To reevaluate transpalatal arch (TPA) biomechanics for molar rotation by transitioning from the traditional linear beam model to a statically indeterminate elastic portal frame analysis. A mathematical model based on Castigliano's second theorem simulated the TPA as a curved portal frame. Different modes of rotational activation were applied to standard and omega-loop designs across variable palatal heights (10 mm to 18 mm) using stainless steel and beta-titanium, incorporating a 2° mechanical clearance to simulate clinical insertion conditions. Palatal height was the primary determinant of TPA stiffness, acting as a lever arm that converts sagittal forces into torsion. In unilateral activation, average to high palatal vaults (palatal height ≥ 14 mm) generated spontaneous neutrality through the interaction between structural compliance and mechanical clearance, where the moment on the nonactivated side dissipates entirely. Regarding stiffness, contrary to clinical assumptions, the omega loop provided modest additional flexibility (less than 5% for unilateral activation, up to 15% for symmetric activation). The TPA functions as a flexible elastic frame, not a rigid linear beam. Vertical legs significantly dampen the force system, rendering the omega loop mechanically redundant. Notably, compensatory bends, generally used for unilateral rotation activation, are often biomechanically unnecessary in high-vaulted patients, as the system naturally reduces the contralateral moment to zero within the mechanical tolerance of the attachment.
A complex spatial light modulator (SLM) based on three-phase covalent macropixels (3PCMPs) enables the generation of three-dimensional computer-generated holograms with inherent suppression of DC and conjugate noise. However, the noise-suppressed complex-field synthesis remains confined to the central viewing zone (VZ), imposing a fundamental limitation on the effective eye-box. We present a design of zone-switching 3PCMP SLM architecture that relocates the domain of noise-suppressed complex-field synthesis across a 3 × 3 array of addressable VZs in a pupil-tracked manner. The proposed scheme integrates a static six-level phase plate with active binary sub-pixel modulation to steer the complex signal toward the selected VZ while preserving suppression of DC and conjugate terms. Wave-optics simulations confirm effective zone switching of the complex signal and show substantially improved holographic image quality across higher-order VZs, enabling a 3 × 3 extension of the effective viewing zone. These results support the feasibility of extending the effective eye-box in complex holographic displays.
Geological carbon storage in mature hydrocarbon fields requires careful evaluation of reservoir architecture, fault-controlled compartmentalization, and sealing integrity to ensure long-term containment of injected CO2. This study presents an integrated geological assessment of the CO2 storage potential of the Jurassic Safa reservoirs in the Obaiyed Field, located within the Shushan Basin of Egypt's Western Desert. The primary objective is to evaluate how structural segmentation and reservoir heterogeneity influence storage capacity and containment behavior in a deeply buried tight sandstone system. The study integrates seismic interpretation, well-log-based petrophysical analysis, core-derived permeability measurements, fault seal evaluation, and three-dimensional static geological modeling to characterize reservoir geometry, petrophysical variability, and structural compartmentalization. Results indicate that the Lower Safa Member represents a heterogeneous tight sandstone reservoir with effective porosity ranging approximately from 8 to 18% and permeability values spanning more than one order of magnitude. Porosity-permeability relationships derived from core measurements confirm significant reservoir heterogeneity and limited flow capacity typical of deeply buried clastic systems. Structural interpretation reveals pronounced fault-bounded compartmentalization that restricts lateral connectivity and promotes pressure partitioning within individual reservoir segments. Although storage capacity within individual compartments is modest, the cumulative storage potential of multiple structurally isolated compartments may provide a substantially larger field-scale storage resource. These findings highlight the potential of structurally segmented Jurassic reservoirs in the Western Desert to serve as viable candidates for geological CO2 storage and demonstrate the suitability of integrated geological screening for evaluating structurally compartmentalized reservoirs as potential CO2 storage sites.
