Intestinal carcinogenicity screening of environmental pollutants using organoid-based cell transformation assay.

The high incidence of colorectal cancer (CRC) is closely associated with environmental pollutant exposure. To identify potential intestinal carcinogens, we developed a cell transformation assay (CTA) using mouse adult stem cell-derived intestinal organoids (mASC-IOs) and assessed the transformation potential on 14 representative chemicals, including Cd, iPb, Cr-VI, iAs-III, Zn, Cu, PFOS, BPA, MEHP, AOM, DMH, MNNG, aspirin, and metformin. We optimized the experimental protocol based on cytotoxicity, amplification, and colony formation of chemical-treated mASC-IOs. In addition, we assessed the accuracy of in vitro study and the human tumor relevance through characterizing interdependence between cell-cell and cell-matrix adhesions, tumorigenicity, pathological feature of subcutaneous tumors, and CRC-related molecular signatures. Remarkably, the results of cell transformation in 14 chemicals showed a strong concordance with epidemiological findings (8/10) and in vivo mouse studies (12/14). In addition, we found that the increase in anchorage-independent growth was positively correlated with the tumorigenicity of tested chemicals. Through analyzing the dose-response relationship of anchorage-independent growth by benchmark dose (BMD) modeling, the potent intestinal carcinogens were identified, with their carcinogenic potency ranked from high to low as AOM, Cd, MEHP, Cr-VI, iAs-III, and DMH. Importantly, the activity of chemical-transformed mASC-IOs was associated with the degree of cellular differentiation of subcutaneous tumors, altered transcription of oncogenic genes, and activated pathways related to CRC development, including Apc, Trp53, Kras, Pik3ca, Smad4 genes, as well as WNT and BMP signaling pathways. Taken together, we successfully developed a mASC-IO-based CTA, which might serve as a potential alternative for intestinal carcinogenicity screening of chemicals.

Toward Robust Self-Training Paradigm for Molecular Prediction Tasks.

Molecular prediction tasks normally demand a series of professional experiments to label the target molecule, which suffers from the limited labeled data problem. One of the semisupervised learning paradigms, known as self-training, utilizes both labeled and unlabeled data. Specifically, a teacher model is trained using labeled data and produces pseudo labels for unlabeled data. These labeled and pseudo-labeled data are then jointly used to train a student model. However, the pseudo labels generated from the teacher model are generally not sufficiently accurate. Thus, we propose a robust self-training strategy by exploring robust loss function to handle such noisy labels in two paradigms, that is, generic and adaptive. We have conducted experiments on three molecular biology prediction tasks with four backbone models to gradually evaluate the performance of the proposed robust self-training strategy. The results demonstrate that the proposed method enhances prediction performance across all tasks, notably within molecular regression tasks, where there has been an average enhancement of 41.5%. Furthermore, the visualization analysis confirms the superiority of our method. Our proposed robust self-training is a simple yet effective strategy that efficiently improves molecular biology prediction performance. It tackles the labeled data insufficient issue in molecular biology by taking advantage of both labeled and unlabeled data. Moreover, it can be easily embedded with any prediction task, which serves as a universal approach for the bioinformatics community.

Improving Sensitivity and Reproducibility of Surface-Enhanced Raman Scattering Biochips Utilizing Magnetoplasmonic Nanoparticles and Statistical Methods.

