1

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

EMS has it!...

P.O. Box 550 • 1560 Industry Rd. Hatfield, Pa 19440 Tel: (215) 412-8400 • Fax: (215) 412-8450 email: info@emsdiasum.com or stacie@ems-secure.com

OUR MAIN INTERACTIVE WEBSITE:

www.emsdiasum.com

Follow us on...

a comprehensive range of Tissue-Tek® products for...

n Tissue Processing • Tissue-Tek VIP 6® AI Tissue Processor • Accessories for the VIP 6 AI, VIP 6, and VIP 5

including VIP® Fixative, Paraffin, and filters

n Embedding • Tissue-Tek TEC 6™ Embedding Console System

and Accessories • Tissue-Tek TEC 5™ Accessories • Nanomolds

n Cassettes • Tissue-Tek® Uni-Cassette® System • Tissue-Tek® Mega-Cassette™ System

n Coverslipping • Tissue-Tek® Film® Coverslipper and Accessories • Tissue-Tek® Glas™ g2 Coverslipper

and Accessories

n Consumables • Coverslipping Film, Coverslips,

and Mountant

n Slide Staining • Tissue-Tek® Prisma® Plus Slide Stainer • Histo-Tek SL 6® Slide Stainer

and Accessories • Tissue-Tek® Manual Slide Staining Set

n Advanced Staining • Tissue-Tek® Quick Ray™ Tissue Microarray System

n Microtomy & Cryotomy • Tissue-Tek® Cryo3® Flex Cryostat and Accessories • Tissue-Tek® Accu-Edge® Cryostat Disposable

Blade Holders • Tissue-Tek® Accu-Cut® SRM 300™ LT

Manual Microtome • Tissue-Tek® Accu-Cut® SRM 200™ Rotary Microtome • Tissue-Tek® Accu-Edge® Disposable Microtome Blades

n Grossing and Trimming • Tissue-Tek® Accu-Edge® Trimming Blades and Handles • Tissue-Tek® Accu-Edge® Scalpels, Scissors,

and Autopsy Knives

n Cytology Preparation • Cyto-Tek® 2500 Cytocentrifuge

n Printing and Storage • Tissue-Tek® Lab Aid® Slide Filing System • Tissue-Tek® SmartWrite® Color Slide Printer

...and much, much more!

TissueTEK Ad_Nov2019_MA.qxp_TissueTEK Ad_Nov2019 12/10/19 3:55 PM Page 1

3

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

3 Editor’s letter4 News10ProfileJan Huisken13 Caliciviruses: A tunnel to initiate infection Michaela J Conley16 Fast volume imaging with bessel beam tomography and machine learning Andres Flores Valle and Johannes D. Seelig18 Lab focus Peng Chen20 4D STEM with a direct electron detector Barnaby D.A. Levin, Chenyu Zhang, Benjamin Bammes, Paul M. Voyles, Robert B. Bilhorn23 Meeting report26 Product focus Digital cameras26 What’s new30 Company spotlight Protochips

Editor Dr Chris Parmenter commed.ma@gmail.comNews editor Dr Rebecca Pool rebeccapool@microscopy-analysis.comSales EMEA Dr. Stefanie Krauth + 49 (0) 6201 606 728 mkrauth@wiley.comSales North America A-K Roland Espinosa201-748-6819 respinosa@wiley.comSales North America L-Z Joe Tomaszewski908-514-0776 jtomaszews@wiley.comSales APAC Yosuke Sato+81-3-3830-1234 ysato@wiley.comAdvertising production Kerstin Kunkel

kkunkel@wiley.comReader Services wiley@vuservice.dePublisher Dr Heiko BaumgartnerDirector of publishing Roy OpieAssistant editor Simon EvansAssistant web editor Felix David

MICROSCOPY AND ANALYSIS ISSN 2049-4424© 2020 John Wiley & Sons, Ltd. Issued in: January, March, May, July, September, NovemberPublished by: John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, UK Tel: +44 (0)1243 770443Fax: +44 (0)1243 770432Email: microscopy-analysis@wiley.co.uk Web: www.microscopy-analysis.comWhile every effort is made to ensure accuracy, John Wiley & Sons, Ltd and its agents cannot accept responsibility for claims made by contributors, manufacturers or advertisers.

DEAR FRIENDS, Welcome to our first issue of 2020! I hope you’re getting stuck into the New Year and the new decade! I can hardly believe it’s been five years since I became editor of M&A! I’m excited to see where our field will be in another five years.

This issue we’ve our usual diverse line-up of scientific editorials covering areas including 4D STEM, Bessel Beam tomography and Cryo-EM of virus structures. There is one inescapable requirement to all our modern microscopies, the image acquisition systems, and this issue our product focus is digital cameras.

The spotlight focusses on John Damiano from Protochips detailing the company's journey in the growth area of in situ microscopy. We also have a fascinating profile with a pioneer of light sheet microscopy Jan Huisken. He walks us though his beginnings in physics, the development of the light sheet through to his latest work, the ‘Flamingo’. We also visit the lab of Peng Chen in Cornell as he and colleagues develop COMPEITS, a super-resolution microscopy technique especially for non-fluorescent samples.

Don't miss the news and What’s New section to catch up on commercial developments. I also report from the November’s Cryo Microscopy Group (CMG) meeting. An issue with something for everyone!

CHRIS

INSIDEJANUARY-FEBRUARY2020 EMEA 46

INTROEDITOR’S LETTERDR CHRIS PARMENTER

2020 vision

Cover story

SCIENTISTS AT THE UNIVERSITY OF CALIFORNIA, Irvine (UCI) have developed a scanning transmission electron microscopy (STEM) method that can visualize the electric charge density of materials with unprecedented sub-angstrom resolution. Using this technique, the researchers were able to uncover the origin of ferroelectricity and the mechanism of electron charge transfer at a ferroelectric interface.

The charge density distribution in materials dictates their structure, chemical bonding, electronic transport, and optical and mechanical properties as these properties are determined by the electron charge rearrangement between nuclei when atoms aggregate together to form bonds. Thus, directly visualizing electron charge distribution has a broad impact on science.

The local electric field and charge density, including contributions from both the electron cloud and nuclear cores, can be determined by electron diffraction imaging in a STEM with a small electron probe. While penetrating through a specimen, the electron beam interacts with the local electric field in its pathway, resulting in a change in the probe’s momentum.

From a diffraction pattern, this change in momentum can be measured and the electric field inside the sample can be deduced at

each probe position. By using a system to synchronize the scanning of the probe with a fast camera (Gatan STEMx™ combined with the OneView® camera), diffraction patterns can be collected at every scanning point to map the local electric field. The charge density can then be determined using Gauss’ law. In an aberration-corrected STEM (JEOL Grand ARM), the electron probe can reach sub-Å size, which means that electric field and electric charge density can be clearly resolved surrounding individual atoms and other similarly sized features, such as atomically sharp interfaces and defects.

As depicted on the cover, polarized materials such as ferroelectrics have an intrinsic electric field which interacts with the electron probe, shifting the probe intensity preferentially in one direction based on the polarization in the material. The group at UCI used this technique to measure how the electric field and charge density changed across a ferroelectric-insulator interface, where the high-resolution charge density information provided key insights into how charge is transferred across the interface and how polarization can appear inside the normally non-polar insulating material. As electron microscopy advances from imaging atoms to probing electrons, it will lead to new understanding and discovery in materials research.

REGISTRATION The journal is free of charge, worldwide, to users who pur chase, specify or approve microscopical, analytical or imaging equipment at their place of work, and marketing exe cutives who make advertising decisions. To register, go to http://www.microscopy-analysis.com/user/register. To amend your address details, go to http://www.microscopy-analysis.com/user/login. The registered address entered must be an organisational address.Subscription charges for non-qualifying readers: $110 per annum (UK) or $195 (Europe); all other countries $280 by airmail.NON-USA returns should be sent to Readerservice Wiley & Sons Ltd, 65341 Eltville, Germany Tel. Germany: 06123/9238-290 Tel. International: 0800/0961137 Fax. Germany: 06123/9238-244 Fax International: +49/6123/9238-244Microscopy and Analysis (ISSN No: 2043-0639 USPS NUMBER 010-289) is published Bi-monthly by Wiley, and distributed in the USA by Asendia USA, 701 Ashland Ave, Folcroft PA. Periodicals postage paid at Philadelphia, PA and additional mailing offices. POSTMASTER: send address changes to Microscopy and Analysis, 701 Ashland Ave, Folcroft PA 19032

Contact DetailsTelephone 1.925.463.0200Email info@gatan.comWeb gatan.com

Contents

4 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

16.8ns13.1ns

5.2ns

9.7ns

Using state-of-the-art TEM, UK- and Germany-based researchers have imaged bonding in molecules at the atomic scale.

Professor Andrei Khlobystov from the University of Nottingham, UK, Professor Ute Kaiser from the University of Ulm, and colleagues, used chromatic and spherical aberration-corrected Sub Angstrom Low-Voltage Electron Microscopy (SALVE), to image metal–metal bonding in dirhenium molecules.

Based on a FEI Titan 80-300 TEM, SALVE uses a custom spherical and chromatic aberration correction system to eliminate the chromatic aberrations at low voltages, providing high-resolution electron energy loss

spectroscopy and energy-filtered TEM.The instrument operates at

acceleration voltages between 20 kV and 80 kV, and at 40 kV the instrument’s resolution is 15 times the diffraction limiting electron wavelength.

“To our knowledge, this is the first time when bond evolution, breaking and formation was recorded on film at the atomic scale,” says Khlobystov. “We are now pushing the frontiers of molecule imaging beyond simple structural analysis, and towards understanding dynamics of individual molecules in real time.”

As described in Science Advances, Khlobystov and colleagues created dirhenium molecules with

unsupported Re–Re bonds using single-walled carbon nanotubes as a ‘nano-test tube’ while simultaneously imaging the structure and dynamics of the single atoms in real time.

“Nanotubes help us to catch atoms or molecules, and to position them exactly where we want,” explains Khlobystov. “In this case we trapped a pair of rhenium atoms bonded together to form Re2. Because rhenium has a high atomic number it is easier to see in TEM than lighter elements, allowing us to identify each metal atom as a dark dot.”

Research is published in Science Advances.

Researchers from the National University of Singapore have synthesised the world’s first one-atom-thick amorphous material.

Atomic resolution imaging revealed the complete absence of long-range periodicity in the monolayer amorphous carbon (MAC), which the researchers hope will settle decades of debate on how atoms are arranged in amorphous solids.

“With MAC, we have shown for the first time that fully amorphous ma-terials can be stable and free-stand-ing in single atomic layers,” says Professor Barbaros Özyilmaz, Head of the NUS Department of Materials

Science and Engineering. “Amor-phous materials are of great techno-logical importance, but surprisingly, they remain poorly understood from a basic science point of view.”

Using laser-assisted chemical vapour deposition, Özyilmaz and colleagues deposited a stable monolayer of amorphous carbon onto a substrate. They then used Raman and X-ray spectroscopy, and transmission electron microscopy to characterise the film.

In past studies, researchers had suggested that these materials could either have a fully-disordered, completely random structure or display nanometre-sized order of

tiny crystallites, surrounded by random disorder.

The newly synthesised MAC films show the latter arrangement with nanometre-sized patches of strained and distorted hexagonal carbon rings in a continuous random network.

As Özyilmaz highlights in Nature: “Extensive atomic-resolution TEM characterisation shows unambig-uously both the complete absence of long-range periodicity and... randomly oriented and strained crystallites comprising only six-member rings embedded in a Zachariasen continuous random network environment.”

TEM captures bonding in molecules at the atomic scale

World’s first monolayer amorphous film

DIRHENIUM molecules within carbon nanotubes

THE AMORPHOUS structure has a widely varying atom-to-atom distance unlike a material with a crystalline structure NUS

US-based researchers have combined extreme speed photography with phase contrast microscopy to image ultrafast phenomena in transparent objects at picosecond resolution.

So-called phase-sensitive compressed ultrafast photography (pCUP) takes an incredible 1 trillion frames per second.

pCUP pioneer, Professor Lihong Wang, Caltech, has used the system to capture a shockwave created by a laser striking water and laser light travelling through a crystal, in a single shot.

Wang and colleagues combined phase-sensitive dark-field imaging with compressed ultrafast pho-tography (CUP), based on the world’s fastest camera that they developed just over a year ago.

As Wang and colleagues explain in Science, CUP uses compressed sensing theory and the streak cam-era technology to achieve receive-only single-shot ultrafast imaging of up to 350 frames per event at 100 billion frames/s.

“Since CUP operates as a passive detector, it can be coupled to many optical imaging systems,” says Wang. “By combining CUP with dark-field micros-copy, we show that pCUP can image ultrafast phase signals... at an improved speed of 1 trillion frames/s.

Wang says the technology is still in its early stages of development but could ultimately have uses in many fields.

Ultrafast camera takes 1tn frames per second

ULTRAFAST PHOTOGRAPHY with phase contrast microscopy captures a shockwave propagating in slow motion Caltech

3.0ns

7.1ns

News

Intuitive

User-friendly Hitachi Tabletop SEM is further enhanced with a new electron gun design

From sample insertion to EDS analysis…

in only minutes!

BSE Image

15 kVAcc

20 kVAcc

SE Image CL Image STEM Image

Variety of detectors available

Improved imaging and analyses

Magni�cation: 10,000x Sample: Tin particles

Magni�cation: 3,000x Sample: Cupper crystal

Magni�cation: 60xSample: Diamond Ring

Magni�cation: 1.000xSample: Kidney

Next-Generation Tabletop Microscope Debuts!

A variety of images acquired using Hitachi’ s unique detectors

Up to 20 kV accelerating voltage allows for enhanced Imaging & EDS performance

Ease of operation and quick and effective high-quality imaging and analysis

Sn Lα1 Ti Kα1 Sn Lα1 Ti Kα1

1mm 1mm 1mm 1mm

Global/Asia+81 3 3504 7111 customercenter.ev@hitachi-hightech.comwww.hitachi-hightech.com/global/science/

Europe+49 2151 643 5310 eminfo@hht-eu.comwww.hitachi-hightech.com/eu/

Americas+1 800 253 3053 microscopy@hitachi-hta.comwww.hitachi-hightech.com/us/

6 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

UK-based Cambridge University researcher, Agavi Stavropoulou-Tatla, has unveiled a stunning image of a mini brain tumor with blood vessels, grown in the lab.

So-called tumoroids are miniature versions of tumors grown in laboratories to offer a robust platform for the development of personalised therapies.

Stavropoulou-Tatla is currently studying for her PhD, and as she says: “This biomimetic model could provide insight into the mechanisms that drive tumor new blood vessel formation and invasion along blood vessels, and serve as a personalised tool for targeted drug testing.”

Stavropoulou-Tatla captured the image using epi-fluorescence microscopy; a Zeiss Axio Observer Z1 inverted microscope with an ORCA-Flash4.0 camera and a 10x lens. The image is the result of the deconvolution of an image z-stack.

The image shows a growing tumouroid that can be used to study the interaction between brain tumor (glioblastoma) cells and blood vessel forming (endothelial) cells. Both types of cell types have been genetically modified to express green and red fluorescent protein respectively.

According to Stavropoulou-Tatla,

this type of brain tumor, glioblas-toma, is able to grow so quickly because it has the power to produce new blood vessels when needed.

“It is highly infiltrative, and this is partially because tumor cells use blood vessels as ‘highways’ for their

migration to different parts of the brain,” she says.

Stavropoulou-Tatla’s initial image won the microscopy prize in the Uni-versity of Cambridge, Department of Engineering ZEISS Photography Competition.

Using deep learning, US-based researchers have devised a method to accelerate neuroimaging by up to sixteen times.

The new imaging method, called point-scanning super-resolution (PSSR) imaging, can restore under-sampled SEM and laser scanning confocal microscopy images, overcoming the resolution, speed, and sensitivity trade-off issues associated with point-scanning image acquisition.

