Cell-cell communication inference and analysis in the tumour microenvironments from single-cell transcriptomics: data resources and computational strategies.

Carcinomas are complex ecosystems composed of cancer, stromal and immune cells. Communication between these cells and their microenvironments induces cancer progression and causes therapy resistance. In order to improve the treatment of cancers, it is essential to quantify crosstalk between and within various cell types in a tumour microenvironment. Focusing on the coordinated expression patterns of ligands and cognate receptors, cell-cell communication can be inferred through ligand-receptor interactions (LRIs). In this manuscript, we carry out the following work: (i) introduce pipeline for ligand-receptor-mediated intercellular communication estimation from single-cell transcriptomics and list a few available LRI-related databases and visualization tools; (ii) demonstrate seven classical intercellular communication scoring strategies, highlight four types of representative intercellular communication inference methods, including network-based approaches, machine learning-based approaches, spatial information-based approaches and other approaches; (iii) summarize the evaluation and validation avenues for intercellular communication inference and analyze the advantages and limitations for the above four types of cell-cell communication methods; (iv) comment several major challenges while provide further research directions for intercellular communication analysis in the tumour microenvironments. We anticipate that this work helps to better understand intercellular crosstalk and to further develop powerful cell-cell communication estimation tools for tumor-targeted therapy.

Deciphering ligand-receptor-mediated intercellular communication based on ensemble deep learning and the joint scoring strategy from single-cell transcriptomic data.

BACKGROUND: Cell-cell communication in a tumor microenvironment is vital to tumorigenesis, tumor progression and therapy. Intercellular communication inference helps understand molecular mechanisms of tumor growth, progression and metastasis. METHODS: Focusing on ligand-receptor co-expressions, in this study, we developed an ensemble deep learning framework, CellComNet, to decipher ligand-receptor-mediated cell-cell communication from single-cell transcriptomic data. First, credible LRIs are captured by integrating data arrangement, feature extraction, dimension reduction, and LRI classification based on an ensemble of heterogeneous Newton boosting machine and deep neural network. Next, known and identified LRIs are screened based on single-cell RNA sequencing (scRNA-seq) data in certain tissues. Finally, cell-cell communication is inferred by incorporating scRNA-seq data, the screened LRIs, a joint scoring strategy that combines expression thresholding and expression product of ligands and receptors. RESULTS: The proposed CellComNet framework was compared with four competing protein-protein interaction prediction models (PIPR, XGBoost, DNNXGB, and OR-RCNN) and obtained the best AUCs and AUPRs on four LRI datasets, elucidating the optimal LRI classification ability. CellComNet was further applied to analyze intercellular communication in human melanoma and head and neck squamous cell carcinoma (HNSCC) tissues. The results demonstrate that cancer-associated fibroblasts highly communicate with melanoma cells and endothelial cells strong communicate with HNSCC cells. CONCLUSIONS: The proposed CellComNet framework efficiently identified credible LRIs and significantly improved cell-cell communication inference performance. We anticipate that CellComNet can contribute to anticancer drug design and tumor-targeted therapy.

CellEnBoost: A Boosting-Based Ligand-Receptor Interaction Identification Model for Cell-to-Cell Communication Inference.

Cell-to-cell communication (CCC) plays important roles in multicellular organisms. The identification of communication between cancer cells themselves and one between cancer cells and normal cells in tumor microenvironment helps understand cancer genesis, development and metastasis. CCC is usually mediated by Ligand-Receptor Interactions (LRIs). In this manuscript, we developed a Boosting-based LRI identification model (CellEnBoost) for CCC inference. First, potential LRIs are predicted by data collection, feature extraction, dimensional reduction, and classification based on an ensemble of Light gradient boosting machine and AdaBoost combining convolutional neural network. Next, the predicted LRIs and known LRIs are filtered. Third, the filtered LRIs are applied to CCC elucidation by combining CCC strength measurement and single-cell RNA sequencing data. Finally, CCC inference results are visualized using heatmap view, Circos plot view, and network view. The experimental results show that CellEnBoost obtained the best AUCs and AUPRs on the collected four LRI datasets. Case study in human head and neck squamous cell carcinoma (HNSCC) tissues demonstrates that fibroblasts were more likely to communicate with HNSCC cells, which is in accord with the results from iTALK. We anticipate that this work can contribute to the diagnosis and treatment of cancers.

