Whole-Mount Immunofluorescence Staining, Confocal Imaging and 3D Reconstruction of the Sinoatrial and Atrioventricular Node in the Mouse

Summary

We provide a step-by-step protocol for whole-mount immunofluorescence staining of the sinoatrial node (SAN) and atrioventricular node (AVN) in murine hearts.

Abstract

The electrical signal physiologically generated by pacemaker cells in the sinoatrial node (SAN) is conducted through the conduction system, which includes the atrioventricular node (AVN), to allow excitation and contraction of the whole heart. Any dysfunction of either SAN or AVN results in arrhythmias, indicating their fundamental role in electrophysiology and arrhythmogenesis. Mouse models are widely used in arrhythmia research, but the specific investigation of SAN and AVN remains challenging.

The SAN is located at the junction of the crista terminalis with the superior vena cava and AVN is located at the apex of the triangle of Koch, formed by the orifice of the coronary sinus, the tricuspid annulus, and the tendon of Todaro. However, due to the small size, visualization by conventional histology remains challenging and it does not allow the study of SAN and AVN within their 3D environment.

Here we describe a whole-mount immunofluorescence approach that allows the local visualization of labelled mouse SAN and AVN. Whole-mount immunofluorescence staining is intended for smaller sections of tissue without the need for manual sectioning. To this purpose, the mouse heart is dissected, with unwanted tissue removed, followed by fixation, permeabilization and blocking. Cells of the conduction system within SAN and AVN are then stained with an anti-HCN4 antibody. Confocal laser scanning microscopy and image processing allow differentiation between nodal cells and working cardiomyocytes, and to clearly localize SAN and AVN. Furthermore, additional antibodies can be combined to label other cell types as well, such as nerve fibers.

Compared to conventional immunohistology, whole-mount immunofluorescence staining preserves the anatomical integrity of the cardiac conduction system, thus allowing the investigation of AVN; especially so into their anatomy and interactions with the surrounding working myocardium and non-myocyte cells.

Introduction

Arrhythmias are common diseases affecting millions of people, and are the cause of significant morbidity and mortality worldwide. Despite enormous advances in treatment and prevention, such as the development of cardiac pacemakers, treatment of arrhythmias remains challenging, primarily due to the very limited knowledge regarding underlying disease mechanisms^1,2,3. A better understanding of both the normal electrophysiology and the pathophysiology of arrhythmias may help to develop novel, innovative and causal treatment strategies in the future. Additionally, to comprehensively study arrhythmogenesis, it is important to localize and visualize the specific cardiac conduction system in animal models such as the mouse, as mice are widely used in electrophysiology research.

The major parts of the cardiac conduction system are the sinoatrial node (SAN), where the electrical impulse is generated in specialized pacemaker cells, and the atrioventricular node (AVN), which is the only electrical connection between the atria and the ventricles⁴. Whenever the electrophysiological properties of SAN and AVN are altered, arrhythmias such as sick sinus syndrome or atrioventricular block can occur which may lead to hemodynamic deterioration, syncope and even death, and thus underline the essential role of both SAN and AVN in electrophysiology and arrhythmogenesis⁵.

Comprehensive studies on SAN or AVN require a precise localization and visualization of both structures, ideally within their physiologic environment. However, due to their small size and location within the working myocardium, without establishing a clear macroscopically visible structure, studying the anatomy and electrophysiology of SAN and AVN is challenging. Anatomical landmarks can be used to roughly identify the region that contains SAN and AVN^6,7,8. In brief, SAN is located in the inter-caval region of the right atrium adjacent to the muscular crista terminalis (CT), AVN is located within the triangle of Koch established by the tricuspid valve, the ostium of the coronary sinus and the tendon of Todaro. Thus far, these anatomical landmarks were mainly used to localize, remove and then study SAN and AVN as individual structures (e.g., by conventional histology). To better understand the complex electrophysiology of SAN and AVN (e.g., regulatory effects of adjacent cells of the working myocardium), however, studying the conduction systems within the physiologic 3D environment is necessary.

