Taking a deeper look...

Multiphoton imaging is taking cellular and tissue imaging to new depths. Jenni Lacey gives us an overview of this powerful tool and introduces a research group at Imperial who are using it to observe complex patterns of neuronal interaction

Multiphoton imaging has become an increasingly popular technique within the biological sciences, providing exceptional imaging of live cells in thick, light-scattering samples. It is a non-linear form of fluorescence microscopy, which uses pulsed long wavelength light to excite dyes within a sample in order that three-dimensional images can be formed. It offers notable advantages over traditional fluorescence imaging and confocal techniques, providing high resolution, rapidly acquired images of structures up to 1.6mm deep into a sample.

This depth penetrat­­ion is all the more impressive when you consider the dense and turbid nature of samples being used, ranging from brain slices and intact organs to live animals. This technique enables researchers to observe discrete, functionally diverse structures and helps to build a comprehensive view of biological processes and cell communication. The impressive resolving power enables researchers to distinguish the finest biological structures such as dendritic spines and synaptic boutons. It can also provide quantitative dynamic information on cell processes.

At present ‘Multiphoton imaging’ is used interchangeably with the term two-photon microscopy, yet, three-photon and four-photon methods can also be achieved, though their applications have not been as widely explored in the life sciences. The foundations for two-photon microscopy were set over 80 years ago when theoretical physicist Maria Goeppert-Mayer first described the theoretical basis in her doctoral dissertation¹. However, it wasn’t until 1990² that the full potential was realised and its application into live cell imaging was discussed by Winfied Denk in his paper published in Science titled ‘Two-photon laser scanning fluorescence microscopy’.

[caption id="attachment_32672" align="alignleft" width="200" caption="Visual cortex (Credit: Kate Smith (Feinberg School of Medicine, Northwestern University) & Javi Munoz-Cuevas (Gallo Research Center, UCSF)"][/caption]

Two-photon excitation relies on the simultaneous absorption of two photons by a fluorophore (dye) elevating it from a ground state to an excited state and resulting in an emitted fluorescent signal. A crucial distinction to make is that two-photon excitation utilises lower energy, longer wavelength light, usually near-infrared; compared with typical fluorescence imaging using light in the ultraviolet or blue/green spectral range.

A femtosecond laser delivers a pulsed beam to the sample and causes two photons to arrive simultaneously (in one quantum event) and summate in energy to excite the dye. Due to the inverse relationship between photon energy and wavelength, the two absorbed photons must have a wavelength approximately twice that required for typical fluorescence excitation. The use of lower energy, long wavelength light helps achieve two of the key benefits offered by this technique; reducing damage to the sample and penetrating deeper into the tissue.

When the theory of two-photon excitation is applied to three dimensional live cell imaging it provides a number of significant advantages over other techniques such as confocal microscopy. It offers outstanding optical sectioning and rapid image acquisition from thick (often difficult to image) specimens.

The impressive optical sectioning can largely be attributed to the discrete localisation of fluorescence excitation; this is limited to just the focal plane. Two-photon excitation relies on defining a single point, where the laser beam will be focused by the microscope optics; it is only at this focal point that the spatial density of photons is great enough to excite a single fluorophore. As a result, a signal can only be generated by this narrow region and there is no emission signal from out-of-focus regions of the sample.

The reduction in out-of-focus fluorescence leads on to, and in part, contributes to the enhanced imaging depth. With no light being absorbed by out-of-focus regions, crucial photons from the excitation beam will reach the desired region of the sample and contribute to a stronger signal. By contrast, confocal microscopy does not prevent excitation of out-of-focus dye so the intensity of photons reaching the region of interest is reduced by prior absorption within the light path. Confocal techniques rely on a pinhole to reject the out-of-focus signal and therefore images suffer from background noise obscuring signals from deep within a sample.

[caption id="attachment_32673" align="alignleft" width="200" caption="Visual cortex (Credit: Kate Smith (Feinberg School of Medicine, Northwestern University) & Javi Munoz-Cuevas (Gallo Research Center, UCSF)"][/caption]

The highly discrete region of excitation with Multiphoton imaging also results in reduced photobleaching of the sample, increasing the sample’s longevity and duration of experiments that can be carried out. The property of the light used also contributes to improved depth penetration compared with other techniques. By utilising longer wavelength light, the excitation beam suffers less from scattering, therefore allowing increased photons to travel deeper into the sample.

