Temperature-induced changes in cardiac output (Q̇) in fish are largely dependent on thermal modulation of heart rate (fH), and at high temperatures Q̇ collapses due to heat-dependent depression of fH. This study tests the hypothesis that firing rate of sinoatrial pacemaker cells sets the upper thermal limit of fH in vivo. To this end, temperature dependence of action potential (AP) frequency of enzymatically isolated pacemaker cells (pacemaker rate, fPM), spontaneous beating rate of isolated sinoatrial preparations (fSA), and in vivo fH of the cold-acclimated (4°C) brown trout (Salmo trutta fario) were compared under acute thermal challenges. With rising temperature, fPM steadily increased because of the acceleration of diastolic depolarization and shortening of AP duration up to the break point temperature (TBP) of 24.0 ± 0.37°C, at which point the electrical activity abruptly ceased. The maximum fPM at TBP was much higher [193 ± 21.0 beats per minute (bpm)] than the peak fSA (94.3 ± 6.0 bpm at 24.1°C) or peak fH (76.7 ± 2.4 at 15.7 ± 0.82°C) (P < 0.05). These findings strongly suggest that the frequency generator of the sinoatrial pacemaker cells does not limit fH at high temperatures in the brown trout in vivo.
- cardiac pacemaker
- thermal tolerance
- fish heart
- single pacemaker cells
- pacemaker action potentials
cardiac output (Q̇), the volume of blood pumped by the heart in a unit of time, is a vital factor for the homeostasis of the vertebrate body, as it determines the rate of convective oxygen transport when metabolic rate of the animal changes under environmental and physiological stresses (7, 19). Under hypoxia and during exercise, Q̇ in fishes is mainly regulated by stroke volume (VS) owing to the changes in end-diastolic volume of the ventricle and subsequent increase in force of ventricular contraction via the Frank-Starling mechanism (11, 32, 40). However, under acute thermal stress, increasing Q̇ is predominately driven by an elevation of heart rate (fH), as VS is largely unaffected. At critically high temperatures, fH and Q̇ typically plateau and decline, and it has therefore been hypothesized that a thermal limitation of cardiac function and hence blood supply to the tissues may limit aerobic performance in fishes (13, 23, 33). This concept is formulated in the hypothesis of oxygen- and capacity-limited thermal tolerance of animals (OCLTT) (9, 29, 34), although recent findings seriously question the general applicability of this hypothesis as an explanation for thermal tolerance of ectotherms (3, 10, 16). Whereas the central role of fH in thermal responses of Q̇ in fishes is well documented, the mechanistic basis of the heat-induced deterioration of fH is still not fully understood. Possible causes for heat-dependent depression of fH in fish include 1) temperature-induced failure of the sinoatrial pacemaker mechanism, 2) disruption of the autonomic homeostatic control of the pacemaker attributable to neural dysfunction (13), 3) deterioration of impulse conduction in the nodal tissues (between sinoatrial pacemaker and atrium or between atrium and ventricle), or 4) inability of atrial and ventricular muscles to follow high pacemaker rates (fPM) because of reduced excitability of the working myocardial cells (36). Some experimental support was recently provided for the reduced electrical excitability of the ventricular myocytes of the brown trout (Salmo trutta fario) (38), whereas other aspects of fH regulation have not yet been critically examined. This is unfortunate because clarification of the cellular and molecular basis of temperature-dependent deterioration of cardiac excitability would increase our understanding on thermal responses of the fish heart and the role of electrically excitable tissues in animal life under the threats of climate change (36).
The aim of the present study was to test the hypothesis that heat-dependent failure of the cardiac pacemaker mechanism is the limiting factor for fH in brown trout at high temperatures. To this end, fPM of single enzymatically isolated pacemaker cells was measured under acute heat challenges using the whole cell perforated patch-clamp technique. Action potential (AP) frequency of pacemaker myocytes was compared with the spontaneous beating rate of the heart in vivo and excised sinoatrial preparations in vitro.