Whole-mount immunofluorescence staining is a method that is used to study anatomical structures in situ while preserving the integrity of the surrounding tissue⁹. Taking advantage of confocal microscopy and image analysis software, SAN and AVN can be visualized with fluorescently labeled antibodies targeting ion channels specifically expressed in these regions.

This following protocol explains the necessary steps to perform a well-established whole-mount staining method for SAN and AVN microscope localization and visualization. Specifically, this protocol describes how (1) to localize SAN and AVN by anatomical landmarks to prepare these samples for staining and microscopy analysis (2) to perform whole-mount immunofluorescence staining of the reference markers HCN4 and Cx43 (3) to prepare SAN and AVN samples for confocal microscopy (4) to perform confocal imaging of SAN and AVN. We also describe how this protocol can be modified to include additional staining of surrounding working myocardium or non-myocyte cells such as autonomous nerve fibers which allows a thorough investigation of the cardiac conduction system within the heart.

Protocol

Animal care and all experimental procedures were conducted in accordance with the guidelines of the Animal Care and Ethics committee of the University of Munich, and all the procedures undertaken on mice were approved by the Government of Bavaria, Munich, Germany (ROB-55.2-2532.Vet_02-16-106, ROB-55.2-2532.Vet_02-19-86). C57BL6/J mice were purchased from Jackson Laboratory.

NOTE: Figure 1 shows the instruments needed for the experiment. Figure 2 shows an illustration of the gross cardiac anatomy. Figure 3 shows the location of SAN and AVN in an adult mouse heart. Figure 4 shows the prepared sample loaded on the confocal microscopy.

1. Preparations

1. Prepare a 4% formaldehyde solution (4% PFA) by diluting 10 mL of 16% formaldehyde solution in 30 mL of PBS.
2. Prepare a 15% sucrose solution and 30% sucrose solution by dissolving 15 g and 30 g of sucrose powder into 100 mL PBS, respectively. After the sucrose powder is fully dissolved, filter the solution using 0.2 µm syringe filter before storing at 4 °C.
3. Prepare a 3%-4% agarose gel, add 3 g or 4 g agarose powder and 100 mL of 1x TAE (diluted from 50x TAE stock solution) into a beaker. Place the beaker on a magnetic stirrer and boil it until the agarose is completely dissolved.
4. Prepare the dissection dish by gently pouring 30 mL of agarose gel (3%-4%) in a 100 mm diameter Petri dish. Leave the Petri dish on the bench at room temperature to cool down and harden.
5. Prepare blocking solution and washing solution according to the recipes provided in the Table 1. Prepare all of the solutions prepared at the day of use. Long term storage is not recommended.
6. Prepare the antibody solutions by diluting them in cold (4 °C) 1x PBS, protect the dilution from light (e.g., by wrapping aluminum foil around) and keep dilutions on ice until used. Dilute the antibodies shortly before incubation. Avoid leaving the diluted antibodies at room temperature.
   NOTE: The appropriate antibody dilution should be tested with comparable tissue samples by performing a conventional immunofluorescent staining. Here, we tested the antibodies by staining frozen sections cut at a thickness of 10 µm.