With the basic theory of this technique described it is then necessary to consider the equipment needed to optimise its implementation; creating a system that draws these principles together. The crucial first stage is the scanhead unit which is used to raster the laser beam across the objective field. This is an electronic control unit designed to scan the laser beam over a defined region focused by the microscope’s optics.

There are different categories of scanners available; Acousto-Optic Deflector (AOD) based, Resonance Scanner or Galvanometer based systems, offering different advantages in speed, resolution and spatial accuracy. The intricacies of each approach help to provide the versatility which makes Multiphoton a popular choice within life science applications.

The second essential element of this system is light collection of the emitted signal from the sample. This is achieved by Photomultiplier Tubes (PMTs) which utilise the internal photoelectric effect; collecting the light signal then converting and amplifying the photoelectron signal. PMTs offer a wide, tuneable spectral response, fast response time and high sensitivity. This makes them ideal for low-level light detection either above or below the sample.

A digital image can then be processed and analysed by a suitable workstation that uses software to assemble three-dimensional images from optical sections. A final element to consider is the collection lens used to focus the light down onto the PMT, this can be achieved either through an objective or condenser for ‘above stage’ or ‘sub-stage’ signal collection.

To understand the true capability of Multiphoton imaging, it is useful to consider a real example of its application. Almost 20 years ago Winfried Denk highlighted the “ubiquitous and rarely mentioned” power that light microscopy offers to electrophysiological studies; in the past decade the combination of these techniques has rapidly grown.

Dr Simon Schultz from Imperial College London has been using Multiphoton imaging since 2003, when he worked as a postdoc in the laboratory of Prof Michael Häusser. At this time, Multiphoton imaging was in its relative infancy but Dr Schultz recognised its potential to help explore questions around neural coding and processing of sensory stimuli within the nervous system. This work has several practical applications, including understanding how brain circuits function after brain disorders or traumatic brain injury. The brain has inbuilt circuits for repair and plasticity and by understanding how these mechanisms work and affect behaviour, we can see how best to take advantage of these for therapeutic benefits.

A second area concerns gaining insight about how the nervous system solves parallel information processing problems. This can be used by engineers and computer scientists working on the development of new distributed computer architectures.

Dr Schultz combines a variety of techniques in his laboratory, including electrophysiology and Multiphoton imaging of calcium signals, to explore the complex patterns of neuronal interaction. Dr Schultz’s team primarily studies regions of the neocortex and cerebellum, recording the electrical signal and observing the individual calcium signals generated by populations of nearby cells. This enables them to describe patterns of signal propagation and neural response to sensory information.

Two photon imaging also allows them to label individual elements deep within the sample and target these cells for electrophysiological recordings. Traditional electrophysiology is performed blind but by additionally filling a pipette with dye and imaging under a Multiphoton microscope, the team is able to guide the pipette to a specific visualised area of the circuit. Combining these techniques neatly creates a picture of network synchrony and spatial patterns of coding.

Dr Schultz describes one of the main advantages of Multiphoton imaging as the ability to spatially, and discretely, identify a neuron’s signal within the cortical circuit, allowing them to actually see the signal within its place of origin. This is a powerful demonstration of the precise optical sectioning offered by this technique. The ability to rapidly acquire images and record functional calcium signals, also demonstrates the speed advantage that Multiphoton offers.

As Multiphoton imaging continues to increase in popularity, coupled with technological advances reducing system cost, this technique is likely to become increasingly prevalent in laboratories around the world. The cross-disciplinary approach of life science studies means there are a wealth of applications still to be explored, but it is clear to see that Multiphoton imaging is a powerful tool for biomedical science.

Author: Jenni Lacey Marketing Executive, Scientifica.

1. Göppert, M. (1929). Über die Wahrscheinlichkeit des Zusammenwirkens zweier Lichtquanten in einem ElementaraktDie Naturwissenschaften, 17: 932.

2. Denk, W. (1990). Two-Photon Laser Scanning Fluorescence Microscopy, Science, New Series, Vol. 248, No. 4951, pp. 73-76