Viral Transgene Expression in Rodent Hearts and the Assessment of Cardiac Arrhythmia Risk

Cardiovascular disease is the leading cause of death worldwide, claiming the lives of 18 million people in 2017 alone¹. Rodents, especially mice and rats, have become the most commonly used model in cardiovascular research due to the ease of handling and the availability of various transgenic overexpression or knockout lines. Rodent models have been fundamental for understanding the disease mechanisms and for identifying potential new therapeutic targets in myocardial infarction², hypertension³, heart failure⁴, and atherosclerosis⁵. However, the use of rodents in studies of cardiac arrhythmias is limited by their small heart size and faster heart rate compared to human or large animal models. Therefore, spontaneous lethal arrhythmias in mice or rats after myocardial infarction are rare². Investigators are forced to focus on indirect secondary changes that might reflect a pro-arrhythmic substrate, such as fibrosis or gene expression, without showing meaningful changes in arrhythmia burden or pro-arrhythmic tendencies. To overcome this limitation, a method that allows a reliable assessment of the susceptibility of mouse and rat hearts to ventricular tachyarrhythmias after genetic modification^6,7 or myocardial infarction² is described in the present protocol. This method combines adrenergic receptor stimulation with programmed electrical stimulation to induce ventricular tachyarrhythmias in isolated, Langendorff-perfused⁸ mouse and rat hearts.

Standard approaches for viral gene transfer in rodent myocardial tissue often involve the exposure of the heart by thoracotomy^9,10,11, which is an invasive procedure and is associated with delayed recovery of the animals after the procedure. This article describes a method of direct intramyocardial injection of virus under ultrasound imaging guidance for the overexpression of transgenes. This less invasive procedure allows for faster animal recovery after viral injection and less tissue injury, as compared to thoracotomy, reduces post-operative pain and inflammation in the animal, and, thus, allows better assessment of the effects of transgenic genes on heart function.

All the methods and procedures described were approved by the animal research ethical review board at the University of Ottawa and the biosafety review committee at the University of Ottawa Heart Institute. The developed safety protocols include that all the procedures dealing with recombinant adenovirus or adeno-associated virus (AAV) were performed in a level II biosafety cabinet. All the items in contact with the virus were thoroughly decontaminated after the experiment. Ctnnb1^flox/flox and αMHC-MerCreMer mice (8-12 weeks old, of either sex) and Sprague-Dawley rats (200-250g, male) were used for the present study. The animals were obtained from commercial sources (see Table of Materials). All the procedures dealing with animals were performed by staff who have been trained and approved by institutional regulatory committees. Appropriate personal protective equipment was used during all procedures.

1. Viral transgene expression in rodent ventricular tissues

NOTE: Store recombinant adenovirus or AAV that expresses the target gene and the corresponding control virus, such as Ad-GFP (titer of 1 x 10¹⁰ PFU/mL) or AAV9-GFP (titer of 1 x 10¹³ GC/mL) (see Table of Materials), in a −80 °C freezer.

1. On the day of virus injection, thaw the virus on ice. Aspirate the virus expressing the target gene and the control virus into two separate syringes (volume = 50 µL) with 30 G 1/2 needles and keep on ice until use.
   NOTE: Any bubbles in the syringe and needle must be carefully removed.
2. Administer buprenorphine to rats or mice (0.05 mg/kg for rats, 0.1 mg/kg for mice, subcutaneous), 1 h later induce anesthesia using 3% isoflurane, and maintain isoflurane at 2% (in 100% oxygen at a flow rate of 0.5-1.0 L/min) for the subsequent procedure.
3. Gently pinch the animal's body (e.g., the tail) with a pair of forceps, and if the animal does not respond to the pinch with any body movements, proper anesthetization is confirmed.
4. Maintain body temperature at 37 °C using an electric heating pad. Shave the hair in the chest region using a hair clipper.
   NOTE: Care should be taken while using an electric heating pad due to potential uneven heat distribution.
5. Apply ophthalmic ointment to both eyes to prevent the drying of the cornea.
6. Keep the animal in a supine position and use a preclinical imaging system (see Table of Materials) for ultrasound imaging of the animal's heart in the short-axis orientation.
   NOTE: The following steps are performed in a level II biosafety cabinet which provides a sterile environment. Standard safety procedures, including those with safe needle handling, are employed.
7. Sterilize the left lower chest region of the animal with alternating rounds of an iodine-based or chlorhexidine-based scrub and alcohol three times in a circular motion.
8. Under ultrasound imaging guidance, insert the 30 G 1/2 needle of the syringe containing the virus as prepared in step 1.1. into the animal's chest via the left lower chest side of the body.
9. Approach the needle tip into the left ventricular front free wall and slowly inject 10-15 μL of the virus. Verify successful injection in ultrasound images by the enhanced brightness near the tip of the needle.
   NOTE: The amount of virus injected at each site is no more than 15 μL to prevent physical damage to the myocardial tissues.
10. Withdraw the needle from the heart and insert it into other regions of the left ventricle for a second and third injection of the same amount of virus.
    NOTE: The total number of injection sites is determined by the experimental design. If focal transgene expression is required, only one injection is needed; if more diffusive transgene expression is required, multiple injection sites (three to five) are usually needed.
11. After the completion of the injections, return the animal to its cage.
12. Carefully monitor the animal until it has regained sufficient consciousness to maintain sternal recumbency and return it to its housing room.
13. Administer buprenorphine HCl (0.1 mg/kg, twice a day for mice, 0.05 mg/kg, twice a day for rats, subcutaneous) to the animal for the following 2 days. Return animals to the vivarium and monitor daily for any unusual signs of pain or distress, and if found, treat animals per the Institutional animal care committee protocols.