“Not only does this approach work, but our training model can be used right away,” says Uri Manor, Director of the Waitt Advanced Biophotonics Core Facility at the Salk Institute for Biological Studies in San Diego.

“It’s extremely fast and easy, and anyone who wants to use this tool will soon be able to log onto 3DEM.org, [the web-based research platform for 3D electron microscopy], and run their data through it,” he adds.

While scanning electron microscopy and laser scanning confocal microscopy are widely used in high resolution cellular and tissue imaging, resolution,

speed, sample preservation, and signal-to-noise ratio are difficult to optimise simultaneously.

With this in mind, Manor, Linjing Fang also from Salk, and colleagues, developed a deep learning-based super-sampling system that can be applied to under-sampled images, which they call point-scanning super-resolution (PSSR) imaging.

Manor first obtained images

from Professor Kristen Harris from the Department of Neuroscience at the University of Texas-Austin that had been acquired on both scanning electron and Airyscan laser scanning confocal microscopes.

Then, working with Jeremy Howard, founder of fast.ai, and Fred Monroe, from the Wicklow AI Medical Research Initiative, they ‘crappified’ the oversampled,

high signal-to-noise ratio images to generate low-resolution analogs, with the image pairs then being used as training data for PSSR models.

These models were then used to restore real-world under-sampled images, creating high-resolution images that were very similar to the ones that had been originally created with greater magnification.

Moreover, researchers were also able to find brain cell features in ‘de-crappified’ versions of the low-resolution images that couldn’t even be detected in the original data.

“Remarkably, our electron microscopy PSSR model could restore under-sampled images acquired with different optics, detectors, samples, or sample preparation methods in other labs,” highlights Manor in bioRxiv. “[The method] enabled previously unattainable 2 nm resolution images with our serial block face scanning electron microscope system.”

The PSSR model could also convert under-sampled confocal images to Airyscan-equivalent spatial resolution and signal-to-noise ratio with around 100x lower laser dose and 16x higher frame rates than the corresponding high-resolution images.

“To image the entire brain at full resolution could take over a hundred years,” says Manor. “With a 16 times increase in throughout, it perhaps becomes 10 years, which is much more practical.”

Grown in the lab: miniature brain tumor with blood vessels

Deep learning restores under-sampled images

A MODEL BRAIN tumor with blood vessels grown in the lab, inset: Agavi Stavropoulou-Tatla Agavi Stavropoulou-Tatla

UK-based electron microscopist, Dr Peter Harris, University of Reading, has recently published an article entitled Microscopy and literature, which takes a look at how microscopy has influenced

literary works.Publishing in Endeavour, Harris shows,

for example, that the publication of Robert Hooke’s Micrographia (1665) influenced Jonathan Swift’s book Gulliver’s Travels (1726).

As he points out, in the chapter ‘A Voyage to Brobdingnag’, Gulliver becomes a kind of ‘human microscope’, observing with his naked eye the spots,

pimples and freckles on the skins of the giant Brobdingnagians in minute detail.

Other examples of microscopy in literature are taken from works by Voltaire, George Eliot, H G Wells and D H Lawrence.

The inspiration for the article came from a conversation that Harris had with Professor Steven Matthews from the English Department at Reading.

While explaining that his job involves running the University’s electron microscopy facility, Matthews had said that the only reference to

microscopes in literature that he could think of was in Lawrence’s The Rainbow, where the heroine, Ursula Brangwen, experiences a revelation about the meaning of life while peering down a microscope.

“This set me wondering whether there were any other examples of microscopy in literature,” says Harris. “It turns out that there are many more than you might think.”

How microscopy has influenced literary legends

RESTORATION RESULTS of manually acquired SEM pairs. Salk Institute

News

TESCAN DynaTOM� Optimized dynamic CT for in situ research and

imaging of delicate samples

� Unique gantry-based design

� Continuous scanning under 10 seconds per rotation

� Software tools for 4D acquisition and reconstruction

� Dynamic screening for synchrotron beamtime

The world’s first dedicated

dynamic micro-CT for your

in situ experimental needs

Metal foam compression

5 mm

8 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Australia-based researchers have discovered new evidence of how hydrogen embrittlement in high-strength steel takes place, and importantly, how to prevent it.

While hydrogen atoms tend to diffuse and accumulate at dislocations and grain boundaries in a steel, cryogenic atom probe microscopy has shown exactly how niobium carbide clusters within its microstructure help to trap the hydrogen, reduce embrittlement and prevent steel fracture.

The latest results point the way to the design of embrittlement-resistant steels for stronger steel tanks and pipelines in hydrogen environments.

Hydrogen embrittlement of high-strength steel is a well-known obstacle for using these steels in sustainable energy production.

While the phenomenon is known to involve the interaction of hydrogen and defects, issues around precisely locating hydrogen atoms throughout the steel microstructure have hindered investigations.

As Dr Yi-Sheng Chen from the Australian Centre for Microscopy and Microanalysis at the University of Sydney writes in Science: “The combination of the low level of interaction between an electron and hydrogen, and the extremely fast diffusion of hydrogen in steels renders it extremely difficult to experimentally determine the location of hydrogen by electron microscopy.”

With this in mind, Chen and colleagues, turned to cryo-transfer atom probe tomography to pinpoint the location of hydrogen at microstructural features in a niobium-bearing low-carbon steels and also study its interaction with second phases such as carbides.

Developing a custom cryo-transfer process, they were able to study needle-shaped steel samples in a CAMECA Local Electrode Atom Probe (LEAP) 4000X Si, under cryogenic conditions. Results showed hydrogen pinned at incoherent interfaces between niobium carbides and the surrounding

steel, providing direct evidence that these boundaries can act as trapping sites.

“These findings are vital for designing embrittlement-resistant steel,” says Chen.

SEM and fluorescence microscopy and have shed new light onto the development of hair bundles, the intricately complex assemblies in the inner ear responsible for hearing.

New results from Dr Jocelyn Krey, at the Oregon Hearing Research Center and the Vollum Institute at Oregon Health & Science University, US, and colleagues, reveal that hearing develops in tandem with form and function. The latest studies point to how scientists might develop techniques to regenerate hair cells and reverse hearing loss.

Hair bundles are precisely arranged cellular structures deep within the spiral cavity of the inner ear. Together, the structures convert vibrational en-ergy into electrical signals in the brain that translate into the sensation of hearing. Once lost – whether by loud noise, toxins, disease or ageing - the bundles do not naturally regenerate in people and other mammals.

However, investigations on mice, by Krey and colleagues, revealed that the development of hair bundles occurs in a feedback loop, in which form follows function and function drives form. Image-scanning microscopy with Airyscan detection and SEM allowed the researchers to examine a stereocilia in a hair bundle.

Analyses showed that stereocilia widen simultaneously with the onset of mechanotransduction, the action of converting mechanical signals in the form of sound into electrical signals measured within the brain.

The stereocilia only elongated to their mature lengths after transduc-tion had been established, indicating that form and function are mutually reinforcing.

“We’ve been looking at these as separate pathways,” says Krey. “But in the course of this research, we observed the change in form occurs at the same time as the conversion of mechanical to electrical signals. So we’re seeing these happen together, and feeding each other in a way we hadn’t seen before.”

The researchers now believe that new techniques to reverse hearing loss should focus on the critical importance of early development.

Research is published in Current Biology.

Steel embrittlement mystery solved

SEM unravels how hearing develops

Surprising beauty found in bacterial culturesUsing fluorescence imaging, researchers at University of California San Diego, US, have discovered that when microbes pair up, stunning floral patterns can emerge.

When Lev Tsimring from the BioCircuits Institute at UCSD, and colleagues, combined non-motile Escherichia coli with motile Acinetobacter baylyi on an agar surface, they noticed that the E. coli ‘catch a wave’ at the front of the growing A. baylyi colony, with unexpected results.

“We were actually mixing these two bacterial species for another project, but one morning I found a mysterious flower-like pattern in a petri dish where a day earlier I placed a droplet of the mixture,” explains Dr Liyang Xiong from Physics at UC San Diego. “The beauty of the pattern struck me, and I began to wonder how bacterial cells could interact with each other to become artists.”

According to the researchers, while E.

coli can’t easily move across agar, A. baylyi can crawl using pili; combine the two and intricate flower-like patterns form over a 24 hour period.

Using an inverted, epifluorescence Nikon TI2 microscope with a Photometrics Cool-Snap cooled CCD camera, the researchers

were able to image the surprising floral formations.

To unravel how these patterns form, Xiong and colleagues developed mathematical models to account for the different physical properties of the two bacteria strains, includ-ing differences in growth rate, motility and effective friction against the agar surface.

Theoretical and computational analysis showed that the pattern formation originates at the expanding boundary of the colony, which becomes unstable due to drag exerted by the E. coli that accumulate there.

Where there is less E. coli accumulation, there is less friction, allowing the boundaries to push out faster. And in regions with more E. coli accumulation and friction, the bound-aries stagnate, creating the ‘petals’ of the flower pattern. The results have important implications in studying growing biofilms.

Research is published in eLife.

ELECTRON microscopy image shows the classic stair-stepping form of mature hair bundles in the inner ear of mice.

CONCENTRATION of hydrogen atoms (red balls) at the crystal boundaries and dislocations in steel University of Sydney

E. COLI and A. baylyi form intricate flower-like patterns when grown together over 24 hours BioCircuits Institute/UC San Diego

ATOM PROBE image showing accumulations of hydrogen (red) at carbon-rich (blue) dislocations in steel University of Sydney

News

9

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Researchers from The Netherlands and Germany have merged two super-resolution microscopy techniques to create a new method to image components within living cells with more precision than ever before.

By doubling the imaging preci-sion compared to standard localisa-tion microscopy, so-called SIMFLUX paves the way to new insights in healthcare and more.

While single-molecule localiza-tion microscopy circumvents the diffraction limit, improvement over state-of-the-art image resolutions of around 20 nm towards 5 nm

is necessary to truly image at the molecular scale.

With this in mind, Professor Sjoerd Stallinga from the Delft University of Technology and colleagues combined single mol-ecule localization microscopy with structured illumination microscopy in a single set-up, SIMFLUX.

Using the new method, they went onto image an artificially produced DNA structure, producing a clear picture of the proteins structures – between 5 to 10 nm in size – that form the cytoskeleton.

As the researchers write in Nature Methods: “We show a near two-fold

improvement in precision over standard localization [microscopy] with the same photon count on DNA-origami nanostructures and tubulin in cells.”

The researchers believe the new method is a major step forward and now hope to image samples in 3D.

“The logical next step would be to make images in three dimensions,” says Bernd Rieger from the Depart-ment of Imaging Physics at Delft. “That’s another big challenge, but we already have some ideas.”

Research is published in Nature Methods.

COMPARISON of Single Molecule Localization Microscopy and SIMFLUX cellular tubulin images TU Delft

Super-resolution method doubles imaging precision

Secret life of the ‘Plastisphere’ exposedUsing novel epifluorescence imaging developed at the Marine Biological Laboratory (MBL), Woods Hole, US, researchers have unravelled the structure of the microbial communities that coat microplastic trash in the world’s oceans.

To study the ‘Plastisphere’ – microscopic communities on plastic marine debris – Linda Amaral-Zettler from MBL and the Royal Netherlands Institute for Sea Research, and colleagues, used confocal laser-scanning microscopy alongside CLASI-FISH, combinatorial labelling and spectral imaging fluorescence in situ-hybridization.

The researchers developed a nested probe set consisting of three existing phylogenetic probes and four newly designed probes, based on fluorophore-labelled oligonu-cleotides, that could target known bacterial groups.

According to the researchers, this nested approach allowed them to identify the spatial organisation of mi-crobes on the plastic samples, differentiating seven major microbial groups simultaneously in one image.

“We now have a toolkit that enables us to understand the spatial structure of the Plastisphere and, combined with other methods, a better future way to understand the Plastisphere’s major microbial players, what they are doing, and their impact on the fate of plastic litter in the ocean,” highlights Amaral-Zettler.

During analyses, the researchers observed diatoms and bacteria colonising the microplastics, dominated in all cases by three phyla: Proteobacteria, Cyanobacteria, and Bacteriodetes.

Spatially, the Plastisphere microbial communities were

heterogeneously mixed, providing the first glimpse of bacterial interactions on marine microplastics.

The technique has previously been used to study microbial communities in the human mouth and in the digestive tract of cuttlefish and vertebrates.

This study customized and extended the technology, and Amaral-Zettler now intends to establish a CLASI-FISH microscopy platform in the Netherlands.

Research is published in Molecular Ecology Resources.

China-based researchers have developed a method that uses ghost imaging to achieve super-resolution microscopy at unprecedented speeds.

Based on stochastic optical reconstruction microscopy (STORM), the new method is poised to capture the details of processes within living cells at higher speeds than ever before.

As the researchers highlight in Optica, the unconventional imaging approach produces nanometre resolution with far fewer imaging frames than traditional super-resolution methods.

“Our imaging method can potentially probe dynamics occurring on millisecond time-scales in subcellular structures with spatial resolution of tens of nanometres – the spatial and temporal resolution at which biological processes take place,” says Zhongyang Wang from the Shanghai Advanced Research Institute, Chinese Academy of Sciences.

Single-molecule, localization-based, wide-field nanoscopy suffers from low time resolution as localisation of a single molecule with high precision requires a low emitter density of fluorophores.

What’s more, reconstructing a super-resolution image demands hundreds or thousands of image frames, even when advanced algorithms, such as compressive sensing and deep learning, are applied.

With this in mind, Wang and colleagues developed a single-frame, wide-field nanoscopy system based on ghost imaging, in which a spatial random phase modulator is applied in a wide-field microscope to achieve random measurement of fluorescence signals.

The random phase modulator converts the fluorescence from a sample into a random speckle pattern.

Coding the fluorescence in this way allows each pixel of a CMOS camera to collect light intensity from the whole object in a single frame.

As the researchers write in Optica, the method enhances the imaging resolution to 80 nm, with the raw image being reconstructed using compressive sensing.

“While STORM requires a low density of fluorescent labels and many image frames, our approach can create a high-resolution image using very few frames and a high density of fluorophores,” says Shensheng Han, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences. Using the method, the researchers imaged a 60 nm ring, resolving the object using only ten frames.

In contrast, STORM approaches needed up to 4000 frames to achieve the same result.

The researchers now hope to speed up the method and achieve video-rate imaging with a large field of view, to acquire 3D and colour images.

Ghost imaging speeds up super-resolution microscopy

THE NEW approach can resolve a 60 nm ring (inset) using just 10 image frames while traditional approaches need up to 4000 frames to achieve the same result. Zhongyang Wang, Chinese Academy of Sciences

SEM IMAGES of microbial communities colonising microplastics collected from the North Atlantic Ocean Cathleen Schlundt

News

10 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

When it comes to light sheet microscopy, biophysicist, Jan Huisken, has been there pretty much from the beginning. In 2004, while working on his PhD in the laboratory of confocal microscopy pioneer, Ernst H K Stelzer, he, Stelzer and colleagues shook the world of biology when they unveiled selective plane illumination microscopy.

While confocal microscopy had always relied on beam scanning to raster across a sample, SPIM drastically reduced the light directed at a sample by shining it in sheets, perpendicular to the detector. The end result was a gentle, high-speed method that could image animal and plant embryos for more than 100 hours, without damaging them.

“We’d been working hand-in-hand with biologists, and so realised how valuable this was for the community at the time,” says Huisken. “We were seeing a huge push from biologists towards imaging larger specimens, including tissues and whole embryos, at a time when many fluorescent proteins were also becoming available.”

“It’s sometimes difficult to describe how things happen, especially when you have a breakthrough development like this turned out to be,” he adds. “But we were right in the middle of these developments and it just felt good seeing the impact that imaging could have on biology.”

Huisken first became interested

in microscopy when he realised astronomy wasn’t what he thought it would be. On completing his degree in Physics at the Ruprecht Karls University in Heidelberg, Germany, he was drawn to the stars. But as he says: “I always thought you’d have your own telescope to experiment with, and then discovered that you were more likely to be sitting in front of a computer controlling a telescope in South America.”

