Studying Muscle Transcriptional Dynamics at Single-molecule Scales in Drosophila.

Skeletal muscles are large syncytia made up of many bundled myofibers that produce forces and enable body motion. Drosophila is a classical model to study muscle biology. The combination of both Drosophila genetics and advanced omics approaches led to the identification of key conserved molecules that regulate muscle morphogenesis and regeneration. However, the transcriptional dynamics of these molecules and the spatial distribution of their messenger RNA within the syncytia cannot be assessed by conventional methods. Here we optimized an existing single-molecule RNA fluorescence in situ hybridization (smFISH) method to enable the detection and quantification of individual mRNA molecules within adult flight muscles and their muscle stem cells. As a proof of concept, we have analyzed the mRNA expression and distribution of two evolutionary conserved transcription factors, Mef2 and Zfh1/Zeb. We show that this method can efficiently detect and quantify single mRNA molecules for both transcripts in the muscle precursor cells, adult muscles, and muscle stem cells.

Deep Learning Methodologies Applied to Digital Pathology in Prostate Cancer: A Systematic Review.

Deep learning (DL), often called artificial intelligence (AI), has been increasingly used in Pathology thanks to the use of scanners to digitize slides which allow us to visualize them on monitors and process them with AI algorithms. Many articles have focused on DL applied to prostate cancer (PCa). This systematic review explains the DL applications and their performances for PCa in digital pathology. Article research was performed using PubMed and Embase to collect relevant articles. A Risk of Bias (RoB) was assessed with an adaptation of the QUADAS-2 tool. Out of the 77 included studies, eight focused on pre-processing tasks such as quality assessment or staining normalization. Most articles (n = 53) focused on diagnosis tasks like cancer detection or Gleason grading. Fifteen articles focused on prediction tasks, such as recurrence prediction or genomic correlations. Best performances were reached for cancer detection with an Area Under the Curve (AUC) up to 0.99 with algorithms already available for routine diagnosis. A few biases outlined by the RoB analysis are often found in these articles, such as the lack of external validation. This review was registered on PROSPERO under CRD42023418661.

Deep Learning for Detecting BRCA Mutations in High-Grade Ovarian Cancer Based on an Innovative Tumor Segmentation Method From Whole Slide Images.

BRCA1 and BRCA2 genes play a crucial role in repairing DNA double-strand breaks through homologous recombination. Their mutations represent a significant proportion of homologous recombination deficiency and are a reliable effective predictor of sensitivity of high-grade ovarian cancer (HGOC) to poly(ADP-ribose) polymerase inhibitors. However, their testing by next-generation sequencing is costly and time-consuming and can be affected by various preanalytical factors. In this study, we present a deep learning classifier for BRCA mutational status prediction from hematoxylin-eosin-safran-stained whole slide images (WSI) of HGOC. We constituted the OvarIA cohort composed of 867 patients with HGOC with known BRCA somatic mutational status from 2 different pathology departments. We first developed a tumor segmentation model according to dynamic sampling and then trained a visual representation encoder with momentum contrastive learning on the predicted tumor tiles. We finally trained a BRCA classifier on more than a million tumor tiles in multiple instance learning with an attention-based mechanism. The tumor segmentation model trained on 8 WSI obtained a dice score of 0.915 and an intersection-over-union score of 0.847 on a test set of 50 WSI, while the BRCA classifier achieved the state-of-the-art area under the receiver operating characteristic curve of 0.739 in 5-fold cross-validation and 0.681 on the testing set. An additional multiscale approach indicates that the relevant information for predicting BRCA mutations is located more in the tumor context than in the cell morphology. Our results suggest that BRCA somatic mutations have a discernible phenotypic effect that could be detected by deep learning and could be used as a prescreening tool in the future.

Artificial Intelligence-Based Opportunities in Liver Pathology-A Systematic Review.

