Researchers at The Rockefeller University have successfully redesigned a three-photon microscopy platform to capture neuronal activity in deep brain regions of living mice. Led by Alipasha Vaziri, PhD, head of the Laboratory of Neurotechnology and Biophysics, the team developed a system known as Three-photon Ultra-Deep Imaging (TrUDI). This advancement allows for recording neural population activity at depths of approximately 2 mm without the need to surgically remove the cortex that typically obscures deeper structures.
The study, titled "Cellular-resolution three-photon ultra-deep imaging (TrUDI) of neuronal population activity at 2-mm depth in the mammalian brain," was published in Nature Methods. By optimizing laser, optical, and acquisition parameters, the new platform generates approximately 250 times more detected fluorescence photons at the same laser power compared to previous approaches. This significant increase in signal enables single-cell-resolution recordings that were previously unattainable at such depths.
Vaziri explained that the team adopted a "clean slate approach" rather than adhering to historical limitations regarding specific lasers or parameters. While three-photon microscopy inherently produces less background fluorescence than two-photon methods, achieving sufficient signal depth requires energetic laser pulses that risk heating and damaging tissue if average power is increased indiscriminately. Previous functional recordings using this technique had only reached approximately 1.2 mm, falling short of the theoretical limit of more than 2 mm.
To overcome these barriers, the researchers utilized computer models to guide coordinated changes to the laser and optics. The TrUDI system matches excitation laser parameters, the optical system, and acquisition settings to both linear and nonlinear limits for safe laser exposure while meeting the spatiotemporal requirements for single-cell resolution calcium imaging. Additionally, the team developed a neuron-targeted excitation scheme that activates the laser only when scanning identified cells.
These systematic optimizations increased the available fluorescence signal for recording transient activity in individual neurons. Tests conducted on mice demonstrated recordings over fields of view measuring approximately 230 by 230 µm² at a rate of 8.6 Hz, successfully reaching depths of about 2 mm into the mouse cortex and hippocampus. This progress expands access to neural circuits that previously required invasive tissue removal for optical imaging.
Despite this breakthrough, Vaziri noted that not every deep brain structure is now accessible. Certain regions of the striatum remain out of reach, and several deep brain nuclei located even deeper still require invasive approaches. The work primarily demonstrates how coordinated optimization can bring three-photon imaging closer to its fundamental depth limit while preserving the integrity of the tissue above the studied regions.