At the time, Huisken was also fascinated by laser physics, so his attentions quickly turned towards Heidelberg’s world renowned European Molecular Biology Laboratory, where he joined Stelzer’s lab to work on optical trapping. EMBL was peppered with light and electron microscopes, but the young researcher was captivated by, as he says, ’the beautiful multi-colour images’ that only fluorescence light microscopy could deliver.

He was part of a project that aimed to hold a sample by optical forces while simultaneously imaging it from different views. As the project evolved, his interest in microscopy grew and he found himself working more and more on light microscopy, and less and less on optical trapping. Come the end of his PhD, SPIM had been developed.

“It was so interesting to see that physicists could make an impact in biology with this kind of instrument development,” says Huisken. “This was part research and part technology

Light revolutionary

Be it imaging the beating heart of a zebrafish or building a light sheet microscope that everyone can use, Professor Jan Huisken is revolutionising the world of bioimaging, reports Rebecca Pool.

and methods development, which is something that Stelzer did extremely well and also something that I learnt from him.”

“At EMBL you couldn’t do anything without it eventually being beneficial to biology, and this is something that I embraced and has become a theme in my later laboratories,” he adds.

With SPIM developed, Huisken realised he wanted to take light sheet microscopy further afield and joined the laboratory of biologist and developmental geneticist, Professor Didier Stainier, at the University of California, San Francisco. According to Huisken, Stainier was running one of the largest zebrafish labs in the world at the time, but biology’s favourite large, transparent organism wasn’t so easy to image with traditional light microscopy.

“I was the only physicist in this lab, and my goal was to bring [SPIM] to a new research field,” he says. “Didier always said, ’I don’t really understand very much of what you are doing, but I trust you’ and that was great – I had all this support and could try so many things that I wouldn’t have necessarily tried in a physics environment.”

Success ensued, and together, Huisken, Stainier and colleagues used the method to image the zebrafish heart and vascular development, unravelling how blood vessels and heart valves form and even using optogenetics to control the organism’s cardiac system.

FLAMINGO: the original version of the microscope reminds Huisken and colleagues of the bird standing on one leg. Michael Weber, Morgridge Institute for Research

Main im

age: David Nevala for the Morgridge Institute

Profile

11

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

“We were imaging the vascular system and heart to a level that nobody had ever seen before,” says Huisken. “We could see the beating heart of a living fish and that was really rewarding.”

Importantly, Huisken also had his eyes opened to the impact that light sheet microscopy could have on the field of developmental biology. Well aware that future research and instrument development would demand a multi-disciplinary team of physicists and biologists, he applied for Group Leader at Max Planck Institute of Molecular Cell Biology and Genetics in Dresden, Germany. Come 2010, he was setting up his first laboratory.

“The Institute had learnt about light sheet microscopy and realised the method was not going to be commercial available any time soon, so hired me to build these instruments and develop them further,” he says.

SPIM development and zebrafish research began apace, and within a few years Huisken and his team had truly pushed back the boundaries of light sheet microscopy. They had successfully confined living zebrafish embryos in tubes for time-lapse imaging over several days, applied the first commercially available tunable lenses for fast, 3D light sheet imaging and then unveiled astonishingly high resolution views of the beating zebrafish heart, in 3D.

“We had a biologist working on sample preparation and imaging, and optics researchers building the microscopes,” says Huisken. “But we were also strongly focused on software development, so we could register movie stacks and reconstruct the heart in multiple colours and in 3D over time, which of course was an enormous effort on the data-processing side.”

“I think this was a great example of how different disciplines can come together to make something happen that simply hadn’t been achievable at all with any other instrument,” he adds.

Indeed, by now, the wonderful world of imaging with light sheet microscopy was gathering momentum. While in the past, biologists typically had to adapt their sample preparation according to the imaging technique, SPIM provided more freedom.

“With our study on the beating zebrafish heart, things became more 3D and dynamic, and very inspirational for a lot of scientists,” points out Huisken. “Researchers were saying, ’Oh yeah, I can look at a freely swimming fish or a worm crawling over a coverslip, or maybe study muscle activity in a fly’.”

“Suddenly researchers could image such events for several days and I think that was a game-changer for a lot of people,” he adds.

Still, despite success, Huisken was only too aware of the cost and

Tackling the data problemWhile light-sheet microscopy is a powerful method for imaging the development of complex biological systems at high spatio-temporal resolution, across long time scales, such experiments can generate terabytes of image data. Given this, finding ways to manage, process and analyse the data is critical.

As Huisken puts it: “Light sheet microscopy is more like video microscopy where you are taking high resolution videos continuously over many hours, or even days - we really have to re-think how we handle microscopy data so we're not drowning in it.”

Huisken and colleagues are fortunate as they can stream data from different microscopes within the Morgridge Institute and store it on a central server. They can then analyse the data using a high-performance computer cluster.

However, the researchers, amongst other scientists, have also developed techniques to compress data and analyse it in real-time, thereby avoiding the accumulation of terabytes of raw data.

“This data issue has created a bit of fear in the community as researchers worry that light sheet microscopy could be difficult to use because of its extremely high data rates,” says Huisken. “We're trying to take this fear away and educate researchers on how to wisely acquire data and process it efficiently.”

A ZEBRAFISH undergoing gastrulation in a light sheet microscope. Cell nuclei are labeled and color-coded depending on depth in the embryo Gopi Shah, Huisken Lab, MPI Dresden

Cleared adult zebrafish Kurt Weiss, Morgridge Institute

The development of the zebrafish vasculature Stephan Daetwyler, Huisken Lab, MPI Dresden

Profile

12 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Meet FlamingoNamed Flamingo, as the original version of the microscope reminds Huisken and colleagues of the bird standing on one leg, the modular SPIM can be constructed in either the classic or inverted geometry for high throughput imaging of conventionally mounted samples. Importantly, it is designed around a central post – rather than being spread out across an optical table – and several instruments can be built for the same price as a single commercial system.

The microscope uses water-dipping optics for illumination and detection, and can image live samples such as fruit fly embryos, zebrafish embryos and larvae, organoids, spheroids and tissue sections. Light sheet uniformity is enhanced via rapid pivoting of the light sheet. And rapid volumetric multi-colour imaging can be achieved via a fast, low-noise camera, multiple laser lines and an emission filter wheel.

Spatial resolution is given by the chosen lens and the pixel size of the camera and the system frame rate and number of planes per second is typically 40 to 400 (depending on the number of pixels). The instruments currently support samples up to some 5 to 6mm in thickness, and depending on sample transparency, tissues as deep as 500 µm can be imaged. (see: involv3d.org/flamingo)

sophistication of SPIM and wanted to develop a light sheet microscope that researchers could build themselves, making the method more accessible for more applications. So while working on zebrafish heart imaging, Huisken and colleagues also developed an open access light sheet microscopy platform called OpenSPIM.

The software and hardware platform was designed so a researcher could construct a customisable microscope for SPIM, and was shared with the scientific community via a public website, http://openspim.org. The initial version implemented a single-sided illumination and single-sided detection light sheet arrangement, was controlled using open source software and enabled on-the-fly image processing, and reconstruction and analysis.

At the time, Huisken and colleagues figured that such a system could be used by many more researchers for the parallel imaging of gene activity in large developing specimens, and so speed up SPIM throughput. And as Huisken says: “This was the very early days of light sheet microscopy, and the platform enabled a lot of people, especially physicists, to quickly build more microscopes which led to more widespread use of the technique.”

Yet, in hindsight, Huisken believes the design was, ultimately, too complex for many biologists that were less familiar with microscope development. “We had a lot of interest from the physics community but at the same time, we never really reached the community that we really wanted to reach – the biologists,” he says.

And so the idea for ’Flamingo’ was born.

A NEW APPROACHBy now, Huisken was ready for another move. While the Max Planck Institute offered an amazing environment for the young group leader, a lack of academic tenure meant the biophysicist couldn’t see a clear career path. However, he was intrigued by the new Morgridge Institute for Research at the University of Wisconsin-Madison, US.

“This looked like it had the perfect environment for what we wanted to do,” he says. “It had a fantastic machine shop and prototyping facility, which I knew would allow us to rapidly produce microscope components, and it also had a great IT infrastructure giving us a place to stream and process the data from light sheet microscopy.”

“These were exactly the two things we needed the most to further light sheet microscopy development,” he adds.

So come 2016, Huisken headed back to the US, joined the Morgridge Institute and set to work developing a new kind of light sheet microscope. The researcher had the idea to develop a standard hardware and software framework that would allow him and colleagues to develop, as he says, ’different flavours’

of light sheet microscope with different configurations for different applications and samples relatively quickly.

“We wanted this platform in the lab to quickly build these different configurations but we also realised that we should build something small that could be taken to the biologist,” he says. “So rather than inviting

biologists to use a single, fancy instrument, we decided to build many instruments and share them with the researchers.”

Which is exactly what Huisken and colleagues have done. The ’Flamingo

microscope’ has been designed as a suitcase-sized, portable SPIM that can be passed between labs. Once shipped to a lab, the instrument can be remotely configured by Huisken or a fellow researcher ready for use, and then shipped onto to the next project (see ’Meet Flamingo’).

According to Huisken, prototype development is complete, and at the time of writing, three instruments had been built with two more underway. “We’ve shown the instrument at several conferences to see if it is just a crazy idea or if biologists actually want it, and the response has been overwhelming,” says Huisken. “It’s been amazing to see how many people are excited about the idea of getting a custom-built SPIM from us, and ultimately these instruments could be used in biological experiments, where commercial instruments don’t meet requirements.”

So what next for Huisken? Alongside colleagues, he now intends to develop new configurations of Flamingo in his lab. And using the same software across each system, he reckons they will be able to swiftly develop new modifications and add-ons for the instruments.

“We are aiming to democratize high-end light microscopy and bring it to labs for free,” he says. “I find this idea of iteratively improving the instruments with biologists very exciting.”

“I think that by going out and collaborating with biologists we are going to have much more of an impact than if we tweaked a single instrument that only exists in my own lab and isn’t accessible to that many researchers,” he adds.

TOP Light sheet microscopy setup in the Huisken lab at the Morgridge Institute for Research, Madison Jan Huisken, Morgridge Institute

CENTRE The Flamingo microscope design and development team (from left): Todd Bakken, Joe Li, Jan Huisken, Rory Power and Susi Power. Courtni Kopietz

ABOVE Central nervous system and vasculature in the brain of a live zebrafish. Sample by Armin Bahl, Harvard University.

Profile

13

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

INTRODUCTION Viruses are intracellular parasites that depend on the host cell machinery to replicate and assemble progeny virus particles. In order to enter cells, viruses must bind to specific receptor proteins on the cell surface and trigger uptake via mechanisms such as endocytosis. Whilst enveloped viruses can enter cells by fusing their envelope with host cell membranes (either at the plasma membrane or in endosomes), the mechanisms by which non-enveloped viruses escape the endosome are poorly understood. Viruses must overcome this barrier to infection in order to deliver their genomes into the cytoplasm of host cells to then subvert the cellular machinery to their own end i.e. to replicate their genomes, produce their proteins and assemble progeny virus particles.

Caliciviruses are a family of viruses that infect many different species including cats, dogs, rabbits, sealions and humans. The most common calicivirus that infects humans is norovirus. Norovirus causes ‘winter vomiting disease’ resulting in diarrhoea and vomiting and is reasonably common throughout the world. Although norovirus infection is trivial to many in developed countries, it causes 70-200,000 childhood deaths every year, most of which are in developing countries[1]. Other caliciviruses also infect humans, such as sapovirus, and some even affect the production of pharmaceuticals. For example, vesivirus 2117 has been found as a contaminant of bioreactors used to produce therapeutics for the treatment of Gaucher disease and Fabry disease (Cerezyme and Fabrazyme, respectively)[2, 3]. Mammalian caliciviruses are also important veterinary pathogens in their own right. For example, feline calicivirus (FCV), is a widespread pathogen of domestic cats, found particularly in animal shelters and necessitating vaccination every one-three years[4]. Highly pathogenic strains

of FCV have also emerged which are termed ‘virulent systemic’ as they cause a much more severe infection than the usual cold-like symptoms and oral ulceration seen with non-virulent isolates[5-10].

Until recently, norovirus was unable to be grown in the laboratory and so some animal caliciviruses have been used as a model to study calicivirus biology[11, 12]. FCV is one of the few caliciviruses for which a protein receptor has been identified and is arguably one of the most well characterised model systems for the study of caliciviruses, particularly one of the vaccine strains known as F9. The receptor for FCV is feline junctional adhesion molecule-A (fJAM-A) and is found at tight junctions between cells as well as on the surface of platelets and some cells of the blood[13, 14]. The JAM-A ectodomain (the portion of the protein which is exposed on the cell surface) is the site of virus binding for both FCV and reovirus[15-17].

Calicivirus particles/capsids are 35-40nm in diameter and have t=3 icosahedral symmetry. The capsids are composed of 180 copies of the major capsid protein VP1. VP1 can be sub-divided into 2 domains: S (shell) domain which forms the contiguous shell of the capsid and the P domains which form the spikes which extend outwards from the capsid shell and include the binding site for fJAM-A and neutralising antibodies[16, 18-20]. VP1 capsomeres are present in three conformations which differ slightly to allow the formation of an icosahedral shell (namely A, B and C). A/B capsomeres/dimers are found at the 5-fold symmetry axes and C/C dimers are found at the 2-fold symmetry axes, resulting in the alternation of A/B and C/C dimers around the 3-fold symmetry axes. Caliciviruses also encode a minor capsid protein, VP2, although until this study, the structure and function of the protein remained unknown despite VP2 having been shown to be essential for productive

infection.Here, we describe the study of FCV

both unbound and decorated with the ectodomain of fJAM-A in terms of structure and the entry pathway utilised by the virus to initiate infection of host cells. We show that upon receptor engagement, the virus undergoes a conformational change allowing the extrusion of a portal-like assembly at a unique three-fold symmetry axis. We hypothesise that this ‘tunnel’ functions as a mechanism of genome delivery and endosomal escape to allow infection of the host cell to progress.

MATERIALS AND METHODSViruses were grown in Crandell reese Feline Kidney (CrFK) cells and purified using several stages of ultracentrifugation. Briefly, samples of infected cells were spun at 74,000 xg to pellet the virus, the virus was layered onto a caesium chloride gradient and spun at 134,000 xg, the virus was then extracted from the tube and pelleted a final time before use in structural and in vitro studies[11]. The soluble ectodomain of fJAM-A was over-expressed in Chinese Hamster Ovary (CHO) cells and purified using an Fc tag via affinity chromatography.

For cryo-electron microscopy, virus alone or equal concentrations of virus and fJAM-A were mixed and added to a C-flat advanced holey carbon grid (ProtoChips) prior to plunging into liquid nitrogen cooled liquid ethane and stored under cryogenic conditions until inspection. Grids were imaged in a Titan Krios electron microscope (ThermoFisher Scientific) equipped with a Falcon III direct detection device (Gatan) at the Astbury Biostructure Laboratory (University of Leeds, UK). Micrographs were collected at a range of defocus values from 1.2-3.5µm with a pixel size of 1.065Å per pixel. Micrographs were processed in Relion for motion correction, CTF estimation, particle picking and extraction, 2D classification, 3D classification, 3D

Caliciviruses: A tunnel to initiate infection

Dr Michaela J ConleyMRC-University of Glasgow Centre for Virus Research, Garscube Campus, 464 Bearsden Road, Glasgow G61 1QH

BIOGRAPHYI am currently a Research Associate at the MRC-University of Glasgow Centre for Virus Research in the structural virology group of Professor David Bhella. My background is in virus entry (including structural, biophysical and molecular techniques). I received my PhD in structural virology from the University of Glasgow in 2018 as well as an MSc (Res) in virology and BSc (Hons) in Microbiology with Virology from the University of Leeds.