2. Assessment of cardiac arrhythmia susceptibility

NOTE: At 4-6 days after adenovirus injection and 1-2 weeks after AAV injection, assess the susceptibility of the animal hearts to cardiac arrhythmias following steps 2.1.-2.2.

1. Perform Langendorff perfusion of the rat or mouse heart.
   1. Prepare Tyrode solution by adding the following to 1 L of water: NaCl, 7.9 g (final = 135 mM); KCl, 0.37 g (5.0 mM); CaCl₂, 0.27 g (1.8 mM); MgCl₂, 0.24 g (1.2 mM); HEPES, 2.38 g (10 mM); and glucose, 1.8 g (10 mM). Adjust the pH to 7.40 with NaOH (see Table of Materials).
   2. Filter the solution with a 0.2 μm filter and bubble with 100% O₂ continuously during steps 2.1.6.-2.2.8.
   3. Administer heparin to the rat or mouse (500 U/kg, intraperitoneal injection) and 10 min later anesthetize the animal with 3% isoflurane. Ensure adequate anesthetization with a gentle pinch on the body.
   4. Use a scalpel to make a 1-1.5 cm vertical incision in the mid-clavicular line for a left thoracotomy and expose the heart.
   5. Collect the heart by cutting with a pair of scissors at the aortic arch level and immediately put the heart into ice-cold Tyrode solution.
      ​NOTE: Caution is exercised not to damage the heart during heart collection.
   6. Cannulate the aorta of the heart with a blunt needle (18 G for rats; 23 G for mice) connected to a modified Langendorff perfusion system (see Table of Materials) in the constant flow rate mode. Adjust the flow rate to keep the perfusion pressure at 70-80 mmHg.
      NOTE: Any bubbles in the perfusion needle must be removed before aortic cannulation. Using a dissecting microscope facilitates the cannulation.
   7. Place the cannulated heart in a silicone elastomer-coated⁹, 10 cm plastic dish with the left ventricle facing up and perfuse the heart^9,12 with O₂-bubbled Tyrode solution at 37 °C.
   8. Verify the correct aortic cannulation at the beginning of perfusion by observing the washout of blood from the heart during the first two to three heartbeats and the changing of the heart color from red to pale.
   9. Place the electrodes of a small animal ECG system (see Table of Materials) around the heart by inserting them into the silicone elastomer coating in the dish (Figure 1). Record ECG using compatible software.
2. Perform adrenergic receptor stimulation and programmed electrical stimulation to induce ventricular tachyarrhythmias.
   1. After successful aortic cannulation, continue to perfuse the heart with Tyrode solution for 20 min to stabilize the heart preparation.
   2. Add 1 μM isoproterenol (see Table of Materials) to the Tyrode solution used for heart perfusion in steps 2.2.3.-2.2.8.
   3. After 10 min of isoproterenol perfusion, stimulate the heart at the apex with two platinum electrodes connected to an electrical stimulator (see Table of Materials).
   4. Start with the stimulation procedure (Figure 2), which includes 10 consecutive stimuli (S1, 5 V, 100 ms intervals) that are followed by an extra stimulus (S2) with an initial interval of 80 ms. Repeatedly reduce the S2 interval by 2 ms each time until the heartbeat can no longer be captured (i.e., reaching the heart's effective refractory period, ERP)¹³.
   5. Monitor any induced ventricular tachyarrhythmias (including ventricular tachycardia and fibrillation) by ECG.
   6. If no arrhythmias are induced by the above procedure, add another extra stimulus (S3) after S2 with the same initial (80 ms) and decrement (by 2 ms) intervals until the ERP is reached.
   7. If ventricular tachyarrhythmias are still not induced, add one more extra stimulus (S4) after S3 with the same initial and decrement intervals until the ERP is reached.
   8. If ventricular tachyarrhythmias are still not induced (as expected in control healthy hearts), stop the electrical stimulation protocol and consider the heart to be non-inducible.

