The next evolution in multiphoton microscopy will further enhance the ability to observe complex and dynamic biological processes from deeper within living tissue with minimal invasion and photodamage.
Multiphoton microscopy (MPM) is regarded as the method of choice for imaging of living, intact biological tissues on length scales from the molecular level through the whole organism. Additionally, multiphoton microscopy is uniquely suited to perform experimental measurements with minimal invasion over long periods of time, thereby providing exquisite detail of inherently dynamic biological processes having time scales from microseconds to days or weeks. As a result, vast quantities of data are becoming available to further enhance our understanding of complex biological interactions.
Compared to similar optical imaging techniques, MPM holds inherent advantages for imaging living tissues by improving depth penetration and reducing photodamage. This is a direct result of employing near infrared (NIR) femtosecond lasers to generate observable nonlinear signals in the visible range. The NIR excitation enhances the ability to image deeper into a sample through a reduction light scattering proportional to the fourth power of the excitation wavelength. In brief, multiphoton excitation (MPE) occurs when two (or more) photons whose sum energy satisfies the transition energy required to promote the fluorophore from ground to excited state simultaneously arrive at the sample as illustrated in Fig. 1a. The fluorescent signal can be generated from exogenous probes (Hoechst, AlexaFluor488, etc.) applied to the tissue or endogenous molecules (NAD(P)H or transfected fluorescent proteins) that are inherently expressed. Additionally, multiphoton imaging is sensitive to second harmonic generation (SHG) where two photons instantaneously convert their energy into a single photon of half the wavelength, as illustrated in Fig. 1b. SHG is not an absorptive and emissive process like fluorescence, which eliminates the fluorophore; however, SHG does require highly ordered molecular structures with particular symmetry. The most common biological structure satisfying these requirements is collagen. Example images of two-photon fluorescence (TPF) and SHG are shown in Fig. 2.
The nonlinear process governing signal generation requires power densities on the order of MW/cm2. Power densities of this magnitude are only reached at the focal plane of the objective lens, confining the observable signal to the plane of focus. Confinement of the signal to the focal plane, known as optical sectioning, greatly reduces overall photodamage by eliminating all signals above or below the focus. Femtosecond lasers have the necessarily high peak power required to maintain low average power, which minimizes sample damage. The nonlinear processes required to generate TPE or SHG are proportional to the square of the intensity (I) of the laser as described in equation 1:
With a constant repetition rate (Frep), intensity can be increased through a reduction in pulse width (τ) or by raising the average power (Pavg). Shorter pulses can, therefore, have advantages in imaging living samples.
The ability to generate images across length scales from the molecular level to the whole organism requires the ability to augment a traditional 'en face' image with depth information to reconstruct a complete three-dimensional representation of the sample. Depth information is conveyed with high spatial resolution through optical sectioning, where sequential images of a single focal plane are taken by adjusting the focus of the microscope further into the sample.
Evolution of multiphoton microscopy
To support increasingly complex experiments, the technology used to capture biological data must continue to evolve. Hence, the Thorlabs Multiphoton Microscope was designed from the ground up to capitalize on technology trends by employing resonant scanner technology for high speed imaging capabilities and low photodamage; full field of view non-descanned detection path employing high sensitivity GaAsP PhotoMultiplier Tubes PMT's positioned directly behind the objective lens; and dedicated near infrared (680–1,400 nm) scan optics that take full advantage of the latest laser and Optical Parametric Oscillators (OPO) sources.
Long-term sample viability requires that the imaging system generate the highest signals while simultaneously minimizing destruction. High speed scanning with a pulsed laser source has many natural advantages in fluorescence microscopy. In single point scanning video-rate imaging, the pixel dwell time is sufficiently short such that only a few (<10) pulses from a typical Ti:Sapphire (80 MHz repetition rate) laser are used to generate the observable optical signal. This allows excitation of the fluorophore to near saturation levels while reducing the probability of photodamage by providing sufficient time for the fluorophore to completely relax to the ground state before re-excitation. A short pixel dwell time equates to fewer photons per pixel being recorded in the image, thereby reducing the signal-to-noise ratio (SNR). This can be overcome through line or frame averaging, which builds the signal over successive integration.
