Page Type: Customer Story

Single Molecule TrackingCustomer Stories

Prof. Ulrich Kubitscheck

The Biophysical Chemistry Workgroup, Rheinische Friedrich-Wilhelms Universität Bonn, Germany

Background

The lab of Prof. Kubitscheck performs a wide range of imaging across different applications including single-molecule tracking and localization, where they examine the export of ribosomal particles through nuclear pore complexes by a number of methods. These methods include indirectly HaloTag labeling proteins that become part of pre-ribosomal subunits and imaging the DEAD-box helicase protein DBP5 that has a role in regulating the export of mRNA and pre-ribosomal particles. DBP5 is known to remove transport factors from mRNA in nuclear pore complexes, but further research is needed in order to uncover the molecular mechanisms and kinetics behind this export process.

Figure 1: DBP5 molecules in a HeLa cell, taken with the Prime BSI. The HeLa nucleus is labeled with NTF2-eGFP (red) and DBP5 molecules labeled with DBP5-Halo-JF549 (green). NTF2-eGFP marks the nuclear envelope, with three DBP5 molecules seen binding to the nuclear pores.

Challenge

As outlined by Prof. Kubitscheck, “The DBP5 protein diffuses around in the cell, attaches to the nuclear pore complexes and dissociates, these processes occur extremely rapidly and the challenge is to catch that in real-time.” This imaging application requires a combination of both speed and sensitivity, in order to capture the complex kinetics of the DBP5 protein while retaining a good signal to noise ratio to observe the samples and perform quantitative analysis. Furthermore, the nuclear pore complexes need to be visualized with high precision.

In addition, imaging was limited to a small field of view (FOV) due to the previous EMCCD solution requiring greater magnification.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

The lab of Prof. Kutischeck previously used EMCCDs for single-molecule imaging, but now have made a switch to sCMOS, making the most of the high speeds and larger field of views while retaining high sensitivity due to the near-perfect 95% quantum efficiency.

As imaging at high speeds with high sensitivity was a necessity, Prof. Kubitscheck turned to the BSI. “With the Prime BSI we can go to very high framerates (200 Hz) and still see a complete cell nucleus, before that we used an EMCCD, but due to the larger pixels and magnification stages we had to use, we could only see a small fraction of the envelope with the restricted field of view.“

Light Sheet MicroscopyCustomer Stories

Prof. Kishan Dholakia

Dr. Stella Corsetti,

Optical Manipulation Group, School of Physics and Astronomy, University of St. Andrews

Background

The Optical Manipulation Group at the University St. Andrews, led by Prof. Kishan Dholakia, looks at the science of light in terms of physics and biomedical research, fundamental and applied, in order to further the agenda for healthcare applications.

Prof. Dholakia wishes to take very high-quality images, in his terms: “A large field of view with high resolution, very quickly, very accurately, using the lowest possible photon dose.” For these reasons, the group uses light-sheet microscopy for the ability to take images with a lower light dose over a large field of view, with an aim to shed light on numerous biological processes.

Figure 1: Taken with the Iris 15, this image shows mouse embryos at the 2-cell stage, stained with mitochondrial stain JC1, and imaged using an open-top light-sheet setup with a resolution of 900 nm.

Challenge

With live samples, the need to avoid photodamage is necessary in order to avoid affecting biological functions, when samples need to be imaged for days. Prof. Dholakia outlined some other issues: “We had challenges in how do we create the field of view, and how do we do Nyquist sampling across this field of view succinctly and accurately.”

Previous CMOS solutions were not able to deliver the desired image quality, meaning that a new solution was needed in order to achieve the large field of view images while retaining enough sensitivity to use lower light levels and preserve live samples for longer, as well as the ability to image with high resolution at low magnifications.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

Prof. Dholakia and the team are now using the Iris 15 sCMOS, with features a rectangular 25 mm field of view, ideal for light sheet imaging. As Prof. Dholakia stated: “What we found exciting was the very small pixel size, which means we can retain the resolution but still get the big picture.” The small 4.25 μm pixel size allows for high resolution at low magnifications with Nyquist sampling, in combination with the large field of view this makes the Iris 15 an ideal solution over other sCMOS cameras.

