APL Bioengineering

Adoptive T-cell therapies rely on the identification and expansion of rare tumor-reactive T cells, yet current enrichment strategies are limited by the low abundance of these cells and complexity of their functional enrichment. Here, we present a microfluidic platform that exploits hydrodynamic shear as a controllable parameter for enriching antigen-specific T cells through peptide-major histocompatibility complex (pMHC)-mediated capture. An eight-channel microfluidic device was engineered to simultaneously interrogate a range of wall shear stresses while maintaining uniform cell delivery, enabling systematic identification of shear conditions that maximize antigen-specific enrichment. Using engineered MART-1-specific Jurkat cells, we demonstrate that T-cell capture is jointly regulated by wall shear stress and pMHC density, with intermediate shear preferentially enriching antigen-specific cells over nonspecific binders. Translation of the optimal operating condition to a high-throughput single-shear device enabled approximately 35-fold enrichment of antigen-specific T cells from peripheral blood mononuclear cells containing only 0.05% target cells. We further show that peptide-MHC complexes isolated directly from melanoma whole-cell lysates support shear-dependent enrichment comparable to recombinant pMHCs. Finally, primary MART-1-specific CD8 T cells enriched using tumor-derived pMHCs retained the ability to recognize melanoma cells and upregulated the activation marker CD137 following antigen-specific stimulation. Together, these findings establish hydrodynamic shear as an orthogonal parameter for antigen-specific T-cell enrichment and provide a framework for integrating force-based selection with tumor-derived pMHCs to isolate functional antigen-specific T cells using tumor-derived pMHCs.

Vascular chips have advanced endothelial mechanobiology by enabling controlled responses to hemodynamic cues, yet disease-relevant wall shear stress (WSS) modeling remains limited. Simplified one-dimensional flow shear systems, designed mainly for physiological mechanobiology, miss the topological organization of pathological flow, whereas patient-specific vascular models capture complex hemodynamics but sacrifice generality and imaging compatibility. Here we develop a programmable vascular chip that converts disease-associated WSS topology into a physiologically parameterized experimental input. The device reconstructs a representative pathological shear-topology field on endothelial layer, supports stationary and physiologically paced oscillatory flow modes, and integrates matched unidirectional-shear references within the same chip. Using this system, we show that oscillatory WSS topology destabilizes endothelial monolayers, drives asymmetric collective emergent behaviors, impairs actin-nuclear mechanotransduction, accompanied by nuclear softening and enhanced perinuclear nanoparticle uptake. Integrated live-cell imaging, fluorescence analysis, Brillouin microscopy, and transport assays enable multimodal phenotyping across collective, subcellular mechanical and functional scales. By making disease-relevant WSS topology experimentally controllable, this vascular-chip framework bridges computational hemodynamics and experimental mechanomedicine, supporting standardized vascular disease modeling and functional screening.

High-grade serous carcinoma (HGSOC) is the deadliest subtype of ovarian cancer, characterized by high metastatic rates. HGSOC is typically diagnosed at late stages, and treatment options are limited, resulting in a 60% recurrence rate. HGSOC cells exhibit metabolic plasticity, dynamically shifting between glycolysis and oxidative phosphorylation (OXPHOS) to meet energy demands for tumor progression. To evaluate therapeutic strategies that target metabolic vulnerabilities, we developed a microphysiological system (MPS) that recapitulates the heterogenous cell states and bioenergetic distribution of HGSOC solid tumors. Our platform utilized HGSOC spheroids embedded in a collagen hydrogel that mimics the extracellular matrix to capture tumor progression in the ovary. We used atovaquone (ATO), an FDA-approved OXPHOS inhibitor, to prototype the capabilities of our platform to investigate metabolic plasticity in HGSOC. Treatment with ATO decreased viability and invasion of HGSOC spheroids. Crucially, ATO exhibited no cytotoxicity toward biomimetic blood vessels, preserving their integrity and permeability. Metabolic imaging revealed that ATO induces an oxidative state in the outer region of the spheroids. At the invasive front, ATO disrupted mitochondrial organization, forcing collective cell migration and eventually inducing breakdown of mitochondrial networks. Furthermore, ATO decreased YAP/TAZ pathway activity in the outer region of the spheroid, providing a potential mechanism for hindered cell invasion. Collectively, our data demonstrates that a low-potency OXPHOS inhibitor like ATO can effectively target metabolic plasticity to suppress HGSOC spheroid progression. Overall, this platform successfully recapitulated metabolic heterogeneity and provided a workflow for safely testing other drugs that target cancer metabolism.

