The metamorphosis of amphibian myocardium: moving to the heart of the matter Free

Amphibians have been known as a classical research subject in animal physiology for more than a century (Locke, 1895; Parker, 1891). However, apart from being a widely used physiological model for more or less fundamental human-centered studies, amphibians themselves are an intriguing experimental object owing to the high diversity of their habitats, lifestyles and anatomy (Wiens, 2007). One of the most intriguing features of amphibian physiology is the peculiarity of their ontogenesis. Most amphibians start their life as larvae (or tadpoles) living in an aquatic environment and using gills for breathing. The transition to adult form and, respectively, to terrestrial habitat, coincides with large rearrangements in an animal's anatomy and physiology – such as resorption of tail, remodeling of respiratory organs and growth of limbs (Brown and Cai, 2007). The rearrangement of the cardiovascular system is among the most striking changes during the metamorphosis. Typically, an adult amphibian heart is five-chambered – it includes the sinus venosus, two atria, ventricle and conus arteriosus – and this heart structure is designed for a two-circuit circulatory system (Burggren, 1988). However, in larval amphibians, the cardiovascular system and the heart itself have somewhat fish-like anatomy with single atrium (Kolesová et al., 2007; Sandoval et al., 2022).

Urodele amphibians are of special interest from this point of view. The Mexican axolotl (Ambystoma mexicanum), also known as the Mexican salamander, is the oldest laboratory animal in a self-sustained laboratory population (Reiß et al., 2015). At the present time, axolotls are widely used in various studies focusing on the processes of regeneration; their ability to regenerate damaged tissue allows them to grow new limbs after amputation (Li et al., 2021) or even heal damaged myocardium after partial resection or cryoinjuries (Dittrich and Lauridsen, 2021). Apart from that, the axolotl is also known for its facultative process of metamorphosis, or facultative paedomorphosis. That means that mature, gilled and fully grown larval animals coexist with metamorphic adults, thus exploiting larval habitats for growth and reproduction (Wilbur and Collins, 1973). An individual animal can spend its whole lifetime as a neotenic larval form, and successfully reproduce, if there are no stimuli triggering the transition to terrestrial lifestyle (Denoël et al., 2005). The latter rarely occurs in the wild – presumably owing to the inability of pituitary in axolotls to produce thyrotropes, which results in hypothyroid state of animals (Crowner et al., 2019). The process of metamorphosis itself in amphibians is driven and controlled by thyroid hormones, and can be artificially induced by thyroid hormones (Brown and Cai, 2007; Prahlad and DeLanney, 1965). Taken together, the process of metamorphosis in axolotls can be fully controlled in laboratory conditions, whereas in other amphibian species it is a naturally occurring developmental pathway, and that makes axolotls a very convenient experimental model.

Though the process of metamorphosis is in general controlled only by thyroid hormones, it includes a complex pattern of changes in gene expression and cell apoptosis. Although axolotls do not meet the need to resorb their tails during the transition to adult form, resorption of the caudal fin rim occurs. Because adult animals are exposed to atmospheric pressure and respiration requirements differing from those for neotenic larvae, the process of metamorphosis includes remodeling of the lungs: a drastic increase in lung volume accompanied by a reduction in the thickness of the lung walls (Coleman and Hessler, 1997). Resorption of the gills and enlargement of the lungs coincides with corresponding changes in the vascular bed: new blood vessels vascularize the metamorphized lungs, increasing the effectiveness in gas exchange, whereas the vessels supplying the gills regress and are replaced by carotid arteries (Kolesová et al., 2007). The changes in respiratory activity and metabolic needs have to be accompanied by remodelling of the heart, from anatomical and physiological points of view. Previous studies have reported changes in the gross morphology of axolotl heart. A comprehensive work by Olejnickova et al. (2022) has demonstrated that metamorphosis in A. mexicanum is related to an increase in the trabecularization of ventricular myocardium and the completion of the atrial septum, with atria being split into two separate compartments. At the tissue level, there is a decrease in the amount of connective tissue (endomysium) surrounding cardiac muscle fibers, though the functional significance of this is unclear (Demircan et al., 2016).

