Periodic Reporting for period 3 - NVS (Nano Voltage Sensors)
Reporting period: 2019-01-01 to 2020-06-30
VSDs and VSPs allow for simultaneous direct visualization of neuronal activity over a large number of neurons in a large field-of-view. However, they suffer from several shortcomings: they could alter membrane capacitance, be phototoxic, suffer from photobleaching, have a short retention time in the membrane, and miss-target the membrane (resulting in nonspecific background labeling). We developed voltage sensing nanoparticles (vsNPs) in the shape of nanorods (vsNRs) that self-insert into the cell membrane and could optically record, non-invasively, action potentials (APs) at the single-particle and nanoscale level, at multiple sites, in a large field-of-view.
This revolutionizing research is expected to lead to new medical treatments and many new answers on how the brain functions. The flexibility in designing semiconductor NPs with exquisite control of sizes, compositions, bandgaps, excited state (exciton) wavefunction and lifetime, makes their development into biomedical reagents highly attractive.
Significant growth potential for voltage imaging of neurons is expected from the highly demanding and rapidly growing fields of Optogenetics and bioelectronic medicine, where voltage sensing and action potential actuation (stimulation) are desired in specific neuronal circuits.
From a pharmaceutical perspective, neural circuits mapping would allow to design better drugs for neuronal disorders ranging from neurodegenerative disorders such as Alzheimer and Parkinson’s to mental disorders, the most prevalent of which is depression. In addition, recording aberrant neuronal signals from peripheral neurons would present the opportunity to correct them, through an external feedback circuit (bioelectronics). This in turn will lead to economic benefits in the form of reduced economic burden and social benefits in the form of a healthier, more productive society.
In a broader sense, the implication of high resolution spatiotemporal recordings of neuron circuits will fuel emerging fields like neuroinformatics, medical informatics, Neuromorphic computing and Neurorobotics. As an example, building computational models from analysis of neuron circuits will allow construction of neural networks. Neural networks can provide robust solutions to problems in a wide range of disciplines, particularly areas involving classification, prediction, filtering, optimization, pattern recognition, and function approximation, all are relevant to Artificial Intelligence and big data applications.
We propose to develop targetable voltage sensing nanoparticles (vsNPs) in the shape of nanorods (vsNRs)) that self-insert into the cell membrane and optically record, non-invasively, action potentials (APs) at the single-particle and nanoscale level, at multiple sites, in a large field-of-view.
In addition, vsNRs could be targeted to specific neurons (or muscle cells), and specific sites within neurons (or muscle cells) by conjugation with specific recognition molecules. Once the recognition molecule binds to a target (a specific membrane protein), the vsNR self- inserts into the cell membrane in close proximity to that target. With single-molecule sensitivity, vsNRs could allow nanoscale recording of pre- and post-synaptic membrane potentials during release of- and signaling by- neurotransmitters, record sub-threshold events, monitor individual ion channel activity, or the release of ions from single Ca2+ stores. With the appropriate imaging methods, vsNRs could be used to optically record fast APs transients, in a high-throughput fashion, simultaneously at many sites, over a large field-of-view, and potentially at a resolution exceeding the diffraction limit. Moreover, deep tissue imaging could be afforded by 2PE using NIR vsNRs and far-field non-linear temporal focusing microscopy. Such capabilities could eliminate the need for a crowded array of contact electrodes and bulky read-out electronics. vsNRs could therefore enable remote, noise-immune, sensitive AP recordings.
We and others have recently demonstrated both theoretically and experimentally that engineered NRs could display large QCSE even at room temperature and on the single particle level, and therefore could act as efficient nanoscale vsNRs by displaying changes in fluorescence quantum yield, fluorescence lifetime and/or peak emission wavelength in response to local field perturbation. Moreover, we have tested the feasibility of a functionalization approach for stable membrane insertion of vsNRs into HEK293T cell membranes and demonstrated feasibility of voltage sensing by vsNRs (see below). Our experimental and theoretical studies suggest that vsNRs could afford simultaneous monitoring of the spiking output of a large number of cells.
Here we seek to optimize all aspects of vsNPs synthesis, functionalization, delivery, targeting and detection in order to provide neuroscientists and physiologists a viable and user-friendly technology for the study of AP signals in the brain and in healthy or diseased heart and muscle tissues. Our aims are: (i) development of synthetic routes for heterostructure nanoparticles to produce ultrasensitive vsNPs and characterization of their performance by advanced spectroscopy; (ii) development of functionalization methods based on ligand exchange with transmembrane (amphiphilic) peptides by self-assembly and optimization of vsNPs insertion into the membrane; (iii) development of optical vsNPS recording methodologies and benchmarking of vsNPs performance; (iv) demonstration of vsNPs utility in model neuronal systems and exploration of their use in studying synapses and the molecular dynamics involved in neuronal integration of excitatory and inhibitory inputs.