MATERIALS AND METHODS
Experiments were conducted on a salmonid fish species, the brown trout (Salmo trutta fario). Fish (body mass 225 ± 24 g, means ± SE, n = 31) were obtained from the local fish farm (Kontiolahti, Finland) and maintained at the animal facilities of the university in 500-liter stainless steel aquaria with continuous flow of aerated (O2 11 mg/l) groundwater at constant temperature of 4°C and under a 15-h:9-h light/dark photoperiod. During the laboratory maintenance (>3 wk), the fish were fed aquarium fish food (Ewos, Finland) five times a week. For the in vitro experiments, hearts were excised from fish stunned by a blow to the head and killed by severance of the spine and destruction of the brain. For attachment of electrocardiogram (ECG) leads, brown trout were anesthetized in neutralized tricaine methanesulfonate (MS-222, 0.3 mg/l; Sigma, Espoo, Finland) and placed dorsal side up on an operating table, and the gills were irrigated with tap water during the operation (38). The experimental protocols conform to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85–23, Revised 1996) and were approved by the Animal Experiment Board in Finland (permission No. STH252A).
Current-clamp experiments on enzymatically isolated pacemaker cells.
AP rate of pacemaker cells was measured in freshly isolated myocytes, i.e., without exposing the cells to cell culture. For isolation of pacemaker cells, the heart was perfused with proteolytic enzymes (collagenase type 1A, 0.75 mg/ml; trypsin type IX, 0.5 mg/ml both from Sigma-Aldrich) and fatty acid free bovine serum albumin from Serva (0.75 mg/ml; Espoo, Finland) as previously described in detail (18, 37). Immediately after completion of the enzymatic digestion, the sinoatrial valve and the adjacent atrial tissue at the base of the valve were excised, minced, and triturated with a Pasteur pipette to release pacemaker myocytes from the sinoatrial ring, the primary pacemaker zone of the fish heart (18, 26, 42, 43, 45). Cells were stored up to 8 h at 5°C in the low-Na+ solution containing (in mmol/l) 100.0 NaCl, 10.0 KCl, 1.2 KH2PO4·2H2O, 4.0 MgSO4·7H2O, 50.0 taurine, 20.0 glucose, and 10.0 HEPES at pH of 6.9 (adjusted with KOH).
A small aliquot of myocyte suspension was placed in the experimental chamber (volume 150 µl; RCP-10T; Dagan, Maryland, MI) and superfused at the rate of ~1.5 ml/min with the external K+-based physiological saline solution containing (in mmol/l) 150.0 NaCl, 5.4 KCl, 1.2 MgCl2, 1.9 CaCl2, 10.0 glucose, and 10.0 HEPES at pH 7.6 (adjusted with NaOH). The β-adrenergic agonist isoprenaline (Sigma) was present at 5 nM in the external solution to simulate the in vivo adrenergic tone (15). Temperature of the external solution was regulated using a Peltier device (TC-100, Dagan), and temperature was continuously recorded on the same file with pacemaker APs. The pipette solution contained (in mmol/l) 140.00 KCl, 4.00 MgATP, 1.00 MgCl2, 0.03 Tris-GTP, and 10.00 HEPES with pH adjusted to 7.2 with KOH. Pacemaker APs were recorded in the current-clamp mode of the perforated whole cell patch-clamp technique using the Axopatch 1-D amplifier and the pClamp 8.2 software package (Axon Instruments, Saratoga, FL). Perforated-patch mode of the patch-clamp technique was used to maintain the integrity of intracellular signaling cascades, including putative Ca2+-dependent regulation of the pacemaker rate. Low resistance access to the cell interior was obtained by inclusion of 25 µM β-Escin (Sigma) in the pipette solution, which perforates the membrane patch in contact with the pipette (31). To this end, a GΩ seal was formed in the voltage-clamp mode, and we waited 10–15 min at the holding potential of −70 mV until access resistance was <20 MΩ. Pipette capacitance, access resistance, and whole cell capacitance were routinely compensated to determine the cell size, which was 27.7 ± 5.1 pF (means ± SE, n = 12 myocytes from 7 fish). Then the recording mode was switched from voltage-clamp to current-clamp mode, and acquisition of spontaneous pacemaker APs was started. The cell preparation consists of a mixture of pacemaker cells and common working atrial myocytes, the number of atrial cells far exceeding the number of pacemaker cells. Therefore, only those cells that were beating spontaneously in a regular manner were accepted for patch clamping. The rate of temperature change under the heat challenge was ~2°C/min. The following AP parameters were analyzed offline using the Clampfit software that measured the following: frequency of pacemaker APs (fPM), AP amplitude, AP duration at 50% repolarization level (APD50), maximal diastolic potential (MDP), and maximal rate of diastolic depolarization (DDMax) (18). Temperature dependency of the variables is expressed as a Q10 value obtained from the equation Q10 = (R2/R1)[10/(T2 − T1)], where T1 and T2 are the temperatures that produce the variable values of R1 and R2, respectively. The break point temperature (TBP) for fHR, fSA, and fPM under heat challenges was defined as the temperature after which a sustained decrease of beating rate started.