However, the information on the functional remodeling of axolotl heart during metamorphosis is quite scarce. The study by Olejnickova et al. (2022) did not report any changes in the basic heart rate of ex vivo axolotl hearts, though there was a noticeable decrease in atrioventricular delay, indicating a possible increase in the velocity of excitation propagation throughout the axolotl heart following metamorphosis. Nevertheless, such major changes in cardiac structure accompanying vast changes in animal lifestyle and metabolic demands have to be associated with physiological remodelling of the myocardium. Thus, the aim of the present study was to investigate possible changes caused by metamorphosis in cardiac electrical activity of axolotls at the cellular level.

Mexican axolotls [Ambystoma mexicanum (Shaw and Nodder 1798)] of either sex weighing 60–100 g (N=13; brown/green wild type) were used in the experiments. Animals were housed in chloride- and chloramine-free tap water in 40 l aquaria (no more than three animals at a time) at +17°C and 12 h:12 h light:dark cycle. The water was changed twice a week; the animals were fed with frozen bloodworms every other day.

The animals were randomly divided into two groups; the first group served as controls, in the second group we induced metamorphosis using 50 nmol l⁻¹ thyroid hormone (tetraiodothyronine or T₄) according to a previously published protocol (Page and Voss, 2009). The process of metamorphosis took up to 45 days. Animals with fully resorbed gills and caudal fin rim, breathing with atmospheric air, were considered to be metamorphic. The metamorphic salamanders were kept in low water to avoid drowning.

Before the experiments, the animals (paedomorphic axolotls or metamorphic salamanders) were anaesthetized in ice-cold water to avoid possible adverse effects of anaesthetics on cardiac electrical activity, and then rapidly decapitated. All experiments conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication no. 85-23, revised 1996) and the EU Directive 2010/63/EU for animal experiments.

Basing on a previously described method to isolate cardiomyocytes from amphibians and other ectotherms (Abramochkin and Kuzmin, 2018; Abramochkin and Vornanen, 2014; Haverinen et al., 2017), we developed a similar method to isolate viable and calcium-tolerant cardiomyocytes from paedomorphic and metamorphic A. mexicanum. After the decapitation, the heart was rapidly excised, mounted onto a Langendorff apparatus and cannulated through the conus arteriosus. At the first stage, the heart was perfused at room temperature (23±1°C) with 20–30 ml of nominally Ca²⁺-free low-Na⁺ solution, containing (in mmol l⁻¹): 100 NaCl, 10 KCl, 1.2 KH₂PO₄, 4 MgSO₄·7H₂O, 50 taurine, 10 glucose and 10 HEPES at a pH of 6.9 (pH adjusted with 5 mol l⁻¹ NaOH). After the first stage, intended to wash out blood from the heart, the perfusion was switched to a solution of the same composition supplied with 0.5 mmol l⁻¹ EGTA. The Ca²⁺ chelator EGTA was used for more effective washout of Ca²⁺ ions from the myocardium of an isolated axolotl heart. We used up to 100 ml of EGTA- containing low-Na⁺ solution to wash out Ca²⁺ from the heart, and then perfused the heart again with up to 50 ml of low-Na⁺ solution to wash out EGTA. Finally, the heart was perfused with low-Na⁺ solution containing proteolytic enzymes: 0.8 mg ml⁻¹ collagenase IA and 0.4 mg ml⁻¹ trypsin from porcine pancreas, supplied with 0.8 mg ml⁻¹ bovine serum albumin. The last stage of enzymatic perfusion lasted for 70–80 min. All the listed perfusion stages were performed at room temperature. After that, the atrial and ventricular tissue were separated, placed into low-Na⁺ solution without any supplements, minced and gently triturated with a Pasteur pipette to release cardiomyocytes into solution. Isolated atrial and ventricular cardiomyocytes were stored separately up to 8 h after isolation at +4°C.