Environmental pollution remains an urgent global challenge that threatens ecosystems and human health. Silicon carbide, with its exceptional chemical inertness, excellent thermal conductivity, and highly tunable electronic structure, has emerged as a powerful dual-phase platform for environmental remediation. This paper critically reviews how precise electronic structure engineering unlocks the catalytic potential of silicon carbide architectures across both aqueous and atmospheric environments. In wastewater treatment, strategies such as constructing heterojunctions and single-atom loading are highlighted for their ability to narrow bandgaps, strengthen internal electric fields, and induce hydrogen spillover effects. These modifications successfully overcome the high dissociation energy barriers of recalcitrant contaminants like per- and polyfluoroalkyl substances. In atmospheric remediation, silicon carbide inherently eliminates thermal runaway during volatile organic compound oxidation and strategically manages temperature-dependent synergistic mechanisms for treating multi-pollutant exhausts. This work meticulously evaluates how complex real-world variables, specifically dynamic pH fluctuations, competitive inorganic anions, and the dual role of moisture, dictate catalytic efficiencies alongside advanced microwave-assisted systems. Furthermore, Density Functional Theory (DFT) calculations are integrated to elucidate the atomic-level thermodynamic barriers and site-specific cleavage pathways driving these enhanced performances. Crucially, this review exposes significant gaps in current research, notably the over-reliance on idealized reaction conditions and static theoretical models without operando validation. By bridging these mechanistic insights with real-world complexities, this paper provides a systematic roadmap for transitioning silicon carbide-based catalysts from laboratory concepts to industrially viable environmental purification technologies.
This paper investigates the path following problem for unmanned ground vehicle (UGV) formation subject to control input and obstacle constraints. A fixed-time Lyapunov-based model predictive control (FLMPC) framework is proposed for multi-vehicle formation tracking and obstacle avoidance. By integrating fixed-time stability theory into the Lyapunov-based model predictive control (LMPC) architecture, the approach incorporates a Lyapunov function derivative constraint and introduces a control input tracking term into the cost function. This design endows the closed-loop system with fixed-time stability while inheriting the rapid convergence property of fixed-time control. To address collision avoidance in mixed scenarios with sudden and static obstacles, a time-triggered avoidance function and a dynamic weighting factor are designed. The resulting scheme enables proactive obstacle avoidance and smooth formation recovery. Moreover, energy-efficient control input design is explicitly embedded in the optimization process, leading to effectively reduced overall energy consumption. Simulation results validate the superiority of the proposed method in terms of convergence speed, obstacle avoidance performance and energy efficiency.
Responses to gait perturbations are critical for maintaining dynamic stability and preventing falls, yet muscle-level mechanisms underlying these responses remain insufficiently understood, particularly during the initial contact (IC) phase. How do individual lower-limb muscle forces adapt during perturbations at IC and in the subsequent recovery step? Twenty-one healthy young women walked on a dual-belt treadmill while perturbations were induced via sudden belt deceleration at IC. Motion capture and ground reaction force data were processed in OpenSim to estimate muscle forces using static optimization. Peak muscle forces (PMF) during perturbed and recovery steps were compared with normal gait using Wilcoxon tests and subsequent FDR correction. During the perturbed step, PMFs increased significantly in 8 of 30 muscles. Following FDR correction, significant differences remained for five muscles, primarily involving proximal and biarticular muscles (e.g., biceps femoris, gastrocnemius, gluteus medius, piriformis, and adductor magnus), indicating a multi-joint stabilization strategy. During the recovery step, several muscles demonstrated increased PMFs, however none remained significant after FDR correction. Gait perturbations at IC elicit increased involvement of a limited set of key stabilizing muscles, particularly across the hip, knee, and ankle joints. While recovery-step adaptations were observed, they should be considered exploratory due to the lack of statistical significance following correction for multiple comparisons. These findings provide novel insight into muscle-level mechanisms of balance recovery and may inform rehabilitation strategies targeting fall prevention.
暂无摘要（点击查看详情）
This study aims to compare the biomechanical performance of three different surgical fixation techniques in the treatment of proximal humerus fractures with medial metaphyseal defects. A total of 24 synthetic humerus bone models were used and equally divided into three groups (n = 8 per group). A standardized unstable proximal humerus fracture model with a medial metaphyseal cortical defect extending to the surgical neck was created using a three-dimensional (3D)-printed osteotomy guide. Group 1 received fixation with a lateral anatomical locking plate. Group 2 was treated with a combination of a lateral anatomical plate and a medial buttress plate (dual plating). Group 3 underwent fixation with an intramedullary nail (IMN); in this group, four specimens had distal locking with an endopin (Group 3a), and the other four with static screws (Group 3b). All specimens were subjected to axial loading until failure. Forces at the onset of failure and complete failure were recorded, and fracture patterns were documented. In Groups 1 and 2, transverse fractures consistently occurred at the level of the most distal screw of the lateral plate. In Group 3, failure was observed either proximally or distally at the nail tip, including butterfly fragment formation and metaphyseal collapse. Group 2 exhibited the highest resistance to axial loading, followed by Group 1 and Group 3, with statistically significant differences between all groups (p = 0.043, p = 0.0003, p < 0.00001). No significant difference was found between subgroups 3a and 3b (p > 0.05). Our study results indicate that double plating provides the greatest axial stability in proximal humerus fractures with medial metaphyseal defects, supporting its use in fracture patterns with medial column deficiency. However, as fixation choice should be guided by patient-specific factors and surgical feasibility, these findings should be interpreted in the context of experimental conditions and loading limitations.
