NADPH-dependent antioxidant pathways have a critical role in scavenging hydrogen peroxide (H2O2) produced by oxidative phosphorylation. Inadequate scavenging results in H2O2 accumulation and can cause disease. To measure NADPH/NADP(+) redox states, we explored genetically encoded sensors based on steady-state fluorescence anisotropy due to FRET (fluorescence resonance energy transfer) between homologous fluorescent proteins (homoFRET); we refer to these sensors as Apollo sensors. We created an Apollo sensor for NADP(+) (Apollo-NADP(+)) that exploits NADP(+)-dependent homodimerization of enzymatically inactive glucose-6-phosphate dehydrogenase (G6PD). This sensor is reversible, responsive to glucose-stimulated metabolism and spectrally tunable for compatibility with many other sensors. We used Apollo-NADP(+) to study beta cells responding to oxidative stress and demonstrated that NADPH is significantly depleted before H2O2 accumulation by imaging a Cerulean-tagged version of Apollo-NADP(+) with the H2O2 sensor HyPer.
Deep learning provides an opportunity to automatically segment and extract cellular features from high-throughput microscopy images. Many labeling strategies have been developed for this purpose, ranging from the use of fluorescent markers to label-free approaches. However, differences in the channels available to each respective training dataset make it difficult to directly compare the effectiveness of these strategies across studies. Here, we explore training models using subimage stacks composed of channels sampled from larger, “hyper-labeled,” image stacks. This allows us to directly compare a variety of labeling strategies and training approaches on identical cells. This approach revealed that fluorescence-based strategies generally provide higher segmentation accuracies but were less accurate than label-free models when labeling was inconsistent. The relative strengths of label and label-free techniques could be combined through the use of merging fluorescence channels and using out-of-focus brightfield images. Beyond comparing labeling strategies, using subimage stacks for training was also found to provide a method of simulating a wide range of labeling conditions, increasing the ability of the final model to accommodate a greater range of candidate cell labeling strategies.
Deep learning provides an opportunity to automatically segment and extract cellular features from high-throughput microscope images. Many segmentation strategies have been developed for this purpose, ranging from the use of fluorescent markers to labelfree approaches. However, differences in the channels available to each respective training dataset make it difficult to directly compare the effectiveness of these strategies. Here we explore training models using subimage stacks composed of channels sampled from larger, 'hyper-labeled', image stacks (e.g. only the brightfield channels). This allows us to directly compare a variety of segmentation and training approaches on identical cells. This approach revealed that fluorescence-based strategies generally provide higher segmentation accuracies, but dipped below label-free models when labeling was inconsistent. The relative strengths of label and label-free techniques could be combined through the use of merging fluorescence channels and using out-of-focus brightfield images. Beyond comparing segmentation strategies, using subimage stacks for training was also found to provide a method of simulating a wide range of labeling conditions during training, increasing the ability of the final model to accomodate a fuller range of experimental setups.
Friedreich's ataxia (FRDA) is a rare inherited neurodegenerative disease. The mutation consists of a GAA repeat expansion within the FXN gene, which down regulates frataxin, leading to abnormal mitochondrial iron accumulation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.