Identification of RNA reads encoding different channels in isolated rat ventricular myocytes and the effect of cell stretching on L-type Ca²⁺current

Abstract

Background

The study aimed to identify transcripts of specific ion channels in rat ventricular cardiomyocytes and determine their potential role in the regulation of ionic currents in response to mechanical stimulation. The gene expression levels of various ion channels in freshly isolated rat ventricular cardiomyocytes were investigated using the RNA-seq technique. We also measured changes in current through Ca_V1.2 channels under cell stretching using the whole-cell patch-clamp method.

Results

Among channels that showed mechanosensitivity, significant amounts of TRPM7, TRPC1, and TRPM4 transcripts were found. We suppose that the recorded L-type Ca²⁺ current is probably expressed through Ca_V1.2. Furthermore, stretching cells by 6, 8, and 10 μm, which increases I_SAC through the TRPM7, TRPC1, and TRPM4 channels, also decreased I_Ca,L through the Ca_V1.2 channels in K⁺ _in/K⁺ _out, Cs⁺ _in/K⁺ _out, K⁺ _in/Cs⁺ _out, and Cs⁺ _in/Cs⁺ _out solutions. The application of a nonspecific I_SAC blocker, Gd³⁺, during cell stretching eliminated I_SAC through nonselective cation channels and I_Ca,L through Ca_V1.2 channels. Since the response to Gd³⁺ was maintained in Cs⁺ _in/Cs⁺ _out solutions, we suggest that voltage-gated Ca_V1.2 channels in the ventricular myocytes of adult rats also exhibit mechanosensitive properties.

Conclusions

Our findings suggest that TRPM7, TRPC1, and TRPM4 channels represent stretch-activated nonselective cation channels in rat ventricular myocytes. Probably the Ca_V1.2 channels in these cells exhibit mechanosensitive properties. Our results provide insight into the molecular mechanisms underlying stretch-induced responses in rat ventricular myocytes, which may have implications for understanding cardiac physiology and pathophysiology.

Background

During the last five decades, scientists have collected a lot of evidence that mechanical stress has a significant impact on the electrophysiological properties of cardiomyocytes. This phenomenon, commonly referred to as mechanoelectric feedback, has been extensively studied [1] and is believed to play a crucial role in the pathophysiology of cardiac arrhythmias [2, 3]. In healthy hearts, this feedback mechanism can involve transmembrane cation fluxes through stretch-activated channels (SACs) [4], which can modulate the membrane potential of cardiac myocytes [5,6,7]. Scientific data have shown that localized stretching of single ventricular or atrial myocytes involves cation flux through cation-nonselective SACs [5], which in turn can affect the membrane potential of these cells. Stretch sensitivity is particularly high in hypertrophied ventricular cardiomyocytes from spontaneously hypertensive rats and in atrial and ventricular cardiomyocytes from patients with end-stage heart failure [8]. In general, the role of cardiomyocyte SACs is not limited to normal heart function but also plays a crucial role in various pathological conditions.

Several studies reported Ca²⁺ as well as Na⁺ and K⁺ permeability in SACs of tissue-cultured embryonic chick cardiac myocytes [9,10,11,12] and rat atrial myocytes [10]. Mechanical stretching has been suggested to increase the amplitude of Ca²⁺ transients by activating SACs, which, in turn, increases their Ca²⁺ permeability [12, 13].

Although the L-type Ca²⁺ channel is considered the main pathway of Ca²⁺ entry during an action potential (AP) in cardiomyocytes [14, 15], there are several studies showing that uniaxial stretching does not affect the L-type Ca²⁺ current (I_Ca,L) in single cardiomyocytes [16,17,18,19]. However, in our previous studies, we observed inhibition of I_Ca,L during uniaxial stretching using a glass stylus [6, 8]. This inhibition was attributed to the intracellular accumulation of Ca²⁺ due to the Ca²⁺ influx through SACs. To date, no studies have been conducted that examined the dependence of the value of I_Ca,L on the degree of axial cell stretching. Although there may be no significant changes in I_Ca,L by uniaxial stretch in a voltage clamp experiment, extra Ca²⁺ may still enter the cell through L-type Ca²⁺ channels during prolongation in the depolarization phase of AP due to delayed voltage-dependent inactivation, which partly contributes to the increase in the amplitude of the Ca²⁺transients [6, 20]. To examine the effects of cell stretching on L-type Ca²⁺ channels in rat ventricular myocytes, we investigated the operation of these channels under different degrees of dosed stretching.

In addition, to understand the function of operative SACs and Ca²⁺ channels, RNA reads encoding all existing channels in adult rat ventricular myocytes were examined. The data obtained showed the maximum number of RNA reads for the Ca_V1.2 channels, which exhibit additional mechanosensitive properties. Among mechanosensitive channels, we identified significant amounts of transcripts for transient receptor potential ion channels, including (melastatin) TRPM7, (canonical) TRPC1, and TRPM4.

Materials and methods

Animals

The experiments were conducted in compliance with the Guide for the Care and Use of Laboratory Animals (8th edition, 2011) published by the US National Institutes of Health, and the experimental protocol was approved by the Ethics Committee of the Russian National Research Medical University. Male outbred Wistar rats (n = 34) weighing between 220 and 270 g (10–12 weeks) were housed in a standard T4 cage in a 12:12 h light : dark cycle and given ad libitum access to food for four weeks prior to the start of the experiment.