ABSTRACTCaliciviruses include the notorious cause of winter vomiting disease, norovirus. Until recently, the mechanism of endosomal escape, required by some families of viruses to infect cells, was not understood for caliciviruses. Cryo-electron microscopy, three-dimensional image reconstruction and focussed classification were used to determine the structure of feline calicivirus alone and bound to its cellular receptor, fJAM-A. A novel portal/tunnel-like assembly was shown to form at a unique three-fold symmetry axis in the viral capsid which is hypothesised to function as a mechanism of endosomal escape and genome delivery, allowing infection to advance.

ACKNOWLEDGEMENTSThanks to Prof David Bhella and Dr Marion McElwee (MRC-University of Glasgow Centre for Virus Research). Funding: PhD studentship from BBSRC WestBIO DTP (BB/J013854/1) and UK Medical Research Council (MC_UU_12014/7).

CORRESPONDING AUTHORMRC-University of Glasgow Centre for Virus ResearchGarscube CampusSir Michael Stoker Building464 Bearsden RoadGlasgow, G61 1QH.Michaela.Conley@glasgow.ac.ukTwitter: @michaelajconley

Microscopy and Analysis 34(1): 13-15 (EU), February 2020

Calicivirus portal-like assembly

14 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

refinement, post processing and local resolution estimation[21]. Maps were visualised in UCSF Chimera[22]. Focussed classification was used to further analyse the data set whilst not imposing any symmetry (previously icosahedral) in Relion[11, 23-25]. Model building was performed using Coot and Phenix from the CCP-EM suite[26-28].

RNA release assays were performed where a constant amount of virus was mixed with fJAM-A of varying concentrations at various pH values alongside a nucleic acid binding dye (Syto9; ThermoFisher Scientific) and fluorescence measured every five minutes for four hours using a plate reader (BMG Labtech) [11].

Negative staining of samples was performed by washing the grids in deionised water following incubation of the sample on the grid for two

minutes. The grids were then washed and incubated in 2% uranyl acetate for two minutes prior to visualisation in a JEOL 1200 EXII transmission electron microscope equipped with a Gatan Orius charge coupled device camera.

RESULTS / DISCUSSIONFeline calicivirus particles were purified and combined with an equal amount of fJAM-A soluble ectodomain at 4°C. Either FCV only or FCV decorated with the soluble ectodomain of fJAM-A were then imaged in a Titan Krios and the resulting micrographs processed as described above. The three-dimensional structure of FCV (see Figure 1A) shows the characteristic cup shaped depressions that can be seen on the surface of all caliciviruses, that are formed by the VP1 P domains[2, 11, 18-20, 29, 30]. Upon

receptor engagement, additional density is visible above each of the P domains of VP1 which corresponds to fJAM-A (see Figure 1B). We have shown that the soluble ectodomain of fJAM-A forms a dimeric complex in solution and, unsurprisingly, our cryoEM data show that two fJAM-A proteins appear to bind to two VP1 capsid proteins although, surprisingly, the ‘usual’ dimeric interface between these two fJAM-A proteins is disrupted/altered upon virus binding. Despite the FCV binding site occurring away from the fJAM-A cis-dimerisation site, fJAM-A proteins appear to bind to VP1 proteins in a head-to-tail arrangement rather than their native arrangement (both in solution and their predicted conformation at tight junctions between cells)[11]. Upon fJAM-A binding to the FCV capsid, the VP1

capsomeres undergo conformational changes including a 15° anti-clockwise rotation of the A/B dimers and the tilting of the C/C dimers away from the two-fold symmetry axes. These conformational changes result in the breakage of icosahedral symmetry and thus a smearing of the density in these regions (see Figure 1D and F).

Due to this loss of icosahedral symmetry in the virus, a relatively novel method of image analysis was utilised known as focussed classification[11, 23-25]. Briefly, focussed classification allows the 3D classification and reconstruction of specified regions of the assembly (rather than the whole complex). Due to the loss of symmetry at the 2-fold symmetry axes in fJAM-A decorated FCV, we focussed the classification on the C/C VP1 dimers (2-fold symmetry axes). Upon inspection of the different classes, one class contained additional density visible at the edge of the area corresponding to the 3-fold symmetry axis which was previously unseen. Subsequent classification was focussed on the 3-fold symmetry axes of the decorated virus and revealed a novel virus feature in a subset corresponding

FIGURE 1 Feline calicivirus undergoes a conformational change upon receptor engagement. Structures of feline calicivirus undecorated (a, c, e) and decorated with the soluble ectodomain of fJAM-A (b, d, f) are shown. Three dimensional maps of the two virus structures are presented (a and b) coloured by radius alongside central sections of the reconstructions (c and d) showing the additional blurred density on each of the capsid spikes (d). Spherical sections for the two structures are also shown (e and f) to highlight to rotation of the spikes (pink boxes) upon fJAM-A binding (f).

A

C

E

B

D

F

Calicivirus portal-like assembly

15

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

to 1 per virus particle.Following focussed classification

of the 3-fold symmetry axes of FCV decorated with fJAM-A, a unique portal-like structure was visible (see Figure 2). This portal-like assembly was only present at a unique 3-fold symmetry axis per virus particle i.e. 1 ‘tunnel’ per virus. The portal-like assembly measured 13nm in length, 7nm in width at the base and 5nm in width at the distal tip[11]. Upon closer inspection, the extrusion of the portal-like assembly is seen to be accompanied by the formation of a small pore in the capsid shell at the centre of the ‘tunnel’ (see Figure 2B) which measures 1nm in diameter[11]. Following focussed classification, all of the virus-receptor complex was of sufficient resolution to permit building of an atomic model into the EM density.

The portal-like assembly is formed of 12 copies of the minor capsid protein, VP2, of which the structure and function was previously undetermined. VP2 is present in 2 different alternating conformers within the portal-like assembly (see Figure 2C and D). The C-terminal helix of one conformer of VP2 contacts neighbouring VP1 proteins via electrostatic interactions (orange in Figure 2C and D) whilst the same C-terminal helix from the other conformer folds into the lumen of the portal-like assembly (red in Figure 2C and D). Kyte-doolittle analysis of the VP2 protein suggests that the N-terminal distal tips of the assembly are highly hydrophobic[11]. Due to the formation of the ‘tunnel’ only upon receptor engagement alongside the formation of a small pore in the capsid shell, we hypothesise that the portal-like assembly functions as a mechanism of genome delivery, escorting the viral RNA across the endosomal membrane into the cytoplasm of the host cell for viral infection and replication to then ensue.

To investigate the entry process further, a constant amount of FCV was incubated with varying amounts of fJAM-A in the presence of a nucleic acid dye which fluoresces upon binding to RNA (or DNA). As endosomal acidification is necessary for FCV entry into host cells, this assay was performed at a range of pH values from 3 to 9 and showed the release of viral RNA only in the low pH conditions (pH 3 or 4) and only when a certain amount of receptor/fJAM-A was present (data not shown)[11, 31]. To further investigate the amount of fJAM-A required to cause the conformational changes in FCV (and therefore release of the viral RNA at low pH), we performed negative staining electron microscopy of FCV with varying ratios of fJAM-A at neutral and low pH. As shown in Figure 3, no intact FCV particles were visible upon incubation with an equal amount of fJAM-A at low pH, however, intact particles are present at neutral pH (under conditions utilised for cryo-EM described above). Upon incubation of FCV with a 1:11 ratio of fJAM-A:VP1, a limited number of

broken and irregular virus particles were visible. However, when the virus was incubated with a 1:12 ratio, mostly intact virus particles were seen (see Figure 3), suggesting that a ratio of 1:11 fJAM-A:VP1 capsid proteins is the minimum required to cause viral disassembly at low pH (and by inference, cause the conformational changes in the capsid; see Figure 1)[11]. As each virus particle is formed of 180 VP1 proteins, a rough estimate of receptor/fJAM-A engagement required for the observed conformational changes can be calculated as 16 receptor molecules per virus particle.

As FCV enters host cells by clathrin mediated endocytosis, it is likely that the virus binds to additional fJAM-A proteins during endocytic uptake following initial binding at the cell surface to sialic acid and fJAM-A[13,

31, 32]. This engagement of additional receptors triggers the conformational change in the capsid (see Figure 1) and formation of the portal-like assembly (see Figure 2) which, we hypothesise, serves as a mechanism of genome delivery across the endosomal membrane upon acidification. Once the viral RNA has been delivered into the host cell cytoplasm, viral replication and host subversion can then begin.

This was the first description of a portal or tunnel-like structure in a mammalian RNA virus that functions as a mechanism of endosomal escape[11]. Other RNA viruses, such as picornaviruses, have been shown to escape the endosome by extruding proteins and inserting peptides into the endosomal membrane to form a pore[33]. Enveloped viruses (such as influenza virus) encode a fusion protein in the lipid bilayer which surround the capsid to mediate fusion of the viral and host membranes (either at the plasma membrane or within the endosome)[34, 35]. Large DNA viruses also encode portal structures, such as herpes viruses for genome delivery into the nucleus, however this is the first description of such a structure in a small RNA containing mammalian virus[11, 25, 36].

SUMMARY AND CONCLUSIONSUpon engagement with its cellular receptor, fJAM-A, feline calicivirus undergoes a conformational change resulting in the rotation and tilting of the virus spikes. This tilting causes breakage of the viral icosahedral symmetry so focussed classification was employed to further characterise these changes. This asymmetric analysis resulted in the discovery of a novel portal or tunnel-like assembly formed by the virus which likely inserts into the endosomal membrane to mediate genome delivery and allow infection to proceed.

Article and references available online at: Microscopy-analysis.com/article/february_20/conley ©John Wiley & Sons Ltd, 2020

FIGURE 2 A portal-like assembly in feline calicivirus is formed upon fJAM-A engagement. The portal-like assembly structure is shown in relation to the whole virus particle; visualised down a two-fold symmetry axis (a) and the unique three-fold symmetry axis where the portal-like assembly forms (b). Modelled structures of the VP2 minor capsid protein are shown; two VP2 proteins in the two different conformations (c) as well as the full portal-like assembly (d).

FIGURE 3 Negative staining of feline calicivirus with comparable amounts of fJAM-A reveals viral disassembly under low pH conditions. Samples of feline calicivirus alone or incubated with varying ratios of fJAM-A:VP1 capsid proteins are shown under both low pH (pH3) and neutral (pH7) conditions. Scale bars of 200nm are depicted.

A

C

B

D

Calicivirus portal-like assembly

16 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

INTRODUCTIONFluorescence microcopy is one of the most widely used tools for investigating the dynamics of biological systems. Such dynamics, for example the propagation of neural signals in the brain, often plays out in all three spatial dimensions. This is challenging for microscopy which typically only monitors a single two-dimensional focal plane at a time. To address this problem, we developed a fluorescence tomography approach that allows recording volume information in a single frame scan from sparsely labeled samples.

Optical microscopes typically record from a single focal plane at a time. This can be limiting for samples that show fast dynamics across multiple focal planes, which is often the case for example when monitoring neural activity in the brain at cellular

resolution. In this situation, neurons are arranged in three dimensions and the dynamics of these cell ensembles evolves over a volume. To monitor such dynamics it can be advantageous to sacrifice some spatial resolution for temporal resolution. We describe in the following one such approach that can reconstruct volumetric information from images recorded in a single frame scan with multiple laser beams.

TOMOGRAPHIC IMAGINGThe developed approach[1] relies on tomography [2], which reconstructs volume information from multiple two-dimensional projections. Tomographic imaging in optical microscopes can be achieved using scattered light or using fluorescence [3, 4]. In the latter case the sample is typically rotated with respect to the imaging path to obtain multiple projections. However, this method is

not compatible with many imaging situations where the sample can only be accessed from one side with a single microscope objective and is not transparent, as is typically the case for in vivo imaging in neuroscience.

TEMPORALLY MULTIPLEXED BESSEL BEAMS For generating projections under these conditions, one can take advantage of Bessel beams. Instead of a focal spot, these beams generate a focal line that is extended along the optical axis while maintaining a lateral resolution that is similar to a focused Gaussian beam; Bessel beams have for example been applied for in vivo imaging of neural activity of sparsely labeled samples [5]. By introducing a lateral offset when passing through the microscope objective these Bessel beams can be tilted by an angle θ with

Fast volume imaging with bessel beam tomography and machine learning

Andres Flores Valle1 and Johannes D. Seelig1

Abteilung Neuronale Schaltkriese, Stiftung Caesar, Ludwig-Erhard-Allee 2, D-53175 Bonn

BIOGRAPHYJohannes Seelig studied physics at University of Basel with a minor in molecular biology from 1997-2002. He obtained his PhD in the field of single molecule microscopy from ETH Zurich in 2006. He was a postdoctoral associate from 2007-11 and later a research specialist at Janelia Research Campus, between 2011-2015, where he worked on neural circuits in Drosophila. Since 2016 he is a “free-floating” Max Planck Research Group Leader at Research Center Caesar.

ABSTRACTNeural activity in the brain propagates at high rates in three spatial dimensions. This is challenging for microscopy which typically only monitors a single two-dimensional focal plane at a time. To address this problem, we developed a fluorescence tomography approach that records volume information at high frame rates.

ACKNOWLEDGEMENTSThis work was supported by the Max Planck Society and the Research Center Caesar.

CORRESPONDING AUTHORDr. Johannes SeeligMax Planck Research Group LeaderResearch Center CaesarBonn, Germanyjohannes.seelig@caesar.de

FIGURE 1 a) Side projections allow undistorted 3D reconstruction. b) Projections under an oblique angle lead to distorted 3D reconstructions. c) Steps in the volumetric reconstruction process.

Microscopy and Analysis 34(1): 16-17 (EU), February 2020

A

B

C

Machine learning aided bessel beam tomography

17

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

respect to the optical axis; scanning across the sample then results in a projection along this tilted beam axis. For volumetric reconstruction, multiple such projections need to be recorded from different angles and to additionally cope with fast sample dynamics, these projections have to be recorded at the same time. This can be achieved using temporal multiplexing where, using pulsed fluorescence excitation together with time-gated detection, signals generated by different beams can be sorted into different channels [6].

BESSEL BEAM TOMOGRAPHYAs described in more detail below, images were reconstructed in the following steps: First, four 2D projections from a volumetric sample were recorded by scanning four temporally multiplexed Bessel beams, each with a different projection angle; then, similar to computed tomography, images were reconstructed using so called back projection, which computationally inverts the projection signal generation process. Compared to reconstruction using side projections (fig. 1 a), which would be best suited for tomography, the oblique projection angles lead to distortions of the reconstructed volume (fig. 1 b). Then, again taking advantage of image reconstruction approaches developed for computed tomography, these distortions were finally corrected using machine learning, by applying a suitable convolutional neural network. These steps are summarized in figure 1 c.

TOMOGRAPHIC RECONSTRUCTIONBessel beam tomography first requires generating four beams with different orientations. Here, the elevation angle was kept constant for all beams (about 13 degrees, limited by the numerical aperture of the objective), and the azimuth was varied in steps of 90 degrees. In a calibration step the exact orientation of the Bessel beams across

the field of view was mapped using fluorescent beads. The fluorescence signal resulting from scanning the Bessel beams across any sample can then be simulated using these calibration measurements similar to computed tomography using so called Radon transforms.

For the volumetric reconstruction of arbitrary samples this signal generation process is computationally inverted and the calibration information is used to calculate the so called back projection of each projected image. The back projection computationally inverts the signal generation process by projecting each image along the corresponding

Bessel beam back through the sample. The intersection of the four back projections then confines the shape and intensity of the volumetric sample. In figure 2 we show how applying back projection using four projections of a volumetric sample (fig. 2a) of 1 um diameter fluorescent beads embedded in agar leads to volume reconstruction (fig. 2b). The distortion in the z-axis is due to the limited projection angle which is constrained by the numerical aperture of the microscope objective.

