Multi-immersion open-top light-sheet microscope for high-throughput imaging of cleared tissues

Adam K. Glaser, Nicholas P. Reder, Ye Chen, Chengbo Yin, Linpeng Wei, Soyoung Kang, Lindsey A. Barner, Weisi Xie, Erin F. McCarty, Chenyi Mao, Aaron R. Halpern, Caleb R. Stoltzfus, Jonathan S. Daniels, Michael Y. Gerner, Philip R Nicovich, Joshua C. Vaughan, Lawrence D. True, Jonathan T.C. Liu

Posted on: 20 March 2019

Preprint posted on 13 February 2019

Article now published in Nature Communications at http://dx.doi.org/10.1038/s41467-019-10534-0

Improvements in optics and materials expand the imaging applications of this fast light-sheet microscope

Selected by Tim Fessenden

Categories: bioengineering, cancer biology, cell biology, developmental biology, pathology

Context

Rich descriptions of normal or diseased tissues using microscopy are circumscribed by the limitations of confocal fluorescence or widefield imaging. These limitations include, among others, sample fixation and embedding techniques, poor light penetrance into a tissue, the maximum axial resolution, and the time needed to scan across a sample. Together these obstacles enact boundaries on the spatial information gleaned from imaging a tissue sample. Moreover, the speed of image acquisition places a constraint on the area of a sample that can reasonably be analyzed. These considerations have collectively taught biologists to think of and analyze small areas of tissues sectioned into thin slices. While the achievements of traditional microscopy of tissue sections need not be defended, simultaneous improvements in tissue preparation, optics, and computation can free biologists and clinicians to image tissue specimens as they are, not as we prepare them to ready them for imaging.

The group of Jonathan T. C. Liu recently published their light sheet microscope optimized for large specimens, namely tissue biopsies¹. Light sheet microscopy uses lower light intensity and yields improved sensitivity because only a thin plane of the specimen is illuminated at a time. But achieving these benefits requires focusing two objectives on the sample in close proximity, which severely restricts the size and format of samples. The Liu group overcame this limitation by constructing a light sheet microscope in an open top format, not unlike a document scanner. They demonstrated exceptionally rapid rendering of thick 3 dimensional human biopsy tissues, leveraging the improved imaging quality of light sheet microscopy.

An open top light sheet microscope compatible with multiple immersion fluids for fast imaging of a variety of samples.

Data

The core of the present work is a systematic exploration of materials and immersion solutions in order to optimize the group’s open-top light sheet microscope for different specimens. This work confronts the sacrifices made in optics to enable the open top, scan bed format. Because the light paths for illumination and emission pass through a plane of glass at oblique angles, the resulting image quality suffers. Further, the refractive index of glass is not matched for many of the solvents used for tissue clearing or expansion microscopy. This limits the type of samples compatible with their original microscope design.

The authors improve this design first with illumination and detection objectives that offer greatly improved axial and lateral resolutions as well as a much greater working distance. As in the preceding design, the authors make use of a solid immersion lens (SIL) on the illumination objective, to match the cylindrically focused illumination beam to the spherically focused emission beam. They also incorporate their recently published method to minimize shadow artifacts when imaging through thick and opaque samples². To achieve this, the incident light is swept within the plane of the light-sheet, a technique they term multidirectional digital scanned light sheet microscopy (mDSLM).

The bulk of this work is in optimizing two materials critical to improve the versatility of this instrument: the fluid in which the detection objective is immersed, and the transparent holder on which the sample rests above the objectives. Using simulations, the authors test 19 immersion fluids that span sample preparations in aqueous (e.g. CLARITY)³, solvent (e.g. iDISCO)⁴, or expansion microscopy⁵ solutions. Against these, they test 13 different glass and 14 different monomer/polymer materials for the sample holding material. For each combination they calculate the maximum thickness of the holding material while retaining acceptable image quality. The resulting matrix identifies the combinations that do not compromise image quality based on the sample type or preparation.

With these results in hand, the authors validate their findings by swapping out the immersion fluid and sample holder material for an array of samples. They demonstrate fast and exceptionally detailed 3D microscopy of human prostate biopsies; cleared mouse brain, lymph node, heart, prostate, and lung samples; and expanded mouse kidney samples prepared at 4x their original dimensions. Leveraging the benefits of the open-top system inherent to its original design, the authors image samples in high throughput fashion, for instance by placing 12 prostate core needle biopsies adjacent to each other on the 10 x 10 cm sample bed.

Implications

In this work the authors describe a conceptually straightforward yet practically challenging solution to a persistent problem in microscopy. That is, how to democratize recently developed microscopy techniques to spread the benefits of improved instruments to more users in more fields? Glaser and colleagues combine light-sheet microscopy with an open-top format, optimized in this work to improve resolution, sensitivity, and imaging depth. In doing so, the authors offer a means to spread the benefits of light-sheet microscopy to minimally processed human biopsies, cleared specimens, and expanded specimens. Many biologists have taken advantage of light-sheet microscopy only by modifying their sample preparation to suit its unique needs. The instrument described here removes this limitation while sacrificing little in terms of image quality.

Many advances in microscopy push at the limits of optics to achieve this or that benchmark. While exciting, the technologies that result from these efforts are often limited to the optimal biological specimens for that instrument, strongly constraining its utility. From the standpoint of adoption and utility, this work offers a contribution to biological imaging that well surpasses that of a narrow technical advance.

References

Glaser, A. K. et al. Light-sheet microscopy for slide-free non-destructive pathology of large clinical specimens. Nat. Biomed. Eng. 1, 0084 (2017).
Glaser, A. K. et al. Multidirectional digital scanned light-sheet microscopy enables uniform fluorescence excitation and contrast-enhanced imaging. Sci. Rep. 8, (2018).
Chung, K. et al. Structural and molecular interrogation of intact biological systems. Nature 497, 332 (2013).
Renier, N. et al. iDISCO: A Simple, Rapid Method to Immunolabel Large Tissue Samples for Volume Imaging. Cell 159, 896–910 (2014).
Chen, F., Tillberg, P. W. & Boyden, E. S. Expansion microscopy. Science 347, 543–548 (2015).

doi: https://doi.org/10.1242/prelights.9291

Read preprint

(No Ratings Yet)

Author's response

Adam K. Glaser and Jonathan T.C. Liu shared

The physical format you have built is particularly well-suited to image large human tissue specimens or entire mouse organs, but does this system also lend itself to live imaging of larger tissues or organoid cultures?

Glaser:

Yes, the beauty of the open-top geometry is that it is highly versatile and customizable. While the focus of our current system has been 3D pathology of larger clinical specimens and mouse organs, the microscope could easily be adapted to accommodate live imaging of larger tissues or high-throughput imaging of organoid cultures in well plates. Similarly, as we allude to at the end of the manuscript, the unobstructed open-top is readily compatible with any accessory technology, including devices for environmental control.

Liu:

Yes, the open-top configuration provides a lot of flexibility to allow for a variety of peripheral and accessory devices such as microfluidic chips, environmental chambers, etc. Organoids are probably ideal in that they are not too large, so the requirements for imaging depth are reasonable, and a multi-well plate could be designed to image hundreds of them in one session with an OTLS system such as ours.