Solutions

Ca²⁺-free physiological salt solution (Ca²⁺-free PSS) containing (in mM: 118 NaCl, 4 KCl, 1 MgCl₂, 1.6 NaH₂PO₄, 24 NaHCO₃, 5 sodium pyruvate, 20 taurine, and 10 glucose, adjusted to pH 7.4 with NaOH (bubbled with carbogen 95% O₂ + 5% CO₂). Enzyme medium containing: Ca²⁺-free PSS supplemented with 10 µM CaCl₂, 0.2 mg ml^− 1 collagenase (Type II, Worthington, 225 U mg^− 1), and 1 mg ml^− 1 bovine serum albumin (Sigma). Before the actual experiments, cells were stored for at least 2 h in a modified Kraftbrühe (KB) medium, containing (mM): 50 L-glutamic acid, 30 KCl, 3 MgSO₄ × 7H₂O, 20 taurine, 10 glucose, 30 KH₂PO₄, 0.5 EGTA, and 20 HEPES, adjusted to pH 7.3 with KOH [6]. The isolated cells were stored in KB medium for up to 8 h. Ventricular cardiomyocytes were perfused with a solution (37 ^oC) containing (mM): 150 NaCl, 5.4 KCl, 1.8 CaCl₂, 1.2 MgCl₂, 20 glucose, and 5 HEPES, at pH 7.4, adjusted with NaOH (K⁺ _out solution). Internal pipette solution containing (mM): 140 KCl, 5 Na₂ATP, 5 MgCl₂, 0.01 EGTA and 10 Hepes/KOH at pH 7.3 (K⁺ _in solution). Later in the text, this configuration is referred to as K⁺ _in/K⁺ _out solutions. In some experiments, we suppressed inward rectifying K⁺ currents by substituting Cs⁺ for extracellular K⁺ and reduced outward rectifying K⁺ currents by replacing K⁺ with Cs⁺ in the intracellular (electrode) solution. Below in the text, this configuration is referred to as Cs⁺ _in/Cs⁺ _out solutions. For this, in a perfusion solution, 5.4 mM KCl was replaced by 5.4 mM CsCl (extracellular Cs⁺_оut solution), while in the patch pipette solution, 140 mM KCl was replaced by 140 mM CsCl (intracellular Cs⁺ _in solution) [5]. With suppressed K⁺ currents, the peak L-type Ca²⁺ current can be approximated by subtracting the late current (I_L) from the most negative current surge [21]. A similar substitution was made to separate the effects of non-selective cation currents I_ns [5, 6].

Isolated cardiomyocyte preparation

We followed a modified version of the previously described cell isolation procedure [5, 6]. Male Wister rats were anesthetized with an intraperitoneal injection of 80 mg kg^-1 ketamine and 10 mg kg^-1 xylazine, with heparin (1000 U kg^-1) added to the anesthetic solution to prevent blood coagulation in the coronary vessels of the excised heart. The chest was opened, and the heart was rapidly excised and attached to a Langendorff apparatus with a constant flow of 1 ml min^-1 at 37 °C to flush the coronary vessels in carbogen-bubbled Ca²⁺-free PSS for 5 min. After this initial perfusion, hearts were retrogradely perfused with the same PSS supplemented with Worthington type II collagenase (0.5 mg ml^-1), 1 mg ml^-1 bovine serum albumin (Sigma), and 10 µM CaCl₂ for 18–20 min. The perfusate was continuously bubbled with carbogen (95% O₂ + 5% CO₂), and the temperature was equilibrated at 37 °C. The enzymes were then washed out with a modified KB medium [22], and the heart was disconnected from the perfusion system. The ventricles were then cut off, chopped, and gently triturated to release the cells into the KB medium. The resulting cell suspension was filtered and stored in KB medium at 22 °C prior to use.

mRNA isolation and purification

mRNA was isolated directly from cardiomyocytes using Oligo-dT25 Dyna beads® (Invitrogen). The mRNA-bound beads were washed in a Tris buffer solution containing lithium dodecyl sulfate (LiDS) or without LiDS (Buffer A: 10 mM TrisHCl (pH 7.5); 0.15 M LiCl; 1 mM EDTA; 0.1% LiDS; Buffer B: 10 mM TrisHCl (pH 7.5); 0.15 M LiCl; 1 mM EDTA). The beads were resuspended in RNase-free water (Sigma, Dorset, UK), and the mRNA was reverse transcribed using cloned AMV reverse transcriptase (Invitrogen). The resulting cDNA was stored at -80 °C until necessary.

RNA sequencing and analysis

The obtained libraries were analyzed for concentration, size distribution, and quality using a dsDNA high sensitivity kit (Invitrogen, Carlsbad, CA, USA) on a Qubit 4 fluorometer (Thermo Fisher Scientific, Inc., Dreieich, Germany) and a high sensitivity D5000 kit (Agilent, Santa Clara, CA, USA) on a 4200 TapeStation. Libraries were normalized according to their molarity, pooled, and then quantified with a library quantification kit for Illumina platforms (Roche, Basel, Switzerland) using a StepOnePlus qPCR machine (Thermo Fisher Scientific, Inc., Dreieich, Germany). The resulting pooled libraries were loaded at 350 pM with 1% PhiX on an S2 FlowCell and sequenced in triplicate using a NovaSeq 6000 next-generation sequencer (Illumina, San Diego, CA, USA) with paired-end reads of 2 × 150 bp.

Raw FASTQ - sequenced reads were first assessed for quality using FastQC v0.11.5 (available online at http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) [23]. The reads were then passed through Trimmomatic v0.36 [24] for quality trimming and adapter sequence removal. The surviving trimmed read pairs were then processed with Fastp [25] to remove poly-G tails and Novaseq/Nextseq-specific artifacts. Following the quality trimming, the reads were assessed again using FastQC. Post - QC and QT, the reads were aligned to the human reference genome GRCh38.p4 using HISAT2 [26] with default parameters, and the resulting SAM alignments were then converted to BAM format and coordinate sorted using SAM Tools v1.3.1 [27]. Finally, the sorted alignment files were analyzed with HTSeq-count v0.6.1p1 [28] using the options (-s no -t exon -I gene_id) for raw count generation.

Unless otherwise stated, all experiments were performed with at least three replicates, and the data are represented as the mean ± standard error (SE).

