Array Tomography Workflow for the Targeted Acquisition of Volume Information using Scanning Electron Microscopy

Electron microscopy analysis is frequently complex. We believe that our protocol facilitates processing and acquisition of the EM data from the large sample surface and intermediate volume. Our protocol helps find target structures in serial sections.

Data recording is limited to what’s necessary. Compared with data acquisition in whole-tissue blocks, recording time is reduced and data analysis easier. Given the overall ease of the approach, we believe that it can be used to address not only scientific questions, but also as a diagnostic tool.

To generate an array for analysis, first clamp the sample onto an ultramicrotome holder and use a razor blade to roughly trim the resin around the sample. Use a diamond trimming tool to fine-trim the sample and use trimming knives of 20-or 90-degree inclination to ensure that the top and bottom surfaces of the block are parallel to the knife’s cutting edge. Mix xylene and glue in a 3:1 proportion and use an eyelash attached to a toothpick to apply the mixture to the top and bottom edges of the trimmed block.

While the mixture is drying, use a wafer cleaving tool to cut a piece of wafer to the appropriate size for the analysis. Clean the wafer in distilled water to rinse away any debris and clean the surface with plasma after the wafer is dry. To prepare an array tomography knife, use foamy sticky tape to attach a needle to the bottom of a histo jumbo knife and place the knife into the ultra microtome holder at zero degrees.

Adjust the edge of the knife parallel to the block surface and bring the trimmed block to the edge of the knife in a position ready for sectioning. Place the clean wafer into the knife basin and fill the basin with water to the same level as the knife edge, then, let the diamond edge of the knife humidify properly, using the attached syringe to add or withdraw water as necessary. For array sectioning, set the microtome to a cutting range of 50 to 100 nanometers and a cutting speed of 6 to 1 millimeters per second, and begin sectioning to obtain a ribbon long enough to cover a targeted Z volume.

Depending on the block size, homogeneity of the tissue, and type of resin, the ribbon will be approximately straight. Use a clean, non-sticky tip of an eyelash to detach the ribbons from the knife edge. Use the eyelash to gently move the ribbon over to the center of the support medium.

Chloroform or a heating pen can be used to stretch the section if necessary. After stretching, retract the syringe to start draining the water. For more delicate ribbons, or a slower water retraction, detach the syringe from the hose to let the water drip.

When the water level reaches the wafer level, gently push the ribbon to the center of the basin and continue draining until all of the remaining water is completely removed from the basin. Let the wafer dry in a clean environment. Drying takes approximately 30 minutes.

When the sample is completely dried, transfer the array to a tightly-closed box to protect it from dirt contamination. And place the box in a 60 degree-Celsius oven for at least 30 minutes. To acquire an overview map of SEM images that reveals section locations on the wafer, use the built-in optical camera to acquire an SEM image that covers a ribbon of sections.

To create a mosaic, click and drag the camera image of the sample and start the automatic acquisition. Acquire overviews at higher resolution to find target structures. If only a few sections need to be imaged, use zoomable viewer to view acquired images in their original locations.

Once a section that should be imaged at high resolution has been identified, click and drag to create an imaging region, then select High-resolution Imaging Settings and store the settings in a template To find particularly small or hard-to-detect rare events, use the high-resolution imaging settings on every 10th section or on one section per ribbon to manually create an imaging region and acquire the images. Then review the images and mark sections that contain the region of interest. To acquire more than 10 consecutive sections, use the section finder to locate all sections automatically.

If the overview images do not show clear regions of interest, acquire higher resolution images of the sections and use the section preview function to create and acquire the images automatically. To determine the optimal imaging settings, activate the live imaging in the microscope control software and navigate to one region of interest, then adjust the imaging settings until the images show clear regions of interest, but without an excessively-long image acquisition, according to the manufacturer’s guidelines. To optimize the positioning of the imaging regions on consecutive sections, zoom to the image of interest and click Start Position Refinement to increase the precision of registered section locations.

To define an imaging region, click and drag on any section while holding down the Alt key and select Create Tile Set Array from the pop-up context menu. The software will create imaging regions in the same relative location in all of the sections that have been found or previously marked. Then set up pixel count, pixel size, tiling layout, and pixel dwell time in each image series as needed.

To configure the auto function, create a separate image series for auto functions as demonstrated, and move the image series to a position on the section that contains high-contrast structures. Set the image series to 1024 x 884 pixels and select a pixel size corresponding to the highest resolution used in the image series. In the list of auto functions, check Auto focus and Auto stigmator.

Select Bisection in the acquisition sequence controls and confirm that the auto function’s image is the first item in the list, then click Acquire All to start the image acquisition. When all of the imaging series have been created and set up, the series will be listed in a job queue. Low-resolution acquisition can be performed manually or automatically directly on selected parts of the section or an entire section using single or mosaic imaging, followed by stitching.

Images from the selected area can then be acquired using high-resolution parameters to visualize mitochondria, nuclei, and microvilli, for example, After the automatic acquisition of resolved mosaic maps has selected, several regions of interest can be cropped out or used to define additional local imaging areas within the regions. Although different specialized cell types in the Drosophila gut are randomly distributed, they can be visually distinguished after screening the images using high-resolution parameters, either from single sections or as a collection of serial images. After the alignment, the stacks can be rendered using different software solutions.

Array tomography analysis allows many sequential sections to be generated on the single wafer, and to be screened using low-resolution parameters to localize the general areas of interest. These areas can be targeted for further analysis using advanced acquisition parameters. For example, mytotic divisions in the notum are not easy to localize on the ultra-structural level, as the cells are relatively large compared to the abscission zone.

Using this method, however, the automatic medium-resolution overview images of the leaps of 20 to 40 sections can be used to localize the dividing cells. The array tomography procedure provides a more straightforward electron microscopy data analysis. We encourage the viewers to invest in mastering the regeneration and familiarizing themselves with the maps-related workflow In our experience, few research topics require the ultra-structural analysis of entire animals or whole organs.

Our method helps with rapid localization of cells and their interaction partners in tissues.