Modifications of the Langendorff Method for Simultaneous Isolation of Atrial and Ventricular Myocytes from Adult Mice

Single CM is a valuable and indispensable tool in cellular level studies of cardiac function and diseases²⁰. Hence, the isolation of viable CMs from the hearts is the initial and most crucial step. Cell quality is one of the significant determinants of conducting successful experiments, especially in optical and electrophysiological experiments. Compared with the CMs of other animals, rodent CMs are more vulnerable to ischemia and hypoxia because of a higher concentration of intracellular sodium ions, which favors calcium influx through the Na⁺/Ca²⁺ exchanger²¹. Moreover, the number of AMs is far lesser than that of VMs; thus, successful isolation is extremely difficult to achieve. The Langendorff method is excellent for isolating mice VMs²², but the success rate in isolating AMs is low, and few reports are available. The proper depth of aorta cannulation is also an essential factor in yielding ideal VMs aside from temperature, enzyme activity, PH, and water quality used for buffer preparation. The principle of the Langendorff method relies on retrograde perfusion of the heart. Upon perfusion, the aortic valve is closed; thereby, the perfusate is forced into the coronary arteries, delivering enzyme solution through the vessel branches, and the myocardial tissue is evenly digested. To achieve this kind of circulation pattern, the aorta must reserve enough length for cannulation and ligation, also the cannula tip must not penetrate the aortic valves or block the CA ostium. Thus, it is reasonable to speculate that the depth of aorta cannulation is also associated with atria perfusion, which affects the digestion efficacy of the atria and the yields of AM in a similar way. The protocol presented here confirmed the hypothesis, and crucial steps for optimizing cell yield along with the suggestions are noted below.

In step 1.9, to better secure the aorta, a blunt 20 G cannula with a notch (or a circumferential groove) from where the distance is 1 mm to the tip is recommended. Based on our experiences, the cannula size that is a little bit larger than the diameter of the aorta was found to prevent the cannula tip from puncturing the aortic valves during cannulation because the aorta, relying on its intrinsic elasticity, can snugly fit into the cannula and produce friction, which acts as a protective factor when moving forward or backward adjusting the cannulation depth. In terms of the notch's position, the heart will easily slip off during cannulation because of gravity if it is too near the cannula tip. Conversely, the space for adjusting the cannulation depth and ligation position will be very confined if too far. Given these circumstances, transecting the aorta between the left common carotid artery (as the green line in Figure 2) to ensure the aorta is long enough and can be cannulated and ligated at the ascending aorta is better in step 3.3. Whereas transecting at the ascending aorta, the reserved length for cannulation at least could secure the cannula tip close to the aortic root and will not penetrate the aortic valves after ligation. The anatomy of the aorta and the measurement of the distance from the cannula tip to the CA ostium indicate that transecting the aorta between the left common carotid artery is an ideal position to achieve a proper aorta cannulation depth.

Cannulation depth was found to be associated with the atria's perfusion, which in turn acts as a depth indicator. Figure 4 shows that atria perfusion is good when the depth is at the ascending aorta, and both atrial appendages are inflated. However, atria perfusion is insufficient when the depth is at (or approaching) the aortic root and both atrial appendages are wizened. The total and viable count of AM yielded from the inflated atrial appendages were higher (Figure 8). These findings indicate that the aorta cannulation depth can affect the perfusion and digestion of the atria and atrial appendages in a certain way and finally affect AM yield and quality. The yield and quality of AM were inferred to be associated with the distribution of the vessels that supply the atria. The study of Fernández et al.²³ has demonstrated various anomalies in the origin and course of mouse CA. They found that the CAs ostia were highly variable and were not all located in the aortic sinus. Some CAs may originate anomalously from above the aortic sinuses, named the high take-off ostium. Some CAs may originate from the same aortic sinus, and the ostium of the atrium vessel is just nearby. The anatomy of the aorta in the present study (Figure 10) is also consistent with the finding of Fernández. This may be the cause why attempts to isolate AM by the Langendorff method have been largely unsuccessful if the cannulation depth is not appropriate. Thus, the cannula tip will have a greater chance of blocking the atrium vessel ostium that is adjacent to the CA ostium if not enough space is available between the cannula tip and the CA ostium. By contrast, the perfusion and cell yield of the ventricle were hardly affected as long as the aortic valves were not penetrated by the cannula tip. This is probably because the CAs that supply blood to the ventricle have bigger ostia and more origins. If one ostium was occluded by the cannula, ventricle perfusion can be compensated by another CA or the collateral circulation, whereas the blood vessel that supplies the atrium is rather small and has no substitute. Thus, the influence of the depth in aorta cannulation is important.