BACKGROUND: Artificial Intelligence (AI)-based Deep Neural Networks (DNNs) can handle a wide range of applications in image analysis, ranging from automated segmentation to diagnostic and prediction. As such, they have revolutionized healthcare, including in the liver pathology field. OBJECTIVE: The present study aims to provide a systematic review of applications and performances provided by DNN algorithms in liver pathology throughout the Pubmed and Embase databases up to December 2022, for tumoral, metabolic and inflammatory fields. RESULTS: 42 articles were selected and fully reviewed. Each article was evaluated through the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool, highlighting their risks of bias. CONCLUSIONS: DNN-based models are well represented in the field of liver pathology, and their applications are diverse. Most studies, however, presented at least one domain with a high risk of bias according to the QUADAS-2 tool. Hence, DNN models in liver pathology present future opportunities and persistent limitations. To our knowledge, this review is the first one solely focused on DNN-based applications in liver pathology, and to evaluate their bias through the lens of the QUADAS2 tool.

An end-to-end pipeline based on open source deep learning tools for reliable analysis of complex 3D images of ovaries.

Computational analysis of bio-images by deep learning (DL) algorithms has made exceptional progress in recent years and has become much more accessible to non-specialists with the development of ready-to-use tools. The study of oogenesis mechanisms and female reproductive success has also recently benefited from the development of efficient protocols for three-dimensional (3D) imaging of ovaries. Such datasets have a great potential for generating new quantitative data but are, however, complex to analyze due to the lack of efficient workflows for 3D image analysis. Here, we have integrated two existing open-source DL tools, Noise2Void and Cellpose, into an analysis pipeline dedicated to 3D follicular content analysis, which is available on Fiji. Our pipeline was developed on larvae and adult medaka ovaries but was also successfully applied to different types of ovaries (trout, zebrafish and mouse). Image enhancement, Cellpose segmentation and post-processing of labels enabled automatic and accurate quantification of these 3D images, which exhibited irregular fluorescent staining, low autofluorescence signal or heterogeneous follicles sizes. In the future, this pipeline will be useful for extensive cellular phenotyping in fish or mammals for developmental or toxicology studies.

Artificial Intelligence in Predicting Microsatellite Instability and KRAS, BRAF Mutations from Whole-Slide Images in Colorectal Cancer: A Systematic Review.

Mismatch repair deficiency (d-MMR)/microsatellite instability (MSI), KRAS, and BRAF mutational status are crucial for treating advanced colorectal cancer patients. Traditional methods like immunohistochemistry or polymerase chain reaction (PCR) can be challenged by artificial intelligence (AI) based on whole slide images (WSI) to predict tumor status. In this systematic review, we evaluated the role of AI in predicting MSI status, KRAS, and BRAF mutations in colorectal cancer. Studies published in PubMed up to June 2023 were included (n = 17), and we reported the risk of bias and the performance for each study. Some studies were impacted by the reduced number of slides included in the data set and the lack of external validation cohorts. Deep learning models for the d-MMR/MSI status showed a good performance in training cohorts (mean AUC = 0.89, [0.74-0.97]) but slightly less than expected in the validation cohort when available (mean AUC = 0.82, [0.63-0.98]). Contrary to the MSI status, the prediction of KRAS and BRAF mutations was less explored with a less robust methodology. The performance was lower, with a maximum of 0.77 in the training cohort, 0.58 in the validation cohort for KRAS, and 0.82 AUC in the training cohort for BRAF.

Stromal p53 Regulates Breast Cancer Development, the Immune Landscape, and Survival in an Oncogene-Specific Manner.