Surface-enhanced Raman scattering (SERS) technology has been widely recognized for its remarkable sensitivity in biochip development. This study presents a novel sandwich immunoassay that synergizes SERS with magnetoplasmonic nanoparticles (MPNs) to improve sensitivity. By taking advantage of the unique magnetism of these nanoparticles, we further enhance the detection sensitivity of SERS biochips through the applied magnetic field. Despite the high sensitivity, practical applications of SERS biochips are often limited by the issues of stability and reproducibility. In this study, we introduced a straightforward statistical method known as "Gaussian binning", which involves initially binning the two-dimensional Raman mapping data and subsequently applying Gaussian fitting. This approach enables a more consistent and reliable interpretation of data by reducing the variability inherent in Raman signal measurements. Based on our method, the biochip, targeting for C-reactive protein (CRP), achieves an impressive detection limit of 5.96 fg/mL, and with the application of a 3700 G magnetic field, it further enhances the detection limit by 5.7 times, reaching 1.05 fg/mL. Furthermore, this highly sensitive and magnetically tunable SERS biochip is easily designed for versatile adaptability, enabling the detection of other proteins. We believe that this innovation holds promise in enhancing the clinical applicability of SERS biochips.

Creating a semi-opened micro-cavity ovary through sacrificial microspheres as an in vitro model for discovering the potential effect of ovarian toxic agents.

The bio-engineered ovary is an essential technology for treating female infertility. Especially the development of relevant in vitro models could be a critical step in a drug study. Herein, we develop a semi-opened culturing system (SOCS) strategy that maintains a 3D structure of follicles during the culture. Based on the SOCS, we further developed micro-cavity ovary (MCO) with mouse follicles by the microsphere-templated technique, where sacrificial gelatin microspheres were mixed with photo-crosslinkable gelatin methacryloyl (GelMA) to engineer a micro-cavity niche for follicle growth. The semi-opened MCO could support the follicle growing to the antral stage, secreting hormones, and ovulating cumulus-oocyte complex out of the MCO without extra manipulation. The MCO-ovulated oocyte exhibits a highly similar transcriptome to the in vivo counterpart (correlation of 0.97) and can be fertilized. Moreover, we found that a high ROS level could affect the cumulus expansion, which may result in anovulation disorder. The damage could be rescued by melatonin, but the end of cumulus expansion was 3h earlier than anticipation, validating that MCO has the potential for investigating ovarian toxic agents in vitro. We provide a novel approach for building an in vitro ovarian model to recapitulate ovarian functions and test chemical toxicity, suggesting it has the potential for clinical research in the future.

Recombinant Human Collagen-Based Bioinks for the 3D Bioprinting of Full-thickness Human Skin Equivalent.

As a major extracellular matrix component within the skin, collagen has been widely used to engineer human skin tissues. However, most collagen is extracted from animals. Here, we introduced recombinant human type III collagen (rhCol3) as a bioactive component to formulate bioinks for the bioprinting of a full-thickness human skin equivalent. Human dermal fibroblasts were encapsulated in the gelatin methacryloyl-rhCol3 composite bioinks and printed on a transwell to form the dermis layer, on which human epidermal keratinocytes were seeded to perform an air-liquid interface culture for 6 weeks. After optimizing the bioink formulation and bioprinting process, we investigated the effect of rhCol3 on skin tissue formation. The results suggest that a higher concentration of rhCol3 would enhance the growth of both cells, resulting in a more confluent (~100%) spreading of the epidermal keratinocytes at an early stage (3 days), compared to the rhCol3-free counterpart. Moreover, in an in vivo experiment, adding rhCol3 in the hydrogel formulation would contribute to the skin wound healing process. Taken together, we conclude that rhCol3 could act as a functional bioink component to promote basic skin cellular processes for skin tissue engineering.

Deep Ensemble Learning with Atrous Spatial Pyramid Networks for Protein Secondary Structure Prediction.