Short pixel dwell times necessitate a highly efficient detection system. The detection optics in the Thorlabs Multiphoton Microscope are designed to collect signal from the full field-of-view of the objective lens independent of the size of the scan field. In this manner, signal that is scattered by the sample but is still “seen” by the objective is collected, improving the total image intensity. This provides a great improvement in signal detection efficiency when imaging deep in highly light scattering samples, especially when combined with low-magnification high-numerical-aperture (that is, 20X 1.0NA) objective lenses. High sensitivity, low noise GaAsP PMTs further improve the total efficiency of the detection path and overall image SNR.
In experiments where imaging to the greatest depths is essential, the dedicated NIR scan optics in the Thorlabs Multiphoton Microscope support the widest excitation wavelength range. Multiphoton microscopy improves depth penetration by using NIR excitation to reduce scattering; however, the signals generated are still in the visible range, making them difficult to detect from deep within a sample. The latest generation of engineered fluorophores and fluorescent proteins look to push the signal generated further into the red (>600nm) region. Many of these fluorophores require excitation wavelengths outside those provided by the Ti:Sapphire laser (680-1080nm) and are accessible with wavelength extending (>1000nm) (OPO) technology.1 Pushing the excitation wavelength further into the NIR benefits SHG imaging in the same manner as TPF by generating signals at >600nm instead of ∼400nm.
Next generation in multiphoton excitation
It is a common practice in MPM to interrogate samples labelled with several fluorescent probes. The cost of femtosecond lasers often limits a multiphoton imaging system to a single laser. Imaging a sample with several fluorophores often requires tuning of the laser to a compromising wavelength to excite the fluorophores simultaneously, though less efficiently. Alternatively, at the expense of imaging speed and photodamage, the laser can be tuned to the optimal excitation wavelength for each fluorophore and the entire specimen rescanned sequentially. To reduce the damaging effects of rescanning the sample or exciting at a compromising wavelength with higher power, a laser with a shorter pulse width and, therefore, broader bandwidth can be employed.
Thorlabs has partnered with IdestaQE to introduce the Octavius-2P, a turn-key 10fsTi:Sapphire laser designed for multiphoton applications. A 10fs laser has the high peak power necessary for deep tissue imaging2 with lower average power (refer to equation 1) to minimize photodamage. The broad bandwidth of the 10fs laser pulse (Fig. 3a) is ideal for exciting multiple fluorophores simultaneously with well-spaced excitation spectra.2 The total fluorescence signal generation is proportional to the overlap of the two-photon excitation spectrum of the fluorophore and the two-photon excitation spectrum of the broadband laser. Broadband laser pulses, therefore, increase the signal generated by the fluorophores without the necessity to have the laser tuned exactly at the excitation peak. It should also be noted that the two photons required to generate the nonlinear signal do not have to be equal in wavelength, allowing the entire laser pulse spectrum to contribute to the excitation of the fluorophore.
Reducing the pulse width of a laser pulse leads to an increased susceptibility of temporal pulse broadening due to dispersion, necessitating a microscope design with minimized dispersion characteristics and pre-compensation.4 As previously stated, the dedicated NIR excitation path was designed to have a minimum amount of dispersion. Additionally, an optional dispersion compensation module (DCOMP-BCU) that relies on chirped mirror technology (multilayer dielectric coatings) can be added to the Thorlabs Multiphoton Microscope. Chirped mirrors compensate for dispersion based on precise specification in the multilayer coating and a fixed number of reflections. Dispersion compensation from chirped mirrors is matched to the microscope and does not require tuning. The advantage over prism-based systems is complete dispersion compensation tailored to the laser spectrum and microscope, compact footprint, and no beam deviation through the module.
Multiphoton microscopy is an ideal technology for imaging living and intact tissues that has led to the rapid adoption of the technique by researchers and an explosion of commercially available systems on the market. The Thorlabs Multiphoton Microscope combined with the Idesta Octavius-2P represents a flexible platform geared towardsfacilitating the latest and most advanced research applications. This has been accomplished by a design that is ideally suited to observing complex biological processes and takes full advantage of current and next-generation technology.