Deep learning tools and modeling to estimate the temporal expression of cell cycle proteins from 2D still images

Differential impact of genetic deletion of TIGIT or PD-1 on melanoma-specific T-lymphocytes

Immune checkpoint (IC) blockade and adoptive transfer of tumor-specific T-cells (ACT) are two major strategies to treat metastatic melanoma. Their combination can potentiate T-cell activation in the suppressive tumor microenvironment, but the autoimmune adverse effects associated with systemic injection of IC blockers persist with this strategy. ACT of tumor-reactive T-cells defective for IC expression would overcome this issue. For this purpose, PD-1 and TIGIT appear to be relevant candidates, because their co-expression on highly tumor-reactive lymphocytes limits their therapeutic efficacy within the tumor microenvironme,nt. Our study compares the consequences of PDCD1 or TIGIT genetic deletion on anti-tumor properties and T-cell fitness of melanoma-specific T lymphocytes. Transcriptomic analyses revealed down-regulation of cell cycle-related genes in PD-1KO T-cells, consistent with biological observations, whereas proliferative pathways were preserved in TIGITKO T-cells. Functional analyses showed that PD-1KO and TIGITKO T-cells displayed superior antitumor reactivity than their wild-type counterpart in vitro and in a preclinical melanoma model using immunodeficient mice. Interestingly, it appears that TIGITKO T-cells were more effective at inhibiting tumor cell proliferation in vivo, and persist longer within tumors than PD-1KO T-cells, consistent with the absence of impact of TIGIT deletion on T-cell fitness. Taken together, these results suggest that TIGIT deletion, over PD-1 deletion, in melanoma-specific T-cells is a compelling option for future immunotherapeutic strategies.

Targeted therapy for capillary-venous malformations

Sporadic venous malformations are genetic conditions primarily caused by somatic gain-of-function mutation of PIK3CA or TEK, an endothelial transmembrane receptor signaling through PIK3CA. Venous malformations are associated with pain, bleedings, thrombosis, pulmonary embolism, esthetic deformities and, in severe cases, life-threatening situations. No authorized medical treatment exists for patients with venous malformations. Here, we created a genetic mouse model of PIK3CA-related capillary venous malformations that replicates patient phenotypes. We showed that these malformations only partially signal through AKT proteins. We compared the efficacy of different drugs, including rapamycin, a mTORC1 inhibitor, miransertib, an AKT inhibitor and alpelisib, a PI3Kα inhibitor at improving the lesions seen in the mouse model. We demonstrated the effectiveness of alpelisib in preventing vascular malformations' occurrence, improving the already established ones, and prolonging survival. Considering these findings, we were authorized to treat 25 patients with alpelisib, including 7 children displaying PIK3CA (n = 16) or TEK (n = 9)-related capillary venous malformations resistant to usual therapies including sirolimus, debulking surgical procedures or percutaneous sclerotherapies. We assessed the volume of vascular malformations using magnetic resonance imaging (MRI) for each patient. Alpelisib demonstrated improvement in all 25 patients. Vascular malformations previously considered intractable were reduced and clinical symptoms were attenuated. MRI showed a decrease of 33.4% and 27.8% in the median volume of PIK3CA and TEK malformations respectively, over 6 months on alpelisib. In conclusion, this study supports PI3Kα inhibition as a promising therapeutic strategy in patients with PIK3CA or TEK-related capillary venous malformations.

Automated identification of protein expression intensity and classification of protein cellular locations in mouse brain regions from immunofluorescence images

Knowledge of protein expression in mammalian brains at regional and cellular levels can facilitate understanding of protein functions and associated diseases. As the mouse brain is a typical mammalian brain considering cell type and structure, several studies have been conducted to analyze protein expression in mouse brains. However, labeling protein expression using biotechnology is costly and time-consuming. Therefore, automated models that can accurately recognize protein expression are needed. Here, we constructed machine learning models to automatically annotate the protein expression intensity and cellular location in different mouse brain regions from immunofluorescence images. The brain regions and sub-regions were segmented through learning image features using an autoencoder and then performing K-means clustering and registration to align with the anatomical references. The protein expression intensities for those segmented structures were computed on the basis of the statistics of the image pixels, and patch-based weakly supervised methods and multi-instance learning were used to classify the cellular locations. Results demonstrated that the models achieved high accuracy in the expression intensity estimation, and the F1 score of the cellular location prediction was 74.5%. This work established an automated pipeline for analyzing mouse brain images and provided a foundation for further study of protein expression and functions.

The invasive margin of early-stage human colon tumors is infiltrated with neutrophils of an antitumoral phenotype

Neutrophils infiltrate several types of cancer; however, whether their presence is associated with disease progression remains controversial. Here, we show that colon tumors overexpress neutrophil chemoattractants compared to healthy tissues, leading to their recruitment to the invasive margin and the central part of colon tumors. Of note, tumor-associated neutrophils expressing tumor necrosis factor α, which usually represents an antitumoral phenotype, were predominantly located in the invasive margin. Tumor-associated neutrophils from the invasive margin displayed an antitumoral phenotype with higher ICAM-1 and CD95 expression than neutrophils from healthy adjacent tissues. A higher neutrophil/lymphocyte ratio was found at later stages compared to the early phases of colon cancer. A neutrophil/lymphocyte ratio ≤3.5 predicted tumor samples had significantly more neutrophils at the invasive margin and the central part. Moreover, tumor-associated neutrophils at the invasive margin of early-stage tumors showed higher ICAM-1 and CD95 expression. Coculture of colon cancer cell lines with primary neutrophils induced ICAM-1 and CD95 expression, confirming our in situ findings. Thus, our data demonstrate that tumor-associated neutrophils with an antitumoral phenotype characterized by high ICAM-1 and CD95 expression infiltrate the invasive margin of early-stage colon tumors, suggesting that these cells can combat the disease at its early courses. The presence of tumor-associated neutrophils with antitumoral phenotype could help predict outcomes of patients with colon cancer.