The present study aims to develop and characterize a controlled-release delivery system for protein therapeutics in skeletal muscle regeneration following an acute injury. The therapeutic protein, a membrane-GPI anchored protein called Cripto, was immobilized in an injectable hydrogel delivery vehicle for local administration and sustained release. The hydrogel was made of poly(ethylene glycol)-fibrinogen (PEG-Fibrinogen, PF), in the form of injectable microspheres. The PF microspheres exhibited a spherical morphology with an average diameter of approximately 100 micrometers, and the Cripto protein was uniformly entrapped within them. The release rate of Cripto from the PF microspheres was controlled by tuning the crosslinking density of the hydrogel, which was varied by changing the concentration of poly(ethylene glycol) diacrylate (PEG-DA) crosslinker. In vitro experiments confirmed a sustained-release profile of Cripto from the PF microspheres for up to 27 days. The released Cripto was biologically active and promoted the in vitro proliferation of mouse myoblasts. The therapeutic effect of PF-mediated delivery of Cripto in vivo was tested in a cardiotoxin (CTX)-induced muscle injury model in mice. The Cripto caused an increase in the in vivo expression of the myogenic markers Pax7, the differentiation makers eMHC and Desmin, higher numbers of centro-nucleated myofibers and greater areas of regenerated muscle tissue. Collectively, these results establish the PF microspheres as a potential delivery system for the localized, sustained release of therapeutic proteins toward the accelerated repair of damaged muscle tissue following acute injuries.

Surgical site infection (SSI) is a common issue post-surgery which often prolongs hospitalization and can lead to serious complications such as sternal wound infection following cardiac surgery via median sternotomy. Controlled release of suitable antibiotics could allow maximizing drug efficacy and safety, and therefore achieving a desired therapeutic response. In this study, we have developed a vancomycin laden PEGylated fibrinogen–polyethylene glycol diacrylate (PF-PEGDA) hydrogel system that can release vancomycin at a controlled and predictable rate to be applied in SSI prevention. Two configurations were developed to study effect of the hydrogel on drug release, namely, vancomycin laden hydrogel and vancomycin solution on top of blank hydrogel. The relationship between the rigidity of the hydrogel and drug diffusion was found to comply with a universal power law, i.e., softer hydrogels result in a greater diffusion coefficient hence faster release rate. Besides, vancomycin laden hydrogels exhibited burst release, whereas the vancomycin solution on top of blank hydrogels exhibited lag release. A mathematical model was developed to simulate vancomycin permeation through the hydrogels. The permeation of vancomycin can be predicted accurately by using the mathematical model, which provided a useful tool to customize drug loading, hydrogel thickness and stiffness for personalized medication to manage SSI. To evaluate the potential of hydrogels for bone healing applications in cardiovascular medicine, we performed a proof-of-concept median sternotomy in rabbits and applied the hydrogels. The hydrogel formulations accelerated the onset of osteo-genetic processes in rabbits, demonstrating its potential to be used in human.

Atrial fibrillation (AF) is a common atrial arrhythmia that impairs quality of life and causes embolic stroke, heart failure and other complications. Recent advancements in machine learning (ML) and deep learning (DL) have shown potential for enhancing diagnostic accuracy. It is essential for DL models to be robust and generalizable across variations in ethnicity, age, sex, and other factors. Although a number of ECG database have been made available to the research community, none includes a Japanese population sample. Saitama Heart Database Atrial Fibrillation (SHDB-AF) is a novel open-sourced Holter ECG database from Japan, containing data from 100 unique patients with paroxysmal AF. Each record in SHDB-AF is 24 hours long and sampled at 200 Hz, totaling 24 million seconds of ECG data.

The developing mouse pancreas is surrounded by mesoderm compartments providing signals that induce pancreas formation. Most pancreatic organoid protocols lack this mesoderm niche and only partially capture the pancreatic cell repertoire. This work aims to generate pancreatic aggregates by differentiating mouse embryonic stem cells (mESCs) into mesoderm progenitors (MPs) and pancreas progenitors (PPs), without using Matrigel. First, mESCs were differentiated into epiblast stem cells (EpiSCs) to enhance the PP differentiation rate. Next, PPs and MPs aggregated together giving rise to various pancreatic cell types, including endocrine, acinar, and ductal cells, and to endothelial cells. Single-cell RNA sequencing analysis revealed a larger endocrine population within the PP + MP aggregates, as compared to PPs alone or PPs in Matrigel aggregates. The PP + MP aggregate gene expression signatures and its endocrine population percentage closely resembled those of the endocrine population found in the mouse embryonic pancreas, which holds promise for studying pancreas development.