EnANNDeep: An Ensemble-based lncRNA-protein Interaction Prediction Framework with Adaptive k-Nearest Neighbor Classifier and Deep Models.

lncRNA-protein interactions (LPIs) prediction can deepen the understanding of many important biological processes. Artificial intelligence methods have reported many possible LPIs. However, most computational techniques were evaluated mainly on one dataset, which may produce prediction bias. More importantly, they were validated only under cross validation on lncRNA-protein pairs, and did not consider the performance under cross validations on lncRNAs and proteins, thus fail to search related proteins/lncRNAs for a new lncRNA/protein. Under an ensemble learning framework (EnANNDeep) composed of adaptive k-nearest neighbor classifier and Deep models, this study focuses on systematically finding underlying linkages between lncRNAs and proteins. First, five LPI-related datasets are arranged. Second, multiple source features are integrated to depict an lncRNA-protein pair. Third, adaptive k-nearest neighbor classifier, deep neural network, and deep forest are designed to score unknown lncRNA-protein pairs, respectively. Finally, interaction probabilities from the three predictors are integrated based on a soft voting technique. In comparing to five classical LPI identification models (SFPEL, PMDKN, CatBoost, PLIPCOM, and LPI-SKF) under fivefold cross validations on lncRNAs, proteins, and LPIs, EnANNDeep computes the best average AUCs of 0.8660, 0.8775, and 0.9166, respectively, and the best average AUPRs of 0.8545, 0.8595, and 0.9054, respectively, indicating its superior LPI prediction ability. Case study analyses indicate that SNHG10 may have dense linkage with Q15717. In the ensemble framework, adaptive k-nearest neighbor classifier can separately pick the most appropriate k for each query lncRNA-protein pair. More importantly, deep models including deep neural network and deep forest can effectively learn the representative features of lncRNAs and proteins.

A variant of TNFR2-Fc fusion protein exhibits improved efficacy in treating experimental rheumatoid arthritis.

Etanercept, a TNF receptor 2-Fc fusion protein, is currently being used for the treatment of rheumatoid arthritis (RA). However, 25% to 38% of patients show no response which is suspected to be partially due to insufficient affinity of this protein to TNFalpha. By using computational protein design, we found that residue W89 and E92 of TNFR2 were critical for ligand binding. Among several mutants tested, W89Y/E92N displayed 1.49-fold higher neutralizing activity to TNFalpha, as compared to that of Etanercept. Surface plasmon resonance (SPR) based binding assay revealed that the equilibrium dissociation constant of W89Y/E92N to TNFalpha was 3.65-fold higher than that of Etanercept. In a rat model of collagen-induced arthritis (CIA), W89Y/E92N showed a significantly better ability than Etanercept in reducing paw swelling and improvement of arthritic joint histopathologically. These data demonstrate that W89Y/E92N is potentially a better candidate with improved efficacy in treating RA and other autoimmune diseases.

Laser-triggered collaborative chemophotothermal effect of gold nanoparticles for targeted colon cancer therapy.

Nanotechnology has shown advantages for cancer treatment. Multimodal nanoparticles (NPs) combining chemotherapy and photothermal therapy are promising and elicit synergetic benefit. However, there were still less multifunctional nanomaterials with good targeting and anti-tumor property applied as the colon cancer therapeutic strategy. In this study, we designed the gold NPs modified with AS1411 and DNA riched of GC intercalation (hairpin DNA) with doxorubicin (DOX) for targeted chemotherapy and NIR laser-triggered chemo-photothermal effect (PTT). We took advantage of PTT effect to realize DOX release from hairpin DNA. We also demonstrated AS1411 based NPs exhibited remarkable targeted binding towards SW480 colon cancer cells in vitro and enhanced uptake inside the cells. Strikingly, AS1411 based NPs exhibited the most efficient cytotoxicity and markedly enhanced inhibition effect on cells proliferation to SW480 cells under laser exposure when compared to the NPs merely with PTT or chemotherapy. Our study appears to provide an alternative nanoplatform with good targeted and chemo-photothermal therapy against colon cancer.

Synthesis and Application of AS1411-Functionalized Gold Nanoparticles for Targeted Therapy of Gastric Cancer.

Gastric cancer therapy is still a big challenge, and nanomedicines bring much more hope. It is essential to develop multifunctional nanoparticles, especially those with high targeted capacity and antitumor effects, to improve gastric cancer therapy. In this study, we constructed AS1411 aptamer-based gold nanoparticles with appropriate size facilitating endocytosis and actively targeted drug delivery for gastric cancer cells via the specific AS1411-nucleolin interaction. The AS1411-based nanoparticles showed obviously increased targeted capacity towards AGS cells compared to random ssDNA-based nanoparticles. Meanwhile, compared to L929 cells, the AS1411-based nanoparticles showed selective drug uptake and delivery for AGS cells. Importantly, the AS1411-based nanoparticles exhibited significantly stronger antitumor effects on AGS cells under laser irradiation compared to chemotherapy alone. Our nanoparticles combined targeted drug delivery and efficient antitumor effects within one single nanoplatform, which are promising to be applied as targeted nanomedicines against gastric cancer.