The secondary structure of proteins is significant for studying the three-dimensional structure and functions of proteins. Several models from image understanding and natural language modeling have been successfully adapted in the protein sequence study area, such as Long Short-term Memory (LSTM) network and Convolutional Neural Network (CNN). Recently, Gated Convolutional Neural Network (GCNN) has been proposed for natural language processing. It has achieved high levels of sentence scoring, as well as reduced the latency. Conditionally Parameterized Convolution (CondConv) is another novel study which has gained great success in the image processing area. Compared with vanilla CNN, CondConv uses extra sample-dependant modules to conditionally adjust the convolutional network. In this paper, we propose a novel Conditionally Parameterized Convolutional network (CondGCNN) which utilizes the power of both CondConv and GCNN. CondGCNN leverages an ensemble encoder to combine the capabilities of both LSTM and CondGCNN to encode protein sequences by better capturing protein sequential features. In addition, we explore the similarity between the secondary structure prediction problem and the image segmentation problem, and propose an ASP network (Atrous Spatial Pyramid Pooling (ASPP) based network) to capture fine boundary details in secondary structure. Extensive experiments show that the proposed method can achieve higher performance on protein secondary structure prediction task than existing methods on CB513, Casp11, CASP12, CASP13, and CASP14 datasets. We also conducted ablation studies over each component to verify the effectiveness. Our method is expected to be useful for any protein related prediction tasks, which is not limited to protein secondary structure prediction.

Tunable Microgel-Templated Porogel (MTP) Bioink for 3D Bioprinting Applications.

Micropores are essential for tissue engineering to ensure adequate mass transportation for embedded cells. Despite the considerable progress made by advanced 3D bioprinting technologies, it remains challenging to engineer micropores of 100 µm or smaller in cell-laden constructs. Here, a microgel-templated porogel (MTP) bioink platform is reported to introduce controlled microporosity in 3D bioprinted hydrogels in the presence of living cells. Templated gelatin microgels are fabricated with varied sizes (≈10, ≈45, and ≈100 µm) and mixed with photo-crosslinkable formulations to make composite MTP bioinks. The addition of microgels significantly enhances the shear-thinning and self-healing viscoelastic properties and thus the printability of bioinks with cell densities up to 1 × 108 mL-1 in matrix. Consistent printability is achieved for a series of MTP bioinks based on different component ratios and matrix materials. After photo-crosslinking the matrix phase, the templated microgels dissociated and diffused under physiological conditions, resulting in corresponding micropores in situ. When embedding osteoblast-like cells in the matrix phase, the MTP bioinks support higher metabolic activity and more uniform mineral formation than bulk gel controls. The approach provides a facile strategy to engineer precise micropores in 3D printed structures to compensate for the limited resolution of current bioprinting approaches.

Comprehensive Study on Enhancing Low-Quality Position-Specific Scoring Matrix with Deep Learning for Accurate Protein Structure Property Prediction: Using Bagging Multiple Sequence Alignment Learning.

Accurate predictions of protein structure properties, for example, secondary structure and solvent accessibility, are essential in analyzing the structure and function of a protein. Position-specific scoring matrix (PSSM) features are widely used in the structure property prediction. However, some proteins may have low-quality PSSM features due to insufficient homologous sequences, leading to limited prediction accuracy. To address this limitation, we propose an enhancing scheme for PSSM features. We introduce the "Bagging MSA" (multiple sequence alignment) method to calculate PSSM features used to train our model, adopt a convolutional network to capture local context features and bidirectional long short-term memory for long-term dependencies, and integrate them under an unsupervised framework. Structure property prediction models are then built upon such enhanced PSSM features for more accurate predictions. Moreover, we develop two frameworks to evaluate the effectiveness of the enhanced PSSM features, which also bring proposed method into real-world scenarios. Empirical evaluation of CB513, CASP11, and CASP12 data sets indicates that our unsupervised enhancing scheme indeed generates more informative PSSM features for structure property prediction.

EPTool: A New Enhancing PSSM Tool for Protein Secondary Structure Prediction.

Recently, a deep learning-based enhancing Position-Specific Scoring Matrix (PSSM) method (Bagging Multiple Sequence Alignment [MSA] Learning) Guo et al. has been proposed, and its effectiveness has been empirically proved. Program EPTool is the implementation of Bagging MSA Learning, which provides a complete training and evaluation workflow for the enhancing PSSM model. It is capable of handling different input data set and various computing algorithms to train the enhancing model, then eventually improve the PSSM quality for those proteins with insufficient homologous sequences. In addition, EPTool equips several convenient applications, such as PSSM features calculator, and PSSM features visualization. In this article, we propose designed EPTool and briefly introduce its functionalities and applications. The detailed accessible instructions are also provided.