Postdoc Dr. Stella Corsetti also outlined some of the advantages of the Iris 15: “The Iris15 camera gives us the advantage to Nyquist sample the resolution over a wide field of view due to the small pixel size.”

Prof. Dholakia also liked the camera dimensions: “The size of the camera is quite advantageous, we have been using other sCMOS cameras but the smaller physical size of the Iris 15 is important, we are interested in building platforms of microscopes that can fit on small breadboards, space is a premium and the Iris 15 fills a need there.”

Whole Live Organism ImagingCustomer Stories

Dr. Simon Berger

Prof. Alex Hajnal

Department of Molecular Life Sciences, University of Zurich, Switzerland

Background

The laboratory of Prof. Alex Hajnal study the model organism Caenorhabditis elegans, the nematode worm, in order to understand biological processes such as organogenesis, the mechanisms involved, and how these relate to other areas such as cancer research. Traditionally, C. elegans imaging is done by sandwiching the worms between glass slides and using drugs/chemicals to immobilize the worms for easier imaging. However, these are not physiological conditions since only an actively moving worm can develop and function normally.

Dr. Simon Berger of the Hajnal lab has developed microfluidic devices to fine-tune the experimental environment, safely securing the worms for efficient imaging while also ensuring they can behave in a normal physiological manner, without affecting normal development. Dr. Berger works with many live worms in a single microfluidic device, imaging through the transparent material and taking 3D z-stacks of worms while they are actively moving.

Figure 1: A single C. elegans worm imaged over time (from left to right 0 to 16 hours). Shown are two markers in gray/white: AJM-1::GFP (apical junction molecule marking cell-cell junctions) and PIP-2::mCherry (marking actin filaments in the worm anchor cell). The process visualized over time here is vulva morphogenesis, which takes place during the fourth larval stage of C. elegans development and is an important process to study gene regulation in a developmental context.

Challenge

When working with live organisms it is vital to keep the light level as low as possible in order to avoid photodamage or bleaching of the sample. However, without a sensitive camera, it can be difficult to collect enough signal while operating at a low light level.

In addition, as the worms are active and moving, it is vital to image at a high speed in order to avoid motion artifacts. High speed imaging would require a lower exposure time, meaning the signal levels are lower still, so this high speed is also dependent on a highly sensitive camera.

Having a large field of view would also allow for as many worms as possible to be captured in one field, streamlining experimentation. Having a small pixel size would allow for optimal resolution at lower magnifications, increasing both the potential field of view and the light level.

Finally, these experiments are set up in a complex way, with advanced triggering needed to control LEDs and the piezo movement in order to capture the worms as fast as possible.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

The Prime BSI back-illuminated sCMOS is the ideal camera for this application, offering high sensitivity, high speeds, and optimized pixel size.

As stated by Dr. Berger: “The advantage of the [Prime] BSI is you have high quantum efficiency, therefore you can image with lower illumination intensities and short exposure times for longer periods of time without damaging the animals.” Dr. Berger also enjoyed the high imaging speeds: “Comparing the Prime BSI to an EMCCD, you also benefit from being able to image much faster at higher framerates, which means even if the worm still moves while you acquire a z-stack it doesn’t matter as much, as you take it faster.”

The 6.5 μm pixel was also a benefit as it allowed for lower magnifications: “Nowadays with the [Prime] BSI I can image more or less anything with 40x and get higher intensities and field of view, this amplifies the benefit of higher quantum efficiency.” Previous 72% front-illuminated sCMOS cameras and EMCCDs were insufficient for this application due to the low sensitivity and low speeds respectively, overall the Prime BSI is the model camera for studying this model organism.

Spinning Disk Expansion MicroscopyCustomer Stories

Prof. Dr. Helge Ewers

Institute Of Chemistry and Biochemistry, Freie Universität Berlin, Germany

Background

The laboratory of Prof. Ewers moves in a number of different research directions, one of these deals with the septin cytoskeleton. Septins are a family of essential, conserved GTP-binding proteins that form heteromeric, non-polar complexes that further assemble into filamentous structures. Septins have conserved functions in cell division and form part of the contractile actomyosin structure at the cleavage furrow in dividing cells. Currently, the Ewers laboratory uses super-resolution microscopy methods to understand the nanoscopic organization of septin structures in cell division.