Compounds

In certain experiments, nifedipine (Sigma-Aldrich), a specific blocker of the L-type Ca²⁺ channel [29], was added to the bath solution at a concentration of 5–10 µM to assess the pharmacological specificity of the observed Ca²⁺ currents [30, 31]. At these concentrations, nifedipine selectively blocks I_Ca,L without affecting the T-type Ca²⁺ current (I_Ca,T), the fast Na⁺ current, the delayed rectifier current (I_K), or the hyperpolarization-activated inward current [29]. Isoproterenol, a ß-adrenergic stimulator, dose-dependently stimulated I_Ca,L in isolated cardiac myocytes, reaching a plateau at 100 nM. The effect of isoproterenol (Sigma-Aldrich) (20 nM) on the peak current amplitudes of I_Ca,L at 0 mV was considered a test of the quality of isolated cells. Gadolinium (Gd³⁺) (Sigma-Aldrich), known as a non-specific blocker of mechanically gated channels in the I_SAC (current through stretch activated channels) [5, 6], was added to the PSS (5 µM GdCl₃), BAPTA (Tocris), and 1,2-bis(2-aminophenoxy)ethane-N,N,N’,N’-tetraacetic acid.

Mechanical stretch of the ventricular myocytes

The mechanical stimulation method used in this study has previously been described in detail. Here, we only report the peculiarities relevant to this study. After whole-cell access with the patch pipette (P), a fire-polished glass stylus (S) was attached to the membrane [5, 6, 22, 32]. When the stylus was newly polished and the surface membrane was clean, the attachment was successful in approximately 70% of the attempts. The stylus was then lifted 2 μm to prevent ‘scratching’ of the lower cell surface on the coverslip during the stretch. A motorized micromanipulator (MP 285, Sutter, Novato, Calif., USA, accuracy 0.2 μm) increased the S-P distance stepwise by up to 12 μm, with P being the fixed point [5, 22]. Stretch and release of stretch could be repeated 3–4 times with the same cell, on average. Our method was shown to stretch the cell surface locally, while the membrane in the line between P and S was stretched as expected (approximately 80% of the entire membrane surface remains unaffected) [5, 22]. The effect of mechanical stretching on the sarcomere pattern was imaged by a slow-scan CCD camera (Princeton Instruments, Trenton, N.J., USA) and evaluated by Meta Morph software (Universal Imaging, West Chester, PA, USA). S and P were positioned 40 μm apart before attaching them to the cell. Cell stretching by 4 μm (increasing the S - P distance) increased local stretching by approximately 6%, by 6 μm about 10%, by 8 μm about 14%, and by 10 μm about 18%. These values were less than expected but close to those previously obtained in isolated mouse cardiomyocytes. Presumably, the extent of local stretch decays from the cell surface to the interior of the cell, where the optical focus was set [6, 22].

To investigate the effect of cell stretch at 4, 6, 8, and 10 μm on I_Ca,L values, measurements were taken. A standard elongation of 6 μm was used to study I_Ca,L on the background of cell elongation under the action of various compounds.

Whole-cell patch-clamp

A total of 113 cells (n = 113) were used in the experiments. Whole-cell patch-clamp recordings of K⁺, Ca²⁺, and non-selective (I_ns) currents were obtained using an Axopatch 200B amplifier and pClamp 10 software (Molecular Devices, San Jose, CA, USA). The data were filtered at 2 kHz, sampled at 5 kHz, and analyzed using the same software. Myocytes were superfused in a small recording chamber (RC-26; Warner Instrument Corp., Brunswick, CT, USA) with a volume of 150 µl, which was mounted on an inverted microscope.

The borosilicate glass patch-clamp electrodes had tip resistances ranging from 1.5 to 2.5 MΩ when filled. Cell access was obtained by rupturing the patch after seal formation. Pulses of 140 ms and 350 ms were applied at 1 Hz, starting from a holding potential (V_hp) of − 45 mV, which caused the inactivation of tetrodotoxin (TTX)-sensitive Na⁺ currents. To evaluate membrane capacitance and access resistance, currents in response to trains of short (5 mV) pulses applied at − 45 mV were taken without compensation for capacitive and leak currents. Cells with similar geometry were selected based on their length and diameter (control rat ventricular cardiomyocytes had an average diameter of 25 ± 6 μm), and, on average, had a membrane capacitance of 150 ± 16 pF (n = 16) and an input resistance of 58 ± 5 MΩ (in 16 representative cells). Glass tools were adjusted to the same 40 μm S -P distance before the application of stretch to minimize the effect of differences in the size of the stretched membrane. Since the area of mechanical stretching was small and unknown, we did not divide the stretch-induced currents by the entire membrane capacitance.

Currents through L-type Ca²⁺ channels, membrane currents at the end of the pulse (late current: I_L), and other currents were plotted as functions of their respective clamp step potentials. The seal resistance remained constant, i.e. it was 1.5 ± 0.3 GΩ before and 1.4 ± 0.4 GΩ during the stretch. Similarly, access resistance and membrane capacitance remained unaffected, indicating that the stretch-induced inward current was due to the activation of an ionic current rather than leakage around the seal. The zero current potential (V₀) for I_L was determined by the intercept of the resulting I/V curve with the voltage axis and corresponded to the resting membrane potential of a non-clamped cell (between − 70 and − 80 mV). Online records of net membrane current were carried out at V_hp = − 45 mV (time course) [6, 22, 33].

Typically, the measurements lasted approximately 30 min, during which time the access resistance and capacitive current remained stable. To obtain the current-voltage relations (I/V curves), a series of 20 pulses of 140 ms (or 350 ms) duration were applied, starting from V_hp = − 45 mV.

Freshly isolated brick-like cardiomyocytes can attach to the glass bottom in two different positions: edgewise, staying on the narrow side, and broadwise, staying on the broad side [34]. However, the response to stretching is identical in cardiomyocytes occupying both positions (edgewise and broadwise). On the other hand, the response to compression differs depending on the cell’s position [34, 35]. For our experiments, we selected cells that stayed on the narrow side (edgewise) and had similar sizes.

In this study, I_Ca,L recordings are presented in both absolute values (e.g., pA) and normalized form (e.g., pA/pF). However, changes in I_Ca,L in response to cell stretching or treatment with compounds are presented as the difference between I_Ca,L under control conditions and that under stretching or treatment conditions.

