Biology subjects

Ardekani, A. M.

Publications and source records attributed to Ardekani, A. M..

9 recordsLinked to original sources

bioRxiv2026

A multiscale modeling framework for transport of PEGylated lipid nanoparticle through the extracellular matrix

Lipid nanoparticles (LNPs) are one of the leading platforms for delivering nucleic acid therapeutics, yet their efficacy is limited by physicochemical interactions with the extracellular matrix (ECM) that trap the particles before they reach target cells. PEGylated nanoparticles mitigate these interactions by forming a protective steric layer on their surfaces. However, there is a lack of a predictive tool that gives mechanistic insights about how PEG surface density governs the underlying interaction and results in enhanced diffusive transport of LNPs through the ECM. Here, we present a multiscale hierarchical computational framework that couples all-atom constant pH molecular dynamics (CpHMD) with a highly coarse-grained model of the complete LNP within a crosslinked hyaluronic acid (HA) network. These atomistic simulations resolve the free energy of interaction between the LNP surface and HA chains across varying PEG lipid compositions, and integrate these free energy profiles to inform the coarse-grained simulations of LNP transport through the matrix. This work highlights that even a slightly PEGylated surface depletes the near-contact shell between the LNP and HA chains, which disrupts their adhesive interactions. These protective PEG layers produce a sharp, non-linear enhancement in LNP diffusivities, with just 1% PEG increasing the diffusivity nearly eight-fold relative to bare LNPs, which remain trapped in the matrix structures. This work provides a quantitative estimate of how PEG surface density governs LNP transport through the ECM, which offers predictive guidance for engineering LNP surface properties in target-specific drug delivery.

biophysics↗

bioRxiv2026

RNA post hoc loading into empty lipid nanoparticles occurs on millisecond timescales during turbulent mixing

Lipid nanoparticles (LNPs) are conventionally produced through mixing of lipids dissolved in ethanol against a buffer containing RNA. An alternative strategy offering improved cold-chain stability involves formulating empty LNPs (eLNPs), removing ethanol, and post hoc loading (PHL) RNA into the aqueous eLNPs. The kinetics of this approach remain unknown. Here, we employ a flowthrough small-angle X-ray scattering (SAXS) setup based on a confined impinging jets (CIJ) mixer to probe PHL kinetics. We show that RNA PHL in a scalable CIJ mixer is efficient and reproducible, and that SAXS data confirms that this process concludes within [~]12 ms under favorable conditions in rapid turbulent micromixing, suggesting a diffusion-limited aggregation mechanism. Favorable conditions were identified as an acidic pH 5.5 buffer combined with turbulent CIJ mixing. In contrast, PHL performed with a neutral pH 7.4 buffer using a CIJ mixer or under laminar flow with a pH 5.5 buffer resulted in inefficient PHL.

bioengineering↗

bioRxiv2026

Multiscale modeling of the subcutaneous administration of peptides

With the increasing prevalence of subcutaneous administration of peptides, understanding their release and absorption is key to designing formulations with desired pharmacokinetics. Though the absorption of monoclonal anti-bodies (mAbs) has been widely explored through computational modeling, that of peptides remains poorly understood, as key features of peptide absorption, including concentration-dependent oligomerization and reversible binding with serum albumin and extracellular matrix, have not been captured. In this work, we present a first-of-its-kind approach to simulating subcutaneous administration of peptides that couples a high-fidelity tissue-level poroelastic model with a systemic compartment pharmacokinetic model. While accounting for competing binding and oligomerization tendencies of peptides, the model not only captures the process of injection but also tracks the absorption over subsequent days. We demonstrate the model using a single-dose administration of semaglutide and validate it against experimentally observed pharmacokinetic parameters. The results show the distribution of the different forms of the injected peptide throughout the body and describe the role of binding in sustaining its release. The model also reveals novel mechanisms, such as albumin-bound monomers enveloping the plume and the balance of oligomerization and binding in early stages of peptide absorption.