Figure 1: Expansion microscopy image of a dividing cell imaged on a spinning disc microscope. Shown is a genome-edited cell expressing SEPT2-GFP (magenta) from the endogenous locus that is immunostained for tubulin (teal), taken with the Prime BSI.

Challenge

One technique used to understand the septin organization in the Ewers laboratory is the new Expansion microscopy technique. Expansion microscopy achieves higher resolution by the physical expansion of the sample rather than optically. When the target molecules are spaced wider than in the native state, regular microscopes can yield a much higher resolution. Expanding the sample into both X- and Y-directions however means that imaging the same cell will take more space on a camera chip. This is a significant challenge if a cell is expanded fourfold into X- and Y-directions, which means a 16-fold increase in area that must be imaged.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

Fortunately, the Teledyne Photometrics Prime BSI camera has a very large field of view. The Ewers laboratory uses this camera on their custom-made spinning disc microscope, where it does not only provide a large field of view, but the 6.5 μm pixel size enables true sub-resolution imaging with expansion microscopy at about the same sensitivity of an EMCCD camera. “Without the large chip,” Dr. Ewers said, “we could not image the entire cell. And without the matched pixel size, we could not realize the gain in resolution through the expansion process. The results are stunning large images of septin structures in our genome-edited cells.”

Patch Clamp ElectrophysiologyCustomer Stories

Dr. Rodrigo de Campos Perin

Laboratory of Neural Microcircuitry (LNMC), École polytechnique fédérale de Lausanne, Switzerland

Background

The EPFL houses the Laboratory of Neural Microcircuitry (LNMC), which is dedicated to researching and unraveling the structure and function of neural microcircuits, particularly in the neocortex. This group is headed by Prof. Henry Markram and includes research scientist Dr. Rodrigo de Campos Perin, who is working with a combination of fluorescence microscopy and patch clamping for intracellular electrophysiological recordings. This work often uses multi-patch electrophysiology in order to record from multiple neurons.

Figure 1: The large field of view of the Prime BSI makes it easy to keep track of multiple patch-clamp electrodes, with six seen in the frame. The field of view of the previous camera solution is shown in the black square, which only shows the tips of four electrodes.

Challenge

When performing patch clamping – especially multi-patching – with live neurons it is vital to have a good view of the field in order to accurately position the multiple electrodes in relation to cells. Dr. Perin noted that: “Keeping four electrodes in the field of view with our previous camera proved challenging in our usual experiments”. This combination of fluorescence and brightfield to position electrodes means that cameras need a high dynamic range to display both high and low signals without saturation, and a high resolution is needed to more accurately determine which cellular features are being patched and recorded.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

Dr. Perin is now using the Prime BSI sCMOS camera for this application, and benefiting from the large field of view and excellent resolution. As mentioned by Dr. Perin: “We recently started using the BSI camera as a solution that allowed us to accelerate and make experiments more reliable in multiple ways.” The BSI enabled imaging of fl­uorescent signals while also positioning patch clamp electrodes, improving the reliability in identifying and then recording fl­uorescently-labeled cells.

In addition, the 11- and 16-bit modes of the BSI offer a high dynamic range, which Dr. Perin said: “Helps us concentrate on experiments without needing to adjust light intensity several times when performing electrophysiological recordings. It also provides the added advantage of more clearly visualizing anatomical features without the image saturation we experienced with our previous camera.”

The large field of view of the BSI now allows for six or more electrodes to be accommodated at high magnification, and makes it easier to target cells of interest even if they are deeper in the tissue, requiring fewer position adjustments and making experiments faster. Finally, the BSI’s 6.5 µm pixel offers excellent resolution, meaning that regions of interest can be expanded without loss of image quality, an approach much preferred to changing objectives or optically zooming as this can mechanically interfere with recordings.

