Making Pathology Research More Efficient

A new, faster strategy uses scanning electron microscopy for ultrastructural imaging of large tissue sections

Electron microscopy is an essential tool for pathologists, providing high resolution imaging of tissue structures, cells, organelles and microbes. However, ultrastructural analyses of large areas of tissue can be challenging and time consuming.

Dr. Mike Reichelt, Principal Scientific Researcher at Genentech, Inc. (USA)

We spoke with Dr. Mike Reichelt, a Principal Scientific Researcher at Genentech, Inc. (USA) who is responsible for the daily operations of the electron microscopy lab within the Pathology Department. Dr. Reichelt also develops and implements innovative strategies to address the needs for Genentech’s research and development organization.

Dr. Reichelt and his colleagues recently published an article describing an efficient workflow for imaging large tissue sections with backscattered electron – scanning electron microscopy (BSE-SEM) for applications in ultrastructural pathology.

What was your motivation for developing this new workflow for electron microscopy imaging of large tissue sections?

We had a major need in our pathology department, namely, the analysis of healthy or diseased tissues at both the histological and ultrastructural level. I was able to develop a strategy focused on the preparation of large (several square millimeters) tissue sections with an emphasis on wide-scale image acquisition to easily correlate low magnification histologic overviews with highly resolved ultrastructural features within the same tissue section.

Mouse kidney glomerulus in center surrounded by various renal tubules.

You use back scattered electron – scanning electron microscopy (BSE-SEM) for your strategy. Can you describe your motivation for using this versus traditional methods?

Traditionally, analysis of tissues at the ultrastructural level has most often been performed with the transmission electron microscope (TEM). While this procedure has been used for decades with great success, it is time consuming and has many hurdles:

  • The small size of TEM grids limits the size of the tissue section.
  • Grid bars further limit and obscure the field of view.
  • The preparation of ultrathin sections requires special training and skilled personnel.
  • It can be extremely challenging and time consuming to correlate the high magnification TEM view with the ROIs defined by light microscopy.

In contrast, these hurdles are eliminated when imaging tissues with modern field emission scanning electron microscopes (FE-SEMs) in combination with sample processing strategies that allow the preparation of large intact tissue sections (several square milimeters). Our new workflow takes full advantage of the very large scan fields possible with modern FE-SEMs that allow for the acquisition of large overview images which can be explored at the ultrastructural level by digitally zooming into the images.

Digital zooming from kidney tissue overview into glomerulus (glo). Right side: podocyte (pc) with foot processes (fp), basement membrane (bm) and capillary space (cs) with erythrocystes (ec).

Since its implementation in our lab, this BSE-SEM-based multiscale imaging procedure has substantially simplified and accelerated our ultrastructural tissue analysis.

Nearly 100% of all ultrastructural tissue imaging projects (at least 20 per year) are now run on ZEISS GeminiSEM 300. With this procedure we have already contributed to four published, peer-reviewed papers since we established the method in 2018 and numerous more manuscripts by our scientific collaborators are in work.

In regards to new methods and technology, where do you see your field going in the next five years?

The three most significant developments in ultrastructural imaging for the next five years that are relevant for research pathology in the biotech industry are:

Number 1:

The implementation of efficient, correlated workflows that combine the power of immunofluorescence microscopy methods for the molecular characterization of cell types, organelles and drug targets or the localization of therapeutic molecules, with the power of electron microscopy imaging to provide ultrastructural resolution and context.

Although several strategies for correlated imaging have been described in the literature, major bottlenecks remain nevertheless, such as the development of optimized tissue processing methods that preserve both antigens and GFP-fluorescence combined with excellent ultrastructure and membrane contrast as well as the implementation of straightforward workflows for the acquisition and correlation of light and electron microscopy images.

ZEISS ZEN Connect, ZEISS Shuttle & Find and ZEISS Atlas are good steps toward this goal.

Example of correlative immunofluorescence and electron microscopy imaging with ZEISS Shuttle & Find. The composite image shows very nicely the blue staining of the nuclei and the green fluorescence of podocalyxin over the podocyte foot processes, which can only be fully resolved in the SEM. Light microscopy images were collected with ZEISS Axio Imager 2 and electron microscopy images were collected with ZEISS GeminiSEM 300.

Number 2:

The interpretation, quantification and analysis of ultrastructural imaging data will be significantly accelerated by novel “machine learning” and “artificial intelligence” tools that will help with the automatic segmentation of ultrastructural features (e.g. organelles).

Number 3:

Investigators will increasingly ask for 3D data sets for more comprehensive evaluations and analyses. I believe serial section array tomography, which is compatible with correlative immunofluorescence and ultrastructural imaging, as well as serial block face imaging of large tissue volumes (3View) will be most relevant for our electron microscopy lab.

Traditionally, a major bottleneck with volume electron microscopy methods has been the extremely time-consuming segmentation of the large image data sets, which made these techniques less relevant in the biotech industry, where innovation needs to be paired with efficiency and high throughput. However, given the progress in recent years in the development of automatized segmentation methods, this former bottleneck appears to be rapidly disappearing, and therefore volume electron microscopy will also become increasingly important.

Learn More

Read the full paper describing Dr. Reichelt’s method here.

Read articles where Dr. Reichelt implemented this method:

  • IRE1α disruption in triple-negative breast cancer cooperates with anti-angiogenic therapy by reversing ER stress adaptation and remodeling the tumor microenvironment. Link
  • Disruption of IRE1α through its kinase domain attenuates multiple myeloma. Link
  • Autophagy of Sensory Neurons in the Trigeminal and Dorsal Root Ganglia of Cynomolgus Monkeys (Macaca fascicularis). Link

Learn more about ZEISS GeminiSEM scanning electron microscopes.

Read Next

Leave a Reply

Your email address will not be published. Required fields are marked *

Visits on this Page:9978