Protocol for three-dimensional whole-mount imaging of the vascular network in the intestinal muscle

Summary

Three-dimensional (3D) imaging of vascular networks is essential for the investigation of vascular patterning and organization. Here, we present a step-by-step protocol for the 3D visualization of the vasculature within whole-mount preparations of the mouse intestinal muscularis propria layer. We then detail the quantitative analysis of the resulting images for parameters such as vessel density, vessel diameter, the number of endothelial cells, and proliferation. The protocol can be easily extended to study cell-cell interactions such as neuro-vascular or immune-vascular interactions.

For complete details on the use and execution of this protocol, please refer to Schrenk et al.¹

Graphical abstract

Highlights

• •3D visualization of the vasculature of the mouse intestinal muscularis propria layer
• •Directions for quantifying vascular parameters of vascular area and vessel diameter
• •Instructions for quantification of endothelial cell proliferation in vivo
• •Technique applicable to study neuro-vascular or immune-vascular interactions

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Three-dimensional (3D) imaging of vascular networks is essential for the investigation of vascular patterning and organization. Here, we present a step-by-step protocol for the 3D visualization of the vasculature within whole-mount preparations of the mouse intestinal muscularis propria layer. We then detail the quantitative analysis of the resulting images for parameters such as vessel density, vessel diameter, the number of endothelial cells, and proliferation. The protocol can be easily extended to study cell-cell interactions such as neuro-vascular or immune-vascular interactions.

Before you begin

The protocol below was developed in our laboratory to investigate the effects of a gain-of-function genetic mutation on the 3D structure and organization of the intestinal vasculature. This protocol was used to investigate functional aspects of the vasculature including vessel permeability and endothelial cell proliferation.¹ Here, we show that this method can be extended to study crosstalk of the vasculature with other cell types, including cells from the enteric nervous system and immune system. While we have used this protocol to investigate pathogenic processes of the intestinal endothelium, we anticipate that this protocol is highly relevant to investigations of intestinal development and function.

Before initiating this protocol, it is necessary to ensure that the appropriate experimental conditions are met to achieve the desired research outcomes. Mouse cohorts should be arranged in advance to ensure that tissues are collected at the same developmental time points in all animals. We also recommend that the appropriate genotype and/or treatment controls are included, to ensure changes to the intestinal phenotypes are not related to technical aspects of the experiment. For example, both tamoxifen and non-tamoxifen treated animals should be included when tamoxifen-inducible mouse strains are used. If pharmaceutical agents are administered to animals, the appropriate vehicle-only control treatment should also be included.

Institutional permissions

Mouse maintenance and all steps of this protocol were conducted according to the procedures approved by the CCHMC Institutional Animal Care and Use Committee under protocol #IACUC2023-0025, and adherence to the NIH guide for the care and use of laboratory animals. A prior approval from the relevant institutional animal ethical committee must be obtained for the performance of this experiment.

Key resources table

All catalog numbers are for items sourced in the United States of America (USA).

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Goat anti-mouse/rat CD31/PECAM-1 Alexa Fluor 488 conjugated (0.5 μg/mL)	R&D Systems	Cat#FAB3628G
Mouse anti-Tubulin β 3 (TUBB3), Alexa Fluor 647 conjugated (2.5 μg/mL)	BioLegend	Cat#801210
Rat anti-mouse TER-119/erythroid cells, Alexa Fluor 647 conjugated (2.5 μg/mL)	BioLegend	Cat#116218
Rat anti-mouse CD41, PE conjugated (1 μg/mL)	BD Biosciences	Cat#BDB558040
Rabbit anti-mouse F4/80, Alexa Fluor 555 conjugated (4 μg/mL)	Cell Signaling Technology	Cat#99651
Rat anti-mouse Ly-6G, PE conjugated (4 μg/mL)	Cell Signaling Technology	Cat#18359
Rat anti-mouse LYVE-1 (5 μg/mL)	R&D Systems	Cat#MAB2125
Goat anti-rat IgG antibody (H + L), biotinylated (7.5 μg/mL)	Vector Laboratories	Cat#BA-9400-1.5
Streptavidin, Texas Red (5 μg/mL)	Vector Laboratories	Cat#SA-5006-1
Experimental models: Organisms/strains
C57BL/6J, 2 weeks–12 months old, either gender	The Jackson Laboratory	Strain #:000664 RRID:IMSR_JAX:000664
Critical commercial assays
Click-iT EdU imaging kit	Thermo Fisher Scientific	Cat#C10337
EdU (5-ethynyl-2′-deoxyuridine)	Thermo Fisher Scientific	Cat#E10187
Software and algorithms
ImageJ	Schneider et al.2	https://imagej.net
Imaris software (v.10.1)	Bitplane	http://www.bitplane.com/imaris
AngioTool software	Zudaire et al.3	https://ccrod.cancer.gov/confluence/display/ROB2/Home
NIS-Elements software – Advanced Research	Nikon	https://www.microscope.healthcare.nikon.com/products/software/nis-elements
Other
Hanks’ balanced salt solution (HBSS)	Thermo Fisher Scientific	Cat#14025092
10% neutral-buffered formalin	Sigma-Aldrich	Cat#HT501320
10× phosphate-buffered saline (PBS)	Fisher BioReagents	Cat#BP 3994
Triton X-100	Sigma	Cat#X100
Bovine serum albumin	Sigma-Aldrich	Cat#A7906
Goat serum	Vector Laboratories	Cat#S-1000-20
DAPI (4′,6-diamidino-2-phenylindole, dihydrochloride)	Invitrogen	Cat#D1306
Anti-fade fluorescent mounting medium	SouthernBiotech	Cat#0100-01
Dissecting dish	Electron Microscopy Sciences	Cat#70540
Wax dissection tray	Electron Microscopy Sciences	Cat#62366
Syringe needles (BD PrecisionGlide needle; 26G × 5/8 Sub-Q)	BD Biosciences	Cat#305115
Superfrost plus microscope slide	Thermo Fisher Scientific	Cat#12-550-15
Microscope cover glass (.17 mm thick)	VWR	Cat#48393-059