Functional Calcium Retinal ImagingCustomer Stories

Prof. Maximilian Jösch

Neuroethology – Life Sciences, Institute of Science and Technology (IST), Austria

Background

Prof. Jösch is the head principal investigator of a functional imaging neuroscience group who are currently studying retinal processing in mice with a novel imaging method. The group wants to understand how the brain receives information about the surrounding world and how the information is processed and computed by the brain, using the retina as a window through which to study the brain. Many neurological diseases can be spotted early in the retina before they occur in the brain, including diseases like Parkinson’s and autism, which allows these diseases to be studied in a very controlled fashion.

They use a custom-built microscope imaging system to interrogate the function of live mouse retinal tissue via calcium (and potentially voltage) imaging.

Figure 1: A video showing calcium imaging in the mouse retina using Genetic Encoded Calcium Indicators, imaged with the Prime 95B.

Challenge

This kind of functional calcium imaging requires very high-speed imaging, with Prof. Jösch wanting to sample across the entire sensor at 100 fps in order to observe the fast dynamic functional processes in the retinal tissue.

As Prof. Jösch states: “We need to be sensitive and we need to be fast, if we are not sensitive enough we would need to have more light, and with more light we are basically destroying the tissue… since we are doing live imaging we want to avoid cooking the sample through direct light exposure.”

As the imaging takes place across live retinal tissue the largest field of view possible is desired but Prof. Jösch and team are more interested in function than anatomy: “Our main concerns are the signal to noise of the functional response so we can extract the information we require.”

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

With the speed, sensitivity, and field of view of the Prime 95B, it makes a good solution for calcium imaging and functional neuroscience. As Prof. Jösch mentions: “We wanted something that was more sensitive and faster, which is why we changed to this system [Prime 95B], and the data looks pretty nice.”

Live Cell Time Lapse ImagingCustomer Stories

Prof. Kurt Anderson

Dr. Matt Renshaw

Crick Advanced Light Microscopy (CALM), Francis Crick Institute, London

Background

Prof. Anderson is the head of the Crick Advanced Light Microscopy (CALM) facility. Prof. Anderson and senior laboratory research scientist Dr. Matt Renshaw oversee over 16 advanced microscopy systems in the CALM facility, including point scanning confocal, spinning disk confocal, multi-photon, light-sheet, TIRF, and more. CALM staff also train scientists to use these specific systems so they can better obtain quantitative imaging data for their experiments, running microscope courses frequently throughout the year.

One of these imaging systems is used for long-term time-lapse (LTTL) imaging of live cells, a microscope in a closed controlled environment system designed for up to 48 hour experiments on live samples. Researchers from all over the Crick Institute use the LTTL system for cell documentation.

Figure 1: Live cells expressing a FUCCI cell cycle sensor in order to visualize the life cycle of each cell. The image is a single time point from a longer time-lapse, taken with the Prime BSI by Jingkun Zeng of the Diffley Lab, Francis Crick Institute.

Challenge

Long term (up to 48 hours) time lapse imaging benefits from a large field of view (FOV) in order to capture as many cells as possible, often taken at low magnifications in order to capture as much data as possible with each frame. By decreasing the magnification, image resolution can be affected. In addition, sensitivity is vital in order to capture quantitative data from cell populations across these long term experiments.

In general, the challenge is that users of the LTTL system want a big field of view in order to track as many cells as possible over the long term, in order to look at mitosis and other cell behavior, while retaining sensitivity, speed, and resolution at these lower magnifications.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

The large FOV and high resolution of the Prime BSI is a good fit for the LTTL system compared to previous CCD or EMCCD solutions.

As the majority of live cell experiments involve samples that are sensitive to light, users can make the most of the high sensitivity of the BSI and drop the exposure and light intensity, decrease the bit depth to 12 bit (increasing the imaging speed) and work in low-light conditions. As mentioned by Dr. Renshaw, “we originally had a CCD solution, the smaller sensor would only capture a quarter of the FOV as the sCMOS, so would need four times the images and exposures to capture the same number of cells.” The CALM facility has access to a number of scientific camera solutions, with Dr. Renshaw also saying “we had a EMCCD with very big pixels, by changing to the Prime BSI with smaller pixels, it allowed users to reduce the magnification from 60 to 40 or 20, letting them take images with a much larger FOV without a loss of sensitivity”.

