Benchtop mesoSPIM: a next-generation open-source light-sheet microscope for cleared samples

Tissue clearing and light-sheet microscopy have revolutionized biological imaging, enabling 3D visualization of large samples with high resolution and speed. The combination of these techniques allows for the study of entire organs and even whole animals without the need for sectioning. Current methods allow clearing of samples up to several centimeters, including whole mice and human organs. However, imaging large samples at high speed and resolution presents a significant challenge, requiring a balance between field of view (FOV) and resolution. The original mesoSPIM, an open-source light-sheet microscope, has been instrumental in making this technology more accessible. This paper addresses the increasing demand for imaging larger samples with higher speed and resolution by presenting a significantly improved version, the 'Benchtop' mesoSPIM.

The combination of tissue clearing and light-sheet microscopy has spurred advancements in various biomedical fields. Several studies highlighted the use of these techniques in neuroscience to visualize neuronal networks in the whole mouse brain and to study specific neuronal populations. In developmental biology, the techniques are used to image whole embryos, and in pathology, they are being increasingly adopted for 3D imaging. Existing light-sheet microscopes offer various FOVs and resolutions, with some focusing on high resolution and others on very large FOVs, often at the cost of resolution. The ExA-SPIM and AMATERAS are two examples of systems pushing the boundaries of high-resolution, large-FOV imaging, however they often present significant cost and complexity. This literature highlights the need for a more accessible, high-performance microscope that balances these competing demands.

The Benchtop mesoSPIM incorporates several key improvements over the previous mesoSPIM version. A new method for testing detection objectives was developed, allowing for objective selection that optimizes light-sheet imaging with large-sensor cameras. This method employs a high-density Ronchi grating and incoherent light source to quantify the modulation transfer function (MTF) across the camera field and focus range, generating 3D MTF graphs. The improved design features a large-sensor sCMOS camera, optimized detection system, and modular design that reduces cost and size, eliminating the need for an optical table. The system supports a wider range of magnifications (0.9x-20x) and sample sizes (3-75mm), and is compatible with various clearing techniques. 3D-printed sample holders provide flexibility and high throughput. The control software, an open-source Python program, supports multiple hardware types and formats, allowing for efficient handling of terabyte-sized datasets. The paper details the testing of various objectives, highlighting the superior performance of Mitutoyo Plan Apo objectives in terms of contrast and field flatness. The impact of chromatic effects on field flatness was also evaluated using different filters and immersion media. Finally, the paper explains the ASLM mode used for achieving uniform axial resolution.

The Benchtop mesoSPIM demonstrates significant improvements over its predecessor. It achieves higher resolution (1.5 µm laterally, 3.3 µm axially) across a much larger FOV. The system's smaller footprint and lower cost make it more accessible. The new objective testing method allows for informed selection of objectives, optimizing performance. The versatility of the system is demonstrated through imaging various samples: mouse brains (single-axon resolution), whole mice, chicken embryos, human cortex samples, and even color centers in crystals used for particle detection. The high-throughput capabilities are highlighted by the use of the SPIM-tower, enabling the simultaneous imaging of multiple samples. The imaging of color centers in gamma-irradiated CaF2 crystals demonstrates the applicability of the system in physics, opening avenues for exploring particle detection. The system is mobile and easily reassembled, facilitating collaboration and use in remote locations, such as underground laboratories.

The Benchtop mesoSPIM addresses the need for a cost-effective, high-performance light-sheet microscope capable of imaging large cleared samples with high resolution and throughput. The improved optical performance, versatility, and ease of use make it a valuable tool for various applications. The development of the new objective testing method is a significant contribution, providing a standardized approach for evaluating and selecting objectives optimized for large-sensor cameras and various immersion media. The results demonstrate the broad applicability of the Benchtop mesoSPIM across diverse fields, from neuroscience and developmental biology to physics, showcasing its potential to accelerate research in these areas. The open-source nature of the project further promotes collaboration and wider adoption of this technology.

The Benchtop mesoSPIM represents a significant advancement in light-sheet microscopy for large cleared tissues. Its improved optical performance, increased throughput, lower cost, simpler assembly, and enhanced portability make it a valuable tool for researchers across various disciplines. The open-source nature of the design and software encourages further development and customization by the community, promising continued innovation in this field. Future research could focus on optimizing the system for even higher resolution and throughput, potentially through the development of custom objectives with higher NA and aberration correction.

The current design uses off-the-shelf air objectives, which results in some spherical aberration due to the index mismatch with the clearing agents. This limits the resolution achievable with thicker samples. The chromatic effects on field flatness, although addressed by the software, might require further adjustments depending on the specific application and combination of objectives and filters. While the system offers high throughput, further automation of sample preparation and image acquisition could enhance efficiency. The currently used components are not all open-source, and the upgrade of previous mesoSPIM versions is dependent on the availability of specific components and requires some expertise in optics and mechanics.