DUCT: Double Resin Casting followed by Micro-Computed Tomography for 3D Liver Analysis

This method allows us to analyze, visualize, and quantify liver vascular and biliary systems architecture in 3D. We can now better understand the spatial aspects of liver disease and regeneration. The main advantage of this technique is that it does not rely on the existence of antibodies or fluorescence imaging.

It also allows for direct visualization of which networks are properly lumenized. This method is suitable for the analysis of tubular systems, hence research of blood vessels, bile ducts, and airways are ideal areas. We used DUCT to understand the structural changes related to biliary regeneration.

Before injecting resin into the biliary system of the mouse, start moving the intestine and pancreas to the left side of the mouse, using a cotton swab wetted with PBS to expose the common bile duct and portal vein. Then make a small transversal incision in the inferior vena cava. Flip the ventral side of the liver towards the heart to expose the visceral surface and the hilar region, and locate the common bile duct that runs from the hilar region across the pancreas, and into the intestine at the sphincter of Oddi.

Use the straight forceps to clear the surrounding tissue area of approximately five millimeters from the common bile duct. Then, by placing a silk suture thread under the common bile duct, tie a loose overhand knot around the common bile duct. Hold the spring scissors flat against the common bile duct to make an oblique incision to the common bile duct at the spot where the common bile duct enters the pancreas and intestine next to the sphincter of Oddi.

Next, mix one milliliter of the yellow resin with 50 microliters of the MV curing agent, and fill a one milliliter luer syringe with the resin curing agent mixture. Connect the filled syringe with the tubing set one, and press the plunger to fill the tubing till the resin curing agent mixture drips from the tip of the tubing. Use the forceps to straighten the area around the common bile duct incision.

Before inserting the tubing into the opening in the common bile duct with the longest edge of the beveled tip downwards towards the dorsal side of the bile duct. Then tighten the silk thread knot to secure the tubing inside the common bile duct. After injecting the resin into the common bile duct, observe the gallbladder and the individual liver lobes.

Massage the liver with a PBS-wetted cotton swab to help spread the resin equally. Resin-filled bile ducts terminal branches will be faintly visible at the liver surface. Stop injecting the resin when dots of resin appear at the surface of the liver or when resistance is met.

Then remove the tubing by pulling out of the common bile duct and quickly tighten the silk knot using forceps to prevent the resin from leaking out. Cut the loose ends of the silk suture, so they do not interfere with the portal vein injection. Before injecting the resin in the portal vein, clear the portal vein from its surrounding tissue about two centimeters from its entry to the liver, using straight forceps.

Then place the silk suture thread under the cleared area of the portal vein to tie a loose overhand knot. In the portal vein distal to the liver, make a longitudinal incision before tying the knot. Next, prepare the green resin curing agent mixture in the syringe and fill the tubing with the mixture as demonstrated previously.

Use the forceps to straighten the portal vein by pulling the surrounding tissue towards the tail side before inserting the tubing with the longest edge of the beveled tip towards the dorsal side of the vessel. Tighten the silk thread to secure the tubing in the portal vein. After injecting the resin into the portal vein, observe the blood vessels filling up with resin, and massage the liver with a PBS-wetted cotton swab to help spread the resin.

When all blood vessels are filled, pull out the tubing from the portal vein, and quickly tighten the silk knot using forceps to prevent the resin from leaking out. For micro-computed tomography or MicroCT scan of the dissected liver from the mouse, place the right medial liver lobe in a 15 milliliter conical tube, and fill the tube with 1%agarose gel to approximately two third of the total volume of the tube to minimize undesired sample motion during MicroCT measurement. Mount the 15 milliliter conical tube with the sample on the rotational stage of a CT device.

Allow the tube to thermally adapt to the measurement chamber for at least one hour. When the sample is thermally adapted, center it in the field of view or FOV, then optimize the source sample and sample detector distances, or SSD and SDD to reach sufficient voxel resolution. Once the parameters are set, start a CT measurement.

Use dedicated tomographic reconstruction software to reconstruct the CT data. To analyze the CT data, load the files by selecting file, import, and command. Use the surface determination function on the top panel to segment the resin in the data using global thresholding.

In the dialogue window, determine the threshold value with the histogram evaluation by setting the position of the red isovalue line to segment only the resin-filled vessels. On the left panel, select the module create ROI from volume or CAD or mesh to create a region of interest or ROI of the resin-filled vessels. In the dialogue window, use the option create ROIs from solid to select and confirm the name of the processed volume.

Eliminate erroneous segmentation of noise clusters in the background region of this ROI. Mark the ROI on the right panel by right-clicking and selecting the module split ROI. In the dialogue window, set the minimum volume voxel parameter to exclude all the noise particles.

The parameter value is experiment and data-dependent, and optimized for each sample to be analyzed. On the left panel use the smoothing module to create smooth, continuous and solid canal masks without artifacts, like the presence of air bubbles or resin leakage in the resin cost ROI. Create a separate ROI as described before for the system filled with the more absorptive resin like the yellow resin used for the common bile duct injection with higher intensity values in the CT data.

Mark and resin mask the new ROI by right-clicking before selecting subtract ROIs, and then subtracting the new ROI from the resin mask ROI to create a new ROI for the remaining tubular system. Export the resulting ROIs for both tubular systems in various formats based on operator references for subsequent processing in different software. Further process the resulting ROIs in a volume graphics software to export the final visualization in the form of an image or a video.

In this study, a successful double resin injection was achieved with well-filled intrahepatic bile ducts, and portal vein vasculature. The representative analysis illustrates the result of the segmented data from a well-injected liver for a P15 mouse, and an adult mouse where resin is visible in the side branches. In the typical example of unsuccessful injection, physical damage to the liver during the surgical procedure resulted in the resin leaking.

Also the premature hardening of the resin in the needle or the tip of the tubing before the injection, and the insufficient transcardial perfusion caused system under-filling. The minor resin leakage could be manually corrected during MicroCT data segmentation. High injection pressure could rupture vessels or ducts, and irreversibly damage the vessel or duct architecture.

Additionally, the bubbles were another very common injection artifact that led to the sparse filling of the tubular networks. When freshly opened resin was used, a clear difference in the contrast between the yellow resin-injected bile ducts and green resin-injected portal veins was observed. After three months of storage, the contrast was sufficient to distinguish the portal vein from the bile duct.

However, the precipitation affected mixing the two resins, which was visible as a heterogeneous opacity in the filled portal vein. When the resin was older than six months, the contrast degraded to a point where it was unfeasible to distinguish the yellow-injected bile duct from the green-injected portal vein, based on their contrast alone. It is crucial to make an oblique incision into the common bile duct.

Otherwise it will be very difficult to insert the tubing. The DUCT technique enlighten new architectural mechanisms of biliary regeneration in a clinically relevant model with distinct features in the hilar and peripheral lobe regions.