Materials and equipment

Blocking and permeabilization buffer (BPB)

Reagent	Percentage (%)	Amount
BSA	5%	5 g
Goat Serum	5%	5 mL
Triton X-100	0.5%	0.5 mL
1×PBS	N/A	94.5 mL
Total	N/A	100 mL

Store at 4°C for up to 2 weeks.

CRITICAL: Triton X-100 is hazardous. Avoid contact with skin and eyes.

EdU for in vivo administration (5 mg/mL)

• •50 mg of EdU.
• •10 mL PBS.

Aliquot and store at −20°C for up to one year.

Step-by-step method details

Intestinal tissue dissection, preparation, and fixation

Timing: 10–20 min/animal plus overnight fixation

In this section, we describe steps for the dissection of the gastrointestinal tract from the mouse and the downstream processing and fixation of the tissue that must occur prior to dissection of the muscularis tissue.

Optional: For proliferation analysis, inject mice with EdU (50 mg/kg; 10 μL/g mouse weight when using 5 mg/mL stock solution) via intraperitoneal injection at a desired time point e.g. 24 h before collection.

• 1.Euthanize mouse by CO₂ inhalation and cervical dislocation, according to the American Veterinary Medical Association guidelines.
• 2.Immobilize the mouse with syringe needles on a dissection tray to facilitate the removal of the intestine (Figure 1A).
• 3.Open the chest skin using scissors and forceps (Figure 1A) and then carefully open the peritoneum (Figures 1B and 1C) to expose the internal organs.
• 4.Carefully push up the liver to facilitate access to the stomach (Figure 1D).
• 5.To remove the gastrointestinal tract, separate the stomach just below the esophagus (Figure 1E) and cut the colon as close to the rectum as possible (Figure 1F).
• 6.Immediately place the gastrointestinal tract in a petri dish containing cold HBSS on ice (Figure 1G).

CRITICAL: It is important to keep the intestine on ice during all steps of preparation and fixation to avoid tissue degradation.

• 7.To unfold the gastrointestinal tract for further processing, carefully remove the mesentery from the intestinal tissue with small scissors (Figure 1H).
• 8.With needles, straighten and secure the GI tissue in a dissection dish (Figure 1I).
• 9.Add 10% formalin until the intestine is fully submerged and fix tissue overnight at 4°C.

Note: Handling of samples in formalin must be performed in a fume hood.

Note: For all antibodies tested herein, 10% formalin is a suitable fixative. Other antibodies may require an alternative fixation protocol.

• 10.On the following day, transfer the tissue into a 50 mL conical tube with 1× PBS for washing.

Pause point: The fixed tissue can be stored in PBS at 4°C for several weeks.

Figure 1.

Dissection and fixation of the mouse gastrointestinal tract (GIT)

(A) Mice are euthanized by CO₂ inhalation and cervical dislocation and fixed with syringe needles to the dissection board.

(B) The skin and (C) peritoneum of the lower abdomen are cut, revealing the internal organs.

(D) The liver is carefully pushed up into the diaphragm to expose the esophagus and stomach.

(E) The esophagus is cut just above where it meets the stomach.

(F) On the opposite end of the GIT, the colon is cut directly above where it meets the rectum.