Cardiomyocytes isolated from paedomorphic axolotls and metamorphic salamanders were used for whole-cell patch clamp experiments to record action potentials (APs) and major ionic currents using a HEKA EPC-800 (HEKA Elektronik, Lambrecht, Germany) or Axopatch 200A (Molecular Devices, Sunnyvale, CA, USA) amplifier. Ionic currents were recorded in voltage-clamp mode, and APs were recorded in current-clamp mode. Isolated cells were placed into a small experimental chamber (RC-26; Warner Instrument Corporation, Brunswick, CT, USA; volume 150 μl) mounted onto an inverted microscope. Ionic currents and APs were recorded at room temperature (23±1°C). Cardiomyocytes were superfused with modified physiological saline; the composition of the solution varied depending on the type of the experiment. Patch pipettes were pulled using PC-10 or PP-830 pullers (Narishige, Tokyo, Japan) from borosilicate glass capillaries without filament (World Precision Instruments, Sarasota, FL, USA) and filled with pipette solution (the composition also depended on the type of the experiment). The resistance of the filled patch pipettes ranged between 2 and 3 MΩ.

In the experiments involving recording APs and potassium currents (I_Kr and I_K1), we used K⁺-based physiological saline of the following composition (mmol l⁻¹): 150 NaCl, 3 KCl, 1.2 MgCl₂, 5 HEPES, 1.8 CaCl₂ at pH of 7.4 (adjusted with 5 mol l⁻¹ NaOH at room temperature). The pipette solution contained (mmol l⁻¹): 140 KCl, 1 MgCl₂, 5 EGTA, 10 HEPES, 4 MgATP and 0.03 Na₂GTP at a pH of 7.2 (adjusted with 5 mol l⁻¹ KOH at room temperature).

To record Ca²⁺ current, we used Cs⁺-based physiological saline of the following composition (mmol l⁻¹): 150 NaCl, 3 CsCl, 1.2 MgCl₂, 5 HEPES and 2 CaCl₂ at a pH of 7.4 (adjusted with 5 mol l⁻¹ NaOH at room temperature). The corresponding Cs⁺-based pipette solution contained (mmol l⁻¹): 130 CsCl, 1 MgCl₂, 5 EGTA, 10 HEPES, 4 MgATP, 0.03 Na₂GTP and 15 tetraethylammonium at a pH of 7.2 (adjusted with 50% CsOH at room temperature).

For I_Na recordings, we used Cs⁺-based physiological saline with NaCl partially substituted with Tris-HCl to reduce the driving force for inward Na⁺ current. The saline contained (mmol l⁻¹): 20 NaCl, 120 Tris-HCl, 3 CsCl, 1 MgCl₂, 5 HEPES and 1.8 CaCl₂ at a pH of 7.4 (adjusted with 5 mol l⁻¹ NaOH at room temperature). The saline was supplied with 20 μmol l⁻¹ nifedipine to block calcium current. The corresponding Cs⁺-based pipette solution contained (mmol l⁻¹): 130 CsCl, 5 NaCl, 1 MgCl₂, 5 EGTA, 10 HEPES, 4 MgATP and 0.03 Na₂GTP at a pH of 7.2 (adjusted with 50% CsOH at room temperature). The reduction in the driving force for Na⁺ along with 70% compensation of the series resistance ensured stable recording of the fast-activating and high-amplitude I_Na.

Access resistance and cell capacitance were routinely compensated after getting access to the cell interior. Ionic currents were elicited using step voltage clamp protocols given in the figure panels.