Nanocrystals exhibit size-dependent structural behaviour because surface and interface effects become increasingly important as the characteristic domain size decreases. In this work, we investigate the size dependence of three diffraction-derived structural parameters - lattice parameter, isotropic Debye-Waller coefficient and microstrain - by combining molecular dynamics simulations of spherical Pd, Fe and Ti nanocrystals with powder diffraction analysis and comparison with representative literature data. The atomistic results show that small nanocrystals are characterized by a mean compressive state together with pronounced surface-stress inhomogeneity, while enhanced atomic displacements are concentrated in the outer coordination shells. On this basis, three simple nanoscale hypotheses are proposed. The lattice parameter is described as an ideal capillarity-driven reference trend with leading 1/D (where D is the particle diameter) behaviour, although comparison with experimental data confirms that this quantity is not universal and may be strongly modified by surface chemistry, defects, non-stoichiometry and morphology. By contrast, the Debye-Waller coefficient follows a more general 1/D decrease, consistent with a surface-shell picture of enhanced vibrational and static disorder. Microstrain arises from surface-stress gradients and, in the small-particle limit, exhibits a natural 1/D2 dependence associated with surface-stress heterogeneity; over broader size ranges, a mixed 1/D + 1/D2 form provides a more effective description. These results support a heuristic surface-driven interpretation of nanoscale structural disorder and clarify the different degrees of generality of the three scaling laws.
The relationship between molecular structure and macroscopic function is a foundational principle in materials science, in which subtle molecular variations produce pronounced differences in strength, stiffness, and elasticity of macromolecular solids. In hydrogels, replacing static covalent bonds with dynamic covalent bonds (DCBs) creates newfound capabilities, including self-healing and recyclability. Herein, substitutional differences in dual DCB imine boronic ester crosslinkers, together with matrix pH, influencing hydrogel properties are investigated. A comparison of ortho- and para-imine boronic esters showed the formation of 3-amino-benzoxaborole heterocycles in hydrogels derived from 2-formylphenylboronic acid. Tautomerization to the heterocycle significantly enhanced hydrogel elasticity, despite a lower crosslinking density than in hydrogels formed with 4-formylphenylboronic acid. Two closed-loop end-of-life (EOL) management pathways are also demonstrated. Reprocessing through self-healing is accomplished, with hydrogels regaining at least 90% of their original rheological properties. A fully circular recycling pathway is also established, recovering all starting materials for reuse, with recycled hydrogels achieving over 100% recovery of rheological properties. Overall, the presence of previously undisclosed 3-amino-benzoxaborole structures is demonstrated, expanding the understanding of formylphenylboronic acids in polymeric materials, and complete closed-loop EOL pathways are designed to inspire greater focus on full EOL processes in materials circularity.
Brown adipose tissue (BAT) activity in humans is linked to better glucose metabolism and cardiometabolic health. In mice, cold exposure increases circulating succinate and enhances BAT thermogenesis, but whether similar succinate responses occur in humans remains undetermined. We analysed data from 33 young (18-25 years), sedentary adults (20 women) from the ACTIBATE cohort (NCT02365129). Plasma succinate was measured enzymatically at baseline and 60 and 120 min during a personalized cooling protocol. BAT volume, activity, and radiodensity were assessed via static 18F-fluorodeoxyglucose positron emission tomography-computed tomography (18F-FDG PET-CT) at the end of the cold exposure. Circulating succinate changed over time during cold exposure, with no significant change at 60 min and a significant increase at 120 min versus baseline (+ 23.8%, P = 0.017), with substantial interindividual variability. Exploratory stratification by median baseline succinate revealed divergent temporal patterns: the low-baseline group showed increases at 60 min (+ 46%) and 120 min (+ 80.5%; both P ≤ 0.021), while the high-baseline group showed reductions at both timepoints (-31.4% and - 2.4%; both P ≤ 0.003). These profiles were not explained by phenotypic or BAT-related parameters. In the full cohort, changes in circulating succinate were not associated with BAT outcomes. Overall, our findings indicate that circulating succinate responses during acute cold exposure are highly heterogeneous and no associated with BAT PET/CT-derived parameters in humans.