Synthesis and Monkey-PET Study of (R)- and (S)-18F-Labeled 2-Arylbenzoheterocyclic Derivatives as Amyloid Probes with Distinctive in Vivo Kinetics.

This study describes an effective strategy to improve pharmacokinetics of Aß imaging agents, offering a novel class of (R)- and (S)-18F-labeled 2-arylbenzoheterocyclic derivatives which bear an additional chiral hydroxyl group on the side chain. These ligands displayed binding abilities toward Aß aggregates with Ki values ranging from 3.2 to 195.6 nM. Chirality-related discrepancy was observed in biodistribution, and (S)-2-phenylbenzoxazole enantiomers exhibited vastly improved brain clearance with washout ratios higher than 20. Notably, (S)-[18F]28 possessed high binding potency (Ki = 7.6 nM) and exceptional brain kinetics (9.46% ID/g at 2 min, brain2min/brain60min = 27.8) that is superior to well-established [18F]AV45. The excellent pharmacokinetics and low nonspecific binding of (S)-[18F]28 were testified by dynamic PET/CT scans in monkey brains. In addition, (S)-[18F]28 clearly labeled Aß plaques both in vitro and ex vivo. These results might qualify (S)-[18F]28 to detect Aß plaques with high signal-to-noise ratio.

Synthesis of Tat tagged and folate modified N-succinyl-chitosan self-assembly nanoparticles as a novel gene vector.

The purpose of this research was to prepare a novel type of Tat tagged and folate modified N-succinyl-chitosan (Tat-Suc-FA) self-assembly nanoparticles, to provide a new vector for tumor gene therapy. In this study, Tat-Suc-FA polymers was synthesized and characterized using (1)H NMR and FT-IR. The copolymer had a mean diameter of 65 ± 22.6 nm, a zeta potential of 40 ± 0.2 mV. The cytotoxicity assay showed that Tat-Suc-FA polymers were less toxic than chitosan in the tested concentration range (from 2 to 500 µg/ml). Tat-Suc-FA/DNA complexes at various weight ratios were formulated and characterized. Particle sizes of Tat-Suc-FA/DNA complexes were between 54 and 106 nm as determined by dynamic light scattering. Accordingly, Transmission electron microscope photo of Tat-Suc-FA/DNA complexes exhibited a spherical and compact morphology. Zeta potentials of these complexes changed as the weight ratio varied (from 3 to 44 mV). Agarose gel electrophoresis assay showed that Tat-Suc-FA could efficiently condense the DNA, when the weight ratio was above 1.5/1. Together, these results suggest that the low toxic Tat-Suc-FA cationic polymers could be considered for use as a novel type of gene delivery vectors.

Improved tumor targetability of Tat-conjugated PAMAM dendrimers as a novel nanosized anti-tumor drug carrier.

The generation 4-poly-amidoamine-dendrimers (PAMAM G4 dendrimer, P) was conjugated to Tat peptide (Tat, T), a cell-penetrating peptide, in search of an efficient anti-tumor drug delivery vehicle for cancer therapy. In this study, we synthesized BODIPY-labeled Tat-Conjugated PAMAM dendrimers (BPTs) as a novel nanosized anticancer drug carriers and systemically investigated their biodistribution and the tumor accumulation in Sarcoma 180-bearing mice. In addition, the uptake and the cytotoxicity to S180 cells of BPTs thereof were evaluated. The unmodified dendrimer (BP) showed a soon clearance from the blood stream and nonspecific accumulation in tumor. In contrast, the Tat-modified dendrimer, BPT(64) with appropriate particle size showed a better retention in blood and could be accumulated effectively in tumor tissue via the enhanced permeability and retention (EPR) effect. Moreover, BPTs with a high Tat modification rate was accumulated more effectively in tumor tissue. In vitro experiments, these BPTs displayed low cytotoxicity on S180 cells and high uptake to S180 cells. These findings indicate that the nanoparticulate system on the basis of Tat-conjugated PAMAM dendrimers is safer and effective in the concentration range (below 20 µg/ml) to be used as a carrier of anti-tumor drugs for tumor targeting by intravenous administration.

