Super-resolution microscopy for the masses

Editorial

Rebecca Pool

Wednesday, August 17, 2016 - 14:15
Shrinking super-resolution microscopy: Nanoimager CEO, Jeremy Warren with co-inventors Bo Jing and Professor Achillefs Kapanidis (right).
 
Earlier this year and following a hefty £1.2 million cash injection, UK university spin-out, Oxford Nanoimaging, unveiled a super-resolution microscope with a difference.
 
Only 21 by 21 by 15 cm in size, the desktop single-molecule fluorescence microscope - 'Nanoimager' - is set to bring super-resolution to beginners and accomplished researchers, alike, at a fraction of the cost of conventional super-resolution microscopy.
 
As Professor Achillefs Kapanidis, Nanoimager pioneer and leader of Biological Physics group, Gene Machines, at the University of Oxford, says: "Users can see conformational changes of single molecules in real-time, in their field of view as they happen, which is nothing short of amazing."
 
"At the same time, you can pick up the microscope, pack it in a suitcase and carry it by hand between labs," he adds. "It's robust, it's not going to be something that only works some of the time as we see with high-end microscopes; it's a work-horse and it's affordable."
 
Nanoimager, measures only 21 by 21 by 15 cm and is set to democratise super-resolution microscopy.
 
Nanoimager is based on the phenomenon of Förster (or fluorescence) resonance energy transfer (FRET) and follows eight years of development by Gene Machines researchers at Oxford, and a lifetime of research from Kapanidis.
 
The researcher first came across FRET during the mid-nineties, while developing fluorescence methods for ensemble and single-molecule analysis of protein-DNA as part of his PhD.
 
The FRET method analyses energy transfer between donor and acceptor fluorophores to calculate distances; a multitude of molecules can be analysed to determine structural dynamics of an overall macromolecule. 
 
Come his post-doctoral fellowship, Kapanidis was hooked, saying: "What I saw here was the ability to do FRET at the single molecule level with one donor and one acceptor."
 
"I thought this was amazing as there were so many things you could do," he adds. "You could see [molecule] dynamics, rates of motions and you could potentially see the entire machine going through these motions during a chemical reaction. This was fascinating and there was no way you could do it with ensemble methods."
 
Moving super-resolution microscopy forward
Kapanidis spent the next few years at Lawrence Berkeley National Lab and the University of California, Los Angeles, developing single molecule FRET, as well as single molecule fluorescence spectroscopy, to study proteins.
 
Then in 2005, he took a Lecturer position in Biological Physics at the University of Oxford and set up his Gene Machines team, with development of single molecule FRET and super resolution imaging methods continuing at speed.
 
"When I started my lab, we built large microscopes that were able to detect single molecules on a surface," points out Kapanidis. "We needed an optical table, the microscope was housed in a laser-controlled room and required a couple of computers to run it... so we wound up with a solutions that took up a lot of space"
 
"But one of the questions that I asked, especially coming to Oxford where space is at a premium, was 'can I build two microscopes that can fit in the space of one," he adds.
 
The race to miniaturisation had started. Soon Kapanidis and his colleagues were scrutinising which components, from conventional single-molecule fluorescence microscope set-ups, could be eliminated, and also, which advances in technologies could be adopted.
 
"Cameras had got smaller as had lasers; for example, you no longer needed air-cooled lasers that took a lot of space and required special infrastructure," he says.
 
"Then we started wondering, 'do I actually need the microscope body or is it just a glorified objective holder'?" he adds. "And then we said, 'well, maybe we need to build an equivalent system that moves away from the existing microscope components."

'We had shown we could squeeze a lot of optical components into a small space.' Achillefs Kapanidis

By 2011, the Gene Machines team had developed an early prototype, a system with a footprint of around 60 by 35 by 35 cm that could be used for two-color imaging.
 
As Kapanidis highlights: "We had shown we could squeeze a lot of optical components, for what we wanted, into a small space. However, [the instrument] was not stable, wasn't robust, could be misaligned relatively easily and was prone to vibrations as well as lateral or focal thermal drifts."
 
Undeterred, development continued with Kapanidis's graduate student, Bo Jing, leading a major redesign.
 
Determined to maintain a small optical path while eliminating vibrations and drift, researchers developed what they call a 'motherboard', a 20 by 20 by 1.5 cm aluminium plate, to which they had bolted detection components, many excitation elements and more.
 
"We had the main optics for steering the beam to the sample, the objective, the sample, the scanning stage, the autofocus optics, and the imaging optics for the camera here," explains Kapanidis. "Everything was either on the top or bottom of the plate... which gave excellent beam stability and cancelled out drift and vibrations."
 
Crucially, the researchers used a gel to isolate the plate from the housing, which as Kapanidis says 'was an ingenious solution and very cost efficient'.
 
And come 2015, a commerical prototype included a closed loop scanning stage, 3D dual-colour imaging and a cooled sCMOS camera, enabling the researchers to apply an algorithm to correct heterogeneities in pixel noise and increase the certainty of localising molecules.
 
Nanoimager captures actin in MDBK cell, stained with Alexa647-phalloidin and images at a resolution of 11 nm.
 
The current commercial prototype system is now described as the most commercially advanced FRET solution.
 
For example, localisation-based super-resolution imaging is claimed to increase the level of detail around ten-fold relative to traditional fluorescence microscopy while automated focusing and sample exploration allow fast imaging.
 
But, as Kapanidis points out, miniaturisation has not come without its compromises; the current commercial prototype will, for example, only house a single camera. However, a wide combination of lasers is feasible while and the user has the ability to scan a wide field of view and change illumination angle; millions of single molecules can be captured in a single acquisition.
 
"The system enables many applications," asserts Kapanidis. "For example, we can do single molecule FRET, single molecule tracking, colocalisation analysis with two or more colours."
 
"It's an affordable instrument and you could imagine researchers having different versions for different applications," he adds.
 
So with the product and the team already in place, the future looks bright for Oxford Nanoimaging. Chief Executive, Jeremy Warren, joined in February this year, bringing a wealth of start-up experience while Nanoimager demonstrations on fixed and live cells at the Micron Oxford Advanced Bioimaging Unit have proven very successful.
 
At the same time, Wellcome Trust investigator, Professor Sivaramesh Wigneshweraraj, from Imperial College London, is using the instrument to investigate the initiation of gene transcription by RNA polymerase, critical for designing antibiotics to block this process.
 
E. coli cells expressing a YFP-labelled component of the Twin-arginine transport (Tat) complex, which transports molecules across the cell membrane. Bacteria illuminated at 532 nm in HiLo mode, 0.5 micron across.
 
Indeed, Oxford Nanoimaging has already received its first two system orders from leading UK laboratories.
 
"We are already a team of 12 people supported by our funds, which we will also use to build additional demonstration instruments and get the components for the first commercial units," says Kapanidis.
 
"I'd like to see instruments based on the Nanoimager concept in clinics, food industry plant and environmentally-controlled stations," he adds. "We could also see teaching versions to help train students in single molecule imaging, single molecule FRET as well as the measuring modalities that are going to dominate biophysical; and biochemical measurements in years to come."

 

Website developed by S8080 Digital Media