Coevolution of tumor cells and adjacent stromal elements is a key feature during tumor progression; however, the precise regulatory mechanisms during this process remain unknown. Here, we show stromal p53 loss enhances oncogenic KrasG12D, but not ErbB2, driven tumorigenesis in murine mammary epithelia. Stroma-specific p53 deletion increases both epithelial and fibroblast proliferation in mammary glands bearing the KrasG12D oncogene in epithelia, while concurrently increasing DNA damage and/or DNA replication stress and decreasing apoptosis in the tumor cells proper. Normal epithelia was not affected by stromal p53 deletion. Tumors with p53-null stroma had a significant decrease in total, cytotoxic, and regulatory T cells; however, there was a significant increase in myeloid-derived suppressor cells, total macrophages, and M2-polarized tumor-associated macrophages, with no impact on angiogenesis or connective tissue deposition. Stroma-specific p53 deletion reprogrammed gene expression in both fibroblasts and adjacent epithelium, with p53 targets and chemokine receptors/chemokine signaling pathways in fibroblasts and DNA replication, DNA damage repair, and apoptosis in epithelia being the most significantly impacted biological processes. A gene cluster in p53-deficient mouse fibroblasts was negatively associated with patient survival when compared with two independent datasets. In summary, stroma-specific p53 loss promotes mammary tumorigenesis in an oncogene-specific manner, influences the tumor immune landscape, and ultimately impacts patient survival. IMPLICATIONS: Expression of the p53 tumor suppressor in breast cancer tumor stroma regulates tumorigenesis in an oncogene-specific manner, influences the tumor immune landscape, and ultimately impacts patient survival.

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

Automatic characterization of fluorescent labeling in intact mammalian tissues remains a challenge due to the lack of quantifying techniques capable of segregating densely packed nuclei and intricate tissue patterns. Here, we describe a powerful deep learning-based approach that couples remarkably precise nuclear segmentation with quantitation of fluorescent labeling intensity within segmented nuclei, and then apply it to the analysis of cell cycle dependent protein concentration in mouse tissues using 2D fluorescent still images. First, several existing deep learning-based methods were evaluated to accurately segment nuclei using different imaging modalities with a small training dataset. Next, we developed a deep learning-based approach to identify and measure fluorescent labels within segmented nuclei, and created an ImageJ plugin to allow for efficient manual correction of nuclear segmentation and label identification. Lastly, using fluorescence intensity as a readout for protein concentration, a three-step global estimation method was applied to the characterization of the cell cycle dependent expression of E2F proteins in the developing mouse intestine.

Imaging Mass Spectrometry Reveals Alterations in N-Linked Glycosylation That Are Associated With Histopathological Changes in Nonalcoholic Steatohepatitis in Mouse and Human.

Nonalcoholic steatohepatitis (NASH) is the progressive form of nonalcoholic fatty liver disease (NAFLD) and is characterized by inflammation, hepatocyte injury, and fibrosis. Further, NASH is a risk factor for cirrhosis and hepatocellular carcinoma. Previous research demonstrated that serum N-glycan profiles can be altered in NASH patients. Here, we hypothesized that these N-glycan modifications may be associated with specific liver damage in NAFLD and NASH. To investigate the N-glycome profile in tissue, imaging mass spectrometry was used for a qualitative and quantitative in situ N-linked glycan analysis of mouse and human NAFLD/NASH tissue. A murine model was used to induce NAFLD and NASH through ad libitum feeding with either a high-fat diet or a Western diet, respectively. Mice fed a high-fat diet or Western diet developed inflammation, steatosis, and fibrosis, consistent with NAFLD/NASH phenotypes. Induction of NAFLD/NASH for 18 months using high caloric diets resulted in increased expression of mannose, complex/fucosylated, and hybrid N-glycan structures compared to control mouse livers. To validate the animal results, liver biopsy specimens from 51 human NAFLD/NASH patients representing the full range of NASH Clinical Research Network fibrosis stages were analyzed. Importantly, the same glycan alterations observed in mouse models were observed in human NASH biopsies and correlated with the degree of fibrosis. In addition, spatial glycan alterations were localized specifically to histopathological changes in tissue like fibrotic and fatty areas. We demonstrate that the use of standard staining's combined with imaging mass spectrometry provide a full profile of the origin of N-glycan modifications within the tissue. These results indicate that the spatial distribution of abundances of released N-glycans correlate with regions of tissue steatosis associated with NAFLD/NASH.

Targeting the KRAS α4-α5 allosteric interface inhibits pancreatic cancer tumorigenesis.