All the data are presented as means±s.e.m. for n cells from N animals. The number of cells tested and animals used in the experiments is indicated in the figure legends. The normality of the sample distribution was tested using a Shapiro–Wilk test. Student's t-test was used to compare current densities and AP configuration parameters. An extra sum-of-squares F-test was used to compare the parameters of I_Na steady-state kinetics in cells from paedomorphic and metamorphic animals. The differences were considered statistically significant at P<0.05.

Collagenase type IA, trypsin from porcine pancreas, nifedipine, tetrodotoxin and tetraiodothyronine were purchased from Sigma-Aldrich. E-4031 was purchased from Tocris Bioscience.

As was indicated in the previous studies and protocols, after 2 weeks of exposure to T₄, the animals started to resorb their external gills and caudal fin rim, indicating successful induction of metamorphosis; axolotls also moulted their epidermis. The animals were considered to be metamorphic after full resorption of their gills, which occurred up to 45 days after the first exposure to T₄. Representative images of paedomorphic and metamorphic animals are shown in Fig. 1A,B.

In the present study, we have for the first time developed a method for enzymatic isolation of viable cardiomyocytes from axolotls. Cardiomyocytes isolated from working atrial and ventricular myocardium of A. mexicanum are shown in Fig. 1C,D. Myocytes looked like thin, elongated cells with clear striation. The cells had either a spindle-like or a spider-like shape. The metamorphosis did not cause any pronounced changes in the shape of cardiac myocytes. Though we did not evaluate the size of axolotl cardiomyocytes directly, we made indirect measurements of cell capacitance during whole-cell patch clamp experiments. The electrical capacitances of atrial cardiomyocytes from paedomorphic and metamorphic animals were 107.6±8.53 pF (n=23) and 132.3±16.68 pF (n=29), respectively (P=0.1822, Student's t-test). The capacitances of ventricular cardiomyocytes from paedomorphic and metamorphic animals were 131.8±11.94 pF (n=26) and 158.0±15.34 pF (n=23), respectively (P=0.1815, Student's t-test). Thus, we can presume that metamorphosis does not cause any changes in the size of cardiac myocytes in axolotls.

APs were elicited at rate of 0.2 Hz and recorded in current-clamp mode in both ventricular and atrial cardiomyocytes isolated from paedomorphic and metamorphic A. mexicanum. Representative examples of APs are shown in Fig. 2A,B. Metamorphosis did not cause any statistically significant changes in the level of resting membrane potential (RMP) or in the maximum velocity of APs upstroke in ventricular or atrial myocardium (Fig. 2C,D). However, T₄-induced metamorphosis produced visible and statistically significant AP shortening at levels of 90% and 50% repolarization, in both ventricular (Fig. 2E) and atrial (Fig. 2F) cardiomyocytes.

Depolarization from a holding potential of −80 mV (the protocol is shown in Fig. 3) caused activation of outward current with pronounced outward rectification in axolotl atrial and ventricular cardiomyocytes. Metamorphosis was followed by an increase in the overall amplitude of the net I_K current (Fig. 3A). The current was partially sensitive to a selective blocker of the rapid delayed rectifier current I_Kr E-4031 (1 μmol l⁻¹) – this component of the net current was considered to be I_Kr. The other component of the net I_K was insensitive to E-4031 and was referred to as the slow delayed rectifier current I_Ks. Both components, I_Kr and I_Ks, were present in ventricular and atrial myocytes from both paedomorphic and metamorphic animals. Metamorphosis caused a strong and robust, statistically significant increase in the amplitude of I_Ks in both ventricular (Fig. 3B) and atrial (Fig. 3C) axolotl myocytes at positive holding potentials; namely, at a holding potential of +60 mV, I_Ks increased by more than 3-fold in atrial cells and by almost 9-fold in ventricular cells from metamorphic animals.

