Locomotor control is facilitated by mechanosensory inputs that report how the body interacts with a physical medium. Effective representation of compliant wing deformations is particularly challenging due to the many degrees of freedom. Structural configurations can constrain the stimulus space, and strategic placement of sensors can simplify computation. Here, we measured and modeled wing displacement fields and characterized spatiotemporal encoding of the wing mechanosensors. Our data show how dragonfly wing architecture prescribes deformation modes consistent across models and measurements. We found that the wing's state under normal flapping conditions is detected by the spike timing of few sensors, with additional sensors recruited under perturbation. The functional integration of wing biomechanics and sensor placement enables a straightforward solution for information transfer.

Quantitative analysis of neuron morphology is essential in order to develop our understanding of circuit organisation and development. The recent acquisition of whole-brain electron microscopy-based (EM) reconstructions of the Drosophila melanogaster nervous system now provide the resolution needed to examine morphology at scale. Utilising these data, together with new computational tools, we extract and analyse the dendrites of all T4 and T5 neurons within one hemisphere (n \~ 6000).T4 and T5 neurons are the first uniquely direction-selective neurons in the visual pathway, and are classified into four subtypes (a,b,c, and d). Each subtype encodes one of four cardinal motion directions (up, down, forwards, backwards). The dendrites of these neurons form in two distinct neuropils, the Medulla (T4) and the Lobula (T5), and are asymmetrically oriented in a direction inverse to the direction of motion which they encode. However, their densely overlapping and compact arbours has made rigorous morphological quantification challenging. The presence of differences beyond their characteristic orientation, both between T4 and T5, as well as within subtypes, has remained poorly understood.Our analysis reveals a high degree of structural similarity across both types and subtypes. Particularly, measures of geometry and graph topology show only minor variation, with no consistent separation between T4 and T5, or their subtypes.These results indicate that, despite forming in different neuropils, and serving distinct motion directions, T4 and T5 dendrites follow closely aligned morphological patterns. This suggests that their arborization may be governed by shared developmental constraints and mechanisms.

Continuous glucose monitors have proven invaluable for monitoring blood glucose levels for diabetics, but they are of limited use for observing glucose dynamics at the cellular (or subcellular) level. We have developed a second generation, genetically encoded intensity-based glucose sensing fluorescent reporter (iGlucoSnFR2). We show that when it is targeted to the cytosol, it reports intracellular glucose consumption and gluconeogenesis in cell culture, along with efflux from the endoplasmic reticulum. It outperforms the original iGlucoSnFR in vivo when observed by fiber photometry in mouse brain and reports transient increase in glucose concentration when stimulated by noradrenaline or electrical stimulation. Last, we demonstrate that membrane localized iGlucoSnFR2 can be calibrated in vivo to indicate absolute changes in extracellular glucose concentration in awake mice. We anticipate iGlucoSnFR2 facilitating previously unobservable measurements of glucose dynamics with high spatial and temporal resolution in living mammals and other experimental organisms.

The timing of spikes dictates a neuron’s impact on downstream circuits and behavior, and spike timing is determined by the membrane potential (Vm). However, due to technical challenges, it has been impossible to analyze the relative timing of Vm dynamics between neurons during behavior. Using large scale membrane voltage imaging, we simultaneously recorded Vm from many individual hippocampal neurons in animals engaged in a virtual spatial task. We found that relative phase of Vm theta oscillations across neurons exhibit gradual or discrete shifts depending on spatial position. This finding extends beyond previous studies showing Vm dynamics in single neurons or spiking activity in multiple neurons, revealing previously unknown evidence for consistent coding of space by spike-independent relative phase of Vm theta dynamics between neurons.

To establish functional connectivity between two candidate neurons that might form a circuit element, a common approach is to activate an optogenetic tool such as Chrimson in the candidate pre-synaptic neuron and monitor fluorescence of the calcium-sensitive indicator GCaMP in a candidate post-synaptic neuron. While performing such experiments in Drosophila, we found that low levels of leaky Chrimson expression can lead to strong artifactual GCaMP signals in presumptive postsynaptic neurons even when Chrimson is not intentionally expressed in any particular neurons. Withholding all-trans retinal, the chromophore required as a co-factor for Chrimson response to light, eliminates GCaMP signal but does not provide an experimental control for leaky Chrimson expression. Leaky Chrimson expression appears to be an inherent feature of current Chrimson transgenes, since artifactual connectivity was detected with Chrimson transgenes integrated into multiple genomic locations. While these false-positive signals may complicate the interpretation of functional connectivity experiments, we illustrate how a no-Gal4 negative control improves interpretability of functional connectivity assays. We also propose a simple but effective procedure to identify experimental conditions that minimize potentially incorrect interpretations caused by leaky Chrimson expression.