A stable filamentous coaxial microelectrode for Li-ion batteries: a case of olivine LiFePO4.

By using a facile spray drying method, we fabricated a stable filamentous coaxial microelectrode of olivine LiFePO4, which delivers an excellent electrochemical performance in Li-ion batteries. Due to its simple structure, the microelectrode is a perfect model to study electrochemical reactions, even through in situ electrochemical measurements.

Platinums sensitize human epithelial tumor cells to lymphotoxin alpha by inhibiting NFkappaB-dependent transcription.

Lymphotoxin alpha (LTalpha) was first identified as a direct anti-tumor factor, whereas increasing evidence has recently shown that in most cases the growth inhibition mediated by LTalpha requires the synergistic action of other factors, such as RNA transcription or protein synthesis inhibitor. In this study, we evaluated the combined effects of LTalpha and ten chemotherapeutic drugs on cell growth in a panel of human epithelial tumor cells, and explored the molecular mechanism of their mutual action. The results showed that platinums (cisplatin, carboplatin, oxaliplatin) are more universally effective than other chemotherapeutic drugs (doxorubicin, epi-doxorubicin, 5-flourouracil, mitomycin, cyclophosphamide, vincristine and vinorelbine) to enhance the response of six human epithelial tumor cell lines (A375, Bcap37, NCI-H157, SW480, BGC-823 and HeLa) to LTalpha. A systemic treatment with a combination of LTalpha and cisplatin in a human Bcap37 breast cancer xenograft nude mice model dramatically improved the therapeutic efficacy of LTalpha. Further analysis revealed that the sensitization of platinums was associated with platinums-induced suppression of nuclear factor-kappaB (NFkappaB) and subsequent downregulation of X-linked inhibitor of apoptosis protein (XIAP), which rescued caspase-3 from inhibition. Our results suggested that a proper combination of bio-agents such as LTalpha and conventional chemotherapeutic drugs such as platinums may be an efficient treatment strategy for human epithelial cancers.

Indomethacin enhances the cytotoxicity of recombinant human lymphotoxin alpha on tumor cells by suppressing NFkappaB signaling.

Indomethacin is widely used to treat inflammatory and prevent adverse events which frequently accrue in biotherapy. In phase I clinical trial of recombinant human lymphotoxin alpha (rhLTalpha), indomethacin was premedicated to alleviate chill and fever. However, it is unknown whether indomethacin influences the therapeutic efficacy of rhLTalpha. In this study we found that pre-treatment with low dose of indomethacin enhanced the cytotoxicity of rhLTalpha and/or cisplatin/adriamycin on human tumor cells. Further investigation demonstrated that indomethacin dose-dependently suppressed the activation of nuclear factor kappa B (NFkappaB) by inhibiting phosphorylation and degradation of IkappaBalpha. In addition, indomethacin decreased the expression of NFkappaB-regulated gene products involved in rhLTalpha-induced anti-apoptosis (XIAP, cFLIP and cIAP-1), which may explain its sensitization of tumor cells to rhLTalpha and/or cisplatin/adriamycin.

[Construction of combined site-directed random mutation libraries of recombinant human lymphotoxin].

To construct the combined site-directed random mutation library of recombinant human Lymphotoxin (rhLT) for in vitro molecular evolution study, and to study the structure and function relationship. The random point mutations at the sites of 46,106 and 130 were individually generated by overlap PCR amplification with the random nucleotide primers. The three point mutations were combined and cloned into pMD-18T vector to construct the combined mutation library. DNA sequencing was used to evaluate the diversity and randomness of the mutation sites. The combined mutation library was re-engineered, inserted in prokaryotic expression vector pBV220, transformed and expressed in Escherichia coli strain DH5alpha. The biological activity of some of the mutants was tested in 1929 mouse fibroblast cells. As much as 1.5 x 10(5) clones were obtained, which represents 4.5 times of the complete mutation libraries at 99% confidence. Sequencing 50 clones revealed no obvious bias in the nucleotide and amino acid mutations at the sites. Among the 30 expressed samples underwent for the bioassay, 70% (21 samples) were inactive, 23.3% (7 samples) had lower activity than rhLT, the remaining 6.7% (2 samples) had higher activity than rhLT. The combined site-directed random mutation library of rhLT has been constructed successfully. In combination with phase display, the library is ready for in vitro molecular evolution study.