IMPROVED RECONSTRUCTION USING MACHINE LEARNINGThe volume reconstructions obtained from these oblique projections leads

to a distortion along the z-axis with a factor of 1/sin(θ), as shown in figure 1b. To correct these distortions we used machine learning where a convolutional neural network (U-net) was trained using simulated data. Using a large set of simulated ground truth data as well as the corresponding simulated projection images such a network can be trained to recover the volume at high resolution based on the projection images. Figure 3 shows the reconstruction of four projections experimentally recorded from a sample of pollen grains (fig. 3a). Applying back projection to the four images leads to a distorted volume, as shown in the first row of figure 3b. Applying the trained U-net to these back projections can partially restore the information in the z-axis (second row in figure 3b) as seen in the comparison with the ground truth recorded using two-photon microscopy with a Gaussian beam in the third row of figure 3b.

CONCLUSIONWe combined scanning microscopy with tilted Bessel beams, temporal multiplexing and tomographic reconstruction aided by machine learning to reconstruct volumetric information recorded in a single scan across a three-dimensional sample. This allowed recording volume information in a single frame scan. In future work optimizing the tilted beams, for example using objectives with higher numerical aperture, as well as training neural networks with larger training data sets specific for the sample of interest will allow improving the spatial and temporal resolution of this imaging approach.

Article and references available online at: Microscopy-analysis.com/article/february_20/seelig ©John Wiley & Sons Ltd, 2020

FIGURE 2 Reconstruction of beads. a) Four recorded projections with each Bessel beam from a volumetric sample composed of 1 µm diameter beads. b) Reconstruction of the sample using back projections compared to ground truth.

FIGURE 3 Reconstruction of pollen grains. a) Four projections recorded with Bessel beams from a sample of pollen grains. b) First row: reconstruction of the sample using only back projection. Second row: reconstruction of the sample after applying a U-net to the back projections to correct distortions. Third row: ground truth recorded with a two-photon microscope with a Gaussian focus.

A

B

A

B

Machine learning aided bessel beam tomography

18 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Late last year, a team of chemists from Cornell University, US, unveiled a super-resolution method to image non-fluorescent reactions at nanometre resolution.

Based on widefield fluorescence microscopy, so-called COMPEITS - COMPetition Enabled Imaging Technique with Super-resolution - relies on competition between fluorescent and non-fluorescent reactions, as part of a single-molecule fluorescent-detection scheme. Crucially, the new method can be used to image a host of non-fluorescent reactions, critical to water decontamination, industrial chemical separation processes and more.

“This method turned out to be actually quite simple to implement,” highlights Professor Peng Chen from the Department of Chemistry and Chemical Biology, Cornell. “[It] can be broadly applied to image various classes of non-fluorescent systems, such as unlabelled proteins, neurotransmitters and chemical warfare agents... and really extends reaction imaging to an almost unlimited number of reactions.”

“We expect it to have profound impacts on many fields including energy science, cell biology, neuroscience and nanotechnology,” he adds.

Without a doubt, super resolution fluorescence microscopy methods, such as stimulation emission depletion microscopy and photoactivated localisation microscopy, have revolutionised biology and other research fields. However, these techniques can only interrogate particles and processes that fluoresce or are fluorescently labelled, which doesn’t help the lion-share of biological and chemical reactions.

Given this, Chen’s group set out to develop a super resolution method that can interrogate non-fluorescent processes. As Chen’s colleague, Dr Xianwen Mao, tells Microscopy and Analysis: “The inability of original super-resolution techniques to study non-fluorescent reactions has been a primary limitation for quite some time as the reactants and products

in so many reactions are completely non-fluorescent.”

“So we wanted to tackle this long-standing challenge and develop a super-resolution technique with much broader applicability,” he adds.

Chen’s research group has spent many years pioneering the application of single-molecule fluorescence imaging to observe chemical reactions in real time. As part of this, the researchers have been looking at single molecule catalysis alongside single molecule biophysical chemistry.

According to Mao, he and colleagues typically use single molecule localisation microscopy and magnetic tweezers to study reactions including single-nanoparticle catalysis, reactivity within single nanocatalysts as well as photoelectrocatalysis, plasmonic catalysis and single polymer growth dynamics. And what’s more, the researchers have developed new super-resolution methods to, say, track single molecule processes, such as DNA kinetics in living cells, and map photoelectrocatalytic reactions, including water-splitting, to sub-particle resolution.

So this time around, the researchers decided to harness their expertise on single-molecule reactions, and

A view of the dark side

‘COMPEITS’ can image non-fluorescent reactions at nanometre resolution – meet the researchers pioneering single-molecule fluorescence microscopy, reports Rebecca Pool.

developed COMPEITS. The new method uses an inverted Olympus IX71 fluorescence microscope with a photoelectrocatalytic microfluidic cell as well as two-laser epifluorescence illumination to drive and image a reaction. Within the flow cell, a catalyst particle is used to catalyse two reactions - the all-important non-fluorescent reaction and its auxiliary fluorogenic partner reaction.

As the fluorogenic reaction proceeds, its product molecules fluoresce and can be imaged at nanometre resolution. However, as the non-fluorescent reaction also gets underway, it competes for the same surface sites on the catalyst particle, suppressing the rate of the fluorogenic reaction. According to the researchers, the extent of this suppression can be imaged using the widefield fluorescence microscope set-up to nanometre resolution, delivering super-resolution spatial information on the non-fluorescent reaction.

“We introduce the competing agents into the photoelectrochemical microfluidic cell [with the photocatalyst] so we can then visualise the reactions,” says Mao. “We can, for example, apply potential to modulate the electric field and change the photoelectrocatalytic efficiencies, and we can also use one of the lasers to generate the charge carriers in the photocatalyst and drive the reactions.”

“This took around two years to develop, and this imaging capability is realised by combining the epifluorescence microscopy with the microfluidic cell, and excitation and imaging lasers,” he adds. “By doing this, we have all of these capabilities.”

The researchers have already used COMPEITS to image the photoelectrocatalytic oxidation of hydroquinone, a micropollutant found in aquatic ecosystems, on single bismuth vanadate photocatalyst particles in exquisite detail. They were able to map, with nanometre precision, the non-fluorescent surface reaction, discovering previously unknown behaviours of the catalyst.

Indeed, they could see that hydroquinone adsorption varied

DURING A FLUOROGENIC oxidation reaction, fluorescence microscopy can identify a catalyst’s most active sites (left, red and yellow). Adding a competing non-fluorescent reaction decreases the reaction rate (centre) and reveals where on the catalyst a “dark” reaction occurs (right). Cornell/Peng

DR XIANWEN MAO and Professor Peng Chen, Cornell, have developed COMPEITS to image non-fluorescent reactions Rocky Ye/Cornell University

Lab focus

19

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

according to which side or facet of the catalyst particle, the reaction was taking place on. What’s more, the researchers could even detect differences in hydroquinone adsorption within facets. For example, they noted that this process was actually weaker at the corners of so-called lateral facets, when two such facets intersected at an edge.

As Mao points out, such adsorption information is crucial for truly understanding catalysed surface reactions, and in this case, can be used to help render hydroquinone non-toxic and learn how to remove pollutants from aqueous environments. And hydroquinone is only the beginning.

Given these results and its ability to provide spatially resolved detail on how unlabelled, non-fluorescent particles behave during reactions, Mao and colleagues are confident COMPEITS can interrogate a range of surface processes.

“Photocatalytic pollutant degradation includes many organic contaminants in water that could be dye molecules, pharmaceuticals and carcinogenic aromatics,” highlights Mao. “However I think COMPEITS is going to be important for other fields beyond catalysis... for example, we could also look at water splitting, necessary for generating hydrogen for the hydrogen economy, as well as carbon dioxide reduction for artificial photosynthesis.”

The technique could be used to scrutinise molecular binding processes for chemical separation processes, which as Mao points out, account for up to 25% of the world’s energy consumption. If researchers could identify the optimal binding sites to predict the best size and shape for an adsorbent material, they could design more energy efficient processes that maximise separation efficiency and enhance molecular selectivity.

The method could also be used to shed new light on other molecular binding processes across many materials systems including graphene-based materials and covalent organic frameworks. Such information could address fundamental questions in materials engineering, such as where molecules preferentially bind, and how binding locations affect optical, electronic and magnetic properties.

So where next for COMPEITS? The researchers are now looking to raise the resolution of COMPEITS even further, and will also explore the use of photoactivatable proteins to trigger reactions.

“Many reactions in biological systems are also non-fluorescent so if we used photoactivatable fluorescent proteins we could also use the method for bioimaging,” says Mao. “Sometimes antibodies bind to these proteins, so perhaps by modifying the imaging set-up we could image this process.”

“So many of these reactions are very simple but completely non-fluorescent, yet we should be able to visualise them,” he adds. “With properly chosen auxiliary systems, essentially you can map out any kind of reaction with COMPEITS.”

Professor Peng Chen and colleagues have used single-molecule imaging to interrogate a host of nanoscale materials and biological systems, employing single-molecule fluorescence imaging, single-molecule FRET, single-molecule tracking, super-resolution localisation microscopy and magnetic tweezers.

More than a decade ago, Chen and colleagues were using single-mol-ecule fluorescence microscopy, on their homebuilt Olympus IX71 total internal reflection microscope, to study the catalysis of individual gold nanoparticles. They also man-aged to pinpoint the reactive sites on single-walled carbon nanotube electrocatalysts.

More recently, Chen and colleagues have used magnetic tweezers, mounted onto an Olym-pus IX71 inverted microscope, to study polymer growth, in real-time and at the single-polymer level. After fixing the polymer chain to a glass plate and then applying a ru-thenium catalyst to trigger growth, the researchers were able to use the tweezers to pull the chain and watch it grow.

Around the same time, the re-searchers used single-molecule su-per-resolution catalysis imaging in correlation with electron microscopy to visualise enhanced catalytic activ-ity at the palladium-gold interface in single bimetallic nanoparticles. As Chen said at the time: “This is ex-perimentally observing something that people knew about but couldn’t see, and now we have a new way of directly seeing it.”

And only late last year, Chen and colleagues won $2 million from the US Department of Energy to study how light absorbing quantum dots can be combined with bacteria for efficient energy conversion. Head-ing up an interdisciplinary team of researchers, including Professors Tobias Hanrath and Professor Buz Barstow also from Cornell, Chen will create a hybrid quantum dot-bac-terial cell system that can harvest sunlight and use carbon dioxide to produce high-value chemicals including plastic precursors, the biofuel butanol and the bio-plastic, polyhydroxybutyrate.

As Chen points out: “This is a sys-tem that others have shown to work, and it’s very promising, but it hasn’t been brought to large scale or with extremely high efficiencies yet.”

“We need to understand how the system works at a fundamental level, and find the underlying factors that can contribute to or limit system performance, so we can potentially engineer or improve it,” he adds.

BACTERIA laying on a thin layer of quantum dots are probed with a laser Cornell

PROFESSOR PENG CHEN imaged the enhanced catalytic activity at the palladium-gold interface in single bimetallic nanoparticles Cornell

PROFESSOR PENG CHEN and colleagues use single-molecule fluorescence microscopy to study individual molecules react with a single nanoscale particle of gold in real time Cornell

Interrogating systems

Lab focus

20 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

INTRODUCTIONScanning transmission electron microscopy (STEM) is a powerful tool for studying specimens at very high (sub-angstrom) spatial resolution. Conventional detectors for STEM, such as bright-field (BF), annular bright-field (ABF), annular dark-field (ADF) and high angle annular dark-field (HAADF) detectors are monolithic, and form images, I(x, y), by integrating the total signal over a range of scattering angles in the diffraction plane as the focused STEM probe is rastered over a two-dimensional (2D) area of a specimen (Figure 1a). In contrast, detectors for four-dimensional STEM (4D STEM) record a pixelated, angle-resolved 2D convergent beam electron diffraction (CBED) pattern, I(kx, ky), for every position, (x, y), in the 2D STEM raster over the specimen (Figure 1b), yielding a 4D dataset, I(kx, ky, x, y). As with any diffraction pattern in TEM, the angles in the diffraction plane sampled in 4D STEM may be controlled by varying the microscope camera length. The size of diffraction discs in the CBED pattern may be controlled by varying the convergence angle of the STEM probe. The choice of camera length and convergence angle will depend on the

types of data that one wishes to extract from the 4D STEM dataset.

The key advantage of 4D STEM compared to conventional 2D STEM imaging is that, in a single scan, one can acquire a 4D dataset that contains all possible 2D STEM images of the sample, as well as a variety of novel signals that cannot be derived from conventional 2D STEM imaging. The speed and sensitivity of 4D data collection can be improved by employing a direct electron detector, which can either be used as a stand-alone detector or can be used alongside conventional STEM detectors. For example, a 4D STEM detector can work in conjunction with a traditional annular detector, which can record a 2D image using the signal from a region of the diffraction plane outside of that recorded by the 4D STEM detector. This flexibility allows users to upgrade their existing STEM instruments with 4D STEM capability.

Here, we give a basic overview of the principles of 4D STEM, including some of the types of images that can be extracted from 4D STEM datasets, as well as an overview of the types of direct detectors that can be used for 4D STEM. A complete review of

the technique, including historical developments may be found in Ref. [1].

4D STEM DETECTORSThe availability of high-performance electron detectors has been one of the main catalysts for the growing popularity of 4D STEM[1]. Early efforts at recording 4D datasets consisting of CBED patterns were limited by the relatively slow readout speeds of CCDs[2-4,5]. Recording 4D STEM data is considerably more practical when using a sensitive detector capable of millisecond or sub-millisecond readout speeds. Although 4D STEM is possible using a fast scintillator coupled camera, direct electron detectors are particularly well-suited to 4D STEM due to their high detective quantum efficiency (DQE) and high signal to noise ratio (SNR) compared to scintillator coupled cameras.

There are two principal types of direct electron detector that have been used for 4D STEM applications, and both types have different strengths and weaknesses. A hybrid pixel array detector (PAD)[6, 7] consists of an array of wide, thick pixels (e.g. 150 x 150 x 500 µm for the electron microscope

4D STEM with a direct electron detector

BARNABY D.A. LEVIN1, CHENYU ZHANG2, BENJAMIN BAMMES1, PAUL M. VOYLES2, ROBERT B. BILHORN1

1 Direct Electron LP, San Diego, CA, USA 2 Department of Materials Science And Engineering, University of Wisconsin Madison, Madison, WI, USA.

BIOGRAPHYBarnaby Levin is a Materials Applications Scientist at Direct Electron. He studied Physics as an undergraduate at the University of Oxford before completing a Ph.D. in Applied Physics at Cornell University in 2017, under the supervision of David Muller. Barnaby completed a Postdoc at Arizona State University, super-vised by Peter Crozier, before join-ing Direct Electron in November 2019. Barnaby’s responsibilities at Direct Electron include sup-porting customers with 4D STEM applications.

ABSTRACTThe recent development of fast, sensitive, pixelated detectors has led to the growing popularity of four-dimensional scanning transmission electron microscopy (4D STEM). 4D STEM involves acquiring a 2D convergent beam electron diffraction pattern, at every pixel of a 2D STEM raster, generating a 4D dataset containing a vast amount of information about the specimen. Here, we describe the basic principles of 4D STEM, including different types of information that can be extracted from 4D datasets and how the technique’s dynamic range requirements can be met using a hybrid integrating and counting Monolithic Active Pixel Sensor (MAPS) Direct Detection Device (DDD®).

ACKNOWLEDGEMENTSThis material is based upon work supported by U.S. D.O.E., Office of Science, Award Numbers DE-SC0018493 and DE-FG02-08ER46547, and made use of UW-Madison microscopy facilities supported by Wisconsin MRSEC (DMR-1720415).

CORRESPONDING AUTHORBarnaby D.A. Levinblevin@directelectron.com

FIGURE 1 Schematic diagrams illustrating conventional STEM and 4D STEM. a) In conventional STEM, monolithic detectors are used to produce images by integrating a signal from different angular regions of the diffraction plane. b) In 4D STEM, an entire 2D CBED pattern is recorded at each probe position of a 2D STEM raster, resulting in a 4D dataset.