However, the amplitude of the rapid delayed rectifier current I_Kr remained unchanged in both ventricular and atrial myocardium (Fig. 3B,C). Thus, it is probably the slow delayed rectifier current responsible for the observed increase in the overall amplitude of potassium current accompanying metamorphosis in A. mexicanum. It is also worth noting that this increase in I_Ks leads to a redistribution of I_Kr and I_Ks proportions: as can be seen from the bar graphs in the right of Fig. 3B,C, in paedomorphic animals, I_Kr seems to be the predominant component of the net potassium current, whereas in metamorphic salamanders, I_Ks becomes the main component of I_K.

A Ba²⁺-sensitive current showing significant inward rectification and referred to as background inward rectifier current I_K1 was present both in ventricular and atrial myocytes isolated from paedomorphic and metamorphic A. mexicanum (Fig. 4). Owing to lack of time-dependent inactivation, this current was elicited using ramp-shaped pulses (protocol shown in Fig. 4). In ventricular myocytes, I_K1 was noticeably more pronounced than in atrial cells. Metamorphosis led to an overall downregulation of I_K1. In ventricular cells, the changes did not reach statistical significance (Fig. 4A). In atrial cardiomyocytes, metamorphosis provoked statistically significant decrease in both inward (at potentials negative to −80 mV) and outward (at potentials positive to −80 mV) components of the current (Fig. 4B); namely, at −60 mV, the amplitude of I_K1 in cells from metamorphic animals was decreased by more than 50% from that in myocytes from paedomorphic axolotls.

Fast sodium current was elicited by depolarizing pulses from the holding potential of −120 mV, whereas other (potassium and calcium) currents were blocked. Representative recordings of fast sodium current are shown in Fig. 5A. In ventricular cells, I_Na was more pronounced than in atrial myocytes (Fig. 5B,C). Metamorphosis did not cause any statistically significant changes in I_Na amplitude in the ventricle (Fig. 5B); however, in atrial myocytes it was upregulated with a statistically significant increase in I_Na – at −20 mV, the peak amplitude was increased by more than 80% in metamorphic animals in comparison with that in paedomorphic animals (Fig. 5C).

The steady-state kinetics of I_Na was evaluated using the double-pulse protocol (shown in Fig. 5). Steady-state activation curves were plotted from the values of I_Na at the first step of the protocol, whereas steady-state inactivation curves were plotted using I­_Na amplitude at the second step. The resulting curves are demonstrated at the right sides of Fig. 5B,C. As can be seen from Fig. 5, metamorphosis did not cause any changes in I_Na steady-state kinetics in the working myocardium of axolotls, and thus the parameters of the steady-state curves did not differ (P>0.05 for each parameter comparison, extra sum-of-squares F-test). The resulting parameters of the curves (k and V₅₀) are shown in Table 1.

Calcium current was elicited by depolarizing double-step pulses in both ventricular and atrial cardiomyocytes of axolotls, while potassium currents were blocked by Cs⁺-based solutions used in this series of experiments (protocol shown in Fig. 6). Representative recordings of I_Ca are shown in Fig. 6A, whereas current–voltage relationships for I_Ca in ventricular and atrial cells are demonstrated in Fig. 6B,C. Owing to the overall shape of the current–voltage curve, with the maximum current at a holding potential of 10 mV, and owing to sensitivity of the current to 10 μmol l⁻¹ nifedipine, the recorded net I_Ca was considered to be represented mainly by the L-type calcium current I_CaL (Mesirca et al., 2015).

Metamorphosis was associated with noticeable and statistically significant upregulation of I_Ca in both ventricular and atrial myocardium. As can be seen in Fig. 6B and C, at holding potentials greater than 0 mV, I_Ca amplitude in ventricular and atrial cells from metamorphic animals was statistically significantly higher than that in paedomorphic axolotls. At a holding potential of 10 mV in both atrial and ventricular cardiomyocytes from metamorphic A. mexicanum, I_Ca was increased more than 2-fold in comparison with that in myocytes from paedomorphic animals, indicating a serious increase in Ca²⁺ influx in the myocardium of metamorphic salamanders.