The endothelial cells (ECs) that line blood vessels continuously sense and respond to the physical forces exerted by blood flow. In vivo, pulsatile arterial flow interacts with vessel curvature, branching and other anatomical features to generate complex local hemodynamic environments that dictate the magnitude, direction, pulsatility, and oscillatory nature of wall shear stress experienced by ECs. Currently, accessible and reproducible in vitro models of complex pulsatile flow that recapitulate in vivo vascular anatomy remain limited. Here, we combine a novel rotational-flow endothelial culture platform with detailed computational fluid dynamics (CFD) modeling to characterize four well geometries designed to generate distinct hemodynamic environments. CFD analyses demonstrate that these geometries intrinsically generate pulsatile flow and produce reproducible spatially distinct regions of wall shear stress magnitude, pulsatility, and oscillatory shear within a single culture well. Endothelial alignment mapping and functional assays reveal region-specific cellular responses to the predicted local flow conditions that closely corresponded to the predicted local hemodynamic environment, linking complex flow patterns to endothelial adaptation. The technical advancements of our modeling efforts should support a faster, cheaper, simpler, and--importantly--validated framework for future investigation into EC mechanobiology under complex flow conditions. HIGHLIGHTSO_LISimple engineered well geometries generate distinct hemodynamic microenvironments, mimicking in vivo vascular structures, using a conventional orbital shaker. C_LIO_LIComputational fluid dynamics (CFD) reveals spatially distinct patterns of wall shear stress, pulsatility, and oscillatory shear applied to ECs within individual culture wells. C_LIO_LIHigh average wall shear stress and elevated oscillatory shear index induces a unique perpendicular alignment of ECs to the dominant flow vector. C_LI

Control of fluids is a hallmark of microfluidic systems and fundamental for the successful application of microfluidic devices. Trigger valves use geometric features to autonomously control the release of fluids in microfluidic devices. Our previous work has adapted geometries used in closed trigger valve systems to enable use in open systems, allowing for open microfluidic devices with up to three trigger valves. Here, we focus on the parallel co-flows produced by sequential release of trigger valves and present a model that predicts their layer widths as a function of the geometric characteristics of the different side channels of each trigger valve. We show layered co-flows with widths as low as 50 microns. Additionally, we expand the use of trigger valves in open microfluidic devices by incorporating 1) varied step heights, 2) devices with up to seven trigger valves, and 3) use of varied fluids and plastics. To validate the implementation and use of these trigger valves in open systems, we have developed a theoretical framework to compare predicted outcomes (i.e., fluid travel distance, velocity, and layering width) with our experimental values. This theoretical work offers applications in various fields, including hydrogel patterning for 3D cell culture, organ-on-a-chip models, at-home sample preparation, and autonomous microfluidic systems for biosensing.