Statistics

Values are presented as means ± SD. To determine significant differences, we used Analysis of Variance (ANOVA) with the Bonferroni test as a post-hoc analysis. For cases where multiple factors were evaluated, a two-way ANOVA was performed. Statistical significance was set at p < 0.05.

Results

The levels of the channel’s gene expression

In this study, we used RNA-seq to measure the expression levels of various genes in freshly isolated rat ventricular cardiomyocytes. Our analysis revealed the presence of RNA reads encoding different ion channels, as summarized in Table 1 [36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92], and illustrated in Fig. 1. Specifically, we found that Na_V channel transcripts were expressed in these cells, with Na_V1.5 showing the highest number of RNA reads (3654.4 ± 581.2), followed by Na_V1.1 (only 179 ± 30), [81, 82]. The expression of other Na_V channels was minimal. Importantly, previous research has shown that Na_V1.5 exhibits mechanosensitivity [81, 82], which may have physiological implications for the function of rat ventricular cardiomyocytes.

Fig. 1

Relative abundance of ion channels encoding different channel genes in adult rat ventricular cardiomyocytes Bars represent the mean ± SEM (n = 10). Red columns and red arrows indicate mechanosensitive channels, while black and gray columns indicate all the rest

Table 1 Identified mechanically gated channels and channels with mechanosensitivity in our experiments and in the literature

Our analysis revealed the presence of three Ca_V subfamilies: Ca_V1.2, Ca_V1.3, and Ca_V1.1, as shown in Fig. 1. Of these, the Ca_V1.2 channels had the highest number of RNA reads (1336.66 ± 71.8), with the transcripts of the other Ca_V channels being an order of magnitude smaller. In particular, previous studies have shown that Ca_V1.2 and Ca_V1.3 both exhibit mechanosensitivity [84] (Table 1; Fig. 1). When considering the L-type Ca_V channels, which are responsible for the L-type Ca²⁺ current, we found that the Ca_V1.2 transcripts were significantly more abundant than the Ca_V1.3 transcripts (17.33 ± 2.03), while the number of Ca_V1.1 transcripts was very low (Table 1). Based on these results, we conclude that Ca_V1.2 is likely the primary channel that contributes to the L-type Ca²⁺ current in rat ventricular cardiomyocytes and that Ca_V1.3 and Ca_V1.1 can be neglected. This finding is consistent with previous research indicating that Ca_V1.2 channels are the main contributors to the L-type Ca²⁺ current [84].

Among the multitude of K_V channel transcripts (Fig. 1), the K_V7.1 or KCNQ1 channel had the highest number of RNA reads (821.00 ± 77.20), consistent with previous studies showing its response to membrane stretch [88] and hypo-osmotic swelling [88, 89] (Table 1). The K_V1.2 channel, which has also been shown to exhibit mechanosensitivity [67], was present in smaller quantities with 212.00 ± 10.02 RNA reads (Table 1; Fig. 1). In particular, the number of K_V1.2 transcripts was 3.5 times less than that of the predominant K_V7.1 (Fig. 1).

Among K₂P channels, TREK-1/K₂P2.1 had the highest number of RNA reads (153.00 ± 11.93) compared to others, consistent with its established mechanosensitivity [75,76,77]. TRAAK/K2P4.1 channels also exhibit mechanosensitivity [76, 93], although the number of detected RNA reads was very small (0.66 ± 0.33) (Table 1; Fig. 1). Interestingly, transcripts of the mechanosensitive channels TREK-2/K₂P10.1, which are expressed in the atrial and ventricular cells of human and murine hearts [76], were not detected in our isolated rat ventricular myocytes.

The RNA-seq technique revealed the presence of several Kir channel transcripts (Fig. 1) in freshly isolated rat ventricular cardiomyocytes. Among them, Kir6.2 (1240.66 ± 57.1) was detected in the highest abundance and demonstrated mechanosensitivity only when associated with the sulfonylurea receptor 2 A (SUR2A) subunit [91] (Table 1; Fig. 1). Additionally, RNA reads were found for Kir6.1 channels (82.66 ± 6.49), which are also mechanosensitive [92]. However, unlike mouse ventricular cardiomyocytes, where Kir2.3 channels have been shown to be mechanosensitive [22], we did not detect any RNA reads for Kir2.3 channels in rat ventricular cardiomyocytes.

We identified three types of stretch-activated K_Ca (SAKCA), or BK_Ca channels, that exhibit mechanosensitivity [78]: calcium-activated potassium channel subunit beta-2 (KCNMB2) (6.6 ± 0.9), KCNMB4 (2.7 ± 0.9), and KCNMB1 (0.3 ± 0.3) (Table 1; Fig. 1). Although ATP-sensitive potassium channels (K_ATP) also exhibit mechanosensitivity in rats [41, 80], we did not investigate their presence in our study. Likewise, we did not analyze the expression of Kv7 (KCNQ) channels [88,89,90].

We observed several transcripts for nonselective cation TRP channels, many of which are known to be mechanically gated (listed in descending order of RNA reads). In particular, mechanosensitive TRPM7 (333.5 ± 31.5) [36,37,38,39, 94], TRPC1 (146.66 ± 6.96) [40, 42, 43], TRPM4 (179.66 ± 5.93) [44,45,46,47, 95], TRPV2 (21.66 ± 2.03) [48, 49], TRPC3 (5.0 ± 3.0) [50,51,52,53,54,55, 96], TRPV4 (5.5 ± 0.5) [56,57,58], TRPV1 (2.00 ± 0.58) [59, 60], TRPM3 (1.3 ± 0.9) [61], TRPC5 (1.3 ± 0.9) [62, 97], TRPA1 (1.0 ± 0.9) [63,64,65], and TRPC6 (1.0 ± 1.0) [22, 47, 66, 67, 97] were among the most abundant.

We also identified two additional stretch-activated channels, Piezo 1 (145.33 ± 8.01) [71,72,73, 98, 99], and Piezo 2 (6.0 ± 2.0) [73, 74], that were present in the ventricular cardiomyocytes of rats.

