Rapid automated 3-D pose estimation of larval zebrafish using a physical model-trained neural network.

Quantitative ethology requires an accurate estimation of an organism's postural dynamics in three dimensions plus time. Technological progress over the last decade has made animal pose estimation in challenging scenarios possible with unprecedented detail. Here, we present (i) a fast automated method to record and track the pose of individual larval zebrafish in a 3-D environment, applicable when accurate human labeling is not possible; (ii) a rich annotated dataset of 3-D larval poses for ethologists and the general zebrafish and machine learning community; and (iii) a technique to generate realistic, annotated larval images in different behavioral contexts. Using a three-camera system calibrated with refraction correction, we record diverse larval swims under free swimming conditions and in response to acoustic and optical stimuli. We then employ a convolutional neural network to estimate 3-D larval poses from video images. The network is trained against a set of synthetic larval images rendered using a 3-D physical model of larvae. This 3-D model samples from a distribution of realistic larval poses that we estimate a priori using a template-based pose estimation of a small number of swim bouts. Our network model, trained without any human annotation, performs larval pose estimation three orders of magnitude faster and with accuracy comparable to the template-based approach, capturing detailed kinematics of 3-D larval swims. It also applies accurately to other datasets collected under different imaging conditions and containing behavioral contexts not included in our training.

Quantifying protein dynamics and stability in a living organism.

As an integral part of modern cell biology, fluorescence microscopy enables quantification of the stability and dynamics of fluorescence-labeled biomolecules inside cultured cells. However, obtaining time-resolved data from individual cells within a live vertebrate organism remains challenging. Here we demonstrate a customized pipeline that integrates meganuclease-mediated mosaic transformation with fluorescence-detected temperature-jump microscopy to probe dynamics and stability of endogenously expressed proteins in different tissues of living multicellular organisms.

Optical Control of Metal Ion Probes in Cells and Zebrafish Using Highly Selective DNAzymes Conjugated to Upconversion Nanoparticles.

Spatial and temporal distributions of metal ions in vitro and in vivo are crucial in our understanding of the roles of metal ions in biological systems, and yet there is a very limited number of methods to probe metal ions with high space and time resolution, especially in vivo. To overcome this limitation, we report a Zn2+-specific near-infrared (NIR) DNAzyme nanoprobe for real-time metal ion tracking with spatiotemporal control in early embryos and larvae of zebrafish. By conjugating photocaged DNAzymes onto lanthanide-doped upconversion nanoparticles (UCNPs), we have achieved upconversion of a deep tissue penetrating NIR 980 nm light into 365 nm emission. The UV photon then efficiently photodecages a substrate strand containing a nitrobenzyl group at the 2'-OH of adenosine ribonucleotide, allowing enzymatic cleavage by a complementary DNA strand containing a Zn2+-selective DNAzyme. The product containing a visible FAM fluorophore that is initially quenched by BHQ1 and Dabcyl quenchers is released after cleavage, resulting in higher fluorescent signals. The DNAzyme-UCNP probe enables Zn2+ sensing by exciting in the NIR biological imaging window in both living cells and zebrafish embryos and detecting in the visible region. In this study, we introduce a platform that can be used to understand the Zn2+ distribution with spatiotemporal control, thereby giving insights into the dynamical Zn2+ ion distribution in intracellular and in vivo models.

ß-Carotene As a Lipophilic Scavenger of Nitric Oxide.

The efficient bleaching following continuous bubbling of gaseous nitric oxide (NO(•)) to ß-carotene (ß-Car) dissolved in n-hexane under anaerobic conditions results from an initial addition of two NO(•) followed by fragmentation coupled with further NO(•) addition as shown by mass spectrometry (MS). Density functional theory (DFT) calculations demonstrated that hydrogen atom transfer (HAT) and electron transfer (ET) from ß-Car to NO(•) are strongly energetically unfavorable in contrast to radical adduct formation (RAF) followed by degradation. The results indicated the lowest energy for addition of the first NO(•) at C7 with an activation free energy of ΔG(≠) = 74.40 kJ mol(-1) and a rate constant of 0.56 s(-1), followed by trans-addition of a second NO(•) at C8 with ΔG(≠) = 55.51 kJ mol(-1). MS confirmed the formation of a dinitrosyl-ß-Car (596.6 m/z), and of a ß-Car fragment (400.4 m/z) formed by C7/C8 bond cleavage and suggested to be of importance for progression of bleaching. Up to eight reaction products with increasing mass of 28 m/z are assigned to continuous addition of NO(•) to the initially formed fragment forming nitroxides. Continuous wave photolysis of sodium nitroprusside (SNP) as a NO(•) source dissolved together with ß-Car in 4:1 (v/v) methanol:tetrahydrofuran gradually bleached ß-Car. Nanosecond laser flash photolysis at 355 nm followed by transient absorption spectroscopy showed a ß-Car derived intermediate with an absorption maximum around 420 nm in agreement with a prediction (425 nm) from time-dependent DFT (TDDFT) for the trans-C7,8 dinitrosyl adduct of ß-Car. The NO(•) adduct of ß-Car decays with a rate constant of ∼10(7) s(-1) at 25 °C.