Glioblastoma (GBM) is a highly aggressive brain tumor with a five-year survival rate of less than 5%. The current standard of care established 20 years ago includes maximal surgical resection and administration of alkylating agent temozolomide (TMZ). GBM is highly invasive, and GBM cells that evade surgical resection can become resistant to TMZ and develop new aggressive secondary tumors. Post-relapse there are few treatment options available to patients. Tissue engineering approaches suggest the opportunity to develop in vitro models of the GBM tumor microenvironment that may accelerate the discovery of novel therapies for GBM. Here, we report the adaptation of hydrogel microdroplets (microgels) to encapsulate GBM cells in a tailorable 3D matrix to assess patterns of growth and to screen TMZ drug response using patient-derived xenograft (PDX) specimens. We exploit a unique aspect of the microgel system to account for the cellular heterogeneity within the tumor microenvironment (TME). We combine cell-laden microgels generated from TMZ-resistant and TMZ responsive variants of the same PDX specimens to create heterogeneous populations with varying levels of drug sensitivity. We demonstrate a range of drug resistance phenotypes as a function of the ratio of TMZ-responsive to resistance cells and identify the population required for TMZ-resistance to overtake take the response. We then investigate the influence of tumor mimetic shifts in hyaluronic acid bioavailability and hypoxia on patterns of TMZ resistance. We show exposure to matrix-bound hyaluronan increases TMZ resistance and the glioma stem cell population in both cell variants. Lastly, we report an increase in TMZ sensitivity but divergent changes in the GSC subfraction for TMZ resistant vs responsive GBM in the presence of hypoxia. Together, we demonstrate the versatility of cell-laden microgel approach to replicate heterogenous tumor populations, model shifts in the tumor microenvironment, and rapidly screen therapeutic response.

Neurological disorders are a major cause of death and disability worldwide. The brain's energy metabolism is essential to its proper function, yet the mechanisms driving neuroenergetic dysfunction remain poorly understood. A key challenge is the limited availability of human-relevant models that can reproduce the complexity of brain physiology. An Organ-on-Chip (OoC) system was developed to mimic the neurovascular unit metabolic coupling by incorporating human isogenic iPSC-derived endothelial-like cells, pericyte-like cells, astrocytes, and a cerebral organoid, representing the main cellular components of the NVU. The novel, customized microfluidic platform enables research on neurovascular coupling by interconnecting a blood-brain barrier-on-a-chip model with a 3D brain parenchymal compartment to mimic physiological conditions.

Angiogenesis drives the formation and remodeling of capillary networks throughout tissue repair, regulating the vascular environment that supports healing and tissue remodeling. Experimental characterization of these processes is commonly performed using CD31-stained histological tissue sections to quantify capillary surface density and morphology throughout healing. However, these measurements provide only two-dimensional characterization of an underlying three-dimensional (3D) vascular network, limiting direct estimation of volumetric capillary density and vascular architecture. To address this limitation, an experimentally informed framework was developed to generate representative 3D capillary networks, enabling estimation of volumetric capillary density from histologically quantified vascular measurements. CD31-stained histological sections obtained from a longitudinal porcine lumpectomy study were analyzed to quantify the percentage of CD31-positive area (%CD31+) and capillary morphology within healthy tissue and healing surgical cavities. Histologically quantified morphology distributions and literature-informed vascular branching characteristics were incorporated into a capillary network generation framework to construct representative 3D vascular networks. Capillary branches were iteratively generated within representative tissue volumes until virtual histological sections reproduced experimental %CD31+ measurements, enabling estimation of volumetric capillary density. Generated capillary networks demonstrated good agreement with experimentally characterized 3D vascular architecture, while simulated histological sections accurately reproduced experimentally quantified capillary counts and vascularization measurements. Application of the framework to the porcine lumpectomy dataset captured temporal changes in vascular remodeling throughout healing, revealing progressive increases in volumetric capillary density and vascular maturation. Collectively, this framework provides an experimentally informed methodology for relating histological vascular measurements to volumetric capillary density estimates, supporting future computational studies of angiogenesis and tissue repair.