2. Organ harvest and tissue preparation

1. Anesthetize the mouse by placing it into an incubation chamber connected to an isoflurane vaporizer. Set the vaporizer to deliver 4-5% isoflurane (96-95% oxygen, respectively).
   NOTE: Full anesthesia is confirmed by the loss of the postural reaction and righting reflex by gently rolling the chamber until the mouse is placed on its back.
2. Put the mouse in a supine position on the surgical table. Place the mouse nose into an anesthesia mask connected to a modified Bain circuit with its inner tube connected to the isoflurane vaporizer. To maintain anesthesia, use 1-2% isoflurane in oxygen at a flow rate of 1 L/min. Scavenge away excess anesthetic vapor from the mouse via the outer tube of the mask and draw through a canister of activated charcoal, which absorbs the excess anesthetic gas.
3. When full anesthesia is achieved, inject fentanyl for analgesia (0.1 µg/25 g body weight i.p.).
4. When the toe-pinch reflex is undetectable, make a clear cut from the jugulum to the symphysis using iris scissors to remove fur and skin. Make another cut from left to right underneath the ribs using iris scissors to carefully open the abdomen.
5. Life the xiphoid a little bit using curved forceps to allow cutting the diaphragm from left to right without injuring any organs. Cut the rib cage in a medial axillary line on both sides using iris scissors to flip it cranially and to allow access to the heart.
6. Cut the inferior vena cava and descending thoracic aorta at the level of the diaphragm using iris scissors. Puncture the heart with a 27 G needlein the area of the apex and then carefully push the needle into the left ventricle (LV). Gently inject 5-10 mL of ice-cold PBS into the LV to perfuse the heart.
   NOTE: The color of the heart should turn from red to gray indicating successful perfusion with PBS.
7. Carefully lift the apex of the heart using a tweezer allowing to cut the large vessels and to remove the heart.
8. Excise the heart by cutting the large arteries and veins as far away from the heart as possible to avoid any damage to the superior vena cava (SVC). Conserve the SVC since it will serve as important landmark during later processing.
9. After heart removal, turn off the isoflurane vaporizer.
10. Put the heart into a dissection dish filled with ice cold PBS under the dissecting microscope. After determination of the left/right and front/back of the heart, turn the heart around with the front of the heart at the bottom of the dish (to expose the large vessels that are located posterior).
11. Immobilize the heart by putting little pinsthrough the apex and the left atrial appendage (LAA) into the agarose at the bottom of the dissection dish (Figure 2A). Using fine tweezers and scissors, carefully remove non-cardiac tissue around the SVC and inferior vena cava (IVC) (e.g., lungs, fat, pericardium) to expose the inter-caval region (Figure 3A).
12. Remove the majority of the ventricles by cutting parallel the groove between the ventricles and the atria with the micro scissor. Preserve a small part of the ventricular tissue for the later loading on the Plexiglas ring.
13. For fixation and dehydration, put the sample (containing the atria, SVC and IVC) in 4% PFA overnight at 4 °C.
14. The next day, transfer the heart to 15% sucrose solution for 24 hours at 4 °C.
15. The next day, transfer the heart to 30% sucrose solution for 24 hours at 4 °C.
    NOTE: For the fixation and dehydration in step 2.13, 2.14 and 2.15, the samples are left at 4°C without further stirring or rocking.

3. Whole-mount immunofluorescence staining

1. Wash the heart in 1% Triton X-100 diluted in PBS and block and permeabilize in blocking solution (Table 1) overnight at 4 °C.
2. Place the heart in a 1.5 mL tube and incubate with rabbit anti-mouse connexin-43 (dilution of 1:200) and rat anti-mouse HCN4 (dilution of 1:200) antibodies diluted with blocking solution for 7 days at 4 °C.
   NOTE: The optimal concentration of primary antibodies should be tested before using the datasheets as orientation. Other antigens of interest could also be stained in this step, as long as the host species of the primary antigen is different from the other ones. A higher concentration of Triton X-100 may help obtain more efficient antibody staining as demonstrated before¹⁰, but concentration might be determined individually.
3. After 7 days, remove the solution containing primary antibodies using a pipetteand wash the heart with 1%Triton X-100 solution 3 times (each time for 1 h at room temperature on the orbital shaker).
4. After washing, incubate the heart = with Alexa Fluor 488 goat anti-rat IgG (dilution of 1:200) and Alexa Fluor 647 goat anti-rabbit IgG (dilution of 1:200) for 7 days at 4 °C.
5. After 7 days, remove all the solution containing the secondary antibodies using a pipette. Then wash the heart using washing solution (Table 1) 3 times (each time for 1 h at room temperature on the orbital shaker).
6. To stain nuclei, incubate the heart in DAPI solution (10 µg/mL) overnight at 4 °C.
7. On the next day, wash the heart with washing solution 3 times (each time for 1 h at room temperature on the orbital shaker).
   NOTE: For the blocking, permeabilization, antibodies incubation and DAPI staining in step 3.1, 3.2, 3.4 and 3.6, leave the samples at steady state in the 4 °C, no stirring is required. The stained tissue could be preserved fully covered with washing solution at 4 °C and protected from light for a few days until imaging at the confocal microscope.