The data presented here and below reveal an intriguing observation: While many ion channels have been studied in terms of their response to membrane stretch, only a few have been thoroughly investigated. However, among those that have been studied, all demonstrate a certain degree of mechanosensitivity. On the basis of this, it is plausible to speculate that mechanosensitivity is a ubiquitous characteristic of ion channels and that all channels, irrespective of their gating mechanism, should respond to membrane stretch and tension.

Definition and analysis of L-type Ca ²⁺ current

In this study, two methods were used to estimate the current through L-type Ca²⁺ channels (I_Ca,L) in K⁺ _in/K⁺ _out solutions. The first method involved evaluating the time course of I_Ca,L by subtracting the late current at the end of the 140 ms pulse from the negative peak current (Fig. 2A). The second method involved determining the current-voltage (I/V) relationship for I_Ca,L by subtracting the current at the same potential in the presence of 5 or 10 µM nifedipine from the maximum point of the negative peak current (Fig. 2B). The value of I_L at 0 mV was almost equal to the current in the presence of nifedipine (Fig. 2A). The negative peak current at 0 mV in control conditions (blue track) and in the presence of 10 µM nifedipine (red track), which selectively blocks L-type Ca²⁺ channels, was used to estimate I_Ca,L. I_Ca,L was calculated by taking the difference between the negative peak current and the current in the presence of nifedipine (or by subtracting the late current at the end of the 140 ms or 350 ms pulse at a frequency of 1 Hz). Since the I/V relation of I_L in control (green triangles) completely coincides with the I/V relation for I_Ca,L in the presence of nifedipine (red circles) in the range from − 80 mV to + 50 mV (Fig. 2B), the calculation method presented in previous study for K⁺ _in/K⁺ _out solutions was applied [21], whereas I_Ca,L was estimated as the difference between the negative current at − 10 mV (blue circle) and either the current in the presence of nifedipine at − 10 mV (red circle) or the late current at − 10 mV in control (green triangles) (Fig. 2B).

It should be noted that in K⁺ _in/K⁺ _out solutions, the positive peak on the I/V relation for I_Ca,L between − 30 and − 70 mV (Fig. 2B: blue circles) is associated with the positive currents in response to voltage steps in this potential range and was first demonstrated by Josephson and Sperelakis (1982), [100]. However, not all subsequent studies on I_Ca,L took this positive peak into account.

Fig. 2

Evaluation of I_Ca,L and I_ns in K⁺ _in/K⁺ _out and Cs⁺ _in/Cs⁺ _out solutions. The membrane potential (V_m) was maintained at V_hp = − 45 mV. A: Raw recordings of I_L and I_Ca,L (blue trace) and recordings in the presence of nifedipine (red trace) in K⁺ _in/K⁺ _out solutions. B: Current-voltage (I/V) relationships for I_L (green triangles) and peak I_Ca,L (blue circles) under control conditions and in the presence of nifedipine (red circles). C: Raw recordings of I_L and I_Ca,L (blue trace) and recordings in the presence of nifedipine (red trace) in Cs⁺ _in/Cs⁺ _out solutions, where K⁺ currents are suppressed. D: I/V curves for I_L (blue triangles) and I_Ca,L (blue circles) under control conditions and in the presence of nifedipine (red circles)

Figure 2 C shows that in Cs⁺ _in/Cs⁺ _out solutions, the negative peak current at 0 mV in control conditions (blue track) and in the presence of 10 µM nifedipine (red track) were used to estimate I_Ca,L by taking the difference between the peak current and the current in the presence of nifedipine or the late current at the end of the pulse. Figure 2D shows the I/V relations for I_Ca,L in control conditions (blue circles) and in the presence of nifedipine (red circles). In this case, I_Ca,L was calculated as the difference between the negative current (blue circles) and the current in the presence of nifedipine (red circles) or when the current equals 0 nA. The I/V relations for I_L measured at the end of the pulse are shown under control conditions (blue triangles) and in the presence of nifedipine (red triangles). Our experiments used an internal pipette solution containing 0.01 mM EGTA, and under these conditions, we measured the I_Ca,L value when cells were stretched by 4, 6, 8, and 10 μm. Previous studies have shown that the reduction of I_Ca,L, even by a stretch of 10 μm, was no longer observed when the cells were pre-dialyzed with 5 mM BAPTA, i.e., control I_Ca,L and I_Ca,L during stretch superimposed [6]. To study I_Ca,L in the context of cell elongation under the action of various compounds, a standard elongation of 6 μm was used in this study.

As demonstrated earlier in K⁺ _in/K⁺ _out solutions, the late current (I_L) at the end of the pulse reflects the cell’s response to stretching. Stretch-induced changes in I_L exhibit an outward rectifying voltage dependence with a reversal potential (E_rev) of − 16 mV [5, 6]. To calculate stretch-induced current (Δ^SI_L), we take the difference between the control current values (^CI_L) and the current values in the presence of cell stretching (^SI_L) at − 45 or − 80 mV (Δ^SI_L(−45) and Δ^SI_L(−80)) [5].

In our earlier work conducted on isolated mouse ventricular cardiomyocytes [5, 6], as well as in studies by our colleagues on the same cells [22], it was found that in K⁺ _in/K⁺ _out solutions, the net current was composed of I_ns (current through stretch-activated non-selective cation channels), I_K1 (inwardly rectifying potassium current), and I_o (presumably the sum of several outwardly rectifying currents, such as K⁺ currents through TREK channels [76, 77] or the outwardly rectifying canonical transient receptor potential-6 (TRPC6) channels [66]. Cell stretch can modify net membrane currents (Δ^SI_L) by modulating both K⁺ and I_ns currents [5]. However, our present studies investigating the RNA transcripts of ion channels and their mechanosensitivity reveal significant differences in the channel spectrum of ventricular cardiomyocytes in adult rats (see below).