High-throughput microscopy is vital for screening applications, where three-dimensional (3D) cellular models play a key role. However, due to defocus susceptibility, current 3D high-throughput microscopes require axial scanning, which lowers throughput and increases photobleaching and photodamage. Point spread function (PSF) engineering is an optical method that enables various 3D imaging capabilities, yet it has not been implemented in high-throughput microscopy due to the cumbersome optical extension it typically requires. Here we demonstrate compact PSF engineering in the objective lens, which allows us to enhance the imaging depth of field and, combined with deep learning, recover 3D information using single snapshots. Beyond the applications shown here, this work showcases the usefulness of high-throughput microscopy in obtaining training data for deep learning-based algorithms, applicable to a variety of microscopy modalities.

Fundamental properties of light unavoidably impose features on images collected using fluorescence microscopes. Accounting for these features is often critical in quantitatively interpreting microscopy images, especially those gathering information at scales on par with or smaller than light’s emission wavelength. Here the optics responsible for generating fluorescent images, fluorophore properties, and microscopy modalities leveraging properties of both light and fluorophores, in addition to the necessarily probabilistic modeling tools imposed by the stochastic nature of light and measurement, are reviewed.

DL4MicEverywhere is a platform that lets users train and implement their models in different computational environments. These environments include Google Colab, personal computational resources such as a desktop or laptop, and HPC systems. This flexibility is achieved by encapsulating each deep learning technique in an interactive Jupyter Notebook within a Docker container, enabling others to replicate analyses consistently across multiple platforms. DL4MicEverywhere (https://github.com/HenriquesLab/DL4MicEverywhere) enables users to install and interact with a large offering of standardized, user-friendly deep learning workflows, away from the limitations of proprietary platforms such as Google Colab and in a secure computational environment with controlled data privacy and resources. DL4MicEverywhere can be launched graphically, via X11 forwarding, or directly through a command line (headless mode), supporting HPC usage. This cross-platform containerization technology boosts the long-term platform’s sustainability and reproducibility, enhancing user convenience⁷.

DL4MicEverywhere features a zero-code interface that handles all the behind-the-scenes complexities, so users no longer need to deal with Docker configuration and deployment through a terminal. The intuitive interface abstracts away these technical details while providing a standardized Docker encapsulation for executing advanced techniques reliably. Researchers can select a notebook, choose computing resources and run the corresponding deep learning-powered analysis with just a few clicks (Fig. 1c–e). This allows users to train and apply models on various computing resources they control, eliminating reliance on third-party platforms. Furthermore, researchers can launch a notebook on local or remote systems with GPU acceleration whenever available, without worrying about complex software dependencies, Docker container management, or loss of access to deep learning frameworks (Fig. 1f–h). Compared to ZeroCostDL4Mic, DL4MicEverywhere doubles the number of deep learning approaches and provides new bioimaging analysis tasks, such as semantic segmentation, interactive instance segmentation, image registration, 3D single molecule localization microscopy, temporal and spatial upsampling, and image generation. The platform is designed to encourage the sharing and reuse of deep learning workflows provided as Jupyter Notebooks, which are then integrated into the BioImage Model Zoo. DL4MicEverywhere is strengthened by automated build pipelines⁸ that allow tracked versioning of ZeroCostDL4Mic notebooks and the seamless integration of new trainable models contributed by the community as user-friendly notebooks independently of the original ZeroCostDL4Mic framework (Fig. 1b). DL4MicEverywhere handles the corresponding testing and building of fully documented and open-source containers, making it easy for researchers to share not just the latest method, but the full software environment required to run it reliably.

Background

The aim of this study was to test the hypothesis that proinflammatory cytokines correlate with knee loading mechanics during gait following a mechanical walking stimulus in subjects 2 years after anterior cruciate ligament reconstruction. Elevated systemic levels of proinflammatory cytokines can be sustained for years after injury. Considering roughly 50% of these patients progress to Osteoarthritis 10–15 years after injury, a better understanding of the role of proinflammatory cytokines such as tumor necrosis factor-α and Interleukin-1β on Osteoarthritis risk is needed.

