Mutations in the fused in Sarcoma ( FUS ) gene induce cytoplasmic FUS aggregations, contributing to the neurodegenerative disease amyotrophic lateral sclerosis (ALS) in certain cases. While FUS is mainly a nuclear protein involved in transcriptional processes with limited cytoplasmic functions, it shows an additional somatodendritic localization in neurons. In this study we analyzed the localization of FUS in motoneuron synapses, these being the most affected neurons in ALS, using super-resolution microscopy to distinguish between the pre- and postsynaptic compartments. We report a maturation-based variation of FUS localization in rodent synapses where a predominantly postsynaptic FUS was observed in the early stages of synaptic development, while in mature synapses the protein was entirely localized in the axonal terminal. Likewise, we also show that at the synapse of human motoneurons derived from induced pluripotent stem cells of a healthy control, FUS is mainly postsynaptic in the early developmental stages. In motoneurons derived from ALS patients harboring a very aggressive juvenile FUS mutation, increased synaptic accumulation of mutated FUS was observed. Moreover increased aggregation of other synaptic proteins Bassoon and Homer1 was also detected in these abnormal synapses. Having demonstrated changes in the FUS localization during synaptogenesis, a role of synaptic FUS in both dendritic and axonal cellular compartments is probable, and we propose a gain-of-toxic function due to the synaptic aggregation of mutant FUS in ALS.

TAR DNA binding protein 43 (TDP-43) is closely related to the pathogenesis of amyotrophic lateral sclerosis (ALS) and translocates to stress granules (SGs). The role of SGs as aggregation-promoting “crucibles” for TDP-43, however, is still under debate. We analyzed TDP-43 mobility and localization under different stress and recovery conditions using live cell single-molecule tracking and super-resolution microscopy. Besides reduced mobility within SGs, a stress induced decrease of TDP-43 mobility in the cytoplasm and the nucleus was observed. Stress removal led to a recovery of TDP-43 mobility, which strongly depended on the stress duration. ‘Stimulated-emission depletion microscopy’ (STED) and ‘tracking and localization microscopy’ (TALM) revealed not only TDP-43 substructures within stress granules but also numerous patches of slow TDP-43 species throughout the cytoplasm. This work provides insights into the aggregation of TDP-43 in living cells and provide evidence suggesting that TDP-43 oligomerization and aggregation takes place in the cytoplasm separate from SGs.

Tuberculosis remains a serious global health problem causing 1.3 million deaths annually. The causative pathogen Mycobacterium tuberculosis (Mtb) has developed several mechanisms to evade the immune system and resistances to many conventional antibiotics, so that alternative treatment strategies are urgently needed. By isolation from bronchoalveolar lavage and peptide optimization, a new antimicrobial peptide named NapFab is discovered. While showing robust activity against extracellular Mtb, the activity of NapFab against intracellular bacteria is limited due to low intracellular availability. By loading NapFab onto dendritic mesoporous silica nanoparticles (DMSN) as a carrier system, cellular uptake, and consequently antimycobacterial activity against intracellular Mtb is significantly enhanced. Furthermore, using lattice light-sheet fluorescence microscopy, it can be shown that the peptide is gradually released from the DMSN inside living macrophages over time. By electron microscopy and tomography, it is demonstrated that peptide loaded DMSN are stored in vesicular structures in proximity to mycobacterial phagosomes inside the cells, but the nanoparticles are typically not in direct contact with the bacteria. Based on the combination of functional and live-cell imaging analyses, it is hypothesized that after being released from the DMSN NapFab is able to enter the bacterial phagosome and gain access to the bacilli.

In cells, proteins encoded by the same gene do not all behave uniformly but engage in functional subpopulations induced by spatial or temporal segregation. While conventional microscopy has limitations in revealing such spatial and temporal diversity, single-molecule tracking (SMT) microscopy circumvented this problem and allows for high-resolution imaging and quantification of dynamic single-molecule properties. Particularly in the nucleus, SMT has identified specific DNA residence times of transcription factors (TFs), DNA-bound TF fractions and positions of transcriptional hot-spots upon cell stimulation. By contrast to cell stimulation, SMT has not been employed to follow dynamic TF changes along stages of cell differentiation. Herein, we analysed the serum response factor (SRF), a TF involved in the differentiation of many cell types to study nuclear single-molecule dynamics in neuronal differentiation. Our data in living mouse hippocampal neurons show dynamic changes in SRF DNA residence time and SRF DNA-bound fraction between the stages of adhesion, neurite growth and neurite differentiation in axon and dendrites. Using TALM (tracking and localization microscopy), we identified nuclear positions of SRF clusters and observed changes in their numbers and size during differentiation. Furthermore, we show that the SRF cofactor MRTF-A (myocardin-related TF or MKL1) responds to cell activation by enhancing the long-bound DNA fraction. Finally, a first SMT colocalization study of two proteins was performed in living cells showing enhanced SRF/MRTF-A colocalization upon stimulation. In summary, SMT revealed modulation of dynamic TF properties during cell stimulation and differentiation.

We investigate the influence of different accuracy-detection rate trade-offs on image reconstruction in single molecule localization microscopy. Our main focus is the investigation of image artifacts experienced when using low localization accuracy, especially in the presence of sample drift and inhomogeneous background. In this context we present a newly developed SMLM software termed FIRESTORM which is optimized for high accuracy reconstruction. For our analysis we used in silico SMLM data and compared the reconstructed images to the ground truth data. We observe two discriminable reconstruction populations of which only one shows the desired localization behavior.