(G) The complete GIT is placed into a Petri dish with cold HBSS on ice.

(H) The mesentery was carefully cut from the intestines so that the GIT could be fully straightened.

(I) Using dissection pins, the tissue was straightened and secured to a dissection dish. The tissue was covered with 10% formalin and fixed at 4°C overnight.

Preparation of the muscularis layer of the intestinal muscle

Timing: 20–30 min/animal

In this section, we describe steps for the isolation of the muscularis tissue layer from the intestine, fixation of the tissue, and storage.

CRITICAL: The dissection of the intestinal muscularis (Figure 2A) should be performed in a petri dish filled with 1× PBS to prevent the tissue from drying out.

CRITICAL: When comparing the vasculature between different groups of mice (e.g., wild type versus mutant mice) it is important to process the same portion of the intestine (e.g., ileum)

• 11.In a petri dish containing 1× PBS, use scissors to dissect about 1–2 cm of the intestine.
• 12.With scissors, open the tissue longitudinally along the mesenteric border.
• 13.Remove the intestinal content by washing the tissue in 1× PBS.
• 14.Move the tissue into a new petri dish filled with 1× PBS under a dissecting microscope.
• 15.Use forceps in one hand to hold the intestine down (Figure 2B). Place the forceps on the other hand in between the muscularis and the submucosal layer (Figure 2C).
• 16.Using the forceps, carefully separate the muscularis from the submucosa without tearing the tissue (Figures 2D and 2E).
• 17.The muscularis tissue (Figure 2F) can be moved to a 24-well plate containing 1× PBS.

Pause point: The muscularis tissue can be stored in PBS at 4°C for several months. Seal the 24-well plate with Parafilm to prevent evaporation.

Figure 2.

Isolation of the muscularis layer from the intestinal tissue

(A) Schematic of all the layers of the intestinal wall including serosa, muscularis propria, submucosa and mucosa.

(B) Using fine forceps in both hands, a ∼2 cm piece of intestinal tissue is held down with the microvilli facing towards the bottom of the dish.

(C) The submucosal layer is held down with one forceps while the other forceps is used to carefully pull up the overlying muscularis layer.

(D and E) The muscularis layer is separated from the submucosa, carefully working section by section.

(F) When the entire muscularis is removed, it appears as a tube of thin tissue. The entire tissue is then stored in PBS in a 24-well plate until further processing.

Schematics in (A) were created with biorender.com.

Whole-mount staining of vascular network and proliferating cells

Timing: About 24 h for a staining with one or more conjugated antibodies. All additional antibodies (primary and secondaries) need overnight (16-20 h) incubation.

In this section, we describe how to perform fluorescent immunostaining of the muscularis tissue to label components of the vasculature. We also include information to perform optional labeling of proliferating cells within the tissue.

Note: Intestinal muscularis samples are stained in 24-well plates. The reagent volumes for all steps are 700 μL/well to cover the tissue and all washing and incubation steps are performed on an orbital shaker. Use BPB for blocking and all primary and secondary antibody incubations. If primary antibodies are from different species, you can incubate specimens with multiple antibodies at the same time.

• 18.Block non-specific antibody binding and permeabilize the tissue by incubating the muscularis specimens in BPB in the 24-well plate, for a minimum duration of 6 h at 20°C (or room temperature) on an orbital shaker.

Optional: If proliferation analysis is desired and mice have been injected with EdU, EdU incorporation can be visualized using the Click-iT EdU imaging kit according to manufacturer’s protocol. Samples can be incubated with the Click-iT reaction cocktail for 30 min at 20°C , protected from light before washing in BPB for 1 h at 20°C . Perform EdU staining before proceeding to antibody staining.

CRITICAL: Click-iT reaction cocktail must be prepared precisely as described in the product manual, including the order in which each component is added to the mixture.

• 19.Transfer specimens into a new well and incubate them in primary antibody (e.g., Alexa 488 conjugated anti-CD31, at the concentration specified in the key resources table) diluted in BPB for 16–20 h at 4°C.

CRITICAL: Wrap plates in aluminum foil to shield them from light from this point on.

Note: When testing the specificity of a new antibody, include negative controls, including isotype IgG control and no primary antibody. It is furthermore important to titer antibodies for optimal concentration to be used. This process involves staining multiple samples of the same tissue with various concentrations of antibody, otherwise using the same staining conditions for each sample. For example, an antibody can be tested at dilutions of 1:50, 1:100, 1:200, 1:500, etc. Then, these are compared to determine which dilution yields the best signal to noise ratio.

• 20.Transfer specimens into a new well and wash them three times in 1× PBS for a minimum of 1 h each wash.