RAS is the most frequently mutated oncogene in human cancer with nearly ~20% of cancer patients possessing mutations in one of three RAS genes (K, N or HRAS). However, KRAS is mutated in nearly 90% of pancreatic ductal carcinomas (PDAC). Although pharmacological inhibition of RAS has been challenging, KRAS(G12C)-specific inhibitors have recently entered the clinic. While KRAS(G12C) is frequently expressed in lung cancers, it is rare in PDAC. Thus, more broadly efficacious RAS inhibitors are needed for treating KRAS mutant-driven cancers such as PDAC. A RAS-specific tool biologic, NS1 Monobody, inhibits HRAS- and KRAS-mediated signalling and oncogenic transformation both in vitro and in vivo by targeting the α4-α5 allosteric site of RAS and blocking RAS self-association. Here, we evaluated the efficacy of targeting the α4-α5 interface of KRAS as an approach to inhibit PDAC development using an immunocompetent orthotopic mouse model. Chemically regulated NS1 expression inhibited ERK and AKT activation in KRAS(G12D) mutant KPC PDAC cells and reduced the formation and progression of pancreatic tumours. NS1-expressing tumours were characterized by increased infiltration of CD4 + T helper cells. These results suggest that targeting the #x3B1;4-#x3B1;5 allosteric site of KRAS may represent a viable therapeutic approach for inhibiting KRAS-mutant pancreatic tumours.

Deep learning provides high accuracy in automated chondrocyte viability assessment in articular cartilage using nonlinear optical microscopy.

Chondrocyte viability is a crucial factor in evaluating cartilage health. Most cell viability assays rely on dyes and are not applicable for in vivo or longitudinal studies. We previously demonstrated that two-photon excited autofluorescence and second harmonic generation microscopy provided high-resolution images of cells and collagen structure; those images allowed us to distinguish live from dead chondrocytes by visual assessment or by the normalized autofluorescence ratio. However, both methods require human involvement and have low throughputs. Methods for automated cell-based image processing can improve throughput. Conventional image processing algorithms do not perform well on autofluorescence images acquired by nonlinear microscopes due to low image contrast. In this study, we compared conventional, machine learning, and deep learning methods in chondrocyte segmentation and classification. We demonstrated that deep learning significantly improved the outcome of the chondrocyte segmentation and classification. With appropriate training, the deep learning method can achieve 90% accuracy in chondrocyte viability measurement. The significance of this work is that automated imaging analysis is possible and should not become a major hurdle for the use of nonlinear optical imaging methods in biological or clinical studies.

The coordination of spindle-positioning forces during the asymmetric division of the Caenorhabditis elegans zygote.

In Caenorhabditis elegans zygote, astral microtubules generate forces essential to position the mitotic spindle, by pushing against and pulling from the cortex. Measuring microtubule dynamics there, we revealed the presence of two populations, corresponding to pulling and pushing events. It offers a unique opportunity to study, under physiological conditions, the variations of both spindle-positioning forces along space and time. We propose a threefold control of pulling force, by polarity, spindle position and mitotic progression. We showed that the sole anteroposterior asymmetry in dynein on-rate, encoding pulling force imbalance, is sufficient to cause posterior spindle displacement. The positional regulation, reflecting the number of microtubule contacts in the posterior-most region, reinforces this imbalance only in late anaphase. Furthermore, we exhibited the first direct proof that dynein processivity increases along mitosis. It reflects the temporal control of pulling forces, which strengthens at anaphase onset following mitotic progression and independently from chromatid separation. In contrast, the pushing force remains constant and symmetric and contributes to maintaining the spindle at the cell centre during metaphase.

A deep learning segmentation strategy that minimizes the amount of manually annotated images.

Deep learning has revolutionized the automatic processing of images. While deep convolutional neural networks have demonstrated astonishing segmentation results for many biological objects acquired with microscopy, this technology's good performance relies on large training datasets. In this paper, we present a strategy to minimize the amount of time spent in manually annotating images for segmentation. It involves using an efficient and open source annotation tool, the artificial increase of the training data set with data augmentation, the creation of an artificial data set with a conditional generative adversarial network and the combination of semantic and instance segmentations. We evaluate the impact of each of these approaches for the segmentation of nuclei in 2D widefield images of human precancerous polyp biopsies in order to define an optimal strategy.