4D STEM using a Direct Detector

A B

Microscopy and Analysis 34(1): 20-22 (EU), February 2020

21

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

pixel array detector (EMPAD) described in Ref. [7] (Thermo Fisher Scientific, Waltham, MA USA) bump-bonded to an application specific integrated circuit. PADs typically exhibit extremely high dynamic range, due to a high saturation value per pixel[7], but their relatively small number of pixels limits the angular resolution of the detector and limits the ability to use the PAD as an “all-in-one” detector that can be used for other TEM imaging applications, where a large number of pixels is advantageous. In contrast, monolithic active pixel sensors (MAPS)[8-11] use relatively thin pixels, which limits beam spreading within each pixel, allowing for smaller pixel sizes (e.g. 6.5 x 6.5 µm in the case of a DE-16 (Direct Electron, San Diego, CA USA)) and a correspondingly much larger number of pixels per sensor. The traditional weakness of MAPS detectors for 4D STEM is their relatively limited dynamic range, due to a relatively low saturation value per pixel. This can be mitigated to some degree by using the large number of pixels on the sensor to spread features of interest in the CBED pattern over many pixels.

In many 4D-STEM experiments, it is desirable to simultaneously record very bright and faint features in the CBED pattern without saturating the detector. For example, one may wish to simultaneously record the bright central (0,0) beam (bright-field disc), as well as the faint signal of scattered electrons in the dark-field region. To maximize the signal-to-noise ratio of in the sparse dark-field region, MAPS detectors can be used to perform electron counting[12,13], which excludes background noise and normalizes the signal of each detected electron. However, isolating and counting each primary electron incident on the detector requires sparsity (i.e. < ~0.1 e– per pixel), which is generally not satisfied in the intense bright-field disc. Hybrid counting and integrating

techniques can provide a method of overcoming the challenge of limited dynamic range for 4D STEM using MAPS detectors[14,15].

Direct Electron has developed a patent-pending hybrid counting technique[15], which is implemented for our DE-16 detector. The workflow of the method is illustrated in Figure 2. During 4D STEM data acquisition, specialized software (DE-4DExplorer) is used to calculate a sparsity map for each frame. The sparsity map is a binary mask corresponding to regions of the frame where the number of primary electrons per pixel is low enough to be processed using electron counting. In these sparse regions (primarily in the dark-field area of the diffraction pattern), individual primary electron events can be distinguished and thus electron counting can be used to improve SNR by effectively eliminating the Landau noise caused by the variable amount of energy deposited on the sensor by each primary electron[16,17]. In bright regions (such as the bright-field disc), there is a sufficient number of primary electrons to statistically diminish Landau noise. However, in these bright areas many primary electrons are recorded by the same or neighbouring pixels, making it impossible to distinguish individual electron events using electron counting algorithms. To maximize the SNR without sacrificing linearity, the hybrid counting technique performs electron counting in the sparse regions of the CBED pattern while using conventional integration in non-sparse regions. The intensity of non-sparse regions is scaled based on the average pixel intensity per primary electron. Thus, the intensity value for each pixel in the final processed frame approximately corresponds to the actual number of primary electrons incident on each pixel in each frame. By applying this hybrid counting technique, the DE-16 can image the

low-angle dark-field regions of the CBED pattern with a high SNR, while simultaneously recording useful information from the more intense bright-field disc.

Note that Direct Electron’s hybrid counting method automatically detects sparse regions in each recorded camera frame without prior information from the user. Electron counting is performed without manually having to define the position and size of the bright and sparse regions in advance. In principle, hybrid counting may therefore be applicable not only to 4D STEM, but also to other techniques that require high sensitivity and high dynamic range, such as electron energy loss spectroscopy (EELS).

4D STEM DATA ANALYSIS4D STEM datasets occupy far more computer memory than a conventional STEM image. For example, a 4D STEM dataset consisting of a 1024x1024 pixel CBED pattern recorded at every pixel of a 1024x1024 STEM raster and stored in 16-bit format will occupy ~2 TB of memory. The acquisition of 4D STEM data also requires very high rates of data transfer compared to 2D STEM imaging. For example, a data transfer rate of ~2 GBs–1 is required for 4D STEM acquisition of 16-bit 1024x1024 pixel CBED patterns at a rate of one frame per millisecond. In practice, the memory occupied by a 4D STEM dataset may be reduced either by limiting the number of spatial pixels in the STEM raster, by using fewer pixels to record patterns in diffraction space, or by downsampling the data after acquisition.

Analysis of 4D STEM datasets typically requires a powerful computer and specialized software. The data presented below has been analyzed using a combination of custom written Python code, as well as libraries from Hyperspy[18] and py4DSTEM[19]. Other software tools currently available for

the analysis of 4D STEM data include LiberTEM[20], Pycroscopy[21], pixStem[22], pyXem[23] and the Cornell Spectrum Imager plugin for ImageJ[24], which was originally written for the analysis of hyperspectral data, but now also includes 4D tools.

One of the simplest operations that can be applied to the 4D dataset is to mask an area of diffraction space and integrate the signal from within the mask in each diffraction pattern to recover a 2D STEM image[6,25,26]. This is illustrated in Figure 3 for a 4D dataset recorded from a sample of SrTiO3 imaged along a [001] direction. Using this masking technique, images equivalent to that of any conventional STEM detector (e.g. BF, ABF, ADF) can be generated, but with the advantage that the 4D STEM dataset allows complete flexibility in choosing detector angles. Indeed, one can vary the shape and position of the mask, to generate images from any kind of “virtual detector”[27]. For example, one can choose to reduce the convergence angle of the STEM probe such that Bragg discs are well separated and do not overlap. In this mode, diffraction patterns are often referred to as nanobeam electron diffraction (NBED) or nanobeam diffraction (NBD) patterns, rather than CBED patterns. One can then generate an image analogous to that of dark-field TEM by placing a circular mask around a diffracted Bragg disc in the NBED pattern, with the mask acting as a virtual objective aperture[28-32].

In addition to reconstructing images using simple masks, 4D STEM enables the reconstruction of differential phase contrast (DPC) images[2,33-35], which were previously only available by using segmented detectors[36], and “Centre of Mass” (COM) images[6,37], which are unique to 4D STEM, and have also been referred to as “first moment” or COM DPC images in the literature[1,35,38]. DPC and COM both

FIGURE 2 Illustration of the hybrid-counting technique. During data acquisition, the DE-4DExplorer software calculates a sparsity map for each frame. This is a binary mask corresponding to regions of the frame where the number of primary electrons per pixel is low enough to be processed by electron counting. Within the masked (yellow) area of the frame, data is recorded in counting mode. In the remaining area, data is collected in integrating mode. The data is then merged to form a hybrid counted frame, exhibiting a high SNR in the dark-field region, whilst simultaneously preserving information from the more intense bright-field disc.

4D STEM using a Direct Detector

22 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

FIGURE 4 a) Image showing the magnitude of COM shifts in a 4D STEM dataset acquired from a sample of SrTiO3. The largest COM shifts occur when the probe is adjacent to an atomic column. The diagram beneath the image illustrates how the magnitude and direction of the shift in the COM are defined with respect to the position of the centre of the probe in vacuum. b) Map showing the direction of COM shifts from the same dataset as in (a). When the probe is positioned adjacent to an atomic column, the COM shifts towards the column. The direction indicated by each colour is shown by the colour wheel beneath the image.

essentially measure the magnitude and direction of the average shift in the 2D momentum of the electron probe for each CBED pattern in the 4D STEM dataset. The DPC and COM imaging modes are sensitive to specimen and crystal grain orientation[6], as well as to electric fields in the specimen[33,39-44], which can cause a shift in the average momentum of the electron probe either by displacing the entire diffraction pattern, or altering the distribution of intensity within the diffraction pattern, depending on the relative length scales of the probe and the source of the field[37]. Magnetic fields can also be detected using DPC and COM when operating the microscope in a Lorentz mode[2,7,34,45,46]. An example of COM imaging is shown in Figure 4. For a diffraction pattern consisting of pixels with coordinates k = (kx, ky), and with the intensity in each pixel defined as I(k), the position of the centre of mass in the diffraction pattern can be defined mathematically as:

kCOM=∫k I(k) dk

A suitable choice of origin for measuring kCOM is the position of the centre of the probe in vacuum when no specimen is present, which under ideal conditions should correspond to the centre of the optic axis and the centre of the detector. The effect of the nuclear potential on the COM is apparent when imaging a specimen at atomic resolution[33,39,47,48]. Figures 4a and 4b show the magnitude and direction respectively of the shift of the COM for the same area of the SrTiO3 sample displayed in the images in Figure 3. The magnitude of the shift of the COM is strongest when the probe is close to, but not directly on top of, the Sr and Ti/O columns, and the direction of the shift is oriented towards the columns. Here, the COM images are measuring the deflection of the negatively charged electron beam by the screened, positively charged atomic nuclei.

As with the generation of conventional STEM images from 4D STEM data, one has complete flexibility

to apply a mask of any shape or size to the 4D dataset when generating DPC or COM images, enabling the DPC or COM associated with specific regions of the CBED or NBED pattern to be calculated[37].

Another application of 4D STEM is mapping local changes in lattice constant and local strain within the specimen[49-51]. Strain information can be extracted from 4D STEM datasets in different ways. When acquiring 4D STEM data containing NBED patterns with well separated Bragg discs, strain is encoded in the disc positions[52], or (for very thin samples) the disc centre of mass[53]. For polycrystalline or amorphous materials, the ellipticity of NBED patterns may also be useful[54].

Further techniques that can make use of 4D STEM datasets include mapping of crystal grain orientations in a specimen[3,4,55-57], where there is some overlap between 4D STEM and the similar technique of nanobeam precession electron diffraction[58], and electron ptychography, in which computational methods are used to reconstruct the projected potential of the specimen, with the potential for spatial resolution beyond that of traditional STEM imaging techniques[59-70]. A number

of other imaging modes besides those mentioned above may also be performed using 4D STEM[1]. As 4D STEM is a relatively new and growing technique, there may be further imaging modes that are yet to be developed that can take advantage of the wealth of information contained in 4D STEM datasets.

MATERIALS AND METHODSThe data presented in this article was acquired on an Aberration Corrected FEI/Thermo Scientific Titan 80-200 S/TEM (Thermo Fisher Scientific, Waltham, MA USA), operated at 200 kV with a STEM probe current of ~20 pA and a probe convergence semi-angle of 24.5 mrad. A DE-16 direct detection camera (Direct Electron, San Diego, CA USA) was used to record 4D STEM data. Each of the CBED patterns in the 4D STEM data was captured using a 1024x1024 pixel centred area of the 4096x4096 pixel sensor. A hardware binning of 2 was applied to the CBED patterns, reducing them to 512x512 pixels in size, allowing the 4D STEM dataset to be recorded at a rate of ~1100 frames per second (fps). Higher frame rates (up to >4200 fps) are accessible by further reducing readout area.

The wedge-shaped single crystal SrTiO3 [100] test specimen was prepared by multistep mechanical polishing at 1.6° tilt, followed by further thinning in a Fischione 1050 ion mill (E.A. Fischione Instruments, Inc., Export, PA, USA). The specimen was gently cleaned using an Ibss GV10x DS Asher cleaner (Ibss Group Inc., Burlingame, CA, USA) before being transferred into the TEM column.

SUMMARY AND CONCLUSIONSIn summary, 4D STEM is a versatile technique that enables a multitude of different types of analyses to be performed on a specimen. These include reconstructing images equivalent to those of any traditional STEM detector of any shape or size, imaging of electric and magnetic fields, mapping strain within a specimen, mapping crystal grain orientations, analyzing medium range order, and generating ptychographic reconstructions with high spatial resolution. Both hybrid PADs and MAPS detectors may be used for 4D STEM. Using a hybrid counting method, implemented on detectors such as the DE-16, dynamic range and sensitivity are sufficient to allow simultaneous imaging of the bright-field disc and low-angle dark-field regions, overcoming a traditional weakness of the MAPS architecture, allowing users to record 4D STEM datasets that take advantage of the angular resolution available to a detector with a high pixel count. 4D STEM methods using pixelated direct electron detectors can complement existing STEM detectors and can readily be implemented on existing STEM instruments.

Article and references available online at: Microscopy-analysis.com/article/february_20/levin ©John Wiley & Sons Ltd, 2020

FIGURE 3 a) A position-averaged CBED (PACBED) pattern, derived from a 4D STEM dataset acquired from a SrTiO3 sample, imaged along a [001] direction. b) Diagram of the crystal structure of SrTiO3 in a [001] projection. c) Example of a mask that can be applied to the 4D dataset to produce a BF image. d) Example of a mask that can be applied to the 4D dataset to produce an ABF image. e) Example of a mask that can be applied to the 4D dataset to produce an ADF image. The crystal diagram in the lower corner of each image indicates atomic column positions. The spacing between adjacent Sr atomic columns (green) is ~3.9 Å.

4D STEM using a Direct Detector

A

B

C D E

A B

23

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Mid-November in the UK Cryo microscopy community is synonymous with the annual Cryo Microscopy Group (CMG) meeting. November 2019 was no different and saw a diverse line up of speakers share their latest achievements at the University of Nottingham with a receptive and enthusiastic audience.

The first speaker Thomas Braun from the University of Basel presented his work on revolutionary cryo-EM grid preparation which uses nanolitre deposition onto EM grids for the purpose of preparing thin films of solutions. The use of a microcapillary technology for the deposition and microfluidics for solution mixing and sample extraction combine to give a very powerful sample preparation technique. Braun presented this using the title single-cell proteomics in which he showed it was possible to extract cellular contents and deposit this onto a TEM grid with a view to analysing the proteins by cryo-TEM and single particle analysis.

Following on from Braun, Nicole Hondow from the LEMAS centre at the University of Leeds, provided the audience with a wonderful overview of her work using cryo techniques for materials systems characterisations. She explained that she’d started using Cryo-TEM to analyse nanoparticle solutions, but then explored Cryo-STEM, EDX and EELS and was able

to use traditional microanalysis techniques on these beam sensitive and hydrated systems. This is something that I’ve noticed happening more and more as the benefits of cryo-EM are fully realised by materials scientists.

She also reported on some Cryo-FIB analyses that the LEMAS team had been working on with colleagues in chemical engineering. Using the FIB they’d been able to analyse fully hydrated colloidal dispersions in three dimensions, resulting in impressive renderings that added insight to these structures.

The CMG always has wonderful support from trade partners and this meeting was no exception, both from microscope manufacturers, cryo-stage manufacturers and accessories companies. As part of the day the organising committee has the technobyte session, a platform for the vendors to update the attendees on their latest hardware/software.

A feature of the CMG for some time has been a focus on the student/early career or those new to cryomicroscopy. Traditional poster presentations have given way to the Freeze Frames competition and this year two entrants braved the podium to show two slides

of their work and explain it in two minutes. Both entrants kept to the slide and time limits (this doesn’t happen every year) and Zubair Nizamudeen was awarded first place with his work on correlating light, electron microscopy and SIMS data with Rachael Xerri as runner-up, who showed off her impressive cryo-TEM of internally complex phases of drug loaded particles.

After the generously long lunch break, which allowed may discussions with the vendors and speakers alike, the delegates returned to the presentations for more science and were treated to a ‘how to’ in high pressure freezing (HPF) by Xavier Heiligenstein, who summarised his long standing work in the areas of HPF and attempts to perform correlative microscopy that were now bearing fruit.

Heiligenstein described his attempts to freeze quicker in order to capture events happing in cells as well as the efforts to which he had gone to freeze thicker samples. Now having left Institute Curie and working for CryoCapCell as its CSO he predicted a bright and less complex future for the field of correlative cryo-microscopy.

Last up on the list of speakers was

David Scurr who sought to educate and amuse with his title of Sub-zero SIMS. For those who aren’t familiar, SIMS is secondary ion mass spectrometry, a well-established surface characterisation technique, but Scurr’s system isn’t any old SIMS – it has a couple of tricks up its sleeve. These tricks come in the form of an Orbitrap and a fully functional Cryo-stage which makes it a cryo-OrbiSIMS!