The optimal voltage sensors for single particle electrophysiology need to have large voltage sensitivity while maintaining small sizes for membrane insertion. Using a high-throughput QCSE screening assay, we successfully demonstrated that the 12 nm long type-II NRs (sample(v)) exhibit much larger voltage sensitivities (ΔF/F and Δλ) compared to the other NPs studied, including spherical QDs and 40 nm long quasi-type-I NRs. These NRs, with both positive and negative ΔF/F and Δλ due to random orientations in the electric field, are capable of reporting not only the field strength but also the field direction. we were able to provide an estimate for the tradeoff between detection probability and false positive rate of an “action-potential-like” voltage transient. realistic parameters currently enable a 50% detection probability with a false positive rate well below 1%, figures which are already close to the scale of what is necessary for electrophysiology applications.
We used the nanocavity-based method for measuring single nanorod quantum yield, and observed that the quantum yield that corresponds to bright states of the same nanorod can be as high as 90%.
Functionalization, membrane insertion, & targeting
We are exploring two leading strategies for particle functionalization:
(1) functionalization by ligand adsorption to as-synthesized hydrophobic NRs and QDs
Lipid coating: We have developed a novel method for membrane insertion of large QDs (>5nm diameter) by delivery via a polar solvent. This is achieved by lipid functionalization of the NP surface at specific lipid:QD ratios and lipid type.
peptide coating: by adsorption of amphipatic and/or hydrophobic trans membrane peptide sequences we prepared water soluble particles which show single particle membrane staining which is peptide dependent.
(2) functionalization by ligand exchange of TOPO coated NRs.
Curvature dependent facet selectivity: We explore facet selective chemistry that is dictated by surface geometry alone and was demonstrated for GNPs. This approach relies on the observation that the pKa of acidic surface ligands depends on ligand density which depends on the curvature (defined as 1/r). For non spherical particles as the NRs we would expect a different pKa for the long axis and for the tips. This should lead to pH dependent chemical reactivity.
DNA Origami: We explored an alternative functionalization method that is based on a library of chemically modified dsDNA molecules, where hydrophobic tags (primarily porphyrins) will act as membrane anchors. Following the work on DNA nanopores, and together with our collaborator, Dr. Stulz (U. Southampton), we have designed and prepared a small library of DNA duplexes for functionalizing the surface of vsNRs for cell membrane insertion.
vsNP-based optical recording methodologies
We have developed several high throughput imaging assays for selecting best performing membrane potential nanosensors. 1. Membrane staining assay: A rapid scan of wells with a high-NA objective is performed in order to test for membrane staining, indicating for some association of the nanoparticles to the membrane. This serves as a fast low content screen, which is used as a preliminary step for subsequent assays. 2. Fluorescence anisotropy: Fluorescence anisotropy microscopy is used for measuring the directionality of anisotropic nanoparticles relative to the membrane. Particles are analyzed and scored according to their directionality relative to the membrane normal.
3. Induced transmembrane voltage: In order to polarize the cell membrane and test for correlation between the fluorescence of membrane-associated nanoparticles and membrane potential, we apply high and short pulses of voltage with a set of external electrodes on a large number of stained cells.
Spherical CHO cells are used, as their induced-polarization is analytically modeled. Nanoparticles that display voltage-sensitivity of fluorescence upon membrane polarization are selected for further characterization and full voltage calibration by patch clamping.
Demonstration of biological applications using vsNRs
We optimized conditions for loading vsNRs into primary dissociated cortical neurons.
For vsNRs imaging in neurons we have built a dedicated setup. The setup is equipped with a dual emission image splitter (Optosplit II, Andor) for vsNR signal acquisition at multiple wavelengths, and a patch-clamp recording hardware (Axon) for validation and calibration of the vsNR signal.
In one set of experiments, we use dichroic mirror with the edge at 640 nm to simultaneously record and effectively separate fluorescence of vsNRs and voltage-sensitive membrane dye.This experimental arrangement is ideal for optimization of the loading and imaging conditions.
In another set of experiments using a dichroic mirror with the edge at the vsNR emission maximum (~600nm) we collect at the same time voltage-related fluctuations of vsNR intensity (F; sum of two channels) and voltage-related shift of the vsNR emission maximum (; ratio between two channels). This setup arrangement is the most sensitive for voltage-related changes of vsNR fluorescence and it is ideal for recordings that require profound sensitivity.
vsNRs recordings suffer from poor signal to noise ratio (SNR) because fluorescent signal of a single NR is affected by intermittency (blinking). We work on development of automated software for analysis of NR fluorescence in neurons, including algorithms that effectively identify and excise blinking periods.