Inward rectifier K+ current (IK1) and Na+ current (INa) were recorded under the same experimental conditions as previously reported for the 12°C-acclimated brown trout (38).
Recording of beating rate of sinoatrial preparations.
The whole atrium and a part of the sinus venosus with the intact sinoatrial valve and the adjacent pacemaker area were gently excised and carefully rinsed in K+-based Tyrode solution to wash out any blood. The preparation (25.8 ± 2.7 mg, means ± SE, n = 10) was placed in a 15-ml experimental chamber filled with constantly oxygenated and thermostabilized external saline solution (for composition see Current clamp experiments on enzymatically isolated pacemaker cells). One corner of the atrial wall was connected to the force transducer (Grass FT03) using a small hook and braided silk thread, whereas the opposite corner of the preparation was pinned to the Sylgard-coated bottom of the chamber (1, 38). The rate of sinoatrial beating (fSA
Recording of ECGs.
In vivo fH was measured from ECGs as previously described in detail (1, 4, 38). To this end, the fish (n = 9) were anesthetized, and two recording electrodes (7-strand Teflon-coated wire, length 40.00 cm, diameter 0.23 mm; A-M Systems, Carlsborg, WA) were inserted from the ventral side close to the pericardium. The wires were secured by two sutures, one to the belly of the fish and another in the front of the dorsal fin. To limit movements, the fish was placed into a respiratory chamber (1 l, O2 fH variability appeared in the ECG. Rising temperature challenges at the rate of 3°C/h (Computec Technologies, Joensuu, Finland) (1) were started from the acclimation temperature of the fish (4°C) and continued until a clear depression of fH
All data in the text and figures, except the original recordings, are presented as means ± SE with n experiments, which refers to the number of animals (in vivo ECG), number of sinoatrial preparations (i.e., animals), number of pacemaker cells (from 7 fishes), and number of atrial myocytes. After we checked the normality of distribution and homogeneity of variances and we made necessary transformation of variables, one-way ANOVA was used to test the statistical differences of Q10 values, TBP, beating rates at TBP, and temperature dependence of atrial ion currents followed by Tukey’s, Dunnett’s (equal variance variances), or Dunnett’s T3 (nonequal variances) post hoc comparisons. These tests were carried out using IBM SPSS statistics (version 21.0) software. Frequency responses to temperature (fH, fSA, and fPM) were fitted with the third-order polynomial equation (y = a + bx + cx2 + dx3), and the fits were statistically compared using the F-test (GraphPad Prism 5.03). For all comparisons, a P value <0.05 was considered statistically significant.
In vivo fH.