Breast cancer cells that disseminate to distant organs can remain dormant (non-proliferative) for years before reactivating and progressing into lethal metastatic disease. Understanding the transition between dormancy and reactivation is therefore critical for early intervention and treatment. In this study, we investigate a comprehensive range of deep learning (DL) architectures to classify dormant versus proliferative breast tumor cells within a 3-dimensional growth factor reduced basement membrane extract (3D BME) system that models tumor dormancy and outgrowth. To capture the underlying spatiotemporal dynamics, we evaluate both spatial and sequence-based learning approaches. We consider convolutional neural networks (EfficientNet, ResNet, DenseNet, MobileNet, VGG, AlexNet), segmentation-based models (U-Net, U-Net++, Attention U-Net, DeepLabV3, HRNet) and transformer-based architectures (Vision Transformer, Swin Transformer, SegFormer). We investigate transfer learning using both fixed and fine-tuned strategies. Experimental results show that classification performance is greatly enhanced through the integration of temporal information. EfficientNet-B7, EfficientNet-B6, DenseNet-169, and DenseNet201 are consistently better than competing architectures for all tested models. EfficientNet-B7 with the use of temporal sequences input reaches an accuracy of 98.86% with a ROC-AUC of 0.998. The results highlight the significance of spatio-temporal feature learning and the value of DL frameworks in automated classification of dormant versus proliferative breast cancer cells in physiologically relevant microenvironments.

Diagnostic assessment of breast cancer biopsies remains reliant on resource-intensive histopathology and molecular profiling, which often lack real-time physiological readouts. Magnetic levitation (MagLev) enables label-free density profiling of single cells, yet its application to intact tissue biopsies has been precluded by size-dependent geometric artifacts and the absence of analytical frameworks for biopsy-scale samples. Here, we report the first application of MagLev to intact invasive breast carcinoma biopsies (200-600 m) for biophysical profiling, generating multivariate biophysical signatures from 203 samples across 17 patients. We developed a physics-based size-correction algorithm (xmc) that isolates biological density from geometric artifact, and demonstrate that tissue viability is predicted not by average levitation height, but by spatial heterogeneity across replicate samples, reflecting the microenvironmental complexity of metabolically active tumors. Multivariate integration using Partial Least Squares (PLS) regression and Factor Analysis of Mixed Data (FAMD) identified nodal status (N) as the strongest biophysical predictor, suggesting that lymphatic dissemination capacity leaves a measurable signature in the primary tumor density profile. Unsupervised patient clustering in PLS-derived latent space recovered three clinically coherent subgroups aligned with molecular subtypes. This 30-minute, low-cost assay provides exploratory biophysical stratification complementary to existing diagnostics, particularly in resource-limited settings.

A wide spectrum of microparticles is inhaled with each breath, deposited on airway surfaces, entrapped in the mucus, and removed by mucociliary transport (MCT). However, the influence of particle size on MCT remains largely unknown. Here, we investigated the MCT of microparticles with a trachea-on-a-chip method that integrates a micro-machined device with a trachea explant from newborn pigs. This method preserves airway structures for mucus secretion and cilia beating (e.g., airway surface epithelia and submucosal glands), maintains physiological air-liquid-interface on the airway surface, and allows tracks motion of microparticles with high resolution. Using this method, we found that, in normal airways, 6 um polystyrene particles clear rapidly, whereas 102 um particles clear slower and require mucus strands for motion. In cystic fibrosis (CF) airways, MCT of microparticles reduces, but particle size-dependence persists. Methacholine increases particle motion in normal airways, but not in CF airways. These findings suggest two distinct MCT processes, in which large particles rely on mucus strands for clearance, small particles can be cleared independent of mucus strands, and CF disrupts both.

Despite major improvements in molecular characterization of breast cancer, current biomarkers still fall short in accurate treatment prediction. Interrogating tumor tissue ex-vivo in its native conformation is a direct strategy for guiding treatment of individual patients but presents a challenge. In this study, we developed a microfluidic tissue array (FTA) using small biopsy samples (< 1mm3) mimicking physiological flow for consistent exchange of nutrients and waste, retaining the tumor native stroma. Cell/patient-derived breast cancer xenograft tissues were maintained over 2 weeks in the array and their response to therapeutic agents, doxorubicin or neratinib, were interrogated. Drug response in the uFTA showed >2-fold reduction in tumor cell viability which corroborated tumor size shrinkage in mice bearing the same tumor load. EdU/Ki67 assays indicated selective retention of cells with higher proliferative capacity after drug treatment, underscoring in vivo clinical relevance . We have also developed a valved-FTA to increase throughput and variety of treatment conditions on the same chip. Together, this FTA can be staged as a powerful, low-cost benchtop theranostic tool for personalized cancer therapeutics compatible with FDA New Approach Methods.