This combination of high spatial resolution and high mass resolution at low temperatures (native state if the sample is correctly prepared) results in Scurr having investigated samples such as cells, bacteria and skin, in which he can investigate the ion fragments from those samples and comment on their chemistry. This can be done in three dimensions as the ion beams of the system allow 3D profiling, meaning this is an incredibly powerful technique with huge-potential.

The meeting closed on schedule and the meeting chair thanked local organisers and helpers, before wishing attendees a safe trip home.

The 32nd CMG will take place in November 2020. The venue will be confirmed later in the year.

CMG gathers in NottinghamMicroscopy & Analysis editor Chris Parmenter reports from the succesful 2019 CMG meeting at the University of Nottingham, UK

THE FREEZE FRAMES competition winner Zubair Nizamudeen, far left, and runner-up Rachael Xerri, second left, with members of the committee

Meeting report

24 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

product focus digital camerasKEY WORLDWIDE AMERICAS ASIA/PACIFIC EUROPE

ACCU-SCOPE INC.(01) 631-864-1000info@accu-scope.comaccu-scope.com/

Excelis™ HD Digital Microscope Camera (Model AU-600-HD)Applications Observe and capture color images of specimens on a microscope

Key features •See full LIVE HD images via HDMI interface •60fps live view via HDMI; 30fps via USB 2.0 •On-board software gets you going

FAST; no PC required •Save images and video to SD card, PC or Mac •Includes CaptaVision+ software for more advanced image acquisition, measurements and analysis (i.e. LIVE EDF, LIVE Stitching, fluorescence mode, and more)

Excelis™ HDS Digital Microscope Camera with HD Monitor (Model AU-600-HDS)Applications Observe and capture color images of specimens on a microscope without the need for a PC

Key features •See full LIVE HD images on included 11.6" HD monitor via HDMI interface •60fps live view via HDMI; 30fps via USB 2.0 •On-board software gets you going FAST;

no PC required •Save images and video to SD card, PC or Mac •Includes CaptaVision+ software for more advanced image acquisition, measurements and analysis (i.e. LIVE EDF, LIVE Stitching, fluorescence mode, and more)

Excelis™ MPX-5C Pro Digital Microscope Camera (Model AU-5C-CMOS)Applications View and capture color images and videos of specimens on a microscope.

Key features •View and capture gorgeous 5MP images with CCD-like quality •35fps live view via USB 3.0 •Big 3.45um

pixels provide excellent resolution and superb sensitivity •Fluorescence imaging with high signal-to-noise, exposure times up to 15 seconds and adjustable gain •Includes CaptaVision+ software for full camera control, advanced image acquisition, customizable user interface, dynamic measurements and more (i.e. LIVE EDF, LIVE Stitching, fluorescence mode, and more)

Excelis™ MPX-20C Digital Microscope Camera (Model AU-20C-CMOS)Applications View and capture color images and videos of specimens on a microscope, ideal for use with stereo microscopes

Key features •View and capture gorgeous 20MP images, ideal for use with stereo microscopes and lower

magnifications •15fps live view at full resolution (20MP); 53fps at 5MP •Includes CaptaVision+ software for full camera control, advanced image acquisition, customizable user interface, dynamic measurements and more (i.e. LIVE EDF, LIVE Stitching, fluorescence mode, and more)

ADVANCED MICROSCOPY TECHNIQUES

1-978-774-5550sales@amtimaging.comamtimaging.com

NanoSprint43 Mark IIApplications Life science, materials scienceKey features 43 Megapixels •10 fps •Super Wide FOV

NanoSprint15 MarkIIApplications Life science, materials science, in-situKey features •15 Megapixels •30fps •Wide FOV •16 Bit

NanoSprint12Applications Life science, materials scienceKey features •12 Megapixels •55fps •Wide FOV

NanoSprint5Applications Life science, materials scienceKey features 5 Megapixels •80fps •Wide FOV

CAIRN RESEARCH LTDj.kerin@cairn-research.co.ukcairn-research.co.uk

Multisplit V2 – 4-way emission image splitterApplications FRET, ratiometric calcium, simultaneous multi depth imaging, super resolutionKey features •4 images on a single camera chip •Support for sensors up to 25mm diagonal •425nm to 875nm coatings on all surfaces •Simple and precise controls for image registration •Supports 25mm filters and ultra flat dichroics

Optosplit II bypass – 2-way emission image splitterApplications Simultaneous multi fluorescent probe imaging, FRET, Simultaneous dual depth imaging, TIRF, Super resolution STORMKey features •1 or 2 images on a single camera chip •Supports 25mm filters •Support for sensors up to 18.8mm diagonal •Single lever switching from split to bypass •425nm to 875nm coatings on all surfaces

TwinCam – multi camera adapterApplications Simultaneous use of two dyes or genetic markers, FRET, ratiometric calcium, simultaneous multi depth imaging, super resolutionKey features •2 cameras on 1 microscope port •Support for sensors up to 18.8mm diagonal •425nm to 875nm AR coatings on all surfaces •25mm filters and standard microscope dichroic sizes •Compact design with integral C-mount input

Triplesplit – 3-way emission image splitterApplications FRET, ratiometric calcium, simultaneous multi depth imaging, super resolutionKey features •1, 2 or 3 images on a single camera •Support for sensors up to 25mm diagonal •425nm to 875nm coatings on all surfaces •Simple and precise controls for image registration •Supports 25mm filters and ultra flat dichroics

DELTAPIX APS4546760205deltapix@deltapix.dkdeltapix.dk

Invenio 4KHDMI 4K microscope cameraApplications Inspection, education, quality control, pathologyKey features •Standalone 4K HDMI camera with large Sony ExmorTM sensor •Precise colour reproduction •Built-in advanced measurement, Built-in 4K video recording •Fast streaming video 30FPS

Invenio 10EIIIApplications Material science, histology, pathology, quality control, fluorescensKey features •2/3” Sony ExmorTM sensor •High-resolution 10 MegaPixels •Precise colour reproduction •Low noise and high sensitivity, Including software with measurement features.

Invenio 12EIIIApplications Darkfield, brightfield, quality control, histology, pathologyKey features •High-resolution 12MegaPixels •Precise colour reproduction •Low noise •Including software with measurement features.

Invenio 20EIIIApplications Material science,quality control, life science, documentation and publishingKey features •1” Sony ExmorTM sensor, High-resolution 20MegaPixels •Precise colour reproduction •Low noise and high sensitivity •Including software with measurement features.

DIRECT ELECTRON4088586297info@directelectron.comdirectelectron.com

DE-16 Camera SystemApplications In situ TEM, 4D STEM, materials science, biological, Cryo-EM

Key features •Direct Detection with on-chip CDS provides the very best signal-to-noise ratio and sensitivity •Ultra-fast

frame rate up to 4237 fps •Global shutter eliminates distortion at high speed •Ultra-high resolution mode with near-perfect MTF and extraordinary DQE •Integrated protection shutter, and Faraday plate

DE-64 Camera SystemApplications Biological Cryo-EM, single-particle TEM, electron tomography, large-volume TEM imaging, MicroEDKey features •Largest field-of-view available in a direct detection camera •Ultra-high resolution mode with near-perfect MTF and extraordinary DQE •Direct Detection with on-chip CDS provides the very best signal-to-noise ratio and sensitivity •Operates in counting or integrating mode •Integrated protection shutter and Faraday plate

DE-DirectViewApplications Biological Cryo-EM, materials science, in situ TEM, 120kV LaB6 Cryo-EM, general purpose electron microscopyKey features •Direct Detection with on-chip CDS provides the very best signal-to-noise ratio and sensitivity •High speed operation •Economical direct detection camera for limited budgets •Can be installed in JEOL film drawer •Integrated protection shutter, and Faraday plate

DE-FreeScanApplications Scan generator, 4D-STEM, compressive sensing, ptychographyKey features •Scan generator for scanning transmission electron microscopy (STEM) detectors •Enables compressive sensing for high-speed low-dose imaging •Enables arbitrary scan patterns, subsampling, compressive sensing, and 4D-STEM •Designed to operate with our DE-16 direct detection camera for synchronized high-speed 4D-STEM •Also compatible with conventional BF/DF STEM detectors

PROMICRA+420 222 952 667promicra@promicra.compromicra.com

QuickPHOTO - Imaging and Measuring SoftwareApplications Life and material science, histology, pathology, cytology, fluorescence microscopy, education, industrial inspectionKey features •Measuring tools, exporting, reports •EDF, HDR, Fluorescence, Stitching add-ons •Supports Canon EOS DSLRs, Lumenera INFINITY and PROMICAM cameras •Time-lapse imaging •Free 30-day trial

Figure Maker module for QuickPHOTOApplications Life science, entomology, botany, mineralogy, archeologyKey features •Effortless creating of composite figures for publications •Single-click insertion of scale bars and labels •Support of metric and imperial units •Full control of figure resolution (DPI) and dimensions according to a journal specification

Deep Focus EDF Module for QuickPHOTOApplications Entomology, botany, mineralogy, archeology, industrial inspectionKey features •Completely focused images of excellent quality •Can be combined with HDR feature •Suitable for any type of microscope •Automation using focus motorization •Compatible also with Canon EOS DSLRs

PROMICAM USB 3.0 CamerasApplications Life and material science, histology, pathology, cytology, fluorescence microscopy, education, industrial inspectionKey features •Resolution 2.8 - 12 MP •CCD & SONY Pregius global shutter CMOS sensors •Up to 4K UHD live view resolution •High live view frame rates •Supported by QuickPHOTO software

SPOT IMAGING SOLUTIONS5867316000Kenn.Zorn@diaginc.com

RT sCMOS Digital CameraApplications Fluorescence microscopy, Brightfield, Histology, Pathology, Cytology and Semiconductor InspectionKey features •Unprecedented speed and sensitivity

Insight sCMOS 5MP CameraApplications Pathology, cytology, microbiology, industrial inspection, supporting both brightfield and fluorescence microscopyKey features •5MP Resolution •Microscopy Optimized Color Correction •High Speed Preview 30 fps •Microscopy Optimized Brightness 0.49s

Insight sCMOS 12MP CameraApplications Pathology, cytology, microbiology, industrial inspection, supporting both brightfield and fluorescence microscopy.Key features •12MP Resolution •Microscopy Optimized Color Correction •High Speed Preview 30 fps •Microscopy Optimized Brightness 0.49s

Product focus

Engage with the microscopy community Deliver your marketing message with an exclusive advertisement on a cover page added to each article downloaded from the Journal of Microscopy.

This position captures audience engagement quickly and effectively.

Clear return on invest thanks to metrics on views and click-through rates.

Contact your local sales rep

Account Management EMEADr. Stefanie KrauthEmail: mkrauth@wiley.comTel: +49 (0) 6201 606 728

Account Management USRoland Espinosa (Accounts A through K)Email: respinosa@wiley.comTel: + 1 201-748-6819

Joe Tomaszewski (Accounts L through Z)Email: jtomaszews@wiley.comTel: + 1 908-514-0776

Account Management APACYosuke SatoEmail: ysato@wiley.comTel: + 81-3-3830-1234

19 -

5774

01

Unbenannt-3 1 25.03.2019 13:03:23

26 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Cressington 208CHigh Precision Carbon Coating with Voltage

Controlled Evaporation Source

Cressington 108CFor Standard SEM Imaging

with EDS Analysis

PELCO easiShaper™ Carbon Rod Shaper

Your single source for all Carbon

Evaporation Supplies

www.tedpella.com sales@tedpella.com

800.237.3526

Microscopy Products for Science and Industry

Linkam has launched an latest update to its cryo-stage for correlative microscopy; CMS196V 3.

The CMS196V 3 is a cryo-correla-tive microscopy system enabling the full workflow of Correlative Light and Electron Microscopy (CLEM).

It is said to be ideal for the correlation of high resolution struc-tural information with biochemical processes within cells.

The latest model can be integrated with a wide range of research grade upright microscopes, offers enhanced sample stability for cryo-imaging, improved sample handling and reduced sample contamination.

The CMS196V 3 maintains the vitrified state of the sample by means of liquid nitrogen cooling and provides proven capabilities to safely handle and transfer cryo-sam-

ples and image them with optical microscopy.

Samples are kept free of contam-ination at all times due to active self-cleaning via the liquid nitrogen cold trap.

The integrated, encoded, moto-rised XY stage enables coordinate mapping required to locate the same sample position in the fluorescence microscope, as well as in the EM. The chamber top-up keeps samples vitrified at a constant –196°C, reducing photo bleaching and maintaining structural integrity of samples.

The sample cassette holder can hold up to 3 grids and ensures contamination-free sample loading, storage and transfer. Cassettes are available for different grid types including FEI, Planchette, Bessey, Polara and custom designs.

Optional Liquid Nitrogen autofill can extend the use of the system for up to six hours, unattended.

The LINK software for the CMS196V 3 provides control and monitoring of the system.

When combined with the optional high sensitivity camera and imaging module, LINK enables fully auto-mated, tiled, image capture.

The system produces a single, tiled, image of the full EM grid at high resolution. This can then be used to navigate the sample and save co-ordinates of areas of interest.

A full SDK is available for users to develop and integrate control of the CMS196V 3 into their own applica-tions. Support for the CMS196 V 3 is is also provided with Zeiss ZEN, Nikon NIS-Elements software as well LabVIEW.

Leica to support EMBL Imaging Centre

Linkam releases update to cryo-stage

Leica and the European Molecular Biology Laboratory have signed an open innovation collaboration agreement in which Leica will support the upcoming EMBL Imaging Centre, currently under construction at the EMBL campus in Heidelberg, Germany.

The signed framework agreement, which was negotiated and concluded by EMBLEM - EMBL’s wholly-owned commercial technology transfer subsidiary - allows for a broad and long-term partnership.

According to Leica, users will be able to test new technologies for

applications in their research much earlier and organisations can feed these experiences back into their technology development, to make sure new products deliver what users need to accomplish their research goals.

As Markus Lusser, president of Leica Microsystems, says: “The direct exchange of developers and researchers will pave the way for breakthrough applications, ones that confirm their relevance for state-of-the-art scientific research right from the start.”

The framework agreement allows

an even stronger interlinking between the early stages of new technology development and forefront scientific work.

As part of the agreement, the EMBL Imaging Centre will house cutting edge microscope equipment from Leica, which will be supported by Leica engineers to support the users of the EMBL Imaging Centre in their research.

The EMBL Imaging Centre is scheduled to open in mid-2021. It will host up to 300 visiting scientists per year, giving them access to the very latest imaging technologies.

THE EUROPEAN MOLECULAR BIOLOGY Laboratory and Leica Microsystems have signed a cooperation agreement Massimo del Prete/EMBL

What’s new

27

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Coxem has added the EM-30N to its EM-30 family of tabletop microscopes. The new microscope features re-designed imaging electronics for improved resolution with less noise at high magnification, and also introduces a new Panorama mode for wide area scanning.

Delivered with both SE and BSE detectors, the EM-30N allows the operator to view either detector individually, side-by-side, or as a composite image for a better under-standing of microstructure and chemistry.

Low Vacuum mode allows the operator to image non-conductive samples without any special pre-treat-ment, while a built-in optical camera helps to manage multi-sample holders and simplify navigation using our “Macro-Micro-Nano” sample views.

Powered by NanoStation software, the EM-30N is said to seamlessly integrate all SEM functions with EDS routines such as compositional particle analysis.

It provides advanced image analysis tools, such as line profile analysis, to more accurately determine particle size when working at the nanoscale. Automated features, such as near real-time automatic brightness/contrast, simplify operation and help novice users achieve consistent results. An optional STEM detector allows the EM-30N to

utilize its 30 kV accelerating voltage capability to perform TEM analysis of samples on standard TEM grids, while the optional CoolStage allows analysis of samples from -25°C to 50°C.