Amphibian metamorphosis is driven and governed by thyroid hormones (Allen, 1938). It was shown that the climax of metamorphosis in anuran amphibians coincides with peak of circulating concentrations of thyroid hormones (Tata, 1997). External thyroid hormones can induce metamorphosis in anuran tadpoles (Gudernatsch, 1912). However, there are a variety of metamorphosis strategies among different amphibian taxa. It should be pointed out that most of our knowledge on metamorphosis in amphibians comes from studies performed on anurans, namely on Xenopus species (Buchholz et al., 2006). Most anurans exhibit spontaneous metamorphosis, whereas for some urodele species, metamorphosis is not an obligatory developmental stage, and thus they develop neoteny, an ability to reproduce in larval stage. Salamanders of genus Ambystoma are known for their facultative neoteny (Rosenkilde and Ussing, 1996). Ambystoma natural populations can be exclusively paedomorphic, fully or partially metamorphic. This trait supposedly is an ecological adaptation allowing to exploit more available resources (Hanken, 1999). Ambystoma mexicanum are known for their so-called ‘metamorphic failure’, which presumably evolved several times in genus Ambystoma independently (Shaffer, 1993). Axolotls have been used in laboratory practice since the 1800s, but they rarely exhibit spontaneous metamorphosis in laboratory conditions, and inducing metamorphosis in axolotls is harder than in similar studies with anuran species (Allen, 1938; Voss et al., 2009). The metamorphic failure in axolotls is probably related to low levels of thyroid-stimulating hormone and, in turn, to low levels of circulating thyroid hormones, though axolotl tissues express functional thyroid hormone receptors and respond to externally administered thyroid hormones (Taurog, 1974). Thus, the use of external thyroid hormones (usually tetraiodothyronine) is an appropriate method to study metamorphosis in axolotls, and 50 nmol l⁻¹ T₄ is considered to be a standard concentration to induce metamorphosis in larval A. mexicanum, as higher concentrations of T₄ do not accelerate the process of metamorphosis (Rosenkilde and Ussing, 1996). It is a question though, whether this externally induced metamorphosis differs from the natural process of metamorphosis, and could it be that adult salamanders that underwent metamorphosis under the influence of external T₄ are in a state of hyperthyroidism? This is hard to examine in axolotls, because A. mexicanum extremely rarely demonstrate spontaneous metamorphosis. However, this hypothesis could be tested in other amphibian species. A few studies have demonstrated that adult anurans respond to external thyroid hormones and demonstrate some signs of hyperthyroidism (e.g. Warren, 1940). It was shown that the hyperthyroidism state can affect cardiac electrical activity in the frog Rana esculenta, with an increase in heart rate and AP shortening (Di Meo et al., 1995).

The present study is first to describe electrophysiological remodeling in the myocardium of axolotls during metamorphosis, and the first to describe the major ionic currents in the axolotl myocardium. The latter is somewhat surprising, as axolotls have been used as a laboratory experimental subject for more than a century and have been proposed as an animal model for some human cardiac diseases (Meyer et al., 2022). Previously, Alanís et al. (1973) described the electrical activity in isolated preparations of axolotl hearts. The overall AP shape – at least, in the ventricular myocardium – was very similar to what we saw in the present study, but atrial APs recorded by Alanís et al. (1973) were significantly shorter. Unfortunately, these data were obtained under very different experimental conditions – the pacing rhythm was much higher and the temperature was lower than in the present study – thus, it is difficult to compare our results with those data directly. It should be also noted that Alanís et al. (1973) focused on the electrical activity of the conducting system in axolotl heart, whereas we examined the electrical properties of the working myocardium. Our results could be also compared with data from previous studies performed on other amphibian species – namely, bullfrog (Rana catesbeiana) and African clawed frog (Xenopus laevis). The overall AP form and duration in axolotls at room temperature are similar to those in the bullfrog. However, we have for the first time distinguished the two components of the delayed rectifier current: in previous works, amphibians were considered to have only the slow delayed rectifier current I_Ks in their myocardium, whereas we have shown that net current I_K is composed of both the rapid and slow delayed rectifier currents I_Kr and I_Ks, respectively (Giles et al., 1989; Ono and Giles, 1991). This adds important details to our understanding of repolarization processes in the amphibian myocardium.