4. Confocal microscopy

1. Prepare Plexiglas rings and fill with plasticine (Figure 1). In the center of the plasticine a little groove is formed in the center for loading the heart. Make a shallow groove, as this would be easier to acquire a flat imaging plane, and to adjust the space and avoid air bubbles between the sample and the coverslips in step 4.4.
2. Use the same heart sample for the imaging of both SAN and AVN sequentially. For SAN imaging, directly load the heart whereas for AVN imaging, perform microdissection before.
   1. SAN imaging
      1. Identify the SAN by anatomical landmarks: it is located on the dorsal side of the heart within the inter-caval region (the region between the superior and inferior vena cava). The crista terminalis (CT) is the muscle streak between the SAN and RAA.
      2. Place the heart into the plasticine groove with the back of the heart facing up. Add PBS onto the heart to displace all air within the cavity until the heart is fully covered with PBS (usually a few drops of PBS are sufficient).
      3. Under the dissection microscope, gently press the RAA, LAA and remaining parts of the left ventricle into the plasticine to fix the heart. Make sure that the whole inter-caval region and the crista terminalis could be clearly seen (not covered by plasticine). The area that will be imaged is shown in Figure 3A.
   2. AVN imaging
      1. After finish the imaging of SAN, recollect the same sample and subsequently use for the AVN imaging. For AVN microdissection, place the heart = on a dissection dish under the dissection microscope.
      2. Orient the heart with the right side facing up (including the remaining parts of the right ventricle). Put pins through the remaining part of the LV free wall (which is now at the bottom) to immobilize the tissue (Figure 2B).
      3. Cut the remaining RV free wall upwards through the tricuspid valve and superior vena cava. Then flip the RV and RA away to expose interventricular and interatrial septum. (Figure 3B)
         NOTE: Pay attention not to damage the coronary sinus (CS), as it is an important anatomical landmark to find the triangle of Koch which then allows to localize the AVN. The CS runs transversely in the left atrioventricular groove on the posterior side of the heart, and the CS orifice opening is located between IVC and the tricuspid valve at the inferior part of the interatrial septum (Figure 3A and B).
      4. Identify the triangle of Koch that can be found on the endocardial surface of the right atrium, bordered anteriorly by the hinge-line of the septal leaflet of the tricuspid valve (TV), and posteriorly by the tendon of Todaro. The base is formed by the orifice of the coronary sinus (Figure 3B). Since this is the target region for AVN imaging, it needs to be clearly visible.
      5. Transfer the heart into the plasticine groove within the Plexiglas ring with the triangle of Koch clearly being exposed (i.e. not covered by plasticine). Gently press the tissue around the Koch triangle into the plasticine to fix the sample. Add PBS onto the heart to displace all air within the cavity until the heart is fully covered with PBS (usually a few drops of PBS are sufficient).
3. Apply silicone to the edges of the Plexiglas rings to allow covering the hearts loaded within the plasticine-filled Plexiglas rings with cover slips.
4. Gently press the back side of the plasticine to squeeze out parts of the PBS and to attach the heart to the coverslip while avoiding any air bubbles within the imaging area.
   NOTE: Make sure that the regions of interest are not folded and covered during pressing the back side of the plasticine. A flat imaging plane without sample overcompression is necessary for conserving the anatomy and for proper confocal imaging of the samples. For SAN, it is important to make sure the inter-caval region is clearly exposed. For AVN, the triangle of Koch should be fully exposed.
5. Put the whole-mount staining samples up-side down on the platform of the confocal microscope. The exposed SAN/AVN attached to the cover slip is now on the bottom of the microscope platform (Figure 4).
   NOTE: To take images, we use the Carl Zeiss LSM800 with Airyscan Unit and the software ZEN 2.3 SP1 black.
6. Choose plate "BP420-480 + LP605" for the excitation of Alexa Fluor 647 conjugated anti-HCN4. Vary the master gain from 650-750.
7. Take an overview image of the whole SAN and AVN region by using Tile Scan function. Then select the HCN4-positive region by clicking and adding a square on the overview image around the area of interest that will be scanned.
8. Check the parameters, including plates and master gain for the remaining channels (Alexa Fluor 488 and DAPI) of the confocal microscope and set as described in step 4.6.
9. Slowly adjust the focus from top to bottom of the sample to preview the whole sample and to set the First and Last for the Z-stack range. Set an optimal interval for the Z-stack based on the thickness of the optical sample. We use an 20x objective, and 0.8-1 µm as the interval for Z-stack.
10. After all the parameters are properly set, scan the whole area of the SAN and AVN.
11. Perform 3D reconstruction of the images using software (e.g., Imaris version 8.4.2).
    1. Select Images Processing | Baseline Subtraction to remove background staining.
    2. Select surface creation onto setting for selected channel and region of interest for processing.

