An exciting project from the lab is now out on bioRxiv. This is a collaboration with electron microscopy guru Stéphane Vassilopoulos from the Myologie Institute in Paris. One of the most striking discovery of super-resolution microscopy so far has been the periodic actin/spectrin scaffold along axons, with actin rings spaced every 190 nm by spectrin tetramers. Since its discovery in 2013, a number of labs, including ours, have used various super-resolution microscopy techniques (SMLM, SIM, STED) to refine our knowledge about this structure. However, actin rings and the axonal periodic scaffold had not been observed by EM until now.
Using mechanical unroofing and platinum-replica electron microscopy (PREM), we were able to visualize the axonal actin rings regularly spaced under the exposed plasma membrane of axons, and to determine their ultrastructure and molecular organization. And we had a big surprise: the actin rings are not made of short, capped actin filaments as it was assumed since their discovery, by analogy with actin inside the erythrocyte submembrane cytoskeleton. Instead, they are braids made of two long (~0.5 to 1 µm) actin filaments!
These actin braids are connected by a dense mesh of spectrins, which we unambiguously identified using immunogold labeling and PREM. Moreover, we probed the stability of the axonal periodic scaffold using actin-targeting drugs, and found similar results by super-resolution and electron microscopy. Finally, we directly demonstrated the identity and organization of the scaffold components (actin, spectrins) by performing correlative SMLM/PREM, visualizing the same sample by super-resolution and electron microscopy.
It was a great pleasure to work with Stéphane! The project benefited from the magic unroofing touch of our Master 2 student Solène, and from important ground work by Angélique and Ghislaine from the team. To read more, please see our bioRxiv preprint and let us know what you think!
We have a new preprint out! Want to do good super-resolution images? We have put together all our SMLM tips and tricks. This is a methods paper that describes our SMLM workflow, using benchmark samples such as microtubules and clathrin-coated pits.
A good image starts with a good sample… that’s why the first part is about how to our fixation and immunolabeling procedures. Then it’s “just” a matter of imaging and processing. Tips for acquiring 3D-STORM images, as well as multi-color DNA-PAINT. Congrats to Angélique and Karoline for their beautiful images. Check our preprint to know everything, and let us know what you think!
We have a new preprint out! It’s a collaboration with the group of Sandrine Lévêque-Fort at ISMO (Orsay, France) based on the PhD work of Clément Cabriel. They previously used supercritical-angle fluorescence to measure the height of fluorophores at the proximity of the coverslip. This technique, called Direct Optical Nanoscopy with Axially Localized Detection (DONALD), could bring the resolution down to 15 nm for 3D localization microscopy. Now they have combined the SAF-based method with cylindrical lens astigmatism to obtain a robust and precise 3D localization of fluorophores over ~1.5 µm above the coverslip, retaining the key advantafges of DONALD: drift-free, tilt-insensitive and achromatic. The new technique, called Dual-view Astigmatic Imaging with SAF Yield (DAISY 😉), allowed to image in 3D the periodic scaffold of adducin and ß-spectrin along axons of cultured neurons, as you can see on the Figure below:
Pumpy (short for Pumpy McPumpface, official name NanoJ-Fluidics) is a pump array made in LEGO, controlled by an Arduino and open-source software (compatible with ImageJ/Fiji and Micro-Manager). NanoJ-Fluidics automates sample fluid exchange right on the microscope stage, allowing complex workflows using your standard chambers, tubing and reagents: live-to-fixed correlative imaging, sequential staining/imaging/washing protocols…
In one application, we used NanoJ-Fluidics to visualize the dynamics of actin in living cells (using SRRF super-resolved processing), and to perform nanoscale imaging of actin using STORM on the same cell after online fixation and labeling with phalloidin.
We also used NanoJ-Fluidics for sequential multiplexed STORM/PAINT imaging, and we obtained 5-channel super-resolved images of actin, intermediate filaments, microtubules, clathrin and mitochondria in cells with minimal intervention during imaging.
If you are interested in making your own, head over to the NanoJ-Fluidics wiki where you will find everything: LEGO parts, assembly instructions, control software and more!
Today the Company of Biologists (a non-profit organization that published journals such as Development or the Journal of Cell Science) launched a new website called PreLights. PreLights is a website where a group of young researchers (the PreLighters) curate biology preprints, highlighting their value and importance. Some highlights also feature feedback from authors. We are happy to support this open science initiative by providing the PreLight banner image: neurons in culture that reflects the connections and interactions preLights wants to promote.
Another work in our fruitful collaboration with Subhojit Roy and his lab (now at UW Madison). In 2015 we could visualize new axonal actin structures by STORM (see our cover): stable clusters every 3-4 µm we called “hotspots” from which “trails” would rapidly assemble and disassemble along the axon. In this new preprint, trails were shown to have a slight anterograde bias (55%) and to polymerize from the surface of the hotspots, pushing the trails away. This suggested that biased dynamic trails assembly could underlie the slow anterograde transport of actin, whose mechanism is still unknown. Modelling done by Nilaj Chakrabarty and Peter Jung (Ohio University) indeed showed that the observed biased assembly and disassembly of trails would lead to a ~0.5 mm/day transport of actin, in line with earlier measurements.
A new preprint is out today on bioRxiv. This collaboration with the Ricardo Henriques and Jason Mercer labs proposes a new metric to measure the quality of super-resolution images. Simply put, it compares the image to a reference diffraction-limited image, allowing to detect artefacts and missing features in the super-resolved image. We used it to determine when to stop a STORM acquisition when visualizing axonal actin rings, and to optimize the dye concentration in a DNA-PAINT experiment. The method, called NanoJ-SQUIRREL (Super-resolution Quantitative Image Rating and Reporting of Error Locations), is available as an easy to use, open-source plugin for the ImageJ/Fiji plugin software. Try it!