of SMS2, another member of the SMS family, but not by that of enzymatically inactive SMS1. An entry assay using pseudotyped vesicular stomatitis virus possessing RuV envelope proteins revealed that sphingomyelin generated by SMSs is crucial for at least RuV entry. In SGMS1-knockout cells, lipid mixing between the RuV envelope membrane and the membrane of host cells occurred, but entry of the RuV genome from the viral particles into the cytoplasm was strongly inhibited. This indicates that sphingomyelin produced by SMSs is essential for the formation of membrane pores after hemifusion occurs during RuV entry. IMPORTANCE Infection with rubella virus during pregnancy causes congenital rubella syndrome in infants. Despite its importance in public health, the detailed mechanisms of rubella virus cell entry have only recently become somewhat clearer. The E1 protein of rubella virus is classified as a class II fusion protein based on its structural similarity, but it has the unique feature that its activity is dependent on calcium ion binding in the fusion loops. In this study, we found another unique feature, as cellular sphingomyelin plays a critical role in the penetration of the nucleocapsid into the cytoplasm after hemifusion by rubella virus. This provides important insight into the entry mechanism of rubella virus. This study also presents a model of hemifusion arrest during cell entry by an intact virus, providing a useful tool for analyzing membrane fusion, a biologically important phenomenon.

Morphological growth dynamics, mechanical stability, and active microtubule mechanics underlying spindle self-organization.

The spindle is a dynamic intracellular structure self-organized from microtubules and microtubule-associated proteins. The spindle's bipolar morphology is essential for the faithful segregation of chromosomes during cell division, and it is robustly maintained by multifaceted mechanisms. However, abnormally shaped spindles, such as multipolar spindles, can stochastically arise in a cell population and cause chromosome segregation errors. The physical basis of how microtubules fail in bipolarization and occasionally favor nonbipolar assembly is poorly understood. Here, using live fluorescence imaging and quantitative shape analysis in Xenopus egg extracts, we find that spindles of varied shape morphologies emerge through nonrandom, bistable self-organization paths, one leading to a bipolar and the other leading to a multipolar phenotype. The bistability defines the spindle's unique morphological growth dynamics linked to each shape phenotype and can be promoted by a locally distorted microtubule flow that arises within premature structures. We also find that bipolar and multipolar spindles are stable at the steady-state in bulk but can infrequently switch between the two phenotypes. Our microneedle-based physical manipulation further demonstrates that a transient force perturbation applied near the assembled pole can trigger the phenotypic switching, revealing the mechanical plasticity of the spindle. Together with molecular perturbation of kinesin-5 and augmin, our data propose the physical and molecular bases underlying the emergence of spindle-shape variation, which influences chromosome segregation fidelity during cell division.

GPHR-mediated acidification of the Golgi lumen is essential for cholesterol biosynthesis in the brain.

The Golgi pH regulator (GPHR) is essential for maintaining the function and morphology of the Golgi apparatus through the regulation of luminal acidic pH. Abnormal morphology of the Golgi apparatus is associated with neurodegenerative diseases. Here, we found that knockout of GPHR in the mouse brain led to morphological changes in the Golgi apparatus and neurodegeneration, which included brain atrophy, neuronal cell death, and gliosis. Furthermore, in the GPHR knockout mouse brain, transcriptional activity of sterol regulatory element-binding protein 2 (SREBP2) decreased, resulting in a reduction in cholesterol levels. GPHR-deficient cells exhibited suppressed neurite outgrowth, which was recovered by exogenous expression of the active form of SREBP2. Our results show that GPHR-mediated luminal acidification of the Golgi apparatus maintains proper cholesterol levels and, thereby, neuronal morphology.

Collective motion of epithelial cells along a wrinkled 3D-buckled hydrogel.

Epithelial cells migrate autonomously by aligning and inducing a collective motion. Controlling the collective motion of epithelial cells in geometrically confined environments is important for understanding physiological processes such as wound healing and self-organized morphogenesis. However, collective migration under a three-dimensional (3D) curved surface resembling living epithelial tissue has not yet been explored. In this study, we investigated the collective motion of a 3D-buckled polyacrylamide (PAAm) gel that mimics the shape of folds and wrinkles of epithelial tissue to understand the geometric effects of collective motion. We found that the velocity correlation in the space near the hydrogel boundary showed a periodic change that correlated with the wrinkled folding of the hydrogel pattern. Furthermore, the characteristic length of the velocity correlation increased proportionally with the wavelength of wrinkled folding. These observations indicated that the hydrogel pattern could steer the collective motion of epithelial cells over long distances. Our study also suggests that the wrinkled design of the hydrogel is a versatile platform for studying the geometric effect of a curved surface on complex epithelial cell dynamics.