Our understanding of endocytic pathway dynamics is severely restricted by the diffraction limit of light microscopy. To address this, we implemented a novel technique based on the subdiffractional tracking of internalized molecules (sdTIM). This allowed us to image anti–green fluorescent protein Atto647N-tagged nanobodies trapped in synaptic vesicles (SVs) from live hippocampal nerve terminals expressing vesicle-associated membrane protein 2 (VAMP2)–pHluorin with 36-nm localization precision. Our results showed that, once internalized, VAMP2–pHluorin/Atto647N–tagged nanobodies exhibited a markedly lower mobility than on the plasma membrane, an effect that was reversed upon restimulation in presynapses but not in neighboring axons. Using Bayesian model selection applied to hidden Markov modeling, we found that SVs oscillated between diffusive states or a combination of diffusive and transport states with opposite directionality. Importantly, SVs exhibiting diffusive motion were relatively less likely to switch to the transport motion. These results highlight the potential of the sdTIM technique to provide new insights into the dynamics of endocytic pathways in a wide variety of cellular settings.
1 The effects of the naturally occurring neurosteroid tetrahydrodeoxycorticosterone (THDOC) on GABAA receptor-mediated miniature, spontaneous and evoked IPSCs was tested using patch-clamp techniques in slices of hippocampus and cerebellum from rats at two developmental stages (≈10 and ≈20 days postnatal). The cells studied were hippocampal granule cells and cerebellar Purkinje and granule cells. 2 Most miniature GABAergic currents (mIPSCs) decayed with two exponentials and neurosteroids caused a ≈4-fold increase in the decay time constant of the second exponential at the highest concentration used (2 μm). Similar effects were seen at high concentrations of THDOC (1-2 μm) in all cell groups tested. No effects were seen on amplitude or rise time of mIPSCs. 3 The effects of THDOC (1 μm) were shown to be stereoselective and rapidly reversible, indicating that the neurosteroid binds to the GABAA receptor, rather than acting genomically. 4 At concentrations of THDOC likely to occur physiologically (50-100 nm), the decay time of IPSCs was also enhanced (25-50 %) in all cerebellar cell groups tested. In contrast, at 100 nm THDOC, seven of 11 hippocampal granule cells were sensitive from the 10 day group but the 20 day hippocampal granule cells showed no significant enhancement in the presence of these lower concentrations of THDOC. 5 The differences in sensitivity of hippocampal and cerebellar cells to THDOC are compared to data reported in the literature on regional development of expression of different receptor subunits in the brain and it is suggested that the progressive relative insensitivity of the 20 day hippocampal cells may depend on increasing expression of the δ subunit of the GABAA receptor and possibly an increase in the α4 subunit.
Digital agriculture services can greatly assist growers to monitor their fields and optimize their use throughout the growing season. Thus, knowing the exact location of fields and their boundaries is a prerequisite. Unlike property boundaries, which are recorded in local council or title records, field boundaries are not historically recorded. As a result, digital services currently ask their users to manually draw their field, which is time-consuming and creates disincentives. Here, we present a generalized method, hereafter referred to as DECODE (DEtect, COnsolidate, and DElinetate), that automatically extracts accurate field boundary data from satellite imagery using deep learning based on spatial, spectral, and temporal cues. We introduce a new convolutional neural network (FracTAL ResUNet) as well as two uncertainty metrics to characterize the confidence of the field detection and field delineation processes. We finally propose a new methodology to compare and summarize field-based accuracy metrics. To demonstrate the performance and scalability of our method, we extracted fields across the Australian grains zone with a pixel-based accuracy of 0.87 and a field-based accuracy of up to 0.88 depending on the metric. We also trained a model on data from South Africa instead of Australia and found it transferred well to unseen Australian landscapes. We conclude that the accuracy, scalability and transferability of DECODE shows that large-scale field boundary extraction based on deep learning has reached operational maturity. This opens the door to new agricultural services that provide routine, near-real time field-based analytics.
The retina processes visual images to compute features such as the direction of image motion. Starburst amacrine cells (SACs), axonless feed-forward interneurons, are essential components of the retinal direction-selective circuitry. Recent work has highlighted that SAC-mediated dendro-dendritic inhibition controls the action potential output of direction-selective ganglion cells (DSGCs) by vetoing dendritic spike initiation. However, SACs co-release GABA and the excitatory neurotransmitter acetylcholine at dendritic sites. Here we use direct dendritic recordings to show that preferred direction light stimuli evoke SAC-mediated acetylcholine release, which powerfully controls the stimulus sensitivity, receptive field size and action potential output of ON-DSGCs by acting as an excitatory drive for the initiation of dendritic spikes. Consistent with this, paired recordings reveal that the activation of single ON-SACs drove dendritic spike generation, because of predominate cholinergic excitation received on the preferred side of ON-DSGCs. Thus, dendro-dendritic release of neurotransmitters from SACs bi-directionally gate dendritic spike initiation to control the directionally selective action potential output of retinal ganglion cells.