Under acute heat challenges, fH increased from 26.90 ± 1.62 bpm at 4°C to 76.50 ± 2.18 bpm at the TBP of 15.70 ± 0.82°C (Fig. 1A). The Q10 value between 4°C and the TBP was 2.61 ± 0.21. Further rises in temperature caused at first depression of fH (bradycardia) (Fig. 1A) and at higher temperatures cardiac arrhythmia in the form of large heart rate variability and sometimes missing QRS complexes, i.e., block of the impulse transmission between atrium and ventricle (not shown) (1, 38).
Beating rate of sinoatrial preparations.
Acute warming increased fSA from 26.90 ± 1.13 bpm at 4°C to 94.30 ± 4.28 bpm at the TBP of 24.10 ± 0.21°C (Fig. 1B). The Q10 value between 4°C and the TBP was 1.72 ± 0.05. Further rises in temperature caused at first bradycardia and then at higher temperatures various forms of cardiac arrhythmia, including biphasic contractions (Fig. 1B, arrow), mechanical alternans (cyclical variation between weak and strong contractions), and periods of complete unexcitability (not shown).
APs of isolated pacemaker cells.
Under acute heat challenges, fPM increased from 18.40 ± 1.50 bpm at 4°C to 191.40 ± 12.11 bpm at the TBP of 24.00 ± 0.57°C (Q10 = 3.20 ± 0.22) (Fig. 1C). The increase in fPM was caused by strong temperature dependence of DDMax (Q10 = 3.90 ± 0.59) and APD50 (Q10 = 0.46 ± 0.02) (Fig. 2, A and B). Differently from the sinoatrial preparations and in vivo ECG, fPM steadily increased with temperature, without preceding bradycardia or any type of arrhythmia, up to the point at which electrical activity suddenly ceased (Fig. 1C). Cessation of beating was often preceded by reduction of AP amplitude (Fig. 2C). Therefore, the TBP of pacemaker activity of isolated pacemaker cells corresponds with the temperature at which APs disappeared. MDP was fairly insensitive to temperature (Q10 = 1.01 ± 0.04) although it tended to depolarize at the highest temperatures (Fig. 2D).
Comparison of fH, fSA, and fPM.
Frequency responses of intact fish hearts (fH), sinoatrial preparations (fSA), and isolated pacemaker cells (fPM) were fitted by a third-order polynomial equation and statistically analyzed using the F-test (Fig. 3). All three groups differed significantly from each other (F8,272 = 92.2; P = 0.0001). AP frequency of the pacemaker myocytes was higher than the contraction frequency of the sinoatrial preparations at temperatures between 18°C and 26°C and higher than in vivo heart rate in the temperature range between 18°C and 24°C (Fig. 3A). The TBP of fH was significantly lower than TBP of fSA and fPM (F2,28 = 77; P < 0.001), which did not differ (P = 1.000) from each other (Fig. 3B). The beating rate of pacemaker cells at TBP was 2.50 and 2.03 times as high as the peak values of fH and fSA, respectively (F2,28 = 54.4; P = 0.0001) (Fig. 3C).
The objective of the present study was to test the hypothesis that failure of the sinoatrial pacemaker mechanism is the limiting factor for the beating rate of the brown trout heart at high temperatures. In enzymatically isolated pacemaker cells, APs are generated by the cellular pacemaker mechanism without influence from surrounding cells with which pacemaker cells may be in contact within the intact tissue. Therefore, changes in fPM inform us about the capability of the pacemaker mechanism of single cells to increase their firing rate in response to rising temperature. In contrast, in the spontaneously beating sinoatrial preparations, fSA is affected by other factors in addition to the cellular pacemaker mechanism, in particular by electrical connection of pacemaker cells with atrial myocytes, which may affect pacemaker rate via local circuit coupling (24, 25, 28, 39). The ability of atrial tissue to follow the fast rate of the primary pacemaker may be also limited at high temperatures (36, 38). In the intact fish, the situation is even more complicated, including impulse transmission between atrium and ventricle (38), neuronal and humoral control of fH (13), and putative effects of metabolic changes of thermal stress on heart function (2).
