Fig 1.
Schematic overview of the THG microscopy setup.
The microscope is powered by a femtosecond laser source (1050nm) and composed of the following elements: half-waveplate (HWP), acousto-optic modulator, piezo-scanning mirrors, lens (L), microscope objective (Obj), long-pass (LP) and short-pass (SP) dichroic mirrors (angled gray rectangles), band-pass filter (BP) and photomultiplier tubes (black rectangles), respectively displayed for THG, SHG in both forward and epi-scattered detection and 2PF.
Table 1.
Demographic and clinical Information of human MS and healthy control (HC) samples.
Fig 2.
Photodamage is in human D384 medulloblastoma followingTHG imaging.
(A) Representative fluorescence images corresponding to the highest pulse exposure condition (5.25 pulses/µs) at 18 s and 90 s, with DAPI (blue) marking nuclei and PI (red) indicating damaged cells. For all conditions consisting of more than one accumulation (an exposure time longer than 1.8 s), a 5-second interval was applied between subsequent exposures. The energy per pulse was kept constant and was estimated to be ~4.5 nJ per pulse based on power measurements at the sample plane. (B) Graphs depicting the percentage of photodamaged cells following various laser pulse frequencies (0.75 to 5.25 pulses/µs) across exposure durations (1.8 s to 90 s), in control conditions (Neg) and following UV-induced damage (Pos). Overall, No significant effects for pulses/µs were observed in the analysis using a one-way ANOVA (F(3,12) = 0.982, p = 0.44). Notably, in the most disruptive condition (5.25 pulses/µs) no significant difference in in cell damage was observed between 1.8s and 90s (t(6) = 1.921, p = 0.1031). Error bars indicate SD. Scale bars are 50 µm.
Fig 3.
Optical Performance of the nonlinear microscopy setup.
(A) A three-dimensional cutaway from a z-stacked THG image. Volumetric rendering (100 × 100 × 100 µm) (THG) image highlighting cortical GM, showcasing the technique’s capability to provide detailed structural insights. Note that these insights do not provide a linear mapping of the myelin concentration, but instead are the result of local changes in the refractive index and nonlinear third-order susceptibility. (B) Comparative analysis of myelin imaging using THG (left) against LFB staining (right) of cortical tissue. Insets highlight myelinic swellings visible in both THG and LFB images, with THG’s sensitivity to lipid bodies (red arrow in THG inset) and swelling (white arrowheads, purple arrows in LFB staining). (C) Comparison of the relation between median THG signal intensity and imaging depth, comparing gray and white matter regions. Normalized to 0 μm depth (n=6). (D) THG image captured at the edge of a glass capillary, accompanied by a superimposed line-scan analysis (left). Intensity profile across a redline and the corresponding line spread function, showcasing the system’s spatial resolution as measured from the capillary edge (right). (E) Full width at half maximum (FWHM) for axial resolution is approximately 1.3 μm, determined through imaging a glass bottom well. Scale bars are 15 μm, unless specified otherwise.
Fig 4.
Subtle alterations in myelin morphology during THG imaging.
(A) Multiharmonic overlayed micrograph of a myelinated axons in human cortical slice culture (DIV12). Lipofuscin in neuronal soma is broadly fluorescent across SHG (red at 525nm) and 2PF (blue at >580 nm) channels (arrowhead). (B) Irregularities in myelin sheath develop quickly (blue and pink arrows) following 1 mM sodium azide and 0.2 mM glutamate stimulation, resulting in the development of myelin swellings during 2-hour acquisitions (white arrow). Time Lapsed z-stacks were drift corrected and the axon in A (yellow outline) was traced and straightened in ImageJ. (C) Post-stimulation, myelin membrane swelling (ab- and adaxonal) becomes distinct (black arrowhead) as shown in the profile plot (white dotted line in B). Scale bars in A are 20 µm. Scale bar in B is 10 µm.
Fig 5.
Inhibition of calpain-cathepsin axis with ALLN annuls the development of myelin swellings in MS corpus callosum.
(A) Schematic illustrating the treatment of corpus callosum slices with N-Acetyl-L-leucyl-L-leucyl-L-methioninal in HEPES buffer followed by exposure to artificial cerebrospinal fluid (aCSF) adjusted to various [Na+] concentrations to simulate conditions that may induce myelin swelling. (B) Representative THG micrograph of myelin during imaging session. Inset: THG signal after CLAHE contrast enhancement. (C) Inverse pixel-level probability map of CLAHE-corrected THG signal (left). Middle: Object-level classification discerned myelinic swellings (blue) from extracellular space (white). Right: Composite visualization merge of the original THG signal imagery with the superimposed object identification data. (D) Quantitative analysis of swelling density across different sodium concentrations in aCSF for non-MS and MS corpus callosum. Scale bars in B and C are 20 µm. n=6 (non-MS) and n=5 (MS).