Plasmonic enhancement of fluorescence has been challenging in in vivo imaging applications. We present a study demonstrating the plasmonic enhancement of fluorescent membrane proteins within their native physiological environment using tailored metallic nanoparticles. This work highlights two schemes to influence the distance between the emitting dipoles and the enhancing nanoparticles, namely the addition of nanoparticles in the buffer solution and the incorporation in the polymer matrix at the bottom of the cells. Incorporating biological structures native to the cellular environment offers opportunities for the optimization of in vivo fluorescence imaging methods and the detection of membrane proteins.

In the present study, the influence of topographic and mechanical cues on neuronal growth cones (NGCs) and network directionality in 3D-engineered cell culture models is explored. Two-photon polymerization (2PP) is employed to fabricate nanopillar arrays featuring tunable effective shear modulus. Large variations in mechanical properties are obtained by altering the aspect ratio of the nanostructures. The nanopillar arrays are seeded with different neuronal cell lines, including neural progenitor cells (NPCs) derived from human induced pluripotent stem cells (iPSCs), I³Neurons, and primary hippocampal neurons. All cell types exhibit preferential orientations according to the nanopillar topology, as shown by neurites creating a high number of oriented orthogonal networks. Furthermore, the differentiation and maturation of NPCs are affected by the topographic and mechanical properties of the nanopillars, as shown by the expression of the mature neuronal marker Synapsin I. Lastly, NGCs are influenced by effective shear modulus in terms of spreading area, and stochastic optical reconstruction microscopy (STORM) is employed to assess the cytoskeleton organization at nanometric resolution. The developed approach, involving laser-assisted 3D microfabrication, neuro-mechanobiology, and super-resolution microscopy, paves the way for prospective comparative studies on the evolution of neuronal networks and NGCs in healthy and diseased (e.g., neurodegenerative) conditions.

Genetically encoded voltage indicators, particularly those based on microbial rhodopsins, are gaining traction in neuroscience as fluorescent sensors for imaging voltage dynamics with high-spatiotemporal precision. Here we establish a novel genetically encoded voltage indicator candidate based on the recently discovered subfamily of the microbial rhodopsin clade, termed heliorhodopsins. We discovered that upon excitation at 530 to 560 nm, wildtype heliorhodopsin exhibits near-infrared fluorescence, which is sensitive to membrane voltage. We characterized the fluorescence brightness, photostability, voltage sensitivity, and kinetics of wildtype heliorhodopsin in HEK293T cells and further examined the impact of mutating key residues near the retinal chromophore. The S237A mutation significantly improved the fluorescence response of heliorhodopsin by 76% providing a highly promising starting point for further protein evolution.

Localized surface plasmons (LSPs) in metal particles are used in medical, chemical, physical, and biological sensing applications. In this paper, we revisit the classical description of LSPs. We use the Drude model and the Quasi-Static approximation to describe the plasmon resonances in terms of the material and the size of the particles embedded in a dielectric host. We then incorporate the Clausius-Mossotti relation to include shape effects in the classical description. Finally, we incorporate surface damping and retardation effects to arrive at a unified, classical description providing an intuitive and realistic model of plasmonic resonances in metal particles.

Voltage imaging using genetically encoded voltage indicators (GEVIs) has taken the field of neuroscience by storm in the past decade. Its ability to create subcellular and network level readouts of electrical dynamics depends critically on the kinetics of the response to voltage of the indicator used. Engineered microbial rhodopsins form a GEVI subclass known for their high voltage sensitivity and fast response kinetics. Here we review the essential aspects of microbial rhodopsin photocycles that are critical to understanding the mechanisms of voltage sensitivity in these proteins and link them to insights from efforts to create faster, brighter and more sensitive microbial rhodopsin-based GEVIs.