Representative Results

By using the protocol outlined above, confocal microscopy imaging of both SAN and AVN can be reliably performed. Specific staining of the conduction system using fluorescent antibodies targeting HCN4 and staining of the working myocardium using fluorescent antibodies targeting Cx43 allows the clear identification of SAN (Figure 5, Video 1) and AVN (Figure 6, Video 2) within the intact anatomy.

Figure 1: Instruments for dissection and Plexiglas ring holder. A. Plexiglas ring filled with plasticine (red arrow shows the groove formed in the center for loading the heart). B. Dissection Petri dish with agarose gel. C. Spring scissors. D. Iris scissors. E. Curved forceps. F. Fine forceps. Please click here to view a larger version of this figure.

Figure 2: Illustration of the gross cardiac anatomy. A. Schematic illustration of the SAN microdissection. Pins are applied through the apex and the left atrial appendage (LAA). SAN is indicated by red dashed circle. B. Schematic illustration of the AVN microdissection. Pins are put through the remaining part of the LV free wall (which is now at the bottom). AVN is indicated by a red dashed circle. SVC, Superior vena cava; IVC, inferior vena cava; CS, coronary sinus; LAA, left atrial appendage; RAA, right atrial appendage; PA, pulmonary artery; PV, pulmonary vein; LV, left ventricle; IVS, interventricular septum; IAS, interatrial septum; OF, oval fossa. Please click here to view a larger version of this figure.

Figure 3: Location of SAN and AVN in an adult mouse heart. A. Location of the sinoatrial node (SAN) in an adult mouse heart. View from the back of the heart. Location of the SAN is indicated by red dashed line within the inter-caval region (black dashed lines). SVC, Superior vena cava; IVC, inferior vena cava; CS, coronary sinus; LAA, left atrial appendage; RAA, right atrial appendage; PV, pulmonary vein; CT, Crista terminalis; LV, left ventricle; RV, right ventricle. B. Location of the atrioventricular node (AVN) in an adult mouse heart. View from the right. The AVN (red dashed circle) is located at the apex of the triangle of Koch (white dashed triangle) near the bottom of the membranous septum. The triangle of Koch is formed by the tendon of Todaro (TT, green dashed line), tricuspid valve (TV, blue dashed line) and the orifice of the coronary sinus (CS, yellow dashed line). SVC, superior vena cava; IVC, inferior vena cava; IVS, interventricular septum; OF, oval fossa. Please click here to view a larger version of this figure.

