Resolution enhancement with a task-assisted GAN to guide optical nanoscopy image analysis and acquisition.

Super-resolution fluorescence microscopy methods enable the characterization of nanostructures in living and fixed biological tissues. However, they require the adjustment of multiple imaging parameters while attempting to satisfy conflicting objectives, such as maximizing spatial and temporal resolution while minimizing light exposure. To overcome the limitations imposed by these trade-offs, post-acquisition algorithmic approaches have been proposed for resolution enhancement and image-quality improvement. Here we introduce the task-assisted generative adversarial network (TA-GAN), which incorporates an auxiliary task (for example, segmentation, localization) closely related to the observed biological nanostructure characterization. We evaluate how the TA-GAN improves generative accuracy over unassisted methods, using images acquired with different modalities such as confocal, bright-field, stimulated emission depletion and structured illumination microscopy. The TA-GAN is incorporated directly into the acquisition pipeline of the microscope to predict the nanometric content of the field of view without requiring the acquisition of a super-resolved image. This information is used to automatically select the imaging modality and regions of interest, optimizing the acquisition sequence by reducing light exposure. Data-driven microscopy methods like the TA-GAN will enable the observation of dynamic molecular processes with spatial and temporal resolutions that surpass the limits currently imposed by the trade-offs constraining super-resolution microscopy.

A postzygotic de novo NCDN mutation identified in a sporadic FTLD patient results in neurochondrin haploinsufficiency and altered FUS granule dynamics.

Frontotemporal dementia (FTD) is a heterogeneous clinical disorder characterized by progressive abnormalities in behavior, executive functions, personality, language and/or motricity. A neuropathological subtype of FTD, frontotemporal lobar degeneration (FTLD)-FET, is characterized by protein aggregates consisting of the RNA-binding protein fused in sarcoma (FUS). The cause of FTLD-FET is not well understood and there is a lack of genetic evidence to aid in the investigation of mechanisms of the disease. The goal of this study was to identify genetic variants contributing to FTLD-FET and to investigate their effects on FUS pathology. We performed whole-exome sequencing on a 50-year-old FTLD patient with ubiquitin and FUS-positive neuronal inclusions and unaffected parents, and identified a de novo postzygotic nonsense variant in the NCDN gene encoding Neurochondrin (NCDN), NM_014284.3:c.1206G > A, p.(Trp402*). The variant was associated with a ~ 31% reduction in full-length protein levels in the patient's brain, suggesting that this mutation leads to NCDN haploinsufficiency. We examined the effects of NCDN haploinsufficiency on FUS and found that depleting primary cortical neurons of NCDN causes a reduction in the total number of FUS-positive cytoplasmic granules. Moreover, we found that these granules were significantly larger and more highly enriched with FUS. We then examined the effects of a loss of FUS function on NCDN in neurons and found that depleting cells of FUS leads to a decrease in NCDN protein and mRNA levels. Our study identifies the NCDN protein as a likely contributor of FTLD-FET pathophysiology. Moreover, we provide evidence for a negative feedback loop of toxicity between NCDN and FUS, where loss of NCDN alters FUS cytoplasmic dynamics, which in turn has an impact on NCDN expression.

Activity-Dependent Remodeling of Synaptic Protein Organization Revealed by High Throughput Analysis of STED Nanoscopy Images.

The organization of proteins in the apposed nanodomains of pre- and postsynaptic compartments is thought to play a pivotal role in synaptic strength and plasticity. As such, the alignment between pre- and postsynaptic proteins may regulate, for example, the rate of presynaptic release or the strength of postsynaptic signaling. However, the analysis of these structures has mainly been restricted to subsets of synapses, providing a limited view of the diversity of synaptic protein cluster remodeling during synaptic plasticity. To characterize changes in the organization of synaptic nanodomains during synaptic plasticity over a large population of synapses, we combined STimulated Emission Depletion (STED) nanoscopy with a Python-based statistical object distance analysis (pySODA), in dissociated cultured hippocampal circuits exposed to treatments driving different forms of synaptic plasticity. The nanoscale organization, characterized in terms of coupling properties, of presynaptic (Bassoon, RIM1/2) and postsynaptic (PSD95, Homer1c) scaffold proteins was differently altered in response to plasticity-inducing stimuli. For the Bassoon - PSD95 pair, treatments driving synaptic potentiation caused an increase in their coupling probability, whereas a stimulus driving synaptic depression had an opposite effect. To enrich the characterization of the synaptic cluster remodeling at the population level, we applied unsupervised machine learning approaches to include selected morphological features into a multidimensional analysis. This combined analysis revealed a large diversity of synaptic protein cluster subtypes exhibiting differential activity-dependent remodeling, yet with common features depending on the expected direction of plasticity. The expanded palette of synaptic features revealed by our unbiased approach should provide a basis to further explore the widely diverse molecular mechanisms of synaptic plasticity.

Morphologic three-dimensional scanning of fallopian tubes to assist ovarian cancer diagnosis.

The majority of high-grade serous ovarian cancers is now believed to originate in the fallopian tubes. Therefore, current practices include the pathological examination of excised fallopian tubes. Detection of tumors in the fallopian tubes using current clinical approaches remains difficult but is of critical importance to achieve accurate staging and diagnosis. Here, we present an intraoperative imaging system for the detection of human fallopian tube lesions. The system is based on optical coherence tomography (OCT) to access subepithelial tissue architecture. To demonstrate that OCT could identify lesions, we analyzed 180 OCT volumes taken from five different ovarian lesions and from healthy fallopian tubes, and compared them to standard pathological review. We demonstrated that qualitative features could be matched to pathological conditions. We then determined the feasibility of intraluminal imaging of intact human fallopian tubes by building a dedicated endoscopic single-fiber OCT probe to access the mucosal layer inside freshly excised specimens from five patients undergoing prophylactic surgeries. The probe insertion into the lumen acquired images over the entire length of the tubes without damaging the mucosa, providing the first OCT images of intact human fallopian tubes.