Optional: At this point, if an unconjugated primary antibody was used in Step 19, perform additional incubation with the appropriate fluorescently labeled secondary antibody. All secondary antibodies should also be diluted in BPB and incubated for 16–20 h at 4°C. We recommend utilizing secondary antibodies that have been highly cross absorbed with animal serum to minimize non-specific binding.

Optional: Furthermore, the staining protocol could be extended with additional primary antibodies to visualize other antigens or cell types present in the tissue.

• 21.Counterstain nuclei with DAPI in PBS for 2 h at 20°C (or room temperature).
• 22.Transfer specimens to a new well and wash three times in 1 mL 1× PBS.
• 23.To mount the muscularis whole-mount preparation, place the tissue on a microscopic slide with one to three drops of mounting media. Use fine forceps to straighten and flatten the tissue and cover it with a 0.17 mm coverslip.
• 24.Label the microscope slide with sample name and specify antibodies used in immunostaining.
• 25.Let the slide dry at 20°C (or room temperature) in the dark for at least 24 h before proceeding to imaging.

Pause point: Samples can be stored in the refrigerator (4°C) for several months, but it is recommended to image shortly after mounting.

Acquisition of 3D images with confocal microscopy

Timing: 30 min to setup the microscope and about 5–10 min to scan one z-stack, depending on the capabilities and processing power of the computer.

In this section, we describe steps for the acquisition of fluorescent 3D images using a confocal microscope of immunostained vascular networks or other targets in the muscularis tissue samples.

Note: The computer that we use to obtain confocal images uses the Windows 10 Enterprise operating system with an SSD, Intel Xeon CPU, 64 Gb RAM, and a Nvidia Quadro RTX 5000 GPU.

• 26.Allow the mounted sample to return to room temperature (15–30 min) and turn on the confocal microscope.

Note: The whole-mount preparations as dissected can be visualized with a 10× or 20× objective (numerical aperture 0.45 and 0.75, respectively) (Figures 3A–3C) which is ideal for vessel quantifications and analysis of vascular patterns. However, a 40× or 60× objective (numerical aperture 1.15 and 1.27, respectively) can also be used for detailed analysis of expression of specific proteins. If available, we also recommend using a resonant scanner to facilitate faster acquisition of images of thick tissues. If a resonant scanner is used, denoising software is required to reduce excess noise in images. In our laboratory, we use the Nikon Denoise.ai software for this purpose.

• 27.In the confocal settings, adjust laser and detector setting for the specific set of fluorophores used to label the sample. Then adjust the image intensity by changing the laser intensity, detector gain and other parameters (for example, laser intensity = 5% and detector gain = 75).

Note: On our microscope, DAPI was excited with the 405 nm laser and detected with a 450/50 emission filter. Alexa 488 conjugated antibodies were excited with the 488 nm laser and detected with a 525/50 emission filter. PE, Alexa 555, and Texas Red conjugated antibodies were excited with the 561 laser and detected with a 595/50 emission filter. Alexa 647 conjugated antibodies were excited with the 647 nm laser and detected with a 700/75 emission filter.

CRITICAL: For comparison of vasculature or specific markers it is important to keep the offset and gain settings constant between all samples that will be quantified or compared.

• 28.Activate the z-stack mode and setup the lower and upper limits. Make sure the z-stack includes the whole muscularis tissue (usually 50–80 μM) and set the z-stack step size as recommended by Nyquist sampling calculations, typically provided by the image acquisition software.
• 29.For quantitative analysis take z-stacks from at least 5 different randomly assigned fields in each sample.

Figure 3.

Representative images of microvessel network within the muscularis

(A) Muscularis whole-mounts were immunostained for CD31 to label blood vessels. The z-stack confocal images were processed using Imaris software to show the volume view of the 3D vessel architecture.

(B) The same confocal images can also be shown in 2D with max intensity projections (max I.P.).

(C) The “surfaces” function in Imaris software was used to reconstruct the CD31 signaling to illustrate the 3D properties of the vascular network. Scale bar: 100 μm.

Image analysis and quantification of vascular parameters

Timing: Several minutes to hours for each image, depending on the degree of optimization needed and the processing capabilities of the computer.

In this section, we describe the steps to perform analysis of the confocal images obtained in the previous steps. These analyses are optional, according to the parameters desired by the user, and provide quantification of various aspects of the vascular phenotype, including 3D reconstruction, vessel diameter, vascular area, vessel branching, and endothelial cell proliferation.

Note: Performing analysis with Imaris and NIS Elements software may require a computer with high processing power. The computer that we use for these analyses uses the Windows 10 Enterprise operating system with an SSD, Intel Xeon CPU, 218 Gb RAM, and a Nvidia Quadro RTX 5000 GPU.