Other noteworthy factors and trouble shootings in the digestion and cell storage process are listed as follows. First, consider perfusing oxygenated Tyrode's solution in step 4.1 to make the muscle contract and pump out residual blood if the blood in the atria has not been rushed out after aorta ligation. This can help avoid the adverse effect of ca²⁺ and other materials released from the damaged erythrocytes. Second, perfusing ca²⁺-free solution ahead of time to dissociate the connection and expand the space between cells can improve the efficacy of enzyme digestion because intercalated disks between CMs are calcium-dependent intercellular junctions. However, time should be limited to 3-5 min to avoid the calcium paradox phenomenon²⁴. A mixed enzyme solution is recommended. Collagenase type II disrupts the extracellular matrix network, and trypsin helps clear the granular material that remains on the cell surface if collagenase II digestion is incomplete. This ensures a smooth cell surface, which is critical to forming a GΩ seal in the patch clamp recording. Nevertheless, trypsin concentration should be controlled in the proper range to avoid over digestion and cell injury because it can degrade the membrane protein. Using collagenase type II alone to improve cell yield may often lead to tissue over digestion, and the isolated CMs will be calcium intolerant after prolonged collagenase exposure²⁵. The use of 2,3-butanedione monoxime (BDM), a substance that prevents spontaneous contraction by inhibiting the myosin ATPase and preventing cross-bridge formation, is still controversial^26,27,28,29. According to previous experience, adding BDM is necessary for this protocol. Enzyme solution is prepared with solution 1 although solution 1 does not contain calcium, and calcium is added to activate the enzyme. The benefit of adding BDM in the perfusion solution includes (1) inhibiting myocytes contraction and reducing oxygen consumption during enzyme solution perfusion and (2) preventing myocytes from hypoxia and improving the quality of the isolated myocytes. Some studies reported that BDM may have a potential adverse influence on cellular electrical properties. However, the results of the whole-cell patch clamp recording of the sodium current did not indicate an undesirable effect. In the cell storage step (step 6), numerous studies have chosen the KB buffer, a calcium-free but high potassium concentration solution, in which cells can maintain a better state because they are in polarized and low metabolic conditions. However, the glycocalyx of the cell membrane will separate from the lipid bilayer in the absence of exogenous calcium for a certain length of time and the membrane permeability will increase, affecting the subsequent functional analysis^30,31,32.

All myocytes isolation techniques can be essentially categorized currently into either chunk (a small piece of tissue) digestion in an enzymatic solution or CA perfusion with enzymatic solution (the Langendorff perfusion)²². Compared with the Langendorff method, the chunk digestion method is easier to perform and is also routinely used for isolating CMs in many laboratories. However, this method normally produces a low yield of CMs of poor quality from adult tissues²². Moreover, cells isolated by this method may not be suitable for conducting comparative experiments. For example, when testing the cell type-specific drug effects between AM and VM, the impact of different isolation conditions cannot be neglected or excluded. This is because the myocardium and the tissue density of the ventricle are much thicker and denser than that of the atrium, resulting in differing digestion times and enzyme concentrations. Furthermore, excessive agitation and pipetting of the tissue during digestion will damage the cells and significantly impact functional studies. Moreover, many previous studies have shown that AMs are more vulnerable to calcium. However, AMs isolated by the current protocol can be tolerant of gradient calcium reintroduction probably because the tissue is easy to rupture at the end of the digestion process. Thus, mechanical damage is less, whereas cells would suffer more mechanical damages as steps of the chunk method need repeated rupture and centrifugal. More recently, Ackers et al.³³ reported a simplified, Langendorff-free method for the isolation of viable cardiac myocytes and nonmyocytes. The VMs and the fibroblasts can be effectively isolated, but the amount of the AMs was not mentioned. However, this protocol has several limitations. First, the distribution of the heart blood vessels may have variations for individual differences and mice strain, and the recommended cannulation depth cannot guarantee a successful AMs isolation for each time. Second, for those who are new to this procedure may take a certain amount of time to practice aorta transection and retrograde aorta cannulation. Finally, this method was not tested in other heart disease models except in healthy and elderly mice. Hence, it will require adjustments in enzyme concentrations and time of digestion because of the heart's fibrosis extent. The disadvantage of different digestion conditions encountered in the chunk method will not be a problem in the Langendorff method in which the enzyme solution is evenly distributed to the tissue by the vessel beds.

In summary, the protocols for the simultaneous isolation of single AM and VM described here have demonstrated that the proper aorta cannulation depth can effectively improve atrium perfusion and AM yield. The CMs isolated by this method are of high quality, possess good calcium tolerance, and have been successfully applied to patch clamp recording and calcium handling (Ca²⁺ release and Ca²⁺ wave measurement by the IonOptix system) in the team³⁴. It is anticipated that the isolation protocol can be used for cell preparation in a series of cellular and subcellular investigations, which will help deepen the understanding of cardiac physiology and pathology. Importantly, it will enable the discovery of more clinically relevant heart disease mechanisms and intervention methods.