Figure 4: Prepared sample loaded on the confocal microscopy. A. The prepared sample (red arrow) for confocal microscopy mounted into the Plexiglas ring with plasticine (black arrow). View from the bottom. B. Plexiglas ring with mounted sample loaded up-side-down on the platform of the confocal microscope (inverse microscope). Please click here to view a larger version of this figure.

Figure 5. Reconstruction of confocal microscopy imaging of SAN in a C57BL6/J mouse stained with anti-HCN4 (red) and Cx43 (white) antibodies. A. The dashed area delineates the SAN, which is oriented along the inter-caval region between the superior and inferior vena cava. B. Magnification of the inlayfrom panel A. C. 3D reconstruction of panel B. RAA, right atrial appendage; RA, right atrium; SAN, sinoatrial node. Please click here to view a larger version of this figure.

Figure 6. Reconstruction of confocal microscopy imaging of AVN in a C57BL6/J mouse stained with anti-HCN4 (red) and Cx43 (white) antibodies. A. Dashed area indicates the AVN. B. Magnification of the inlay from panel A. C. 3D reconstruction of panel B. IVS, interventricular septum; IAS, interatrial septum; AVN, atrioventricular node. Please click here to view a larger version of this figure.

Figure 7. Negative control image. A. negative control whole-mount staining for SAN. The dashed circle showed the SAN region confirmed by the anatomy landmarks. B. negative control staining for SAN only with second antibodies. C. DAPI staining for SAN. D. negative control whole-mount staining for AVN. The dashed circle showed the AVN region confirmed by the anatomy landmarks. E. negative control staining for AVN only with second antibodies. F. DAPI staining for AVN. RAA, right atrial appendage; RA, right atrium; SVC, superior vena cava; IVS, interventricular septum; IAS, interatrial septum; Please click here to view a larger version of this figure.

Video 1: 3D Reconstruction of the SAN Please click here to download this video.

Video 2: 3D Reconstruction of the AVN Please click here to download this video.

Compound	Final concentration	g or mL/100 mL required
Fixing solution
Formaldehyde (16%)	4%	25 mL
PBS (1x)		75 mL
1% Triton solution
Triton X100	1%	1 mL
PBS		99 mL
Blocking solution
Triton X-100	1%	1 mL
BSA	0.50%	0.5 g
Normal goat serum	20%	20 mL
PBS (1x)		89 mL
Washing solution
Tween 20	0.10%	0.1 mL
BSA	0.50%	0.5 g
PBS (1x)		99.9 mL
TAE (50x)
Tris-base	24.20%	24.2 g
100% acetic acid	5.71%	5.71 mL
0.5 M EDTA	0.05 M	10 mL
dH2O		Add up to 100 mL

Table 1.

Discussion

Cardiac anatomy has traditionally been studied using thin histological sections¹¹. However, these methods do not preserve the three-dimensional structure of the conduction system and thus, only provides 2D information. The whole-mount immunofluorescence staining protocol described here allows to overcome these limitations and can be routinely used for SAN and AVN imaging.

In comparison to standard methods such as conventional immunohistochemistry that require paraffin-embedding, sectioning and antigen retrieval, the whole-mount methodology is advantageous for addressing the exact 3D localization and morphology of SAN and AVN, and to examine the relationship with the surrounding tissue since the tissue morphology is considerably preserved with only minimum tissue dehydration and physical rupture. Also, the immunoreactivity (i.e., the binding of antibodies to tissue antigens) is also highly maintained.