Microbiota-derived metabolites have emerged as key regulators of longevity. The metabolic activity of the gut microbiota, influenced by dietary components and ingested chemical compounds, profoundly impacts host fitness. While the benefits of dietary prebiotics are well-known, chemically targeting the gut microbiota to enhance host fitness remains largely unexplored. Here, we report a novel chemical approach to induce a pro-longevity bacterial metabolite in the host gut. We discovered that wild-type Escherichia coli strains overproduce colanic acids (CAs) when exposed to a low dose of cephaloridine, leading to an increased life span in the host organism Caenorhabditis elegans. In the mouse gut, oral administration of low-dose cephaloridine induced transcription of the capsular polysaccharide synthesis (cps) operon responsible for CA biosynthesis in commensal E. coli at 37 °C, and attenuated age-related metabolic changes. We also found that low-dose cephaloridine overcomes the temperature-dependent inhibition of CA biosynthesis and promotes its induction through a mechanism mediated by the membrane-bound histidine kinase ZraS, independently of cephaloridine's known antibiotic properties. Our work lays a foundation for microbiota-based therapeutics through chemical modulation of bacterial metabolism and highlights the promising potential of leveraging bacteria-targeting drugs in promoting host longevity.

Memories are believed to be stored in synapses and retrieved by reactivating neural ensembles. Learning alters synaptic weights, which can interfere with previously stored memories that share the same synapses, creating a trade-off between plasticity and stability. Interestingly, neural representations change even in stable environments, without apparent learning or forgetting-a phenomenon known as representational drift. Theoretical studies have suggested that multiple neural representations can correspond to a memory, with postlearning exploration of these representation solutions driving drift. However, it remains unclear whether representations explored through drift differ from those learned or offer unique advantages. Here, we show that representational drift uncovers noise-robust representations that are otherwise difficult to learn. We first define the nonlinear solution space manifold of synaptic weights for fixed input-output mappings, which allows us to disentangle drift from learning and forgetting and simulate drift as diffusion within this manifold. Solutions explored by drift have many inactive and saturated neurons, making them robust to weight perturbations due to noise or continual learning. Such solutions are prevalent and entropically favored by drift, but their lack of gradients makes them difficult to learn and nonconducive to future learning. To overcome this, we introduce an allocation procedure that selectively shifts representations for new stimuli into a learning-conducive regime. By combining allocation with drift, we resolve the trade-off between learnability and robustness.

Enzyme-based self-labeling tags enable the covalent attachment of synthetic molecules to proteins inside living cells. A frontier of this field is designing cell-permeable multifunctional ligands that contain fluorophores in combination with affinity tags or pharmacological agents. This is challenging since attachment of additional chemical moieties onto fluorescent ligands can adversely affect membrane permeability. To address this problem, we examined the chemical properties of rhodamine-based self-labeling tag ligands through the lens of medicinal chemistry. We found that the lactone-zwitterion equilibrium constant () of rhodamines inversely correlates with their distribution coefficients (log), suggesting that ligands based on dyes exhibiting low and high log values, such as Si-rhodamines, would efficiently enter cells. We designed cell-permeable multifunctional HaloTag ligands with a biotin moiety to purify mitochondria or a JQ1 appendage to translocate BRD4 within the nucleus. We found that translocation of BRD4 to constitutive heterochromatin in cells leads to apparent increases in transcriptional activity. These fluorescent reagents enable affinity capture and translocation of intracellular proteins in living cells, and our general design concepts will facilitate the design of multifunctional chemical tools for biology.

Preprint: https://doi.org/10.1101/2022.07.02.498544 
Preprint: https://doi.org/10.32388/0xcyuc 

Localisation microscopy often relies on detailed models of point-spread functions. For applications such as deconvolution or PSF engineering, accurate models for light propagation in imaging systems with a high numerical aperture are required. Different models have been proposed based on 2D Fourier transforms or 1D Bessel integrals. The most precise ones combine a vectorial description of the electric field and accurate aberration models. However, it may be unclear which model to choose as there is no comprehensive comparison between the Fourier and Bessel approaches yet. Moreover, many existing libraries are written in Java (e.g., our previous PSF generator software) or MATLAB, which hinders their integration into deep learning algorithms. In this work, we start from the original Richards-Wolf integral and revisit both approaches in a systematic way. We present a unifying framework in which we prove the equivalence between the Fourier and Bessel strategies and detail a variety of correction factors applicable to both of them. Then, we provide a high-performance implementation of our theoretical framework in the form of an open-source library that is built on top of PyTorch, a popular library for deep learning. It enables us to benchmark the accuracy and computational speed of different models and allows for an in-depth comparison of the existing models for the first time. We show that the Bessel strategy is optimal for axisymmetric beams, while the Fourier approach can be applied to more general scenarios. Our work enables the efficient computation of a point-spread function on CPU or GPU, which can then be included in simulation and optimisation pipelines.