When perfused following the protocol described here (Figure 1), an isolated rat or mouse heart beats rhythmically and stably for at least 4 h. If the experimental design requires a longer period of heart perfusion, it is helpful to add albumin into the perfusion solution to reduce the occurrence of myocardial edema after long-time perfusion¹⁴. The inclusion of isoproterenol in the perfusion solution mimics the activation of the sympathetic nervous system which occurs in many arrhythmogenic conditions such as myocardial infarction and heart failure. During the programmed electrical stimulation, the successful pacing of the heart is verified by (1) the 1:1 capture of the heartbeat during the consecutive S1 stimuli and (2) the prolonged (or wider) QRS complex during S1 pacing (Figure 3 and Figure 4). The latter is because the heart is being paced at the apex and the slower conduction of the activation in the ventricular tissues increases the QRS complex duration.

As demonstrated in the published data^2,7 (Figure 3 and Figure 4), these combined adrenergic and electrical stimulations induced no ventricular tachyarrhythmias (defined as at least three consecutive premature ventricular complexes, PVCs) in healthy mouse hearts (sham-operated wild-type hearts in Figure 3) or in control rat hearts (control Ad-GFP-injected rat hearts in Figure 4). By contrast, the same protocol induced ventricular tachyarrhythmias in 77% of wild-type mouse hearts after myocardial infarction (Figure 3B) and in three out of four rat hearts after intramyocardial injection of Ad-Wnt3a (Figure 4). This demonstrates the high fidelity of this ventricular arrhythmia-inducing approach.

Successful intramyocardial transgene expression after virus injection can be verified by upregulated mRNA and protein levels of the transgene in the myocardial tissues identified by real-time quantitative RT-PCR, western blot, or immunohistochemistry. If the virus expresses a fluorescent reporter gene, such as GFP, successful viral transduction can also be verified in isolated, live, single cardiomyocytes by their expression of GFP⁹.

Figure 1: Langendorff perfusion of an isolated rat heart. The heart is collected from the animal, and the aorta is cannulated with a blunt 18 G needle. The needle is connected to a 10 mL syringe filled with 37 °C Tyrode solution at a pressure of 70-80 mmHg. The heart is placed in a silicone elastomer-coated, 10 cm plastic dish with the left ventricle facing up. Electrodes (red, green, and black colors) of a small animal ECG system are placed around the heart by inserting them into the silicone elastomer coating in the dish. Please click here to view a larger version of this figure.

Figure 2: Adrenergic and electrical stimulations of the Langendorff-perfused heart. (A) Rat or mouse heart is perfused on a Langendorff system as shown in Figure 1, and 1 µM isoproterenol is added to the perfusing Tyrode solution to stimulate the β1 adrenergic receptors of the heart. The heart is then stimulated at the apex by two platinum electrodes connected to a stimulator. (B) Representation of the programmed electrical stimulation. For details, see step 2.2. The figure is modified with permission from Wang et al.². Please click here to view a larger version of this figure.

Figure 3: Ventricular tachyarrhythmias induced in mouse hearts after myocardial infarction. (A) Representative ex vivo ECG (Lead II) showing programmed electrical stimulation (PES)-induced ventricular tachycardia (VT, defined as three or more consecutive premature ventricular complexes, PVCs) in a wild-type (WT) mouse heart (at 8 weeks after myocardial infarction, MI) when stimulated with one extra stimulus (S2) but only one single PVC in a cardiac-specific β-catenin knockout (KO) mouse heart (at 8 weeks after MI) when stimulated with three extra stimuli (S4). Isoproterenol (1 µM) was included in the perfusing Tyrode solution during the PES stimulation. Cardiac-specific β-catenin (Ctnnb1) knockout mice were generated by cross-breeding Ctnnb1^flox/flox mice (with exons 2 to 6 floxed) with αMHC-MerCreMer mice. At the age of 8-12 weeks, Ctnnb1^flox/flox;αMHC-MerCreMer^+/− mice (KO) and littermate Ctnnb1^flox/flox_;αMHC-MerCreMer^−/− mice (used as control wild-type mice) received daily subcutaneous injections of tamoxifen (20 mg/kg/day) for 5 consecutive days before they were assigned to either the MI or sham group. (B) Summary of PES-induced PVCs and VTs in WT and KO hearts at 1 week or 8 weeks after MI. An arrhythmia score was assigned to each heart according to the criteria in the left table. At 8 weeks after MI, VT was successfully induced in 77% of WT hearts but only in 18% of KO hearts. Data were analyzed by two-way ANOVA and a Bonferroni post-hoc comparison. The error bars indicate standard error (SE). The figure is reproduced with permission from Wang et al.². Please click here to view a larger version of this figure.

