Multi-photon microscopy without scanning for faster than video-rate fluorescence imaging of live cells

During the last twenty years, there has been an explosion in new microscopy techniques which exploit the high peak intensities from laser sources for excitation of fluorescent dyes used as markers in live cells. These methods, which are based on nonlinear optics, offer several advantages for the biologist over more traditional imaging techniques. These include imaging of deeper tissue thanks to longer excitation wavelengths, avoidance of damaging short-wavelengths, and an overall reduction in photo-bleaching. However, it has been generally accepted that these nonlinear microscopy methods must use a laser focused to a tiny spot which is then scanned around the specimen. This limits the capture rate of information to around 1 frame/second. This is a major limitation to the method for studying live cells, since rapid and important changes in the intra-cellular biochemistry are often missed.

A few methods for increasing the imaging speed of nonlinear microscopy have been demonstrated, but only one is commercially available (which is essential when the technology is to be used in a biology research laboratory). This technique involves splitting a single high-intensity laser beam into up to 64 lower intensity 'beamlets' which are then scanned around the specimen, but this unfortunately can result in a 'patchwork quilt' effect which introduces unwanted artifacts into the images and can render interpretation and analysis difficult.

To provide the advantages of nonlinear microscopy but at fast capture speeds, we propose to capitalize on innovations in sensor technology and use a less well-focused laser beam, which will illuminate the full image field. This 'wide-field' method is known to biologists, but in a linear (single-photon) rather than nonlinear (two-photon) approach, and therefore is a simple adaptation to existing instrumentation that is familiar to the end-user. The key difference in our technology over a conventional fluorescence microscope will be the light source, which we will change from a light-emitting diode to a high peak intensity laser (which we already have in our laboratory). We will also use small modifications to the microscope and add a sensitive scientific camera detector. Our calculations show that nonlinear excitation of fluorescence is possible at capture speeds of up to 100 frames/second. We will test this new technology with non-biological specimens initially, and then apply the method to two different cell types to study both fast and slow calcium signalling events.

If we are successful, this technology is almost certain to change how cell biologists obtain images of their specimens which, in turn, will likely have a long-term impact on pharmacology and the development of new medicines.

Since the first experimental report of two-photon excitation microscopy (TPEM), it has generally been assumed that point-scanning of high peak intensity lasers are needed to generate sufficient fluorescence signal to create an image. The point-scanning TPEM method has proven successful for imaging live tissue with localised photobleaching, but the need to scan the beam prevents the study of fast events. Researchers have used a complex beamsplitter arrangement to provide up to 64 individual scanning beams to increase the speed of TPEM, but this system requires careful optical alignment and discrepancies between the peak intensity of independent beams can lead to artifacts in the data. Consequently, single-photon excitation microscopy in either wide-field epi-fluorescence mode or confocal single-line scanning (XT) mode is most commonly used for imaging fast events in live cells. In such work, however, photo-bleaching of fluorescent molecules limits both the image quality and duration of recordings, and often it is necessary to include fast shutters in the beam path to reduce the dose of light applied to the specimen.

We propose to develop a simple technology to perform fast TPEM of live cells by using a wide-field illumination, similar to that used in single-photon epi-fluorescence microscopy. Our calculations show that using a modified epi-fluorescence microscope, an ultra-short pulsed near-infrared laser and an sCMOS camera (all commercially available) wide-field two-photon excitation of fluorescence may support imaging over a field of >100 microns in diameter at rates of up to 100 frames/second. We suggest that this method will offer many of the advantages of point-scanning TPEM, but with an imaging speed comparable to single-photon epi-fluorescence imaging. If successful, this may be a transformational technology for biomedicine because of reduced cost and complexity of the instrumentation needed and the ability to study dynamic cellular events more easily.

To explain who will benefit from this research and why, we will consider each group of beneficiaries in turn:

The academic beneficiaries of the proposed technology development project are primarily biomedical researchers, optical physicists and materials scientists.

Almost all areas of biomedical research where microscopy is already used will benefit, and the benefit will be felt immediately since the same specimens and reagents may be used. However, reduced photo-toxicity is anticipated by using the proposed wide-field two-photon microscope, and this will prolong the duration of imaging experiments to make possible longer-term pharmacological studies in live tissue without compromising specimen integrity.

The development of this new imaging method provides opportunity for optical physicists and material scientists to further improve what will be possible through advances in laser technology and fluorophore design, which will be to the benefit of biomedical research.

Public sector beneficiaries will, in the longer-term, include healthcare professionals. Locally, the new South Glasgow Hospitals campus will house a new biomedical imaging facility and we are already in discussions with the NHS regarding new technologies which may offer improvements over existing methods. If successful, this technology development may be made very rapidly available to local biomedical researchers.

Private sector beneficiaries are anticipated, once the phase of academic development is complete. The UK laser research and development industry is buoyant, and the PI has excellent links to the key manufacturers of ultra-short pulsed lasers. Through organisation of the EMBO Practical Course in Advanced Optical Microscopy, the PI also has good relationships with manufacturers of microscopes and photo-detectors both in the UK and overseas. Despite the fact that the initial work on two-photon excitation in microscopy was done overseas, this work is likely to steer the focus of biomedical optics towards the UK. Moreover, the pharma industry will benefit from new technologies to aid in screening of new biomarkers and new medicines.

The general public will benefit ultimately through discoveries in fundamental biomedical science and through the use of wide-field two-photon microscopy for the in vitro, and possibly in vivo, evaluation of therapeutics.