• 30.
  Confocal images obtained in step 29 can be analyzed for (a) 3D-reconstruction, (b) vessel diameter, (c) vascular area, number of vessels, vessel length and branching, or (d) percentage of proliferating endothelial cells (EC).

  • a.
    Image analysis: 3D reconstruction using Imaris software (Figure 3C).

    • i.
      Convert the original image file (.nd2) to the Imaris (.ims) file type using the Bitplane Imaris Converter software.
    • ii.
      Open the file in the Bitplane Imaris software.
    • iii.
      To apply 3D surface rendering to images, use the surface module in Imaris. Click >add new surface>create.
    • iv.
      Carefully adjust parameters such as intensity thresholds to best match the vessels in the image. Color and Materials of reconstruction can be chosen to best represent the given vascular network.
    • v.
      After processing the images can be exported as Tiff or PNG files (do not export to JPEG). A d.p.i of >600 is recommended.
  • b.
    Image analysis: Vessel diameter using ImageJ (Fiji) software (Figure 4A).

    • i.
      Import z-stacks obtained at Step 29 into Fiji and convert them to maximum intensity projections: Image>Stacks>Z-Project.
    • ii.
      Calibrate the image by clicking Analyze>Set scale. The ratio of pixel value to unit of measurement can be found in the Nikon raw images files under File>Image properties.
    • iii.
      Open the ROI manager to allow measurement of multiple blood vessels per image: Analyze>Tools>ROI manager.
    • iv.
      Add a grid on top of the maximum intensity projection. Analyze>Tools>Grid.
    • v.
      Measure vessels that intersect with the grid. Using the straight-line tool, manually draw a line through the diameter of the blood vessel (90° angle to the vessel wall). To add a measurement to the ROI manager press > Ctrl+T. Repeat this measurement for each vessel intersecting with a grid.
    • vi.
      To measure the diameter, in the ROI manager select all measurements. Click on Analyze>Measure (or Ctrl+M).
    • vii.
      Copy the measurements into a spreadsheet and process them for analysis.
  • c.
    Image analysis: Vascular area, number of vessels, and vessel branching using AngioTool software (Figure 4B).

    • i.
      Import the maximum intensity projection Tiff file obtained in Step 30b(i) into the AngioTool software. Click on >Open image.
    • ii.
      Carefully adjust parameters such as vessel diameter, intensity, and particle criteria. Modify the outline overlays to best match the vessels in the image.
    • iii.
      Under the saving preferences panel, click on the Excel icon to select the path under which the data will be saved.
    • iv.
      Click on >settings panel. Under “calibration control” define the correlation between pixels in the image and mm.
    • v.
      Click on >Run analysis.
    • vi.
      Extract the data of interest from the exported spreadsheet (e.g., vascular area or branching points).
  • d.
    Image analysis: EdU⁺ proliferating EC using General Analysis 3 in NIS Elements software (Figure 5).

    • i.
      Open Nikon image file (.nd2) in NIS Elements software and create a new General Analysis 3 file.
    • ii.
      Create a threshold of the channel containing the vessel-specific immunostaining signal (e.g., CD31), adjust parameters as necessary.
    • iii.
      Create a mask of the general nuclear staining (e.g., DAPI) on the vessel-threshold created in Step ii.
    • iv.
      In parallel, create a mask of the EdU signal on the vessel-threshold created in Step ii.
    • v.
      Perform the “BrightSpots” function to identify puncta of positive signal under both the DAPI and EdU masks. Adjust parameters as necessary to ensure that individual nuclei are identified.
    • vi.
      Perform the “ObjectCount” function, which yields the number of positive objects identified in the “BrightSpots” step.
    • vii.
      Use the “Filename” function to retrieve the image name of any of the original channels. On the output of this function, use “Filter records” and define “z = 1” to remove excess data from z-stacks.
    • viii.
      Use “JoinRecords” to attach the image name to the DAPI and EdU counts obtained from the “ObjectCount” steps.
    • ix.
      Use “AppendRecords” to join all data into a single file. Further refine this data with “CompactColumns” to remove any repeating information.
    • x.
      Finish the workflow with the “ExportCSV” function. A screenshot of all steps as would be shown in the General Analysis 3 software is shown in Figure 5C.

      CRITICAL: We recommend creating an initial analysis workflow on an example image taken from the set of images to be analyzed. Once a workflow has been established, test the workflow on 1-2 additional images from each experimental group, adjusting parameters as necessary to ensure accurate quantification of all test images. Once this optimization has been performed, batch analysis can be performed on all experimental images.