Live Cell Organelle Transport DynamicsCustomer Stories

Prof. Viki Allan

Mr. Daniel Han

Division of Molecular & Cellular Function, The University of Manchester

Background

Prof. Viki Allan is the leader of a group at the University of Manchester that studies cell dynamics, with a particular focus on intracellular transport and the endocytic pathway. Membrane organelles move around cells along microtubule tracks, driven by proteins such as kinesin and cytoplasmic dynein, and this movement of material within a cell is vital for cell function. Prof Allan and the team are studying these intracellular structures through imaging of living cells.

Figure 1: A living human MRC5 cell stained with LysoBrite Red (a lysosome marker) and imaged at 100x magnification with a widefield light microscope. The image is shown in reverse contrast. The insets show a magnified version of the area indicated with the black square at different time points. A moving lysosome is indicated with the yellow arrowheads, and the time between frames is shown.

Challenge

There are a number of challenges involved in this imaging research, mainly due to the intracellular cargo and transport networks being very small, having a dim signal, and moving rapidly.

Many intracellular membrane organelles are smaller than the diffraction limit for imaging (~200 nm) and can be difficult to resolve. These samples are labeled with low levels of GFP in order to avoid altering organelle behavior, meaning that they don’t exhibit a bright signal, making detector sensitivity a must. In addition, these transport structures can move at a rate of 8 µm/s (typically ~2-5µm/s), which makes them a challenge to track when imaging at high magnification. In order to track these objects successfully, a large field of view and a high framerate camera would be necessary.

Using a camera with the right pixel size could satisfy the Nyquist criterion and achieve a high resolution at a lower field of view, combined with high sensitivity and high speeds. The choice of camera is important when facing these challenges.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

Even considering the numerous challenges, the Prime 95B sCMOS offers a versatile platform for imaging and has been used by Prof. Allan and the team with great success.

As stated by Prof. Allan: “With [the Prime 95B] we can go at faster framerates than with our EMCCD, and we can use full frame to see as much of the cell as possible.” The Prime 95B is back-illuminated and features a near-perfect 95% peak QE: this and the large 11 µm pixel size make the Prime 95B highly sensitive and a great fit for low-light imaging. In addition, the pixel size is smaller than previous EMCCD solutions, allowing Prof. Allan and group to image at a lower magnification while maintaining a high resolution, therefore increasing the available field of view and improving light throughput. This larger field of view and the high speeds of the Prime 95B result in more efficient particle tracking, imaging as much of the cell as possible. Overall, the Prime 95B met the challenges presented by the sample, resulting in a recent publication (Han et al., 2020).

References

Han D., Korabel N., Chen R., Johnston M., Gavrilova A., Allan V.J., Fedotov S., Waigh T.A. (2020) Deciphering anomalous heterogeneous intracellular transport with neural networks, eLife 9:e52224, https://doi.org/10.7554/eLife.52224

mesoSPIMCustomer Stories

Dr. Fabian Voigt, Postdoctoral Researcher

Prof. Fritjof Helmchen

Brain Research Institute, University of Zürich

Background

The mesoSPIM initiative (mesospim.org) is an open science project aimed at making light-sheet microscopes for imaging large cleared tissue samples more accessible to the imaging community. The acronym mesoSPIM stands for “mesoscale selective plane illumination microscopy” and the project was started in 2015 by Dr. Fabian Voigt and Prof. Fritjof Helmchen at the Brain Research Institute of the University of Zürich.

The idea was to create a highly modular and versatile light-sheet microscope capable of quickly scanning cleared organs and entire organisms. The mesoSPIM can image a mouse brain with an isotropic resolution of 6.5 µm within 8 minutes and is compatible with all clearing techniques. Instructions and software to set up and operate the instrument are freely available, and 10 mesoSPIM instruments have been set up around the globe at a wide variety of institutions such as the Wyss Center Geneva and the Sainsbury Wellcome Center in London, with several more instruments under construction.