The current management of acute pancreatitis (AP) primarily relies on supportive measures, such as fluid resuscitation, nutritional support, and infection control. However, these approaches do not adequately address the core drivers of the disease. This limitation arises from an incomplete understanding of the progression from local injury to systemic inflammation and repair, which is a dynamic process with underlying immunoregulatory mechanisms that remain insufficiently characterized. Macrophages are the key effector cells involved throughout the disease course and exhibit pronounced spatiotemporal heterogeneity during this progression. Therefore, they are central to decoding the evolution of the disease and facilitating precise interventions. In terms of spatial dynamics, tissue-resident macrophages (TRMs), which are derived from embryonic sources, and monocyte-derived macrophages (MDMs) recruited from the bone marrow serve functionally complementary roles. Their relative dominance shifts in conjunction with disease progression rather than remaining static. Temporally, the macrophage phenotype undergoes a programmed evolution, beginning with an early phase dominated by M1 proinflammatory responses, transitioning through an intermediate phase where injury and repair coexist, and culminating in a late phase characterized by M2-dominated reparative coordination. This evolution is accompanied by corresponding metabolic reprogramming. Recent single-cell and spatial multiomics studies have unveiled a functional continuum that goes beyond the traditional M1/M2 dichotomy, revealing a rich diversity of cellular subsets and their spatial niches. This insight shifts targeting strategies from broad anti-inflammatory interventions toward more precise modulation aimed at specific phases, subsets, and regions. This review systematically examines the design principles, strengths, and limitations of three classes of intervention-molecular targeting, bioactive natural products, and nanoscale delivery-and identifies the obstacles that continue to impede their clinical translation. Building on this analysis, we propose a dual-dimensional (spatiotemporal) strategy of precise modulation, integrating single-cell and spatial omics, chronobiological principles, and the traditional Chinese medical concept of yin shi zhi yi (adapting treatment to timing), with the aim of shifting AP therapy from symptomatic support toward cause-directed repair.
Autism spectrum disorder (ASD) is characterized by persistent deficits in social communication and the presence of restricted and repetitive behaviors. While ASD has a neurodevelopmental origin, it remains a lifelong condition, yet little is known about how its behavioral and neural features evolve across adulthood. Here, we investigated behavioral, synaptic, and structural alterations across the transition from early to mature adulthood in Cntnap2 knockout mice, a widely used model of ASD. Using a longitudinal behavioral approach combined with electrophysiological recordings and morphological analysis, we show that KO mice exhibit increased stereotyped and repetitive behaviors and reduced exploratory activity at both ages. However, detailed analysis of behavioral patterns revealed age-dependent differences, with early adult KO mice displaying increased behavioral persistence that later evolved into distinct patterns of behavioral sequences. These behavioral changes were associated with alterations in inhibitory synaptic transmission in the dorsolateral striatum (DLS), including changes in spontaneous inhibitory postsynaptic current (sIPSC) frequency and temporal structure. In parallel, mature adult KO mice showed structural remodeling of spiny projection neurons, characterized by increased distal dendritic arborization and age-dependent organization of dendritic spines. Together, our findings demonstrate that ASD-related alterations are not static but evolve across adulthood, revealing a multi-level reorganization of behavioral, synaptic, and structural features. These results highlight the importance of considering adulthood stages in ASD and provide new insights into the dynamic nature of the condition.
