Chemotactic signals are relayed to neighboring cells through the secretion of additional chemoattractants. We previously showed in Dictyostelium discoideum that the adenylyl cyclase A, which synthesizes the chemoattractant cyclic adenosine monophosphate (cAMP), is present in the intraluminal vesicles of multivesicular bodies (MVBs) that coalesce at the back of cells. Using ultrastructural reconstructions, we now show that ACA-containing MVBs release their contents to attract neighboring cells. We show that the released vesicles are capable of directing migration and streaming and are central to chemotactic signal relay. We demonstrate that the released vesicles not only contain cAMP but also can actively synthesize and release cAMP to promote chemotaxis. Through proteomic, pharmacological, and genetic approaches, we determined that the vesicular cAMP is released via the ABCC8 transporter. Together, our findings show that extracellular vesicles released by D. discoideum cells are functional entities that mediate signal relay during chemotaxis and streaming.

Automated segmentation of cellular electron microscopy (EM) datasets remains a challenge. Supervised deep learning (DL) methods that rely on region-of-interest (ROI) annotations yield models that fail to generalize to unrelated datasets. Newer unsupervised DL algorithms require relevant pre-training images, however, pre-training on currently available EM datasets is computationally expensive and shows little value for unseen biological contexts, as these datasets are large and homogeneous. To address this issue, we present CEM500K, a nimble 25 GB dataset of 0.5 × 10⁶ unique 2D cellular EM images curated from nearly 600 three-dimensional (3D) and 10,000 two-dimensional (2D) images from >100 unrelated imaging projects. We show that models pre-trained on CEM500K learn features that are biologically relevant and resilient to meaningful image augmentations. Critically, we evaluate transfer learning from these pre-trained models on six publicly available and one newly derived benchmark segmentation task and report state-of-the-art results on each. We release the CEM500K dataset, pre-trained models and curation pipeline for model building and further expansion by the EM community. Data and code are available at https://www.ebi.ac.uk/pdbe/emdb/empiar/entry/10592/ and https://git.io/JLLTz.

In the liver, mitochondria are exposed to different concentrations of nutrients due to their spatial positioning across the periportal and pericentral axis. How the mitochondria sense and integrate these signals to respond and maintain homeostasis is not known. Here, we combine intravital microscopy, spatial proteomics, and functional assessment to investigate mitochondrial heterogeneity in the context of liver zonation. We find that periportal and pericentral mitochondria are morphologically and functionally distinct; beta-oxidation is elevated in periportal regions, while lipid synthesis is predominant in the pericentral mitochondria. In addition, comparative phosphoproteomics reveals spatially distinct patterns of mitochondrial composition and potential regulation via phosphorylation. Acute pharmacological modulation of nutrient sensing through AMPK and mTOR shifts mitochondrial phenotypes in the periportal and pericentral regions, linking nutrient gradients across the lobule and mitochondrial heterogeneity. This study highlights the role of protein phosphorylation in mitochondrial structure, function, and overall homeostasis in hepatic metabolic zonation. These findings have important implications for liver physiology and disease.