Control over network dynamics at different length scales is a feature of natural materials challenging to replicate in synthetic hydrogels. Hydrogel viscoelasticity is commonly controlled by tuning the kinetics of reversible crosslinks; however, this strategy inherently links the resulting macroscale and nanoscale dynamics of the individual network components. Taking inspiration from biological materials that feature lipids as structural elements, we introduce Lipid Network Crosslinked (LINC) hydrogels that exploit the mobility of individual lipids within self-assembled liposomes as covalent, network-crosslinking points. These mobile, covalent crosslinks increase hydrogel stress relaxation rates over 20-fold compared to polymer-only hydrogels with equivalent crosslinking chemistries and stiffnesses. We demonstrate that liposome design parameters, including degree of surface functionalization and tail saturation, provide a means to independently control the macroscale storage moduli and stress relaxation behavior. Finally, as an application where control over network dynamics at different length scales is critical, we placed cell-adhesive ligands onto more mobile or less mobile network elements. Human neural progenitor cells cultured within LINC hydrogels of identical macroscale viscoelasticity significantly altered their phenotype in response to nanoscale ligand dynamics. These results establish LINC hydrogels as biomimetic materials that leverage nanoscale lipid mobility within a macroscale polymeric network to control dynamics at multiple length scales.

Understanding how neurodegenerative diseases initiate and propagate through neural circuits remains a fundamental challenge in neuroscience. The earliest stages occur years before symptoms emerge, making them inaccessible to study in patients. Microfluidic platforms, where neurons communicate across chambers through microchannels accessible only to their axons, have opened new experimental avenues. However, existing models lack the complexity and precision needed to track how individual circuit components respond to focal pathological changes over time. Here we present a 33-chamber cortical network-on-chip integrating human iPSC-derived excitatory neurons, inhibitory neurons, and astrocytes in a six-layer feedforward architecture recapitulating the laminar structure of the neocortex. Amyloid-{beta} is applied globally, while progerin-induced accelerated ageing in a single chamber establishes a defined disease core. Continuous recordings using high-density microelectrode arrays reveal progressive, layer-dependent changes in firing dynamics and network topology. Machine-learning-based feature analysis identifies a multiparametric electrophysiological signature distinguishing healthy from disease-affected chambers, enabling studies of the earliest timepoint at which pathology becomes detectable. This establishes a scalable framework for mechanistic studies of neurodegeneration and identification of electrophysiological biomarkers of disease progression.

Cell states are increasingly conceptualized as attractors of high-dimensional dynamical systems, yet quantitative approaches for integrating phenotypic information into this framework remain limited. Here, we take an image-based approach that combines unsupervised machine learning (ML) with timelapse imaging to extract and characterize the temporal dynamics of morphological features. Using a cell line with endogenously tagged VE-cadherin, we acquired brightfield and fluorescence timelapse images of human induced pluripotent stem cell-derived endothelial cell (hiPSC-EC) monolayers, which adopt distinct phenotypes at two different magnitudes of shear stress in terms of their morphology, behavior, and VE-cadherin organization. To quantify these phenotypic cell states without segmentation, we trained a diffusion autoencoder to predict VE-cadherin signal from brightfield images. We identified interpretable ML-based features representing cell orientation, elongation, and local density. Treating these variables as dimensions of a morphological state space, we estimated a data-driven vector field and found that the two observed phenotypic cell states correspond to stable fixed points of the inferred dynamical system. Mapping measured cell migration coherence onto this space further distinguished the states. Imaging cells across intermediate shear stresses revealed a regime of bistability in which both states coexist, indicating that the shear-stress-dependent transition between endothelial cell states occurs as a bifurcation of the inferred dynamical system. Finally, we applied this method to study an N-terminal truncation of VE-cadherin, finding that mutated cells preserve alignment and coherent migration, but exhibit altered morphology and increased migration speed. This work demonstrates the applicability of a dynamical systems approach to quantitatively characterize morphological aspects of cell state from interpretable ML-based features.