We established a practical sample procession method for confocal microscopy by using a plastic ring mounted holder filled with plasticine^12,13. Plasticine is ideal since it can be individually adjusted to the size of the tissue, it is hard enough to reliably hold the tissue without excessive pressure of the tissue. Most importantly, it is not fluorescent, avoiding auto-fluorescent background and therefore not interfering with the fluorescent signal of the antibodies used.

We used the confocal laser-scanning microscope (LSM-Zeiss) with the Airyscan detector, which can offer a distinct advantage in obtaining images with high quality. Using a widefield imaging microscope can only offer tissue-level information but lacks cellular resolution. In additional, the widefield microscope carries a risk of high background¹⁴. By using an Airyscan detector for confocal LSM, improved resolution and signal-to-noise ratio (SNR) can be acquired, compared to the traditional one pinhole-and-detector confocal imaging system¹⁵.

The shapes and position of SAN and AVN are presented as 3D reconstruction and additional virtual section planes derived from 3D reconstruction can enhance the interpretation of histological sections, whereas conventional histology allows only a 2D assessment of the anatomy with the inherent risk of non-detection of small but relevant structures. Moreover, 3D reconstruction provides the opportunity to examine anatomical structures of interest transmurally and within the intact microenvironment, for example intrinsic autonomic nervous plex innervating the heart¹⁶. Whole mount in situ staining and 3D reconstruction allows investigation of the cellular environment as well as the specific cell types and their orientation. Furthermore, regional and cell-type specific protein expression can be visualized and may further support western blot and proteomics analyses¹⁷.

SAN and AVN are specialized structures within the heart with a different expression pattern of ion channels and connexins that can be used for reliable identification^10,18. Hyperpolarization-activated cyclic nucleotide-gated cation channels (HCN) establish the diastolic depolarization in pacemaker cells and are therefore specifically expressed within the conduction system with HCN4 being the predominant isoform in the mouse¹⁹. Studies have shown that HCN4 is one of the detectable and stably expressed HCN isoforms throughout the SAN and AVN^11,20. Connexins directly connect neighboring cells and allow the passage of ions from cell to cell; they are therefore essential for the conduction of the electrical impulse through the heart²¹. Connexin expression patterns vary between different regions in the heart with Cx43 being the most abundant isoform that has been found in almost all parts of the heart except for the SAN and AVN²². Combining microdissection to allow exposure of specific regions of interest with anti-HCN4 and anti-CX43 antibodies as well as in silico approaches for three-dimensional reconstruction of confocal images allow reliable identification of SAN and AVN and to comprehensive investigation of the SAN/AVN morphology as well as their interaction with the surrounding tissue.

The SAN and AVN might be easy to distinguish after staining with a specific antibody. To localize the SAN and AVN by their anatomical landmarks, however, some practice is needed before the correct microdissection can be performed.

Using the above protocol, a number of other antigens can also be applied. The protocol also works well on both perfusion-fixed and immersion-fixed tissue. However, the incubation time may be adjusted for optimal fixation since certain antibodies show reduced immunoreactivity when the antigen is over-fixed. As the whole-mount staining approach needs 7 days for primary and secondary antibody incubation, it is important to completely submerge the samples in adequate volumes of antibody-containing solutions. The dilution ratio of each antibody should be tested before the experiment. For optimal staining, any irrelevant tissue should be removed. For the protocol, we only preserved little parts of the ventricles for better orientation and remove all the surrounding fat tissue and pulmonary veins, as surrounding (irrelevant) tissue might bind antibodies as well, especially when the target antigens are globally expressed.