    • xi.
      Save the workflow. Open “Batch GA3 Analysis” from the “Image” menu. Select the workflow and the images to be processed. Run the batch analysis.
    • xii.
      For each image, this analysis will identify the total number of EC nuclei (from the DAPI channel) and the number of EdU+ EC nuclei (from the EdU channel). Express data as a ratio of EdU+ EC nuclei over total. Average the values of all images taken from an individual mouse.

Figure 4.

Analysis of vascular parameters in ImageJ and AngioTool

(A) (Left panel) Representative max I.P. of CD31-stained vessel network of the muscularis. ImageJ software is used for vessel diameter quantification. Scale bar: 100 μm. A grid is placed on top of the image. Parameters such as size and color of the grid can be adjusted (inset). (Right panel) Enlarged area from grid is shown. Using the “line tool” the diameter of each vessel crossing two sides (right and upper side) of a square is measured and shown in yellow.

(B) (Left panel) AngioTool’s graphical user interface for analysis of vascular networks where parameters such as vessel diameter and intensity can be chosen. (Right panel) Representative image after analysis and skeletonization is shown. The vasculature is outlined in yellow, while the skeleton of the vasculature is shown in red and branching points are highlighted with blue dots.

Figure 5.

The muscularis vasculature is more proliferative in young mice compared to adult mice

(A) Young (2 weeks old) and adult (8 weeks old) mice were administered EdU 24 h before analysis. Muscularis whole-mounts were then labeled for EdU incorporation (white), CD31 antibody staining (red), and DAPI positive nuclei (blue). Scale bar: 100 μm.

(B) Images at greater magnification were reconstructed using Imaris software. A threshold of CD31 signal was used to create a mask (red outlines) that was applied to the EdU and DAPI channels. Scale bar: 100 μm.

(C) The General Analysis 3 tool in NIS Elements was used to create a workflow to quantify EdU+ and total EC nuclei. Both EdU and DAPI channels are masked on a CD31 threshold. The “BrightSpots” function is then used to identify positive nuclei, which are then counted, added to a single records file, and exported as a CSV file.

(D) The number of EdU+ EC (CD31⁺) nuclei in adult and young mice (n = 3 per group) was counted and is represented as a percentage of total EC nuclei.

Expected outcomes

The adult mouse intestine contains a complex and highly developed vascular network running throughout the intestinal wall.^4,5 These vessels provide blood to the various layers of the gastrointestinal tract, including the luminal adsorptive mucosa layer, the submucosa, and the muscularis propria (muscularis) which is the layer that provides peristaltic forces (Figure 2A). While excellent protocols have been described to investigate the vasculature within the intestinal submucosa and villi,^6,7,8 the vascular network within the intestinal muscularis is not well studied. Herein we describe a detailed protocol for the investigation of the three-dimensional (3D) structure and organization of blood vessels within the muscularis layer of the intestinal wall using confocal laser scanning microscopy. We have found that this tissue contains an organized network of microvessels (Figure 3), providing a reproducible system in which we have successfully investigated the effects of a genetic activating mutation on microvessel growth, structure, and function.¹

We show that imaging of the isolated muscularis tissue can be performed without tissue clearing with the use of a standard laser scanning confocal microscope, therefore providing a cost and time effective alternative to study the microvasculature of mice of all ages. Tissue clearing has been successfully used to enable staining and imaging of the vasculature of whole tissues that otherwise would be inaccessible for 3D imaging and analysis.^9,10 However, tissue clearing has some limitations that inhibit the ability of this technique to be used for efficient high-throughput analysis of many samples. It is time-intensive, requires the preparation of chemicals that are not standard in most molecular biology laboratories, may shrink or expand the tissue, destroy fluorescent signal, cause morphological deformation or destruction, and may interfere with established staining protocols.^11,12 Furthermore, depending on the thickness of the tissue, these protocols often require the use of a light sheet microscope.

The muscularis tissue obtained with our technique yields tissue that is thin but sturdy, allowing for efficient antibody and laser penetration. It can be flat mounted on standard microscope slides, so that the immunostaining and imaging process can easily be accomplished by any researcher familiar with traditional analysis of tissue sections.

We have used this intestinal whole-mount protocol to obtain 3D images of the vasculature within the intestinal muscularis by immunostaining with antibodies specific for markers of blood vessels. We have found Platelet endothelial cell adhesion molecule (PECAM-1), also known as cluster of differentiation 31 (CD31), immunostaining to be effective for staining of all vessels within the tissue. Figure 3 shows a representative 3D microvascular network (CD31+) that is revealed in a typical z-stack confocal image obtained using this method. Using 3D imaging analysis software, the vascular network can be visualized and reconstructed (Figure 3C).