We demonstrate a new sample-loading strategy ('spot-loading') for Spores-on-a-Chip microfluidic sensors that enables multiplexed experimentation. The previous limitation of one biological sample per device is overcome through controlled sample loading at an intermediate step of the chip fabrication process. As a proof-of-concept, we use dual spore chips to compare the germination behaviours of two spore strains in two distinct microenvironments.

Purpose: Hyperthermia is emerging as an adjunct strategy in pediatric oncology, yet its translation is limited by poor understanding of how different modalities impact complex tumor microenvironments. Neuroblastoma (NB), the most common extracranial solid tumor in children, displays profound heterogeneity that hampers therapeutic predictability. Here, we performed the first systematic head-to-head comparison of photothermal therapy (PTT) and magnetic hyperthermia (MH) in tissue-engineered NB (TE-NB) models. Methods: TE-NB scaffolds incorporating five NB cell lines were loaded with magnetic nanoparticles (MNPs) and subjected to PTT (808-nm laser, 130 W/cm2, 10 min) or MH (285 kHz, 20 mT, 60 min). Constructs were analyzed at 24 h, 48 h, and 5 d post-treatment for DNA content, cell viability, proliferation (Ki67 immunohistochemistry), and apoptosis (caspase-3/7 staining). Results: MH produced consistent MNP-dependent heating with minimal background, while PTT was dominated by nonspecific medium absorption. Both modalities modulated proliferation within 24 h, but effects varied sharply by cell line and scaffold region, reflecting microenvironmental heterogeneity. By 48 h, PTT often triggered paradoxical increases in proliferation, whereas MH disrupted scaffold integrity, reduced DNA content, and suppressed Ki67 expression. Notably, neither modality induced sustained caspase-3/7 activation, indicating that cytotoxicity proceeds via non-apoptotic pathways. Conclusion: Our findings position MH as a superior modality for uniform heating and proliferation control in 3D NB models, but also highlight that hyperthermia should be considered a context-dependent modulator rather than a binary cytotoxic agent. By integrating patient-specific TE-NB platforms, precision hyperthermia could define individualized therapeutic windows, optimize combinations with pro-apoptotic or immunomodulatory agents, and accelerate translation of hyperthermia strategies for children with NB.

Vascular endothelial cells (ECs) form a monolayer lining blood vessels and serve as a barrier between blood and tissues. Clathrin-mediated endocytosis (CME) is a major internalization pathway that involves a physical conformational change of the plasma membrane to form a vesicle and is therefore sensitive to the local environment. ECs are subjected to a myriad of fluid shear stress (FSS) rates from circulating blood, which we hypothesize affects CME. To test this, we used simultaneous two-wavelength axial ratiometry (STAR) microscopy, which provides nanoscale axial resolution, to determine the frequency and morphology of clathrin-coated vesicles as they form. Human umbilical vein endothelial cells (HUVECs) were transfected with dual-tagged clathrin light chain a (CLCa-iRFP-EGFP) and cultured under 10 dyn/cm2 FSS. CME activity was elevated in cells cultured under flow and assayed in static or flow conditions compared to statically cultured and imaged controls, indicating that FSS-induced changes to CME were maintained shortly after flow cessation. Single vesicle analysis showed cells cultured in FSS had a slight preference for vesicle formation with a flat-to-curved clathrin transition compared to control. Next, to assess the impact of different FSS rates, we cultured HUVECs at 20 and 40 dyn/cm2 FSS. We found total CME frequency was elevated compared to control at 20 dyn/cm2, but not 40 dyn/cm2. HUVECs cultured at both 20 and 40 dyn/cm2 had vesicles with increased lifetime and enhanced stability, as well as a higher proportion of vesicles formed through a flat-to-curved transition of clathrin.