Linking single-cell genomic or transcriptomic profiles to functional cellular characteristics, in particular time-varying phenotypic changes, could help unravel molecular mechanisms driving the growth of tumour-cell subpopulations. Here we show that a custom-built optical microscope with an ultrawide field of view, fast automated image analysis and a dye activatable by visible light enables the screening and selective photolabelling of cells of interest in large heterogeneous cell populations on the basis of specific functional cellular dynamics, such as fast migration, morphological variation, small-molecule uptake or cell division. Combining such functional single-cell selection with single-cell RNA sequencing allowed us to (1) functionally annotate the transcriptomic profiles of fast-migrating and spindle-shaped MCF10A cells, of fast-migrating MDA-MB-231 cells and of patient-derived head-and-neck squamous carcinoma cells, and (2) identify critical genes and pathways driving aggressive migration and mesenchymal-like morphology in these cells. Functional single-cell selection upstream of single-cell sequencing does not depend on molecular biomarkers, allows for the enrichment of sparse subpopulations of cells, and can facilitate the identification and understanding of the molecular mechanisms underlying functional phenotypes.

Voltage imaging and optogenetics offer new routes to optically detect and influence neural dynamics. Optimized hardware is necessary to make the most of these new techniques. Here we present the Octoscope, a versatile, multimodal device for all-optical electrophysiology. We illustrate its concept and design and demonstrate its capability to perform both 1-photon and 2-photon voltage imaging with spatial and temporal light patterning, in both inverted and upright configurations, in vitro and in vivo.

Total internal reflection fluorescence (TIRF) microscopy is an important imaging tool for the investigation of biological structures, especially the study on cellular events near the plasma membrane. Imaging at cryogenic temperatures not only enables observing structures in a near-native and fixed state but also suppresses irreversible photo-bleaching rates, resulting in increased photo-stability of fluorophores. Traditional TIRF microscopes produce an evanescent field based on high numerical aperture immersion objective lenses with high magnification, which results in a limited field of view and is incompatible with cryogenic conditions. Here, we present a waveguide-based TIRF microscope, which is able to generate a uniform evanescent field using high refractive index waveguides on photonic chips and to obtain cellular observation at cryogenic temperatures. Our method provides an inexpensive way to achieve total-internal-reflection fluorescence imaging under cryogenic conditions.

Photoactivated genetically encoded voltage indicators (GEVIs) have the potential to enable optically sectioned voltage imaging at the intersection of a photoactivation beam and an imaging beam. We developed a pooled high-throughput screen to identify archaerhodopsin mutants with enhanced photoactivation. After screening ~10⁵ cells, we identified a novel GEVI, NovArch, whose one-photon near-infrared fluorescence is reversibly enhanced by weak one-photon blue or two-photon near-infrared excitation. Because the photoactivation leads to fluorescent signals catalytically rather than stoichiometrically, high fluorescence signals, optical sectioning, and high time resolution are achieved simultaneously at modest blue or two-photon laser power. We demonstrate applications of the combined molecular and optical tools to optical mapping of membrane voltage in distal dendrites in acute mouse brain slices and in spontaneously active neurons in vivo.

Voltage imaging in cells requires high-speed recording of small fluorescent signals, often leading to low signal/noise ratios. Because voltage indicators are membrane bound, their orientations are partially constrained by the plane of the membrane. We explored whether tuning the linear polarization of excitation light could enhance voltage indicator fluorescence. We tested a panel of dye- and protein-based voltage indicators in mammalian cells. The dye BeRST1 showed a 73% increase in brightness between the least and most favorable polarizations. The protein-based reporter ASAP1 showed a 22% increase in brightness, and QuasAr3 showed a 14% increase in brightness. In very thin neurites expressing QuasAr3, improvements were anomalously large, with a 170% increase in brightness between polarization parallel versus perpendicular to the dendrite. Signal/noise ratios of optically recorded action potentials were increased by up to 50% in neurites expressing QuasAr3. These results demonstrate that polarization control can be a facile means to enhance signals from fluorescent voltage indicators, particularly in thin neurites or in high-background environments.