These images can be further analyzed with free image software tools such as Fiji and AngioTool to quantify parameters of the vascular network such as vessel diameter (Figure 4A), vascular area or branching points (Figure 4B).

When combined with in vivo EdU administration, proliferating cells both within and outside of the vasculature can be identified (Figure 5A). We used this technique to demonstrate that the intestinal vasculature in young mice (2 weeks of age) is more proliferative than the vasculature of adult mice (Figures 5B–5D). NIS Elements software can be used to perform detailed 3D analysis to identify only those proliferating cells that are contained within the blood vessels (Figure 5C). While not included in this analysis, we anticipate that an additional staining step could be added to the protocol to label ERG+ (ETS-related gene) endothelial cell nuclei. Colocalization of EdU and ERG signal would specifically identify proliferating endothelial nuclei, facilitating quantification.

In addition, we have found that the muscularis of the colon contains lymphatic vessels that are positive for both CD31 and LYVE-1 (Lymphatic Vessel Endothelial Hyaluronic Acid Receptor 1, a lymphatic-specific marker) (Figure 6A). We have also shown that this protocol is compatible with genetically encoded fluorescent reporters. For example, we used the Rosa26-floxed stop-tdTomato reporter mouse and crossed them with the endothelial-specific, tamoxifen-inducible Cre-driver line Cdh5 (PAC)-iCreERT2 (iCdh5-Cre).¹³ Co-immunostaining with CD31 shows that the tdTomato signal is retained throughout the processing of the muscularis tissue and is colocalized with the CD31 staining (Figure 6B).

Figure 6.

Representative images of vascular staining in the muscularis whole- mounts

(A) Colon muscularis was immunostained for the lymphatic marker Lyve-1 (cyan) together with CD31 (red).

(B) Small intestinal muscularis tissue of Cdh5-CreER^T2-LSL-tdTomato mice shows overlap of tdTomato (cyan) signal with CD31 antibody staining (red).

(C) Co-staining of CD31 together with βIII-tubulin (cyan) reveals the neuronal network (myenteric plexus) alongside CD31 positive vessels (red) within the tissue. Scale bar: 100 μm.

Furthermore, combining the vascular immunostaining with other markers allows for analysis of interactions with the neural and immune systems. We have shown that staining of the muscularis with neuron-specific markers (such as βIII-tubulin/Tuj1) reveals the myenteric plexus within the enteric nervous system (Figure 6C).

Moreover, components of the blood and coagulation systems can be visualized within the vasculature. Here, we show the possibility to visualize TER119+ erythrocytes and CD41+ platelets (Figure 7A). These assessments have been proven particularly useful to detect vascular leakage and quantify erythrocyte extravasation and platelet accumulation, respectively. This technique could likewise be applied to investigate vascular inflammation by examining the expression of vascular inflammation markers or immune cell occurrence. We have successfully visualized such immune cells within the capillary bed including monocytes/macrophages expressing the F4/80 glycoprotein (Figure 7B) and Ly6G+ (Lymphocyte antigen 6 complex locus G6D) inflammatory cells/neutrophils (Figure 7C).

Figure 7.

Representative immunostaining for vascular endothelial cell interacting with blood and immune components

Muscularis tissue was immunostained for CD31 (red) to show the vasculature. To illustrate components of the blood and immune systems, muscularis whole-mounts were additionally stained for (A) CD41 (yellow, platelets), Ter119 (cyan, erythrocytes), (B) F4/80 (white, monocytes), and (C) Ly6G (white, neutrophils). Scale bar: 50 μm.

One major advantage of this technique compared with whole-mount staining of other tissues such as the retina^14,15 or ear skin^16,17,18,19 is the abundance of intestinal tissue that can be collected from a single mouse. The intestinal tissue can be cut into several segments to use for antibody staining of many targets for each animal, greatly reducing the number of animals. Conveniently, after fixation the muscularis tissue withstands long-term storage for staining and analysis at later times.

We expect that this protocol will be used to investigate the vasculature in the intestine for researchers from different fields, for the study of both vessel patterning and interactions. It can be used by researchers specifically interested in gastrointestinal development and function, but also used as a reproducible model system of generalized microvasculature.

In summary, the whole-mount imaging technique we describe here is simple, reproducible, and can provide a large amount of qualitative and quantitative data. The protocol is versatile and can be applied to the study of vasculature and other systems throughout development and in disease models.

Limitations

One potential limitation of this protocol relative simplicity of the vasculature within this tissue. From our experience, the muscularis tissue consists almost solely of capillary networks and larger veins and arteries are rarely seen. This may be an important experimental consideration if the presence of these larger vessels is important to the overall goals of the experiment.