pharmacology and toxicology↗

bioRxiv2026

Anomalous diffusion of nanoparticles in semidilute hyaluronic acid solutions

Efficient drug delivery using nanoparticles (NPs) critically depends on their ability to diffuse through biological tissues to reach target cells at therapeutic concentrations. The extracellular matrix (ECM) poses a key barrier to such transport, which directly influences bio-distribution, cellular uptake, and overall therapeutic efficacy. A key regulator of this transport is hyaluronic acid/hyaluronan (HA), a major ECM polysaccharide that forms a hydrated, viscoelastic network. Increased/reduced hyaluronan concentration can elevate/decrease ECM bulk and effective viscosity. Increase in effective viscosity at the nanometer/micrometer length scales can hinder NP mobility through steric obstruction and hydrodynamic drag. There is a large variability in the HA molecular weights and concentrations, especially across age, tissue/organ, and pathological conditions. This work aims to study the diffusion of different NP types in the mixtures of HA polymers with variable molecular weights using the dynamic light scattering technique (DLS). Furthermore, we perform coarse-grained molecular dynamics (CG-MD) simulations for a model system to complement our findings from the dynamic light scattering experiments. We observe NP undergo anomalous diffusion, which is strongly dependent on the ratio of particle size/HA network mesh size, especially for higher molecular weight mixtures. This is strongly influenced by the effective viscosity, which is defined at the local environment experienced by the NPs. Our work highlights developing a simplified predictive framework coupled with simulations for a target-specific extracellular matrix environment.

biophysics↗

bioRxiv2026

Modeling lipid nanoparticle transport in extracellular matrix: Effects of particle size and rigidity

Lipid nanoparticles (LNPs) traverse through multiple biological barriers, such as crosslinked mesh structures in the extracellular matrix, before reaching their target sites. The physicochemical properties of LNPs determine their ability to penetrate complex biological environments such as the brain extracellular matrix. Their deformation in polymeric matrices affects transport, making it crucial to understand these factors for effective therapeutic delivery. Here, we develop a highly Coarse-Grained (CG) model of the LNP and its surrounding polymeric matrix, simulated as a uniform grid of cross-linked hyaluronic acid (HA) chains. The model for highly coarse-grained LNP was developed from a one-particle-thick membrane model that maintains mechanical features of lipid membranes, such as fluidity, topological changes, and hydrodynamic effects. Here, we investigate the collective influence of lipid nanoparticle size and its bending rigidity on the diffusive transport through the biological matrix. Our work highlights the role of particle to matrix size ratio in understanding deformation-assisted diffusive transport of LNPs in the matrix environment. Our study provides a tool to disentangle the effects of particle size and their bending rigidity on the transport through complex environments. Furthermore, this study systematically complements the rational design of lipid nanoparticle-based drug delivery platforms. SIGNIFICANCELipid nanoparticles (LNPs) are emerging as a powerful platform for delivering nucleic acid-based therapeutics, especially to hard-to-reach tissues like the brain. Their ability to protect and transport charged molecules, such as mRNA or siRNA, offers promising strategies for treating neurological disorders and advancing precision medicine for aging demographics around the world. In the brain parenchyma, LNPs must navigate the dense and heterogeneous extracellular matrix (ECM), composed primarily of crosslinked hyaluronic acid and proteoglycans. Understanding how particle size, deformability, and shape parameter affect the transport through this complex environment is critical for optimizing drug delivery platforms.

biophysics↗

bioRxiv2025

Effects of mixing technique and ethanol removal on lipidnanoparticle physicochemical properties

Optimizing the production of lipid nanoparticle (LNP) therapeutics is necessary for drug delivery efficiency, stability, and scalability. A small but growing body of literature has begun to recognize that LNP properties (e.g., size, shape, and internal structure) depend on the flow conditions during mixing for antisolvent precipitation, in which LNPs are formulated. Here, we use different mixers, varying flow patterns (e.g., laminar or turbulent mixing) and flow rate ratios (FRR), i.e., 3:1 and 1:1, to prepare a standard LNP formulation. We then characterize the resulting formulations using small angle x-ray scattering (SAXS) to provide insights into particle shape/morphology, internal organization (L and HII phases) of yeast RNA (yRNA), and structural differences/similarities that arise from the different mixing methods. The effect of ethanol removal on the LNPs structure, formulated from each mixing technique, is also discussed. We observed the 3:1 FRR mixers outperform the 1:1 configurations in certain desired LNP physiochemical properties. The differences observed in the LNPs produced across the two configurations are discussed. Furthermore, we use computational fluid dynamics to explain the turbulent mixing schemes among the 3:1 and 1:1 mixers.