Low response rates to immune checkpoint inhibitors (ICIs) in solid tumors are often driven by insufficient tumor-infiltrating CD8 T cells and immunosuppressive tumor microenvironment (TME). Although interleukin-2 (IL-2) potently expands and activates CD8 T cells, its clinical use is limited by rapid clearance, dose-limiting toxicity, and regulatory T cell (Treg) stimulation. Engineered IL-2 variants have not yet achieved meaningful clinical efficacy. Here, polymer-modified mesoporous silica nanoparticles displaying dense, unmodified wild-type IL-2 on their surface (IL2-NP) are developed, conferring proteolytic stability and tumor retention. IL2-NP enables avidity-mediated CD8 T cell binding and enhances proliferation and effector function without increased Treg binding or proliferation. Intratumoral IL2-NP expands CD8 T cells, increases CD8/Treg ratios, and reprograms TME through dendritic cell activation and M1-like macrophage polarization. IL2-NP induces regression of both treated and untreated distant colorectal tumors in a CD8 T cell-dependent manner. IL2-NP synergizes with ICIs and leads to complete tumor regression and immunological memory that protect against rechallenge. Treatment is well tolerated, with strong efficacy also observed in triple-negative breast and metastatic ovarian cancer models. Overall, intratumoral IL2-NP elicits robust systemic antitumor immunity, offering a promising strategy to enhance ICIs, cancer vaccines, and adoptive T-cell therapies. Graphical abstractThis work introduces a nanoparticle platform that overcomes major shortcomings of IL-2 immunotherapy by presenting wild-type IL-2 at high density on the nanoparticle surface, thereby increasing binding avidity to effector T cells. The resulting IL-2 nanoparticles enhance cytotoxic T cell expansion, reprogram the tumor microenvironment, and augment responses to immune checkpoint blockade to achieve robust ant-tumor immune response in mouse tumor models. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=182 SRC="FIGDIR/small/738558v1_ufig1.gif" ALT="Figure 1"> View larger version (82K): org.highwire.dtl.DTLVardef@12f8c8corg.highwire.dtl.DTLVardef@b46b1forg.highwire.dtl.DTLVardef@e4efc5org.highwire.dtl.DTLVardef@3993e6_HPS_FORMAT_FIGEXP M_FIG C_FIG

Innate immune agonists are promising therapeutic agents to induce immune responses against cancer. However, these agents have been limited by toxicity associated with systemic accumulation and activity in off-target cells. In this work, a targeted nanoparticle (NP) platform to encapsulate and protect the Toll-like receptor 3 (TLR3) agonist polyinosinic-polycytidylic acid (poly(I:C)) and promote its specific delivery to antigen presenting cells (APCs), macrophages and dendritic cells, for activation of this cell population was designed. To determine NP physiochemical properties that promote APC delivery, we developed a library of NP surface chemistries formed by electrostatic adsorption of polyanion coatings onto liposomes using layer-by-layer (LbL) assembly and screened the particles on APCs and off-target cells. Dextran sulfate was identified as a promising coating to enhance specific APC delivery. We applied these design parameters to develop a poly(I:C)-loaded NP for an APC-targeted immunotherapy. In a model of metastatic ovarian cancer, the LbL NP prolonged poly(I:C) retention in the peritoneal space--with 2-fold remaining 24-48hr after administration compared to free poly(I:C)--ultimately reducing systemic accumulation and associated toxicities. Compared to free drug, the NP reduced the increase in serum levels of TNF, IL-6, and CXCL10 by 9-, 4-, and 31-fold respectively. NP-treated mice experienced lower weight loss and recovered more quickly at a higher poly(I:C) dose, indicating a widening of the therapeutic window. The NP formulation enhanced accumulation of poly(I:C) in the tumor 2-fold and activation of the target APC population compared to free drug, and ultimately slowed tumor growth and extended survival in combination with doxorubicin chemotherapy. Overall, this work demonstrates a modular NP delivery strategy to improve the delivery, safety, and therapeutic window of a TLR3 agonist.