Methods

Serum proinflammatory cytokines concentrations were measured in 21 subjects 2 years after unilateral ACLR from blood drawn at rest and 3.5 h after 30 min of walking. An optoelectronic system and a force plate measured subjects' knee kinetics. Correlations were tested between inflammatory marker response and knee extension and knee adduction moments.

Findings

Changes in proinflammatory cytokines due to mechanical stimulus were correlated (R = 0.86) and showed substantial variation between subjects in both cytokines at 3.5 h post-walk. Knee loading correlated with 3.5-h changes in tumor necrosis factor-α concentration (Knee extension moment: R = -0.5, Knee adduction moment: R = -0.5) and Interleukin-1β concentration (Knee extension moment: R = -0.44). However, no significant changes in concentrations were observed in tumor necrosis factor-α and Interleukin-1β when comparing baseline and post walking stimulus conditions.

Interpretation

The significant associations between changes in serum proinflammatory markers following a mechanical stimulus and gait metrics in subjects at risk for developing Osteoarthritis underscore the importance of investigating the interaction between biomarkers and biomechanical factors in Osteoarthritis development.

Inflammatory bowel diseases (IBD) and familial Mediterranean fever (FMF) are auto-inflammatory diseases with common clinical and biological features. We aimed to determine their association and characterize the natural history in patients with both diagnoses.

Utilizing data from the epi-IIRN cohort, which includes 98% of Israel's population, we calculated the adjusted prevalence of FMF among IBD patients vs non-IBD controls. Case ascertainment of IBD was determined according to validated algorithms and for FMF by ICD-9 codes and colchicine purchase.

In total, 34 375 IBD patients (56% Crohn's disease [CD] and 44% ulcerative colitis [UC]) were compared with 93 602 matched controls. Among IBD patients, 157 (0.5%) had FMF compared with 160 (0.2%) of non-IBD controls (OR = 2.68 [95%CI 2.2–3.3]; P < 0.001). Pediatric-onset IBD had a higher prevalence of FMF compared with adult-onset IBD (30/5243 [0.6%] vs 127/29 132 [0.4%]), without statistical significance (OR = 1.31 [0.88–1.96]; P = 0.2). FMF was more prevalent in CD (114/19 264 [0.6%]) than UC (43/15 111 [0.3%]; OR = 2.1 [1.5–3.0], P < 0.001). FMF diagnosis preceded that of IBD in 130/157 cases (83%). FMF was associated with a more severe disease activity in UC patients at diagnosis, but not in CD patients. Outcomes were comparable between patients with CD+FMF vs CD alone; however, in patients with UC+FMF, time to biologic treatment was shorter.

FMF is more prevalent in IBD patients than in the general population, particularly in CD. The diagnosis of FMF precedes the diagnosis of IBD in most cases, and may be associated with a more severe course in UC.

Implantable cardiac pacemakers are crucial therapeutic tools for managing various cardiac conditions. For effective pacing, electrodes should exhibit flexibility, deformability, biocompatibility, and high conductivity/capacitance. Laser-induced graphene (LIG) shows promise due to its exceptional electrical and electrochemical properties. However, the fragility of LIG and the non-stretchability of polyimide substrates pose challenges when interfacing with the beating heart. Here, we present a simple method for fabricating robust, flexible, and stretchable bioelectronic interfaces by transferring LIG via water-responsive, nonswellable polyvinyl alcohol (PVA) gels. PVA solution penetrates the porous structure of LIG and solidifies into PVA xerogel as the solvent evaporates. The robust PVA xerogel enables the smooth transfer of LIG and prevents stretching of the LIG network during this process, which helps maintain its conductivity. When hydrated, the xerogel becomes a stable, nonswellable hydrogel. This gives the LIG-PVA hydrogel (LIG-PVA-H) composites with excellent conductivity (119.7 ± 4.3Ω sq⁻¹), high stretchability (up to 420%), reliability (cyclic stretch under 15% strain, with ∼ 1-time resistance increase), and good stability in phosphate buffered saline. The LIG-PVA-H composites were used as biointerfaces for electrocardiogram signal recording and electrical pacing on rat hearts ex vivo and in vivo, using commercial setups and a custom-built implantable wireless device. This work expands the application of LIG in bioelectronic interfaces and facilitates the development of electrotherapy for cardiac diseases.