To distinguish the effects on I_ns, we used two methods: first, we eliminated inward rectifying K⁺ currents by substituting extracellular K⁺ with Cs⁺ and second, we decreased outward rectifying K⁺ currents by replacing intracellular K⁺ with Cs⁺. In this case, the current through stretch-activated channels (I_SAC) was calculated as the difference between the control current values (^CI_ns) and the current values on the background of cell stretching (^SI_ns) at − 45 or − 80 mV (I_SAC(−45) and I_SAC(−80)). We used the same approach to measure the differential current in response to specific compounds (ΔI_X) (where “X” represents the compound) at − 45 or − 80 mV [5, 6].

Under our experimental conditions, isolated cardiomyocytes maintained their fundamental properties, as demonstrated by the significant increase in I_Ca,L peak current amplitudes at 0 mV in response to 20 nM isoproterenol in K⁺ _in/K⁺ _out and Cs⁺ _in/Cs⁺ _out solutions (not shown), which is consistent with previous literature reports for this compound [101].

Local stretch increased net currents I _L and decreased I _Ca,L in K ⁺ _in /K ⁺ _out solutions (time-course and voltage-dependence)

Figure 3A.1 illustrates the effect of a 6 μm cell stretch on time-dependent net membrane currents in K⁺ _in/K⁺ _out solutions. At control, the registered holding current (the current at V_hp = − 45 mV (I_hc)) was + 0.44 nA (label C, Fig. 3A.1), and stretch changed this current (^SI_hc) to + 0.27 nA (label S, Fig. 3A1). The stretch-induced difference of the holding current Δ^SI_hc was (-) 0.17 nA ((-) 0.19 ± 0.01 nA, n = 14) at − 45 mV. The minus sign (-) emphasizes that cell stretch leads to more negative values of the initial net holding current at the level of V_hp = − 45 mV. The pulse to 0 mV induced the L-type Ca²⁺ current (I_Ca,L) that activated and inactivated over time. The stretch attenuated the L-type Ca²⁺ current (Fig. 3A.1), which was estimated as the difference between the negative peak current I_max and the current at the end of the pulse in control (I_L) (from |1.55| nA to |1.11| nA during the stretch). The module for I_Ca,L was used because the current value was calculated as a distance (Fig. 3A.1, B.2, C.2) from the starting point (the value of I_Ca,L on the background of nifedipine = I_L, in the positive region) to the point I_max, in the negative region. The time course of the stretch - induced difference of the L-type Ca²⁺ current in the control and during the stretch (Δ^SI^tc_Ca,L) was (+) 0.44 nA ((+) 0,44 ± 0.07 nA, n = 7). The plus sign indicates that cell stretching at the V_hp = − 45 mV led to a shift of negative I_Ca,L to more positive values compared to the initial values, resulting in a decrease in the negative area.

The I/V curves in Fig. 3A.2 show the voltage dependence of I_L and I_Ca,L, and their modulation by a 6 μm stretch. Before stretching, the I_L - I/V curve was N-shaped and crossed the voltage axis (zero current potential V₀) at − 78 mV (− 75 ± 3 mV, n = 7; equivalent to the resting potential of the non-clamped cell). The modest stretch shifted the net currents I_L to more negative values: Δ^SI_{L(− 45)} is (-) 0.15 nA ((-) 0.16 ± 0.02 nA, n = 7) at − 45 mV and Δ^SI_L(−80) is (-) 0.24 nA ((-) 0.26 ± 0.03 nA, n = 7) at − 80 mV. The V₀ also shifted to -70 mV (− 66 ± 3 mV, n = 7) with the stretch.

During cell stretching, although the changes in the ^SI_{L(− 45)} current occur in the positive range, their values are lower than the control values of I_L(−45). The minus (-) sign of Δ^SI_L(−45) emphasizes that cell stretch leads to a change in the curve toward more negative values. Similarly, for I_L(−80), which is usually in the negative range, cell stretch further increases this negative current ^SI_L(−80), leading to a negative differential current Δ^SI_L(−80) that only shows the direction of its change. The I/V curves recorded before and during the stretch crossed each other close to 0 mV, and at positive potentials, the late current increased with the stretch. The I/V curve of I_Ca,L decreased during the 6 μm stretch by Δ^SI_Ca,L = (+) 0.38 nA ((+) 0.32 ± 0.04 nA, n = 7) compared to the control values.

Upon stretching to 8 μm, the holding current ^SI_hc decreased to values close to 0 (label S, Fig. 3B.1), compared to the control holding current (label C, the beginning of the blue tracesin this figure). In this case, the stretch-induced difference in the holding current Δ^SI_hc was (-) 0.41 nA ((-) 0.42 ± 0.02 nA, n = 7) at V_hp = − 45 mV. The pulse at 0 mV induced I_Ca,L, and the stretch attenuated I_Ca,L (Fig. 3B.1, negative current wave from blue in control to red during the stretch) from |1.55| nA in control to |0.94| nA. During time course registration, the stretch-induced difference of L-type Ca²⁺ current in control and during the stretch, Δ^SI^tc_Ca,L equals (+) 0.61 nA ((+) 0,60 ± 0.04 nA, n = 7).

Stretching by 8 μm shifted the I/V relation to more negative currents than a 6 μm stretch (Fig. 3B.2): At − 45 mV, the stretch-induced difference in current was Δ^SI_L(−45) = (-) 0.29 nA ((-) 0.37 ± 0.06 nA, n = 7), while at − 80 mV, it was Δ^SI_L(−80) = (-) 0.57 nA ((-) 0.60 ± 0.09 nA, n = 7), and V₀ was depolarized to -56 mV (− 54 ± 3 mV, n = 7). The I/V curve of I_Ca,L during the stretch showed a further decrease in ^SI_Ca,L, resulting in an increase in the differential current Δ^SI_Ca,L to (+) 0.52 nA ((+) 0.52 ± 0.02 nA, n = 7).