Coxem introduces new tabletop microscope

Cressington 208HRHigh Resolution Coating

with a choice of 15 Metals Including Iridium

Rotary StagesWide Selection

of Sample Holders

Sputter Targetsfor All Coater Brands

Your single source for all Specimen

Coating Supplies

www.tedpella.com sales@tedpella.com

800.237.3526

Microscopy Products for Science and Industry

Thermo Fisher Scientific has unveiled the Aquilos 2 Cryo-FIB, a DualBeam system dedicated to the preparation of thin, electron-transparent samples from biological specimens.

It is said to simplify the cryo-electron tomography workflow by reducing sample preparation time, minimizing the risk of contamination and providing a more precise and detailed view of specific points of interest compared to earlier models.

When used as part of the cryo-ET workflow, the Aquilos 2 allows researchers to prepare more and thinner samples for a clearer view of the inside of cells.

“Sample preparation is manual and can be difficult even for experienced users,” says Trisha Rice, vice president and general manager of life sciences at Thermo Fisher. “We’ve improved that in the Aquilos 2 by automating several steps during the process, allowing researchers to focus more on deepening their understanding of a wide range of diseases and speeding the path to drug discovery and treatment.”

The Aquilos 2 delivers optimal sample preparation and includes an integrated workflow with dedicated hardware and software that improves productivity and throughput compared to the earlier model.

The Thermo Scientific Auto Slice and View software, integrated into the Aquilos 2, allows scientists to acquire 3D images in cryogenic conditions by sequential slicing and then capturing multiple images of the interior of a vitrified cell.

The software allows researchers to remove parts of the cell, capture the 3D data from that cell part, and stop capturing data at exactly the right point before moving it to a higher resolution cryo-transmission electron microscope (cryo-TEM).

Additionally, researchers can further increase productivity with the Aquilos 2 by using the Thermo Scientific Cryo AutoTEM software, a guided solution that allows them to mark several points of interest and automatically prepare multiple samples in unattended, even overnight, runs.

Dedicated hardware keeps vitrified samples frozen at all times and protected from contamination.

Finally, the Aquilos 2 includes the Thermo Scientific EasyLift NanoManipulator, which allows researchers to prepare samples from targeted regions of a cell. Site-specific regions, such as labeled proteins, can be extracted and placed in the inside autogrids for tomogram acquisition in a cryo-TEM.

AQUILOS 2 Cryo-FIB from Thermo Scientific

Thermo Scientific delivers Aquilos 2 Cryo-FIB

What’s new

Microscopy Products for Science and Industry

Pelcotec™ CDMS-XY Calibration Standards - Magnification,

Resolution, LM, SEM, FESEM

AdhesivesConductive Tape - Carbon Tabs

Carbon, Silver & Gold Adhesives

Tools for Microscopy Scribes, Tweezers, Cutters

Stub Grippers, Vacuum Pick-Up

Filaments / CathodesTungsten - LaB6

FIB SuppliesLift Out Grids, Storage,

Probes, FIB Mounts

Cryo-EM Tools & AccessoriesLarge Dewars, Foam Dewars,

Grid Boxes & Grippers

SEM Specimen Mounts

TEM Grids / Support Films

Providing Microscopy Supplies and Specimen Preparation Equipment to Our Valued Customers for Over Half a Century

Don’t Let a Fingerprint Ruin Your Day—Use Evactron® Cleaning

WWW.EVACTRON.COM 1-650-369-0133

Fingerprints are one of the major sources of contamination in vacuum systems. Evactron dual-action turbo plasma cleaning™ expels these adventitious hydrocarbons with: Oxygen radicals plus UV active desorption Hollow cathode plasma radical source No kinetic sputter etch damage or debris A compact and efficient plasma source Cleaning in minutes for days of perfect imaging and analysis

Need clean samples and chamber surfaces? Let us show you the

“fastest way to pristine”™!

29

Mic

ro

sco

pyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Bruker has launched the Luxendo TruLive3D Imager light-sheet imaging system, which features dual-sided illumination and efficient light collection for fast imaging of biological samples in native 3D environments.

According to the company, the TruLive3D Imager leverages the general benefits of single-plane illumination mi-croscopy (SPIM) to enable rapid 3D imag-ing with minimal light exposure, confocal resolution and excellent contrast in 3D.

An extended sample chamber enables multi-sample experiments and an inte-grated environmental chamber ensures long-term live observations of even the most delicate samples such as stem cells, primary human cells and organoids.

“Our primary breast cancer mouse model allows us to specifically induce and control cancer initiation and progression in 3D or-ganoids,” says Dr Martin Jechlinger, Group Leader at the European Molecular Biology Laboratory (EMBL) in Heidelberg, Germany.

“We have been using light-sheet imaging successfully in the past and the new Luxendo system allows us to scale our experiments to hundreds of organoids per imaging session and marks a quantum leap for our research,” he adds.

As Bruker points out, the Luxen-do TruLive3D Imager system

maintains the ease of use and stability of the InVi SPIM and

is optimized to allow fast 3D multi-sample

imaging of live specimens.

The optical concept, with dual-sided

illumination and single-lens detection

from below, enables fast acquisition, high-resolution

imaging, and minimal shadowing effects, while a wide-field imaging option facilitates sample positioning.

The large sample chamber (75 mm in length) can accommodate up to 100 samples into the chamber trough and is ideal for multi-position imaging of small embryos, such as Zebrafish, Drosophila, and mouse, or 3D spheroids.

To facilitate sample mounting for inverted light-sheet microscopes – InVi SPIM and TruLive3D Imager – Luxendo’s new TruLive3D Dish series is optimised for light-sheet imaging of cells, 3D cell culture systems, organoids, and small animals.

The sterile, dual-well dishes are ready to use and customisable, and samples can be grown or placed into holders as with standard petri dishes.

The new dishes also integrate into the large chamber of the TruLive3D Imager, which can fit up to three dishes to maxi-mise throughput.

UK researcher, Dr Philippe Laissue from the University of Essex, has won first place in the Nikon Small World in Motion Competition, one of the world’s largest scientific video competitions.

Laissue’s beautiful, winning video shows a coral-reef building polyp emerging in low light from Staghorn coral. The photosynthetic algae that live inside the coral, in a mutually beneficial relationship, can also be seen.

To image corals, Laissue built a bespoke light-sheet fluorescence microscope to allow gentle observation of the emerging polyps.

“Dimming the light has enabled me to show the coral’s dynamics close up, and illustrate the beauty

and otherworldliness of these ancient organisms,” says Laissue. “At the same time, we can collect important information about what is happening on the cellular level when corals react to different environmental conditions.”

“This helps us to better understand corals and their development, thus contributing to finding the best strategies to protect and conserve them,” he adds.

Laissue also carries out research at The Marine Biological Laboratory, University of Chicago on local cold water coral, Astrangia poculata, and is continuing development of the light-sheet microscope he used for the winning video here.

Researchers from the National University of Singapore (NUS) have joined an international team of scientists to map an entire human brain at sub-cellular resolution using synchrotron x-ray microscopy.

Image acquisition will take place at 1mm3 per minute and at 0.3 micron resolution, with researchers intending to complete the map by 2024.

The work will link the synchrotron facilities in the Asia Pacific region under a collaboration called Synchrotron for Neuroscience Asia Pacific Strategic Enterprise – SYNAPSE – which is expected to involve more than 1000 researchers from Singapore, Japan, South Korea and Taiwan,

Australia and China.According to project co-founder, Professor

Low Chian Ming from the NUS Yong Loo Lin School of Medicine, the overall data acquisition and processing speed is more than ten times faster than current methods such as super-resolution microscopy or magnetic resonance imaging.

“Globally, brain mapping has gained impetus due to the growing impact of brain diseases,” says Low. “The images captured with unprecedented speed, clarity and granularity by SYNAPSE will form an extensive human brain map.”

“Our findings could potentially contribute to effective treatment for increasingly

important neurodegenerative pathologies such as Alzheimer’s disease and other forms of dementia,” he adds.

SYNAPSE will complement its structural map with subcellular and molecular information from other imaging techniques such as infrared spectromicroscopy and cryo-electron tomography.

Launched on 15th January 2020 at the NUS Shaw Foundation Alumni House Auditorium, the founding members of the initiative signed a Memorandum of Understanding (MOU) committing to work together to complete their brain map project by 2024.

As mapping a human brain will generate

a huge amount of data, a second MOU has been signed by the SYNAPSE members to implement a High Performance Computing network to rapidly process, store, mobilise, access, and analyse such data.

Singapore, which will leverage the petascale supercomputing resources at the National Supercomputing Centre (NSCC), will be the data hub of SYNAPSE.

The data hub will link all the SYNAPSE partners via the high-speed 100G international network connections of the Singapore Advanced Research and Education Network (SingAREN).

Image:Yong Loo Lin, School of Medicine, NUS

Asia-Pacific effort to map human brain in four years

Bruker introduces Luxendo TruLive3D Imager

CORAL-REEF building polyp captured using custom light-sheet fluorescence microscopy

Coral growth wins Nikon Small World

Philippe Laissue, University of Essex/Nikon Small W

orld

What’s new 29

30 M

icr

osc

opyand

An

aly

sis

Ja

nuar

y/Fe

brua

ry 2

020

Protochips started out as a semiconductor fabrication business but now supplies supports and systems for in situ electron microscopy – why the change?Protochips was founded by myself and David Nackashi in 2002 as a semiconductor fabrication services provider for small companies and academic groups, and through this work, we were introduced to several customers from the electron microscopy market. As electrical engineers with semiconductor experience, we realised that our ability to design and manufacture custom semiconductor products met a growing need for thin-film sample supports with unique properties. Given this, Protochips pivoted from a fabrication services company to a product company focused on providing in situ systems to the electron microscopy market using our semiconductor technology. Since then, we've released the Fusion, Poseidon, and Atmosphere product lines and, most recently, the AXON software platform for in situ electron microscopy.

How did you develop the in situ sample supports?Although myself and David Nackashi are not microscopists, we knew that in situ capabilities like heating, biasing, and confining liquid and gas had to be miniaturised to fit in the TEM. We realised that one way to do that was to integrate these capabilities onto custom micro-electromechanical system (MEMS) chips, known as E-chips. Nanoscale semiconductor films can support a sample and provide a 'window' against the vacuum of the microscope while also being thin enough to allow electrons and x-rays to pass right through. Additional processing also turns these windows into heaters – and by stacking two chips we can contain a liquid or gas in a closed cell, creating an environment much more relevant to samples than the high vacuum of a TEM column. In

S/TEM imaging of samples in liquid. However, using the same holder with a different pair of chips – this time with microfabricated electrodes on and off the window – enables electrochemistry experiments. Also, using the same holder with yet another pair of chips – this time with a heater on the chip – enables studies of samples in heated liquid. These capabilities provide incredible flexibility for researchers.

As well as sample supports, you provide three different in situ cells – what are these?Poseidon Select is a closed-cell system [using Poseidon E-chips] for imaging samples in native hydrated environments within the TEM. This liquid flow cell can be used to understand relatively unexplored phenomena such as aggregation and self-assembly of nanomaterials, failure mechanisms of lithium ion batteries and the interactions of different biological materials, such as cells with viruses

Meanwhile, Fusion Select is designed for in situ heating and electrical characterization within the TEM, and is used in material sciences and semiconductor markets. Researchers use the system to understand how, says, chemical composition, crystallinity, and defects change under extreme conditions, including rapid temperature changes up to 1200°C.

Lastly, Atmosphere is a closed-cell system for imaging samples at atomic resolution under different gaseous environments within the TEM. This

system targets catalysis markets as well as corrosion and fuel cell markets and can recreate realistic operating environments.

You are also developing sample substrates – Cryo-Chips – for cryo-EM; why? Cryo-electron microscopy is a growing field and we currently sell a holey carbon grid called 'C-flat' as a sample support in this market. Cryo-Chips are a semiconductor replacement for the holey carbon grid. Our prototype has interesting properties that may make it attractive to structural biologists, so we’re evaluating its potential as a new product.

You've just launched the AXON software platform for in situ analysis – what is this?'AXON' is designed to connect the in situ systems with the microscope and the detector for a seamless workflow that enables researchers to observe material dynamics at a level of detail that we believe was previously impossible. We believe the software is a game-changer in the in situ market – it brings all the focus to the sample and will also make the technique much more accessible to a wider market.

How has the electron microscopy market changed in the last decade?The in situ market has grown significantly over the last decade. More people are using electron microscopy as part of their research work, and more of these people are seeing the strong value that in situ studies can provide.

The market is also shifting towards workflow solutions. Rather than focusing only on resolution or other specifications, users want the entire workflow to drive the highest-quality data. They want sample preparation to be quick, easy, and reliable, and they also want the process of imaging, analysis, and data publication to be efficient. We believe our products help to support these goals. For example. our electrical E-chip for FIB lamella has been designed to convert a risky sample preparation process into one that is easy and reliable, providing quality results in less time. Also, AXON software helps to lower the barriers to using in situ systems by automating tedious tasks such as compensating for sample movement. Such workflow-centric product solutions have followed market demand.

What are your plans for the future?We believe that the in situ market will continue to grow because most people want to see their samples in their natural environment, not the vacuum of the microscope column. Given this, we're particularly excited about the future for our software, AXON. Workflow solutions that bring together all the major components of the instrument – the microscope, the detectors and the TEM holders – have the power to transform in situ microscopy.

In situ systems with a difference

From sample holders to closed cells, Protochips develops in situ systems to make life easier for the electron microscopist. Co-founder and chief technology officer, Dr John Damiano, tells Microscopy and Analysis about his company

FOUNDING FATHER John Damiano

this way we built the platform for a strong business that provides a variety of capabilities to users without the need to customise products.

Have you an example of how E-chips can be used in in situ electron microscopy?Our Poseidon holder uses pairs of chips with windows to enable basic

OBSERVATION of TiOx overlayer under different conditions using Protochips Atmosphere system Nano Lett. 2016, 16, 4528-4534.

Company spotlight

TESCAN AMBER X Ĭ High throughput, large area FIB processing up to 1 mm.

Ĭ Ga-free microsample preparation.

Ĭ Ultra-high resolution, field-free FEG-SEM imaging and analysis.

Ĭ In-lens SE and BSE detection.

Ĭ Resolution optimization for high-throughput, multi-modal FIB-SEM tomography.

Ĭ Superior field of view for easy navigation.

Ĭ Essence™ easy-to-use, modular graphical user interface.

A unique combination

of Plasma FIB and field-free

UHR FE-SEM for multiscale

materials characterization

1 mm cross-section through a Li-ion battery electrode

200 µm

TESCAN AMBER X is based on the S8000 platform.

PLEASE CONTACT US FOR MORE INFORMATION...

EMS has it!

P.O. Box 550 • 1560 Industry Rd. Hatfield, Pa 19440 Tel: (215) 412-8400 • Fax: (215) 412-8450 email: info@emsdiasum.com or stacie@ems-secure.com

OUR MAIN INTERACTIVE WEBSITE:

www.emsdiasum.com

The premier holey carbon sample support for automated S/TEM imaging and metrology

CD-FLAT ™

Designed for Metrology CD-Flat is designed for CD-TEM metrology of large specimens, like 3D NAND FIB lamella.

Dedicated Service Our best-in-class service team is ready to assist you with dedicated support and fast response times.

Fast Delivery Available now! New orders ship within 4 weeks, and regular customers can place standing orders to ship on a monthly schedule.

Follow us on...

Specifications

EMS CATALOG NUMBER: CDFT823-50 Copper Grid, 300 Mesh, 3.05 mm OD Holey Carbon, 40 nm thick 8/2 Hole Pattern (8 um diameter, 2 μm spacing between holes) Ships in packages of 50

NEW 8/2 HOLE PATTERN...

The C-flat ™ advantage in a new patternCD-flat™ — the premier ultra-flat holey carbon grid is ready to use with no additional cleaning or handling steps, and is now available, featuring a NEW 8/2 hole pattern.

CD-Flat Ad_Nov2019_MA.qxp_CD-Flat Ad_Nov2019_MA 1/3/20 8:38 AM Page 1