The present results indicate that at high temperatures fPM of isolated pacemaker cells is much higher than either fSA of sinoatrial preparations in vitro or fH of intact fish in vivo. This strongly suggests, contrary to the working hypothesis, that in brown trout the maximum fH is not limited by the frequency generator of single pacemaker cells. The steady increase of fPM at increasing temperatures was due to a heat-induced decrease in APD50 and increase in DDMax, similar to the mammalian sinoatrial pacemaker (41). The ion current basis of diastolic depolarization of the fish cardiac pacemaker is not known, but shortening of APD50 is probably due to the delayed rectifier K+ current (IKr), which is known to be expressed in rainbow trout (Oncorhynchus mykiss) pacemaker cells (18). It is also notable that at temperatures below 14°C fPM was generally lower than fH and fSA. The lower beating rate of isolated brown trout pacemaker cells at low temperatures is consistent with the lower beating rate of isolated mammalian sinoatrial cells in comparison to the intact sinoatrial node (6, 21). In endotherms, the low rate of isolated pacemaker cells is probably due to the absence of interaction between pacemaker cells. Small aggregates of cultured mouse and chicken pacemaker cells beat faster and more synchronously than single enzymatically isolated pacemaker cells, i.e., electrical coupling (“mutual entrainment”) between the cells determines the overall beating behavior of the intact cardiac pacemaker, making it faster and more regular (5, 22, 44). Possibly a similar mechanism is working in the fish cardiac pacemaker.
If fH is not restricted by heat sensitivity of the rate generator of pacemaker cells, the limitation must reside in the interaction between pacemaker cells or between nodal cells and the working atrial and ventricular myocytes or in neuronal and humoral control of fH. The present findings provide some cues for further testing of those issues. At high temperatures (above 16°C), the beating frequency of sinoatrial preparations was much lower than the AP rate of isolated pacemaker cells, i.e., the multicellular sinoatrial preparation was bradycardic in comparison to single pacemaker cells. This is a dramatic difference in thermal response between cell and tissue levels. Frequency of atrial contractions is dependent on firing of atrial APs, which are elicited by signals arriving from the sinoatrial pacemaker. Pacemaker APs must depolarize membrane potential of atrial cells to the threshold, a value of membrane potential at which the density of depolarizing sodium current (INa) exceeds the density of repolarizing potassium currents (IK) (14). According to the hypothesis of temperature-dependent depression of electrical excitation (TDEE), the voltage threshold for excitation in working cardiac myocytes increases at high temperatures as an outcome of discordant temperature dependencies of INa and IK1 (36, 38). In 12°C-acclimated brown trout, INa is sensitive to high temperatures and steeply declines at temperatures above 20.7°C, whereas IK1 is heat resistant and increases up to 32°C (38). In atrial myocytes of the 4°C-acclimated brown trout, the atrial INa declines at the temperatures above 17.04 ± 1.75°C (Fig. 4A), and therefore it could be anticipated that atrial rate is limited by INa-IK1 antagonism. However, in atrial myocytes of brown trout, the density of IK1 is much smaller (−1.51 ± 0.34 pA/pF at −120 mV at 4°C) (Fig. 4B) than in ventricular myocytes (38) and therefore less powerful in antagonizing INa. If the failure of atrial excitation was limiting the beating rate, this should appear as missing beats (doubling of diastolic interval between beats) at high temperatures. However, this clearly was not the case. Instead there was a “real” bradycardia with steady depression of beating rate (fSA) with increasing temperature. This suggests that the strong temperature-dependent increase of fPM is slowed down by interaction of pacemaker cells with other cells, in particular with the numerous atrial myocytes, within the intact sinoatrial pacemaker. The vertebrate cardiac pacemaker, including that of fishes, comprises in addition to pacemaker cells a large number of fibroblasts, atrial myocytes, and nerve cell bodies that can affect pacemaker rate (18, 27, 35). In mammalian sinoatrial pacemakers, atrial myocytes are electrically coupled with pacemaker cells via gap junctions enabling regulation of pacemaker activity by atrial myocytes (17). The negative resting membrane potential of atrial myocyte can “spread” to pacemaker cells and hyperpolarize them, thereby slowing down the fPM (24, 25, 39). Assuming an analogous interaction between pacemaker and atrial cells for the fish sinoatrial tissue, because of the high temperature resilience of the fish IK1 (38), the hyperpolarizing effect of atrial myocytes will increase with increasing temperature and more strongly restrain fPM at higher temperatures.