The maximum stretch of 10 μm caused a shift in the holding current ^SI_hc to the negative region with values equal to − 0.16 nA (label S, the beginning of the red traces in Fig. 3C.1, compared to label C, the beginning of the blue traces). The stretch-induced difference in the holding current Δ^SI_hc was (-) 0.61 nA ((-) 0.74 ± 0.10 nA, n = 7) at − 45 mV. This stretch resulted in a further reduction of I_Ca,L (Fig. 3C.1: negative current wave from blue in control to red during the stretch) from |1.55| nA in control to |0.77| nA. When registering the time course, Δ^SI^tc_Ca,L was equal to (+) 0.78 nA ((+) 0.77 ± 0.04 nA, n = 6).

Stretching the cell by 10 μm caused a larger shift in the I/V relationship to more negative currents compared to an 8 μm stretch (Fig. 3C.2): Δ^SI_L(−45) was (-) 0.57 nA ((-) 0.71 ± 0.09 nA, n = 6) at − 45 mV, while Δ^SI_L(−80) was (-)1.33 nA ((-)1.41 ± 0.15 nA, n = 6) at − 80 mV, and depolarized V₀ to − 35 mV (− 39 ± 3 mV, n = 6). During this stretch, the I/V curve of I_Ca,L showed a larger decrease in ^SI_Ca,L, resulting in an increased differential current Δ^SI_Ca,L of (+) 0.66 nA ((+) 0.65 ± 0.05 nA, n = 6).

Fig. 3

Reduction of I_Ca,L in K⁺ _in/K⁺ _out solutions during local stretching of cardiomyocytes by 6, 8, and 10 μm. V_hp = − 45 mV. A: (6 μm stretch). A.1 - The time course of the membrane current. The holding current at V_hp in control (beginning of the blue traces - label C) and during stretching (beginning of the red traces - label S). A pulse from − 45 to 0 mV induces I_Ca,L, which decreases during stretching (indicated by a negative blue current wave compared to a negative red wave). A.2 – I/V curve of I_L before (blue triangles) and during (red triangles) stretching, as well as I_Ca,L before (blue circles) and during (red circles) stretching. B: (8 μm stretch). B.1 - The time course of the membrane current before and during stretching, which results in a greater reduction of I_Ca,L compared to A.1. Notations as in A.1. B.2 – I/V curves of I_L and I_Ca,L before and during stretching. Notations as in A.2. C: (10 μm stretch). C.1 – The time course of the membrane current in control and during stretching. I_Ca,L decreases with increasing stretching. Notations as in A.1. C.2 – I/V curves of I_L and I_Ca,L before and during stretching. Notations as in A.2

In K⁺ _in/K⁺ _out solutions, Gd³⁺ eliminates stretch-induced reduction in I_Ca,L at all levels of cell stretch, as previously demonstrated [5, 6].

The comparison of the mean values of differential currents Δ^SI_hc, Δ^SI_L(−45), Δ^SI_L(−80), Δ^SI^tc_Ca,L and Δ^SI_Ca,L during cell stretch at 6, 8, and 10 μm is presented in Fig. 4. It should be noted that Δ^SI_hc, Δ^SI_L(−45), and Δ^SI_L(−80) are mechanosensitive and are based on the activation of stretch-activated channels (SAC).

Fig. 4

Comparison of the mean values of the differential currents Δ^SI_hc, Δ^SI_L(−45), Δ^SI_L(−80), Δ^SI^tc_Ca,L and Δ^SI_Ca,L during cell stretching at 6, 8, and 10 μm in K⁺ _in/K⁺ _out solutions. Error bars represent ± SD for n experiments (n = 7)

Local stretch increased current through nonselective cation channels I _ns and, at the same time, reduced I _Ca,L in Cs ⁺ _in /Cs ⁺ _out solutions (time course and voltage dependence).

In Fig. 5A.1, the effect of 6 μm stretch on time-dependent membrane currents in Cs⁺ _in/Cs⁺ _out solutions is presented. At the control stage, the holding current was 0 nA (labeled as C, Fig. 5A.1), while stretch resulted in negative holding current values (labeled as S, Fig. 5A.1). The difference in the holding current induced by stretch was (-) 0.12 nA (0.09 ± 0.01 nA, n = 7) at − 45 mV. The pulse to 0 mV resulted in an L-type Ca²⁺ current of − 1.49 nA (− 1.62 ± 0.13 nA, n = 18), or − 7.3 pA/pF (− 7.1 ± 0.2 pA/pF, n = 18). The stretch of 6 μm reduced the L-type Ca²⁺ current, ^SI_Ca,L (as shown by the negative current wave in Fig. 5A.1, from blue to red). During the recording of the time course, the stretch-induced difference of L-type Ca²⁺ current in control and during the stretch (Δ^SI^tc_Ca,L) was (+) 0.41 nA ((+) 0.35 ± 0.05 nA, n = 7).

Figure 5 A.2 displays the voltage dependence of L_ns, I_Ca,L and its modulation by a 6 μm stretch, as shown in the I/V curves. Before stretching, the I/V curve of I_ns was linear and intersected the voltage axis (zero current potential V₀) at − 40 mV (− 38 ± 3 mV, n = 7). After a modest 6-µm stretch, I_ns shifted to more negative values, and the resulting I_SAC = ^SI_ns - ^CI_ns (in Cs⁺ _in/Cs⁺ _out solutions) was (-) 0.10 nA ((-) 0.09 ± 0.02 nA, n = 7) at − 45 mV and (-) 0.14 nA ((-) 0.12 ± 0.01 nA, n = 7) at -80 mV, while V₀ changed to − 5 mV (− 6 ± 3 mV, n = 7). At potentials close to 0 mV, the I/V curves recorded before and during the stretch intersected, while at positive potentials, I_ns exhibited an increase due to the stretch. Furthermore, the I/V curve of ^SI_Ca,L during the stretch decreased from − 1.27 nA in the control to − 1.00 nA, resulting in a Δ^SI_Ca,L of (+) 0.27 nA ((+) 0.34 ± 0.06 nA, n = 7) when compared to the control values.