In addition, this protocol very specifically isolates the vasculature of the muscularis layer but excludes the submucosal and mucosal layers. While this facilitates quantitative analysis of the isolated muscularis capillary plexus and circumvents the need for intensive tissue clearing,^20,21,22 it may be an important consideration if these structures are of experimental interest.

Furthermore, while the intestine provides abundant tissue for immunostaining of multiple targets, there may be variations in vascular, immune, and neural phenotypes throughout the various regions of the gastrointestinal tract. Therefore, care must be taken to ensure that similar areas of the intestine are examined between multiple animals. It may be necessary in some experiments to examine tissue from multiple parts of the intestine to ensure reproducibility of observed phenotypes across the tissue.

Troubleshooting

Problem 1

Increased fragility of intestinal tissue obtained from young mice (less than two weeks of age) (related to Steps 15 and 16).

The muscularis tissue in intestines of young mice might be extremely thin and fragile compared to tissue from older mice. This tissue may be more susceptible to tearing during the dissection process which may cause issues with downstream imaging and analysis.

Potential solution

Fixation of the whole intestine prior to dissection of the muscularis may contribute to increased attachment of this layer to the underlying tissue, causing increased tearing in delicate tissues. To solve this problem, we recommend performing dissection of the muscularis on freshly removed intestines. The muscularis layer can then be pinned and fixed in formalin. The remainder of the staining and analysis protocol can then be performed without modification.

Problem 2

Taking confocal z-stack images with high resolution objectives (40× and higher) requires much longer acquisition times (related to Step 28).

The muscularis whole-mounts are typically between usually 50–80 μM thick. At higher resolution and therefore smaller confocal step-sizes, the time it takes to acquire images throughout the full thickness of the tissue may be time- and cost-inhibitory.

Potential solution

To solve this problem, we recommend using a confocal microscope equipped with a resonant scanner, which greatly reduces the time it takes to capture a single image and reduces photobleaching. In addition, a computer that is equipped with faster GPU, CPU, and SSD are recommended to reduce the time needed to process images for acquisition and storage.

Problem 3

There is no or low signal during imaging (related to Step 27).

Potential solution

• •Fixation method (Step 9): The formalin fixation method used in our protocol may influence binding of some antibodies. We recommend testing different fixation times and reagent methods for each antibody.
• •Storage (Step 17): Excessive long-term storage of samples may result in degradation of antibody epitopes. If antibody staining is not working as expected, test the staining protocol on fresh tissues.
• •Antibody concentration (Step 19): The concentration of antibody may be too low. We recommend performing antibody titrations to optimize the optimal antibody concentration.
• •Antibody incubation (Step 19): The amount of time that the tissue is incubated in antibody may be too low. We recommend increasing the amount of time that the sample is incubated with antibody.
• •Confocal microscope settings (Step 27): Incorrect microscope settings may prevent detection of signal. Confirm that the appropriate excitation and emission wavelengths are being employed for the fluorescent dyes used. Also perform optimization of laser strength and photodetector gain to optimize true signal versus background. Using negative control samples (isotype and secondary-only antibodies) are important for determining this difference.
• •Antibodies that recommend storage at freezing temperatures should be stored in aliquots to minimize freeze-thaw cycles that can result in antibody degradation and reduced performance.

Problem 4

There is high background signal during imaging (related to Step 27).

Potential solution

• •Blocking (Step 18): Blocking before antibody incubation may be insufficient or the blocking serum may react with the secondary antibody. We recommend optimizing blocking reagents (protein or serum) for each antibody and ensuring that any animal serum used matches the host species of the secondary antibodies.
• •Antibody concentration (Step 19): The concentration of antibody may be too high. We recommend performing antibody titrations to optimize the optimal antibody concentration.
• •Antibody incubation (Step 19): The amount of time that the tissue is incubated in antibody may be too high. We recommend decreasing the amount of time that the sample is incubated with antibody.
• •Secondary antibody (Step 20): Using high-quality secondary antibodies will minimize potential off-target binding and background fluorescence. Pre-absorption of the secondary to mouse serum or cross-absorption with many types of animal serum is highly recommended. In our experience, Invitrogen (through Thermo Fisher Scientific) sells highly-crossed absorbed secondary antibodies with Alexa Fluor Plus fluorophores that typically yield excellent results.
• •Washing (Step 20): There may be insufficient washing of excess antibody. We recommend using longer washing times and increasing the number of washes.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Elisa Boscolo (elisa.boscolo@cchmc.org).

Technical contact

Questions about the technical specifics of performing the protocol should be directed to and will be answered by the technical contact, Sandra Schrenk. (Sandra.schrenk@cchmc.org).

Materials availability

This study did not generate new unique reagents.

Data and code availability

This study did not generate and analyze datasets or codes.