Testing independence between two random sets for the analysis of colocalization in bioimaging.

Colocalization aims at characterizing spatial associations between two fluorescently tagged biomolecules by quantifying the co-occurrence and correlation between the two channels acquired in fluorescence microscopy. Colocalization is presented either as the degree of overlap between the two channels or the overlays of the red and green images, with areas of yellow indicating colocalization of the molecules. This problem remains an open issue in diffraction-limited microscopy and raises new challenges with the emergence of superresolution imaging, a microscopic technique awarded by the 2014 Nobel prize in chemistry. We propose GcoPS, for Geo-coPositioning System, an original method that exploits the random sets structure of the tagged molecules to provide an explicit testing procedure. Our simulation study shows that GcoPS unequivocally outperforms the best competitive methods in adverse situations (noise, irregularly shaped fluorescent patterns, and different optical resolutions). GcoPS is also much faster, a decisive advantage to face the huge amount of data in superresolution imaging. We demonstrate the performances of GcoPS on two biological real data sets, obtained by conventional diffraction-limited microscopy technique and by superresolution technique, respectively.

Two Distinct E2F Transcriptional Modules Drive Cell Cycles and Differentiation.

Orchestrating cell-cycle-dependent mRNA oscillations is critical to cell proliferation in multicellular organisms. Even though our understanding of cell-cycle-regulated transcription has improved significantly over the last three decades, the mechanisms remain untested in vivo. Unbiased transcriptomic profiling of G0, G1-S, and S-G2-M sorted cells from FUCCI mouse embryos suggested a central role for E2Fs in the control of cell-cycle-dependent gene expression. The analysis of gene expression and E2F-tagged knockin mice with tissue imaging and deep-learning tools suggested that post-transcriptional mechanisms universally coordinate the nuclear accumulation of E2F activators (E2F3A) and canonical (E2F4) and atypical (E2F8) repressors during the cell cycle in vivo. In summary, we mapped the spatiotemporal expression of sentinel E2F activators and canonical and atypical repressors at the single-cell level in vivo and propose that two distinct E2F modules relay the control of gene expression in cells actively cycling (E2F3A-8-4) and exiting the cycle (E2F3A-4) during mammalian development.

A quantitative approach for analyzing the spatio-temporal distribution of 3D intracellular events in fluorescence microscopy.

Analysis of the spatial distribution of endomembrane trafficking is fundamental to understand the mechanisms controlling cellular dynamics, cell homeostasy, and cell interaction with its external environment in normal and pathological situations. We present a semi-parametric framework to quantitatively analyze and visualize the spatio-temporal distribution of intracellular events from different conditions. From the spatial coordinates of intracellular features such as segmented subcellular structures or vesicle trajectories, QuantEv automatically estimates weighted densities that are easy to interpret and performs a comprehensive statistical analysis from distribution distances. We apply this approach to study the spatio-temporal distribution of moving Rab6 fluorescently labeled membranes with respect to their direction of movement in crossbow- and disk-shaped cells. We also investigate the position of the generating hub of Rab11-positive membranes and the effect of actin disruption on Rab11 trafficking in coordination with cell shape.

Transcriptome regulation and chromatin occupancy by E2F3 and MYC in mice.

E2F3 and MYC are transcription factors that control cellular proliferation. To study their mechanism of action in the context of a regenerating tissue, we isolated both proliferating (crypts) and non-dividing (villi) cells from wild-type and Rb depleted small intestines of mice and performed ChIP-exo-seq (chromatin immunoprecipitation combined with lambda exonuclease digestion followed by high-throughput sequencing). The genome-wide chromatin occupancy of E2F3 and MYC was determined by mapping sequence reads to the genome and predicting preferred binding sites (peaks). Binding sites could be accurately identified within small regions of only 24 bp-28 bp long, highlighting the precision to which binding peaks can be identified by ChIP-exo-seq. Forty randomly selected E2F3- and MYC-specific binding sites were validated by ChIP-PCR. In addition, we also presented gene expression data sets from wild type, Rb-, E2f3- and Myc-depleted crypts and villi within this manuscript. These represent comprehensive and validated datasets that can be integrated to identify putative direct targets of E2F3 and MYC involved in the control of cellular proliferation in normal and Rb-deficient small intestines.