Radioiodinated benzyloxybenzene derivatives: a class of flexible ligands target to ß-amyloid plaques in Alzheimer's brains.

Benzyloxybenzene, as a novel flexible scaffold without rigid planarity, was synthesized and evaluated as ligand toward Aß plaques. The binding site calculated for these flexible ligands was the hydrophobic Val18_Phe20 channel on the flat surface of Aß fiber. Structure-activity relationship analysis generated a common trend that binding affinities declined significantly from para-substituted ligands to ortho-substituted ones, which was also quantitatively illustrated by 3D-QSAR modeling. Autoradiography in vitro further confirmed the high affinities of radioiodinated ligands [125I]4, [125I]24, and [125I]22 (Ki=24.3, 49.4, and 17.6 nM, respectively). In biodistribution, [125I]4 exhibited high initial uptake and rapid washout property in the brain with brain2 min/brain60 min ratio of 16.3. The excellent in vitro and in vivo biostability of [125I]4 enhanced its potential for clinical application in SPECT imaging of Aß plaques. This approach could also allow the design of a new generation of Aß targeting ligands without rigid and planar framework.

In vivo biodistribution for tumor targeting of 5-fluorouracil (5-FU) loaded N-succinyl-chitosan (Suc-Chi) nanoparticles.

5-Fluorouracil-loaded N-succinyl-chitosan nanoparticles (5-FU-Suc-Chi/NP) were prepared by emulsification solvent diffusion. Biodistribution and tumor targeting were evaluated after i.v. administration of 5-Fu-Suc-Chi/NPs in Sarcoma 180-bearing mice. Also, pharmacokinetic profiles were evaluated after intravenous injection of 5-Fu-Suc-Chi/NP via the tail vein to rats. Our experimental results showed the 5-FU-Suc-Chi/NPs could be sustained at a high level in the blood for a very long time, implying its long systemic retention in the circulation. 5-FU-Suc-Chi/NPs were distributed mainly in tumors and liver, with small quantities being found in kidney and spleen. 5-FU-Suc-Chi/NPs accumulated only slightly in the heart and lung, and lowered the toxic effect of 5-FU in the heart and lung. Pharmacokinetic analysis in plasma showed the area under plasma concentration-time curve (AUC), elimination half-life (t(1/2)), and residence time (MRT) were increased 2.5-fold, 10.98-fold, and 10.8-fold for 5-FU-Suc-Chi/NP compared with that of free 5-FU, respectively. These results indicate that a long half-life in the circulation and tumor targeting of 5-FU-Suc-Chi/NPs are possible.

Preparation, in vitro and in vivo evaluation of (99m)Tc-Annexin B1: a novel radioligand for apoptosis imaging.

To develop a radiopharmaceutical for apoptosis imaging, Annexin B1, a new Ca2+-dependent phosphatidylserine (PS)-binding protein, was directly radiolabeled with (99m)Tc. This procedure yields up to 96% of radiochemical purity and higher radiolabeling efficiency. The preparation has been found to be sufficiently stable in vitro. Binding assay with human activated platelets indicated that (99m)Tc-Annexin B1 retained its PS binding activity. Biodistribution in mice revealed that (99m)Tc-Annexin B1 rapidly cleared from the blood and predominantly accumulated in the kidney. The increase in hepatic uptake in anti-Fas antibody treated mice correlated to histologic evidence of fulminant hepatic apoptosis. These data suggest that (99m)Tc-Annexin B1 can be used as a novel radiotracer to detect apoptosis in vivo.