biophysics↗

bioRxiv2025

Evaluation of Paper-based Loop-Mediated Isothermal Amplification

Colorimetric loop-mediated isothermal amplification (LAMP) on microfluidic paper-based analytical devices (PADs) offers a low-cost, disposable, and equipment-free alternative to liquid LAMP assays. However, amplification on PADs is consistently slower, by 5-46%, than reactions in tubes. To identify the origin of this delay, we evaluated heat transfer, diffusion in porous cellulose, and nonspecific adsorption of LAMP components across both high- and low-copy input regimes. Our results show that once thermal equilibrium is reached, reduced effective diffusion is the dominant contributor to the kinetic lag at low copy numbers, whereas nonspecific adsorption becomes the primary barrier at higher template concentrations. Pre-coating the paper with bovine serum albumin (BSA) mitigates adsorption. It narrows the tube-to-paper gap, thereby accelerating amplification of the SARS-CoV-2 ORF7ab synthetic gene by an average of 6 minutes, from 1E3 to 1E5 copies per reaction. These findings provide a mechanistic basis for the copy-number-dependent behavior of PAD LAMP and offer simple, low-cost strategies to improve the speed and reliability of PAD nucleic acid assays.

bioengineering↗

bioRxiv2024

Rheological and Lipid Characterization of Minipig and Human Skin Tissue: A Comparative Study Across Different Locations and Depths

Understanding the rheology of minipig and human skin is crucial for enhancing drug delivery methods, particularly for injections. Despite many studies on skins viscoelasticity, especially the subcutaneous layer, comparative analyses across different clinical sites are scarce, as is data on the impact of hydration or lipid levels. This study employs shear rheology and lipid analysis to evaluate viscoelasticity and lipid content across three anatomical locations --breast, belly, and neck and three different depth layers in Yucatan minipigs. It reports on how viscoelastic properties change with frequency, time, and strain, noting strain-stiffening and shear-thinning at high strain amplitudes. Human male and female abdominal tissues are also compared to minipig tissues, highlighting distinct viscoelastic traits and lipids role in them. The findings suggest the existence of species, anatomical location, tissue depth, and sex-based rheological differences. We also concluded the minipig male tissue is a more accurate model for human male subcutaneous tissue than for females.

biophysics↗

bioRxiv2020

Swarming bacterial fronts: Dynamics and morphology of active swarm interfaces propagating through passive frictional domains

Swarming, a multicellular mode of flagella-based motility observed in many bacteria species, enables coordinated and rapid surface translocation, expansion and colonization. In the swarming state, bacterial films display several characteristics of active matter including intense and persistent long-ranged flocks and strong fluctuating velocity fields with significant vorticity. Swarm fronts are typically dynamically evolving interfaces. Many of these fronts separate motile active domains from passive frictional regions comprised of dead or non-motile bacteria. Here, we study the dynamics and structural features of a model active-passive interface in swarming Serratia marcescens. We expose localized regions of the swarm to high intensity wide-spectrum light thereby creating large domains of tightly packed immotile bacteria. When the light source is turned off, swarming bacteria outside this passivated region advance into this highly frictional domain and continuously reshape the interphase boundary. Combining results from Particle Image Velocimetry (PIV) and intensity based image analysis, we find that the evolving interface has quantifiable and defined roughness. Correlations between spatially separated surface fluctuations and damping of the same are influenced by the interaction of the interphase region with adjacently located and emergent collective flows. Dynamical growth exponents characterizing the spatiotemporal features of the surface are extracted and are found to differ from classically expected values for passive growth or erosion. To isolate the effects of hydrodynamic interactions generated by collective flows and that arising from steric interactions, we propose and analyze agent-based simulations with full hydrodynamics of rod-shaped, self-propelled particles. Our computations capture qualitative features of the swarm and predict correlation lengths consistent with experiments. We conclude that hydrodynamic and steric interactions enable different modes of surface dynamics, morphology and thus front invasion.

bioengineering↗