The heart rate in vivo was even more depressed by high temperatures than the intrinsic beating rate of sinoatrial preparations in vitro, suggesting further limitations in the intact animal in addition to the atrial bradycardia. These could involve temperature-dependent changes in neuronal and humoral control of the pacemaker, e.g., temperature-dependent changes in parasympathetic control of heart rate. In the Antarctic Pagothenia borchgrevinki, acute increases in temperature are associated with increased cholinergic control of heart rate (12). However, in rainbow trout, acute increase in temperature from 9°C to 16°C did not significantly change parasympathetic control of heart rate (8), suggesting that vagal inhibition might not be a central factor in heat-dependent depression of fH in salmonid fishes. Previous studies conducted on 12°C-acclimated brown trout and summer- and winter-acclimatized roach (Rutilus rutilus) indicate that acute high temperatures cause missing QRS complexes in ECG, indicating a failure in impulse transmission between atrium and ventricle as an additional factor for depression of fH (1, 38). Missing QRS complexes could be due to unexcitability of ventricular myocytes to the signal arriving from the atrium along the nodal pathway or conduction failure of the nodal AP. With consideration of the similarity of nodal APs in primary and secondary pacemaker tissues of fish hearts (20, 30) and the high heat resilience of the sinoatrial pacemaker APs (present study), complete failure of atrioventricular nodal conduction does not seem likely. More likely, unresponsiveness of the ventricle could be due to either the elevated voltage threshold of ventricular myocytes (INa-IK1 antagonism) to excitation and/or decreased amplitude of the nodal AP as formulated in the TDEE hypothesis (36). Indeed, the reduction of the sinoatrial pacemaker AP amplitude at temperatures above 13°C is consistent with the idea that, in addition to the increased excitation threshold of ventricular myocytes, the reduced size of atrioventricular nodal AP could be contributing to missing ventricular beats (36, 38). The reduced amplitude of pacemaker AP found here for the brown trout myocytes is also reported for the mammalian sinoatrial pacemaker at temperatures exceeding the normal body temperature (41).
Perspectives and Significance
Heat resistance of heart rate in the cold-acclimated brown trout is the lowest in the intact fish and increases toward lower levels of biological organization and is clearly the best at the cellular level. Each level of increasing biological complexity brings additional limitations for the high temperature tolerance of heart rate, indicating that interactions between cells and different types of cardiac tissue (and possibly their neuronal and humoral control) are significant in thermal responses of the heartbeat. The heat-induced bradycardia of sinoatrial preparations suggests that, in particular, the interaction of pacemaker cells with atrial myocytes depresses beating rate at high temperatures. The present findings are consistent with the TDEE hypothesis, which suggests that antagonism between the inward INa and the outward IK is crucial for excitability of the fish heart at high temperatures, in particular in impulse conduction between atrium and ventricle (36). Future studies should clarify the pacemaker mechanism of the fish heart at the level of individual ion currents/channels to see which molecular processes are the most vulnerable/tolerant to high temperatures. Studies on knockdown or knockout models for the most heat-sensitive ionic mechanisms in model species might reveal the role of electrical excitability in thermal tolerance of fishes and its putative role in physiology of ectotherms under the present scenarios of climate change.
This study was supported by a grant from the Academy of Finland to M. Vornanen (Project No. 14955) and a grant from Russian Science Foundation to D. Abramochkin (Project No. 14-15-00268).
No conflicts of interest, financial or otherwise, are declared by the authors.
We thank laboratory technician Anita Kervinen for skillful technical assistance and Dr. Hannu Huuskonen for help in statistical analysis of the data.
- Copyright © 2017 the American Physiological Society