The fast sodium current I_Na in the axolotl myocardium is also worth noting. The maximum of I_Na amplitude in the current–voltage relationship curve is shifted towards more positive potentials in comparison with that in mammals (Abramochkin et al., 2022), and the same feature was shown for I_Na in bullfrog cardiomyocytes (Clark and Giles, 1987). However, another study on cardiomyocytes from X. laevis demonstrated I_Na with the conventional shape of the current–voltage curve (Vornanen et al., 2011). Thus, it is hard to say whether the shift of the I_Na maximum is a feature of the amphibian myocardium or whether it depends on the experimental conditions.

One of the most visible changes associated with metamorphosis that we observed in the present study was a significant shortening of both atrial and ventricular APs. The shortening of APs is a well-known consequence of delayed rectifier upregulation induced by changes in environment (Filatova et al., 2019). We suggest that the upregulation of the slow delayed rectifier current I_Ks underlies the observed changes in AP waveform. Though we also reported an increase in amplitude of calcium current, the increase in I_Ks amplitude caused by metamorphosis was far more drastic. It is tempting to speculate that these electrophysiological changes must be linked to an increased heart rate resulting in higher metabolic needs of metamorphic animals. Indeed, the shift to a terrestrial lifestyle is unequivocally associated with at least partial separation of oxygen-poor venous and oxygen-rich arterial blood, and the latter allows higher metabolic rates and underlies the success of terrestrial lifeforms (Jensen et al., 2020; Pough, 1980). However, in axolotls and other amphibian species, metamorphosis is not associated with an increase in heart rate, at least, if we speak about basic heart rate provided by unregulated activity of cardiac pacemaker (Fritsche, 1997; Olejnickova et al., 2022). Thus, it is not the changes in basal heart rate that push these changes in AP waveform during metamorphosis.

The possible explanation might be within the fact that the autonomic regulation of the heart changes as well during metamorphosis in amphibians. It has been shown for different anuran species (Rana temporaria and R. catesbeiana) that although the receptor machinery for autonomic regulation exists in the amphibian myocardium from early developmental stages, the autonomic – and especially sympathetic – innervation of the heart develops later: for example, in R. temporaria sympathetic fibers become fully visible only after metamorphosis (Burggren and Doyle, 1986; Protas and Leontieva, 1992). For axolotls, it has been shown that T₄-induced metamorphosis coincides with an increase in the number of vagal preganglionic neurons, suggesting that metamorphosis in A. mexicanum is also associated with further development of autonomic cardiac control (Taylor et al., 2001). The slow delayed rectifier current I_Ks, in contrast, is one of the main players in mediating sympathetic effects in the heart: phosphorylation of I_Ks channels by protein kinase A followed by enhancement of the I_Ks current underlies AP shortening caused by adrenergic stimulation (Terrenoire et al., 2005; Walsh and Kass, 1991). Thus, one can speculate that the observed upregulation of I_Ks and AP shortening caused by metamorphosis in axolotls are aimed to create a functional reserve for sympathetic regulation in adult salamanders.

One of the most surprising findings was that metamorphosis was followed by a slight downregulation of background inward rectifier current I_K1 in the axolotl heart. As can be seen, it was not accompanied by any changes in the level of RMP in atrial or ventricular cells.