Whole-mount in situ staining, confocal imaging and 3D reconstruction is an invaluable and powerful technique for researchers from various fields. However, the method also has some limitations. (i) It requires specialized equipment such as a confocal microscope that is not commonly available at every institution. (ii) As mentioned above, microdissection of specialized anatomic structures such as the SAN or AVN but also confocal imaging and post-imaging processing requires intensive practicing and experienced personnel. (iii) Success of the method highly relies on the quality of the antibodies used and may therefore be very challenging in some cases. Also (iv) 3D reconstruction can be difficult to achieve and may require intensive troubleshooting to optimize the procedure. For example, optimal intervals have to be set during image acquisition since sections spaced too far apart will leave gaps and might not allow an adequate 3D reconstruction of the anatomy. Sections that are too close may cause oversampling and bleaching due the excessive illumination and will take a long time to complete one image. Finally (v) the major limitation of confocal microscopy is the imaging depth which means that if sample thickness is beyond the maximum operating depth of the confocal microscope, the further fluorophore cannot be detected.

Materials

Name	Company	Catalog Number	Comments
Anesthesia
Isoflurane vaporizer system	Hugo Sachs Elektronik	34-0458, 34-1030, 73-4911, 34-0415, 73-4910	Includes an induction chamber, a gas evacuation unit and charcoal filters
Modified Bain circuit	Hugo Sachs Elektronik	73-4860	Includes an anesthesia mask for mice
Surgical Platform	Kent Scientific	SURGI-M
In vivo instrumentation
Fine forceps	Fine Science Tools	11295-51
Iris scissors	Fine Science Tools	14084-08
Spring scissors	Fine Science Tools	91500-09
Tissue forceps	Fine Science Tools	11051-10
Tissue pins	Fine Science Tools	26007-01	Could use 27G needles as a substitute
General lab instruments
Orbital shaker	Sunlab	D-8040
Magnetic stirrer	IKA	RH basic
Pipette,volume 10 µL, 100 µL, 1000 µL	Eppendorf	Z683884-1EA
Microscopes
Dissection stereo- zoom microscope	VWR	10836-004
Laser Scanning Confocal microscope	Zeiss	LSM 800
Software
Imaris 8.4.2	Oxford instruments
ZEN 2.3 SP1 black	Zeiss
General Lab Material
0.2 µm syringe filter	Sartorius	17597
100 mm petri dish	Falcon	351029
27G needle	BD Microlance 3	300635
50 ml Polypropylene conical Tube	Falcon	352070
5ml Syringe	Braun	4606108V
Cover slips	Thermo Scientific	7632160
Eppendorf Tubes	Eppendorf	30121872
Chemicals
0.5 M EDTA	Sigma	20-158	Components of TEA
16% Formaldehyde Solution	Thermo Scientific	28908	use as a 4% solution
Acetic acid	Merck	100063	Components of TEA
Agarose	Biozym	850070
Bovine Serum Albumin	Sigma	A2153-100G
DPBS (1X) Dulbecco's Phosphate Buffered Saline	Gibco	14190-094
Normal goat serum	Sigma	NS02L
Sucrose	Sigma	S1888-1kg
Tris-base	Roche	TRIS-RO	Components of TEA
Triton X-100	Sigma	T8787-250ml	Diluted to 1% in PBS
Tween 20	Sigma	P2287-500ml
Drugs
Fentanyl 0.5 mg/10 mL	Braun Melsungen
Isoflurane 1 mL/mL	Cp-pharma	31303
Oxygen 5 L	Linde	2020175	Includes a pressure regulator
Antibodies
Goat anti-Rabbit IgG Alexa Fluor 488	Cell Signaling Technology	#4412	diluted to 1:200
Goat anti-Rat IgG Alexa Fluor 647	Invitrogen	#A-21247	diluted to 1:200
Hoechst 33342, Trihydrochloride, Trihydrate (DAPI)	Invitrogen	H3570	diluted to 1:1000
Rabbit Anti-Connexin-43	Sigma	C6219	diluted to 1:200
Rat anti-HCN4 (SHG 1E5)	Invitrogen	MA3-903	diluted to 1:200
Other
Plexiglass ring			Self-designed and 3D printed
Plasticine	Cernit	49655005
Silikonpasten, Baysilone	VWR	291-1220
Animals
Mouse, C57BL/6	The Jackson Laboratory