Hypotension is a common adverse effect of spinal anesthesia and poses increased risks for patients with diabetes mellitus. While various tests exist for diagnosing diabetic autonomic neuropathy, dynamic pupillometry is a simple and cost-effective option. This study aimed to assess whether dynamic pupillometry can predict post-spinal hypotension in diabetic patients. In this observational study, 75 patients with diabetes mellitus who were scheduled for surgery under spinal anesthesia underwent a pupillary examination the day prior. Both static and dynamic parameters were recorded. On the day of surgery, 3 ml of 0.5% hyperbaric bupivacaine was administered for spinal anesthesia, and instances of intraoperative hypotension and the use of vasopressors were noted. Post-spinal hypotension occurred in 77% of the 75 patients studied, with 27 experiencing mild hypotension (20-30% drop in MAP) and 31 having severe hypotension (>30% drop). The severe hypotension group had a significantly lower baseline pupil radius (3.86 ± 0.52 mm) compared to the no hypotension group (4.37 ± 0.57 mm, P = 0.006). The area under the receiver operating characteristic curve for baseline pupil radius was 0.720, with a threshold of less than 4.01 mm showing 70.7% sensitivity and 70.6% specificity for predicting post-spinal hypotension. Baseline pupil radius measured using pupillometry may serve as a simple noninvasive marker to identify diabetic patients at risk of post-spinal hypotension.
Precise coastal chlorophyll-a prediction is essential for marine ecosystem assessment and harmful algal bloom warning. Nonetheless, satellite-based chlorophyll-a measurements generally contain large contiguous gaps. Traditional ordered space-time processing tends to extract and propagate invalid features, and typical static graph representations are not sufficient to model global and time-varying spatial correlations in chlorophyll-a fields, which restricts forecasting precision. To overcome these issues, we introduce the BiTGraph (Biased Temporal Convolution Graph Network), an end-to-end framework for marine chlorophyll-a forecasting under high missing rates. The framework forecasts and handles missing values jointly, reducing information distortion from separate interpolation. BiTGraph includes a multi-scale instance partial temporal convolution module (MSIPT) and a biased graph convolution module (Biased GCN). MSIPT integrates partial temporal convolution with a dynamic mask-update mechanism to learn reliable temporal features in the presence of missingness and increase effective temporal information spread. Biased GCN learns an adaptive graph with missingness-pattern-aware bias, allowing effective cross-node information transfer and feature fusion. Through a hierarchical architecture with alternating MSIPT and Biased GCN, BiTGraph can support joint spatiotemporal completion and forecasting with extreme missingness. Experiments on Bohai Sea and the South China Sea data demonstrate that, at missing rate r=0.2-0.6 and 15-step long-horizon forecasting, BiTGraph consistently achieves lower MAE and RMSE than the best baseline models, with average MAE/RMSE/MAPE at r=0.6 of 0.594/1.081/16.15% (Bohai Sea) and 0.212/0.905/24.44% (South China Sea), offering an effective solution to coastal chlorophyll-a prediction with highly incomplete information.
Motion artifacts present a major challenge for intravital imaging of tissues undergoing physiological movement or mechanical loading. Blurring or artificial changes in image intensity due to shifting on the z-axis reduce data reproducibility and reliability. Existing post-acquisition approaches can partially compensate for motion, but they increase experimental complexity and are not well suited for use in mechanically loaded bone. Therefore, we developed a novel method for real-time correction of axial motion during mechanical loading of bone by synchronizing the movement of the objective to the actuator. Synchronization was achieved by linking the position of the actuator piezo motor to the objective piezo motor with a user refined reduction via potentiometer. Applying axial motion correction effectively removed artificial changes in fluorescent intensity in a static fluorescent marker up to 3000με in bone as measured by similarity and average intensity before and during loading. This improvement was reflected in the improved accuracy in capture of a dynamic fluorescent calcium indicator (GCaMP6f) in osteocytes. Our system provides a user-friendly, robust framework that can be easily adapted to other mechanically loaded tissues, improving data collection and expanding the utility of two-photon imaging across a variety of biological applications.
This study aims to evaluate the biomechanical effects of proximal fibular osteotomy (PFO) on tibiofemoral and proximal tibiofibular load transfer under simulated neutral and varus alignment conditions using a lower-extremity finite element analysis (FEA) model. A patient-specific three-dimensional FEA model of the lower extremity was developed from computed tomography and magnetic resonance imaging data of a healthy volunteer. Neutral, 3° varus, and 5° varus alignments were simulated before and after PFO under a simplified physiological axial load of 900 N. Tibiofemoral compartment forces, proximal tibiofibular joint (PTFJ) force, distal tibiotalar reaction force, and peak von Mises stresses within the cartilage-meniscus complex were evaluated. Proximal fibular osteotomy markedly reduced PTFJ force across all alignment conditions. Under neutral alignment, PTFJ force decreased from 122.94 N to 21.84 N, while distal tibiotalar reaction force increased from 787.55 N to 897.89 N, indicating redistribution of load through distal pathways. Tibiofemoral compartment forces changed minimally. Under 5° varus alignment, medial tibiofemoral force decreased only from 608.45 N to 603.78 N following PFO. Peak medial meniscal and cartilage stresses increased with varus alignment and remained largely unchanged after osteotomy. In this single-subject static FEA model, PFO substantially reduced proximal tibiofibular loading, but produced only limited changes in tibiofemoral load distribution under simulated varus alignment. Within the present model, load redistribution occurred predominantly through distal load-transfer pathways rather than substantial unloading of the medial tibiofemoral compartment. These findings suggest that the biomechanical effect of PFO may primarily involve alteration of fibular load transmission rather than meaningful reduction of medial knee loading.