Figure 1: A full-frame max projection image of an iDISCO cleared APP/PS1 mouse brain taken with the Iris15 and benchtop mesoSPIM with a 1.2x objective (pixel size 3.5 µm). The sample was stained for amyloid beta plaques (green) and arteries (magenta), prepared by Anna Maria Reuss, University Hospital Zurich.

Challenge

For the next generation mesoSPIM, Dr. Voigt and team set out to make the setup more compact and cost-efficient by turning it into a benchtop instrument. Unlike its bigger brother, the benchtop mesoSPIM does not need a dedicated optical table which drastically lowers the required budget. In shrinking the system down to a 45 x 60 x 50 cm box, Dr. Voigt had to find an alternative to the zoom macroscope utilized in the original mesoSPIM – with an overall length of 600 mm and a weight in excess of 15 kg, it would be too large and too heavy for a benchtop mesoSPIM.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

One option was to replace the combination of a zoom macroscope and a 4 MP sCMOS camera in the original mesoSPIM with a camera with a higher pixel count and a set of fixed-magnification objectives.

Therefore, we tested the Photometrics Iris 15 and found its large detector size and rectangular sensor ideal for our applications: For example, in combination with a 1.2x telecentric machine vision objective, the microscope has a FOV of 17.9 x 10.5 mm – perfect for imaging a whole mouse brain processed using the iDISCO clearing protocol. Owing to the large pixel count, fine features can be imaged without the need for time-consuming tiling acquisitions.

In addition, the Iris 15 offers a Programmable Scan Mode which is essential to achieving near-isotropic resolution in our datasets: By making use of a technique called Axially Scanned Light-sheet Microscopy (ASLM), the mesoSPIM is capable of <7 µm axial resolution across FOVs up to 20 mm. Lastly, the small form-factor of the Iris 15 greatly helps us to build a portable next-gen light-sheet microscope.

Synchrotron Beamline MeasurementsCustomer Stories

Dr. Ji-Gwang Hwang

Department of Accelerator Physics, Helmholtz-Zentrum Berlin

Background

Dr. Hwang is participating in the Athena_e project, a collaboration between Helmholtz Centers for the technical evolution of plasma wake-field accelerators in conjunction with normal-conducting accelerating structures. Helmholtz-Zentrum Berlin (HZB) operates an advanced synchrotron light source, BESSY II, as well as the low energy storage ring Metrology Light Source (MLS). HZB and Dr. Hwang have a role in the development of high-sensitivity beam profile monitors with a micrometer spatial resolution.

Figure 1: Experimental data of interferometry fringes at 2.9×104 photons, QE of 95%, read noise 1.2 e- RMS.

Challenge

Achieving a spatial resolution of a few micrometers is only feasible by performing profile detection with interferometric techniques (slits and wavelength filters), but these techniques can reduce light intensity produced by the synchrotron by two orders of magnitude, making detection a challenge. Any detector needs to be carefully evaluated, especially for quantum efficiency (QE), read noise, and any nonlinear behavior.

The influence of QE and read noise on the determination of beam size as a function of the number of photons was estimated quantitatively by a Monte-Carlo based numerical simulation. The results showed that a high QE significantly improved the resolution compared to the read noise at low-intensity levels. In addition to this, the background noise compensation is important since the photons per pixel are from 4 to 20 which has a nonlinear response due to statistical noise. Thus a new state-of-the-art camera that has a back-illuminated sensor and low readout noise is highly demanded.

“With the Prime BSI the signal to noise ratio is much better, and simultaneously the field of view is huge compared to the previous EMCCD.”

Solution

Dr. Hwang chose the Prime BSI sCMOS for their detector needs, testing the camera on an infrared beamline at MLS.

Various filtering techniques including the BSI’s internally available despeckle function, were applied to filter speckles and singular noise. As mentioned by Dr. Hwang: “The Prime BSI is capable of retrieving interferometry fringes with a peak intensity of ~3 photons/pixel. The measurement limit owing to the read noise and QE of the detector is quite consistent with our estimation predicted by the numerical simulation.”