X10468

The imaging setup is illustrated in Figure 1a. A 488-nm laser source (TOPTICA, Germany) was used for illumination. The beam was expanded with an objective (Obj_B, RMS10X, 0.25NA, 10X, Olympus) and lens (L1, f: 150 mm), and apodised using an aperture. The beam was filtered spectrally using a notch filter (λ_c: 480 nm, Δλ: 17 nm) and spatially using a pinhole (15-μm diameter) at the objective focus. A half-wave plate controlled polarisation to maximise the efficiency of the spatial light modulator (SLM, X10468, Hamamatsu, Japan). Phase and amplitude were controlled by using the SLM in diffraction mode. The desired phase was projected onto the SLM, and the desired intensity was modulated by a phase ramp, realising a blazed grating. The first-order diffraction was spatially filtered using an aperture (SF). The beam was relayed by lenses (L2–L5, f: 250, 100, 50, 75 mm) and the illumination objective (Obj_I, 54-10-12, 0.367NA 4X, Navitar). The light sheet was generated by a galvo scanner (Galvo, Thorlabs, NJ) in the Fourier plane. Fluorescence was collected by the detection objective (Obj_D, 54-10-12, 0.367NA 4X, Navitar) and relayed to the camera (CAM, Iris 15, Teledyne Photometrics, AZ) by a tube lens (TL, f: 200 mm). Fluorescence was filtered using a bandpass filter (BP, λ_c: 532 nm, Δλ: 50 nm) and a notch filter (NF, λ_c: 488 nm).

Samples were attached to the bottom of a 60-mm plastic dish with cyanoacrylate glue and immersed in an imaging medium. Images were acquired by using a custom-made setup, as previously described^{57 (link),58 (link)}. A water immersion objective (XLUMPLFLN20XW 20 × /1.0 Olympus) was used for all the experiments performed in this study. We incorporated an SLM (X10468, Hamamatsu Photonics K.K.) into the two-photon microscope to allow electrical correction of aberrations^{12 (link)}. We predicted a wavefront aberration according to the following equation and cancelled it by applying a reverse wavefront: $$\mathrm{\Phi }\left(\rho \right)=-\left(\frac{2\pi \text{WD}}{\lambda }\right)\left(\left(1+\eta \right)\left(\sqrt{n_2^2-{\left(\text{NA}\rho \right)}^2}\right)-\left(\sqrt{n_1^2-{\left(\text{NA}\rho \right)}^2}\right)\right),$$ where λ is the wavelength of the excitation beam, ρ is the normalised pupil radius, η is the factor for changing the depth of the focal spot, n₁ and n₂ are the RIs of water and tissue clearing solutions, respectively, and NA and WD are the numerical aperture and the working distance of the objective used.
DiO, DiI, DiD, and DiR were excited with excitation wavelengths of 915 nm, 880 nm, 850 nm, and 915 nm, respectively. To compare the DiO and DiD imaging depths, the excitation wavelength was set to 820 nm, and the laser power was adjusted at the surface of the sample so as to ensure that the signal was not saturated; the power was kept constant during each imaging session.