Figure 4: Ventricular tachyarrhythmias induced in rat hearts after intramyocardial injection of a virus expressing Wnt3a. (A) PES induced ventricular arrhythmias in hearts of rats (200-250 g, male, Sprague-Dawley) at day 4-6 after intramyocardial injection of an adenovirus expressing Wnt3a (Ad-Wnt3a, bottom) but not in hearts injected with control adenovirus expressing GFP (Ad-GFP, top). Hearts were isolated and perfused on a Langendorff system. Isoproterenol (1 µM) was included in the perfusing Tyrode solution during the PES stimulation. Ex vivo ECG was continuously recorded during PES stimulation. Note that 11 consecutive S1 stimuli (blue color) were used in this experiment. (B) Summary of studies in Panel (A): Ventricular tachycardia (VT) was induced by PES in three out of four hearts with Ad-Wnt3a injection but in none of the four hearts with control Ad-GFP. The figure is reproduced with permission from Lu et al.⁷. Please click here to view a larger version of this figure.

Several steps are critical for the success of the Langendorff-perfused, isolated heart preparation. Firstly, it is important to avoid any damage to the heart during heart collection (e.g., due to accidental squeezing or cutting with the scissors). Secondly, it is critical to put the collected heart into cold Tyrode solution as soon as possible because this will stop the heartbeat and reduce the oxygen consumption of the heart. Thirdly, the needle insertion into the aorta must not be too deep—ideally, the tip of the needle is close to the aortic valve level so that the heart is well-perfused through the coronary arteries. Lastly, bubbles in the perfusion needle need to be removed prior to aortic cannulation—it is usually helpful to turn on the pump for a few seconds right before cannulation to remove the bubble at the tip of the needle.

The Langendorff-perfused heart preparation has several advantages over the in vivo studies in animals. Firstly, it is possible to accurately control the concentration and duration of drugs applied to the heart (e.g., isoproterenol, as used in this study). Secondly, it allows easy access to the different regions of the heart for either stimulation (e.g., electrical pacing at the apex in this manuscript) or physiological signal recording (e.g., optical mapping of the ventricular tissues after loading with a Ca²⁺-sensitive or voltage-sensitive dye, which is not described here). In addition, cardiac function, such as ventricular contractility, can be readily measured by inserting a balloon into the left ventricle and connecting the balloon to a pressure transducer⁸. Thirdly, the intrinsic properties of the heart, such as heart rate and ventricular contractility, can be investigated without the complicating factors in in vivo studies, such as the autonomic nervous system and circulating hormones. However, the limitation of the Langendorff-perfused heart is that it lacks the cross-talk among different organs or tissues in vivo^15,16,17 via either circulating factors or the autonomic nervous system, which may be critical players in some studies.

The ex vivo ECG recording of the Langendorff-perfused heart, as described here and in the previous publications^2,6,7,9, has the advantage of contactless recording and poses no disturbance to the function of the heart as compared to other approaches such as optical mapping, which requires the loading of Ca²⁺-sensitive or voltage-sensitive dyes and the use of an excitation-contraction uncoupler to reduce mechanical heart movement (such as blebbistatin)^12,18. However, the optical mapping has the advantage of providing more detailed information on the electrical activities of the heart, such as the origin of the heartbeat, the pattern of myocardial activation, and the myocardial conduction velocity.

The method of direct intramyocardial injection of a virus, as described here, can also be used for the delivery of other therapeutic materials^{19,20,21,22,23,24}, such as biomaterials, exons, modified mRNA, and stem cell-derived cardiomyocytes. The ultrasound imaging-guided intramyocardial injection has several advantages over the traditional thoracotomy-based injection approach. Firstly, it is less invasive and allows for faster recovery of the animals after the injection procedure. This reduces procedure-associated effects on the animals (e.g., those caused by post-operative pain and chest tissue inflammation when an invasive thoracotomy is used). Secondly, ultrasound imaging verifies successful virus injection in the heart, which increases the consistency and reproducibility of the results. However, the limitation of the ultrasound-guided virus injection approach is that the locations of the virus injection sites cannot be controlled as precisely as in the thoracotomy-based approach, which allows visual localization of the different heart regions.