Hubert M. Taïeb

Background: Acute digestive conditions in dairy cattle are common, yet diagnosis, treatment, and prognosis remain challenging. Hypothesis/Objectives: Describe the prevalence and characteristics of acute digestive conditions in dairy cows admitted to a referral hospital and evaluate the prognostic value of blood L-lactate concentrations during the first 24 h in both medical and surgical cases. Animals: Two hundred thirty-seven dairy cows with a final diagnosis of an acute digestive condition admitted between 2016 and 2021. Methods: Retrospective study. Clinical data, treatments, outcomes, and L-lactate measurements at admission, 6-12, and 12-24 h were extracted. Associations with survival were assessed using generalized linear models and receiver operating characteristic analysis. Results: Acute digestive conditions accounted for 13% of admissions for cattle, were most common in the first months of lactation, and 71 combinations of clinical signs were observed. Forty-five percent of cases were managed medically. Non-obstructive conditions (62%) predominated. Hyper-L-lactatemia at admission was negatively associated with short-term survival (odds ratio [OR], 0.88; 95% confidence interval [CI], 0.82-0.94; P < .001). L-lactate at 6-12 h showed the strongest association (OR, 0.47; 95% CI, 0.29-0.76; P < .01) and highest predictive performance (area under the curve [AUC], 0.83). Percentage change between admission and 6-12 h was also associated with survival (OR = 0.95; 95% CI, 0.92-0.98; P < .01). Conclusions and clinical importance: We documented the variability of acute abdominal conditions in dairy cows and the prognostic value of serial L-lactate measurements, which may support clinical decision making based on assessment of early changes.

Volumetric imaging using X-ray microcomputed tomography (µCT) is a mainstream method for 3D structural studies of bones and teeth. Quantitative analysis of mineralized features in general relies on accurate segmentation. The larger the image and the higher the resolution, the greater the incentive to use (semi)automated high-throughput segmentation approaches, primarily using deep learning-aided methods. In this study we pretrained and evaluated two U-Net segmentation models, called Universal Bone and Universal Jaw, using a diverse 3D image library from contributing colleagues. Our assumption was that pretraining a U-Net on skeletal and dental features would result in higher accuracy and throughput, in comparison with a naïve U-Net trained by an end-user on a limited local dataset. The Universal Bone binary segmentor pretrained on a relatively heterogeneous subset of images showed satisfactory generalization capacity, and was applicable to unseen images with minimal tuning. The Universal Jaw four-class segmentor (background, bone, dentin, enamel) was pretrained on a narrower range of images and required rather extensive retraining on manually retouched image segmentations. The performance of pretrained segmentation U-Nets is defined not only by the number of segmentation classes (the fewer the better), similarity of the unseen image spacing to the training set (the closer the better), but even more so by class balance. For particularly large 3D images, extensive manual retouching and re-training was nevertheless valuable attributable to the high throughput of the pretrained segmentor. Despite achieving partial success in pretraining specialized segmentation models, we advocate for future concerted efforts to merge and expand image collections, share pretrained models and utilize expert segmentations for benchmarking.

Solid cancers frequently relapse with distant metastasis, despite local and systemic treatment. Cellular dormancy has been identified as an important mechanism underlying drug resistance enabling late relapse. Therefore, relapse from invisible, minimal residual cancer of seemingly disease-free patients call for in vitro models of dormant cells suited for drug discovery. Here, we explore dormancy-inducing 3D engineered matrices, which generate mechanical confinement and induce growth arrest and survival against chemotherapy in cancer cells. We characterized the dormant phenotype of solitary cells by P-ERK ^low :P-p38 ^high dormancy signaling ratio, along with Ki67 ⁻ expression. As underlying mechanism, we identified stiffness-dependent nuclear localization of the four-and-a-half LIM domain 2 (FHL2) protein, leading to p53-independent high p21 ^Cip1/Waf1 nuclear expression, validated in murine and human tissue. Suggestive of a resistance-causing role, cells in the dormancy-inducing matrix became sensitive against chemotherapy upon FHL2 down-regulation. Thus, our biomaterial-based approach will enable systematic screens for previously unidentified compounds suited to eradicate potentially relapsing dormant cancer cells.

Resected tumors frequently relapse with distant metastasis, despite systemic treatment. Cellular dormancy has been identified as an important mechanism underlying such drug resistance enabling late relapse. Nonetheless, hurdles associated with detection and isolation of disseminated cancer cells (DCCs) in disease-free patients urge the need for in vitro models of dormant cells suited for drug discovery. Here, we explore dormancy-inducing 3D-engineered matrices, which generate mechanical confinement and induce growth arrest and survival against chemotherapy in cancer cells. We characterized the dormant phenotype of solitary cells by P-ERK ^low :P-p38 ^high dormancy signaling ratio, along with Ki67-expression. As underlying mechanism, we identified stiffness-dependent nuclear localization of the four-and-a-half LIM domains 2 (FHL2) protein, leading to p53-independent high p21 ^Cip1/Waf1 nuclear expression, validated in murine and human tissue. Suggestive of a resistance-causing role, cells in the dormancy-inducing matrix became sensitive against chemotherapy upon FHL2 downregulation. Thus, our biomaterial-based approach will enable systematic screens for novel compounds suited to eradicate potentially relapsing, dormant cancer cells. Teaser: Using semi-synthetic bioengineered hydrogels, we reveal a mechanosensitive and drug protective mechanism of dormant cancer cells in tissues

Using microfluidics, we isolate cancer cells under fluid flow mimicking sinusoidal capillaries. With deep-learning and FUCCItrack, we analyze 2D/3D time-lapse multi-channel images to study cell cycle dynamics, motility, volume, and morphology.

Swimming microorganisms often experience complex environments in their natural habitat. The same is true for microswimmers in envisioned biomedical applications. The simple aqueous conditions typically studied in the lab differ strongly from those found in these environments and often exclude the effects of small volume confinement or the influence that external fields have on their motion. In this work, we investigate magnetically steerable microswimmers, specifically magnetotactic bacteria, in strong spatial confinement and under the influence of an external magnetic field. We trap single cells in micrometer-sized microfluidic chambers and track and analyze their motion, which shows a variety of different trajectories, depending on the chamber size and the strength of the magnetic field. Combining these experimental observations with simulations using a variant of an active Brownian particle model, we explain the variety of trajectories by the interplay between the wall interactions and the magnetic torque. We also analyze the pronounced cell-to-cell heterogeneity, which makes single-cell tracking essential for an understanding of the motility patterns. In this way, our work establishes a basis for the analysis and prediction of microswimmer motility in more complex environments.

Beyond the more conventional single-cell segmentation and tracking, single-cell cycle dynamics is gaining a growing interest in the field of cell biology. Thanks to sophisticated systems, such as the fluorescent ubiquitination-based cell cycle indicator (FUCCI), it is now possible to study cell proliferation, migration, changes in nuclear morphology and single cell cycle dynamics, quantitatively and in real time. In this work, we introduce FUCCItrack, an all-in-one, semi-automated software to segment, track and visualize FUCCI modified cell lines. A user-friendly complete graphical user interface is presented to record and quantitatively analyze both collective cell proliferation as well as single cell information, including migration and changes in nuclear or cell morphology as a function of cell cycle. To enable full control over the analysis, FUCCItrack also contains features for identification of errors and manual corrections.