However, it should be noted that I_K1 amplitude does not necessarily directly correlate with RMP levels in cardiac cells. Rather, I_K1 stabilizes RMP, buffering excessive depolarizing stimuli (Dhamoon and Jalife, 2005; McLerie, 2003). Metamorphosis and the shift to a terrestrial lifestyle in amphibians are probably associated with a rise in the ambient temperature of the environment. Because axolotls are ectotherms, this means changes in core temperature as well, and as I_K1 is a highly temperature-dependent current, such a change in lifestyle will lead to a pronounced increase in I_K1 amplitude (Paajanen and Vornanen, 2004; Abramochkin and Vornanen, 2017). An excessive increase of I_K1 following the rise of temperature in the terrestrial environment might reduce cardiac excitability. Thus, one can speculate that the observed downregulation of I_K1 in the axolotl myocardium serves to prevent such adverse effects.

Fast sodium current I_Na provides cardiac excitability, and its amplitude correlates with the velocity of excitation propagation in the heart (Shaw and Rudy, 1997). A previous study performed by our colleagues has demonstrated that T₄-induced metamorphosis in axolotls is associated with a decrease in atrioventricular delay, indicating an increase in the velocity of excitation propagation in the myocardium of metamorphic axolotls (Olejnickova et al., 2022). Though it might be not of great importance in a basal state, it could also create a functional reserve for sympathetic regulation of the heart in metamorphic animals, along with the aforementioned upregulation of I_Ks, as faster propagation of excitation will increase the maximum heart rate possible under adrenergic stimulation.

Metamorphosis in axolotls was associated with clearly visible upregulation of calcium current in both atrial and ventricular cardiomyocytes. It is widely accepted that the Ca²⁺ current in the heart triggers Ca²⁺ release from the sarcoplasmic reticulum, and the amplitude of Ca²⁺ influx is directly correlated with the contractility of the myocardium (Cleemann and Morad, 1991; Lederer et al., 1989). As mentioned earlier, metamorphosis and the shift to terrestrial habitats in amphibians are associated with remodelling of the cardiovascular system from a fish-like single circulation to two separate circuits and with an increase in metabolic demands. Such changes require higher systemic pressures and, in turn, higher cardiac contractility (Joyce and Wang, 2020). Though there are no data on metamorphic changes in blood pressure in axolotls, previous studies on anurans showed that blood pressure increases during amphibian development (Burggren et al., 1992; Fritsche, 1997; Hou and Burggren, 1995). Some studies have even shown that this increase in systemic blood pressure coincides with a decrease in peripheral resistance, which clearly indicates the role of increased cardiac contractility (Burggren et al., 1992; Fritsche, 1997). Thus, our finding that metamorphosis in axolotls upregulates I_Ca in heart explains and supports the data from previous works. One could even suggest that the effects of metamorphosis on cardiac electrical activity in axolotls are secondary to the massive remodelling of the whole cardiovascular system.

In contrast, the aforementioned decrease in AP duration in metamorphic salamanders will undoubtedly lead to a decrease in the duration of AP plateau phase, and, hence, to a decrease in calcium influx and, consequently, a decrease in the contractility of cardiomyocytes (Bouchard et al., 1995). Thus, the observed upregulation of I_Ca might be a mechanism counterplaying the possible decrease in cardiac contractility resulting from AP shortening, rather than a sign of increased heart contractility in metamorphic animals.

Gross anatomical changes in the amphibian heart caused by metamorphosis are strictly associated with functional changes in myocardial electrical activity at the cellular and molecular levels. Adaptation to air breathing and the subsequent compartmentalization of the cardiovascular system require higher systemic blood pressure and more robust autonomic control in metamorphic animals. Our findings support these statements, reinforce the connection between the gross structural organization and functional characteristics of the heart at a cellular level and, finally, provide a more detailed explanation for some previously described phenomena in amphibian physiology. Though our data broaden the horizons for comparative and functional research in axolotls even further, there are still certain knowledge gaps that require further study.

The equipment used in the present study was provided by Moscow State University within the framework of the federal project ‘The development of infrastructure for science and education’ (agreement no. 355).