Heart transplantation remains the definitive treatment for end-stage heart failure. Preservation of donor hearts is a critical determinant of graft function and post-transplant outcomes. Traditional static cold storage (SCS) has limitations in ischemia tolerance and functional assessment, particularly for marginal and donation after circulatory death (DCD) hearts. Emerging strategies, including normothermic ex vivo perfusion (NEHP) and hypothermic oxygenated perfusion (HOPE), enable extended preservation, functional evaluation, and improved utilization of high-risk donor organs. This review summarizes current preservation techniques, discusses mechanistic strategies to mitigate ischemia-reperfusion injury, highlights emerging trends and controversies, and identifies future research directions to optimize outcomes in heart transplantation.
Programmable integrated photonics aims to replicate the versatility of field-programmable gate arrays in the optical domain. However, scaling these systems has been prevented by the high power consumption and thermal crosstalk of conventional volatile phase shifters. Here we introduce a non-volatile field-programmable photonic gate array, implemented on a hybrid silicon-barium titanate platform, which overcomes the power scaling limitations of previous technologies. Unlike traditional thermo-optic devices that require constant power to maintain a state, our device utilizes ferroelectric domain switching to provide non-volatile memory, allowing optical circuits to be programmed and retained without any holding power or electrical bias. The hexagonal waveguide mesh integrates 58 programmable unit cells and 116 actuators, achieving nanosecond-scale switching speeds of 80 ns while reducing static power consumption to negligible levels (560 nW per π phase shift). To validate this platform, we configured the mesh to perform diverse signal processing functions, including tunable filtering, 4 × 4 linear unitary transformations and optical routing. This work establishes non-volatile ferroelectric silicon photonics as a scalable, heat-free platform essential for the next generation of energy-efficient photonic computing.
Chimeric antigen receptor T-cell (CAR-T) therapy has achieved remarkable success in hematologic malignancies, yet demonstrates limited efficacy in solid tumors, including hepatic cancers. T-cell exhaustion and insufficient persistence represent major obstacles. We hypothesized that optimizing both manufacturing processes and CAR structural design could reduce exhaustion and enhance therapeutic outcomes in immunocompetent mouse models of primary and metastatic liver malignancies. We systematically compared antibody-based versus bead-based activation methods, evaluating their effects on T-cell exhaustion phenotypes. Retroviral vector (RVV) production was optimized for murine T-cell transduction, assessing vector stability and T-cell phenotypes. Different cytokine conditions were tested for their impact on T-cell expansion, memory phenotypes, and anti-tumor efficacy using GPC3-targeted CAR-T cells in hepatocellular carcinoma models. Through structural prediction and electrostatic field simulation, we identified that high positive charge patches (PCP) in the EpCAM-targeting G8.8 scFv caused CAR clustering and tonic signaling. We generated charge-optimized variants and evaluated their therapeutic efficacy in an immunocompetent colorectal cancer liver metastasis model. Antibody activation showed superior homogeneity and expansion despite initially higher exhaustion markers, which equilibrated by day 10 without affecting viability. RVV harvested at 72 hours post-transfection yielded optimal titers. RVV remained stable through freeze-thaw cycles. IL-7 supplementation to IL-2 significantly enhanced memory phenotypes, reduced exhaustion, and improved tumor control in GPC3-CAR-T therapy. Electrostatic optimization of G8.8 scFv substantially reduced tonic signaling, decreased T-cell exhaustion, and enhanced anti-tumor efficacy in the MC38-EpCAM model. Systematic optimization of manufacturing conditions and structure-based charge engineering of CAR constructs synergistically enhance therapeutic efficacy against liver malignancies, providing a translatable framework for improving solid tumor CAR-T therapy.