Noncatalytic PTEN missense mutation predisposes to organ-selective cancer development in vivo.

Inactivation of phosphatase and tensin homology deleted on chromosome 10 (PTEN) is linked to increased PI3K-AKT signaling, enhanced organismal growth, and cancer development. Here we generated and analyzed Pten knock-in mice harboring a C2 domain missense mutation at phenylalanine 341 (Pten(FV)), found in human cancer. Despite having reduced levels of PTEN protein, homozygous Pten(FV/FV) embryos have intact AKT signaling, develop normally, and are carried to term. Heterozygous Pten(FV/+) mice develop carcinoma in the thymus, stomach, adrenal medulla, and mammary gland but not in other organs typically sensitive to Pten deficiency, including the thyroid, prostate, and uterus. Progression to carcinoma in sensitive organs ensues in the absence of overt AKT activation. Carcinoma in the uterus, a cancer-resistant organ, requires a second clonal event associated with the spontaneous activation of AKT and downstream signaling. In summary, this PTEN noncatalytic missense mutation exposes a core tumor suppressor function distinct from inhibition of canonical AKT signaling that predisposes to organ-selective cancer development in vivo.

Redeployment of Myc and E2f1-3 drives Rb-deficient cell cycles.

Robust mechanisms to control cell proliferation have evolved to maintain the integrity of organ architecture. Here, we investigated how two critical proliferative pathways, Myc and E2f, are integrated to control cell cycles in normal and Rb-deficient cells using a murine intestinal model. We show that Myc and E2f1-3 have little impact on normal G1-S transitions. Instead, they synergistically control an S-G2 transcriptional program required for normal cell divisions and maintaining crypt-villus integrity. Surprisingly, Rb deficiency results in the Myc-dependent accumulation of E2f3 protein and chromatin repositioning of both Myc and E2f3, leading to the 'super activation' of a G1-S transcriptional program, ectopic S phase entry and rampant cell proliferation. These findings reveal that Rb-deficient cells hijack and redeploy Myc and E2f3 from an S-G2 program essential for normal cell cycles to a G1-S program that re-engages ectopic cell cycles, exposing an unanticipated addiction of Rb-null cells on Myc.

Actin grips: circular actin-rich cytoskeletal structures that mediate the wrapping of polymeric microfibers by endothelial cells.

Interaction of endothelial-lineage cells with three-dimensional substrates was much less studied than that with flat culture surfaces. We investigated the in vitro attachment of both mature endothelial cells (ECs) and of less differentiated EC colony-forming cells to poly-ε-capro-lactone (PCL) fibers with diameters in 5-20 µm range ('scaffold microfibers', SMFs). We found that notwithstanding the poor intrinsic adhesiveness to PCL, both cell types completely wrapped the SMFs after long-term cultivation, thus attaining a cylindrical morphology. In this system, both EC types grew vigorously for more than a week and became increasingly more differentiated, as shown by multiplexed gene expression. Three-dimensional reconstructions from multiphoton confocal microscopy images using custom software showed that the filamentous (F) actin bundles took a conspicuous ring-like organization around the SMFs. Unlike the classical F-actin-containing stress fibers, these rings were not associated with either focal adhesions or intermediate filaments. We also demonstrated that plasma membrane boundaries adjacent to these circular cytoskeletal structures were tightly yet dynamically apposed to the SMFs, for which reason we suggest to call them 'actin grips'. In conclusion, we describe a particular form of F-actin assembly with relevance for cytoskeletal organization in response to biomaterials, for endothelial-specific cell behavior in vitro and in vivo, and for tissue engineering.