When subjected to an 8 μm stretch, the difference in the holding current at − 45 mV changed to values close to (-) 0.17 nA ((-) 0.16 ± 0.301 nA, n = 7; labeled S, Fig. 5B.1, compared to label Cin the figure). The pulse from − 45 to 0 mV induced I_Ca,L, but the 8 μm stretch attenuated ^SI_Ca,L (as seen in Fig. 5B.1, where the negative current wave shifts from blue to red during the stretch), from − 1.49 nA in the control to − 0.86 nA. When the time course was registered, Δ^SI^tc_Ca,L was calculated to be (+) 0.63 nA ((+) 0.64 ± 0.06 nA, n = 7).

The 8 μm stretch caused a shift in the I/V relation to more negative currents compared to the 6 μm stretch (as seen in Fig. 5B.2): I_SAC(−45) was (-) 0.16 nA ((-) 0.16 ± 0.02 nA, n = 7) at − 45 mV, and I_SAC(−80) was (-) 0.25 nA ((-) 0.18 ± 0.04 nA, n = 7) at − 80 mV. During this stretch, the I/V curve of I_Ca,L further decreased ^SI_Ca,L from − 1.27 to − 0.54 nA, resulting in an increase in differential current Δ^SI_Ca,L to (+) 0.73 nA ((+) 0.62 ± 0.09 nA, n = 7) compared to control values.

The longest 10 μm stretch caused a shift of the holding current to the negative region (label S, Fig. 5C.1, compared to label C). The stretch-induced difference in the holding current was (-) 0.35 nA ((-) 0.34 ± 0.02 nA, n = 7) at − 45 mV. Furthermore, this stretch reduced ^SI_Ca,L even more (Fig. 5C.1), from − 1.49 nA in the control to − 0.74 nA. In the case of time course registration, Δ^SI^tc_Ca,L equals (+) 0.75 nA ((+) 0.75 ± 0.05 nA, n = 7).

Fig. 5

Reduction of I_Ca,L in Cs⁺ _in/Cs⁺ _out solutions with K⁺ currents suppressed during local stretching of cardiomyocytes by 6, 8, and 10 μm. V_hp = − 45 mV. A: (6 μm stretch). A.1 – The time course of the membrane current. The holding current at V_hp in control (beginning of the blue traces - label C) and during stretching (beginning of the red traces – label S). A pulse from − 45 to 0 mV induces I_Ca,L, which decreases during stretching (indicated by a negative blue current wave compared to a negative red wave). A.2 – I/V curve of I_ns before (blue triangles) and during (red triangles) stretching, as well as I_Ca,L before (blue circles) and during (red circles) stretching. B: (8 μm stretch). B.1 - The time course of the membrane current before and during stretching, which results in a greater reduction of I_Ca,L compared to A.1. Notations as in A.1. B.2 – I/V curves of I_ns and I_Ca,L before and during stretching. Notations as in A.2. C: (10 μm stretch). C.1 – The time course of the membrane current in control and during stretching. I_Ca,L decreases with increasing stretching. Notations as in A.1. C.2 – I/V curves of I_ns and I_Ca,L before and during stretching. Notations as in A.2

Stretching by 10 μm shifted the I/V relation to more negative currents than an 8 μm stretch (Fig. 5C.2). Specifically, I_SAC(−45) was (-) 0.32 nA ((-) 0.34 ± 0.05 nA, n = 7) and I_SAC(−80) was (-) 0.53 nA ((-) 0.48 ± 0.10 nA, n = 7). The 10-µm stretch further reduced ^SI_Ca,L to − 0.38 nA (Fig. 5C.2, red circles), resulting in an increase in differential current Δ^SI_Ca,L to (+) 0.89 nA ((+) 0.79 ± 0.09 nA, n = 7).

In Cs⁺ _in/Cs⁺ _out solutions, Gd³⁺ eliminates the reduction of stretch-induced I_Ca,L as well as Δ^SI_L(−45), Δ^SI_L(−80), as previously demonstrated in our studies [5, 6] (not shown in Fig. 5).

Fig. 6

Comparison of the mean values of the differential currents I_SAC(hc), I_SAC(−45), I_SAC(−80), Δ^SI^tc_Ca,L and Δ^SI_Ca,L during cell stretching at 6, 8, and 10 μm in Cs⁺ _in/Cs⁺ _out solutions. The ± SD is shown for n experiments (n = 7)

Figure 6 compares the mean values of I_SAC(hc), I_SAC(−45), I_SAC(−80), ^SI_Ca,L, and Δ^SI_Ca,L during cell stretching at 6, 8, and 10 μm in Cs⁺ _in/Cs⁺ _out solutions. I_SAC(hc), I_SAC(−45), and I_SAC(−80) are also mechanosensitive currents based on the work of SAC. Additionally, I_Ca,L is likely to be conducted through Ca_V1.2 channels and is also mechanosensitive in ventricular rat cardiomyocytes.

Gd ³⁺ eliminates the L-type Ca ²⁺ current via Ca _V 1.2 channels in K ⁺ _in /K ⁺ _out, Cs ⁺ _in /K ⁺ _out, K ⁺ _in /Cs ⁺ _out and Cs ⁺ _in /Cs ⁺ _out solutions

Fig. 7

Effect of 5 µM Gd³⁺ on I_L and I_Ca,L in K⁺ _in/K⁺ _out (A), Cs⁺ _in/K⁺ _out (B), K⁺ _in/Cs⁺ _out (C), and Cs⁺ _in/Cs⁺ _out (D) solutions. Control curves in blue: I_L – triangles and I_Ca,L – circles. Curves in the presence of Gd³⁺ in red: I_L – triangles and I_Ca,L – circles. Note: C: Switching cell perfusion from K⁺ _in/K⁺ _out solution (I_L – black triangles and I_Ca,L – black circles) to K⁺ _in/Cs⁺ _out solution (I_L – blue triangles and I_Ca,L – blue circles), followed by the application of Gd³⁺ (I_L – red triangles and I_Ca,L – red circles). While Gd³⁺ exerts only a minor effect on I_L, it eliminates I_Ca,L in all solutions tested

Figure 7. A, B, C, and D illustrate that I_Ca,L values remained constant (1.48 ± 0.06 nA, n = 24) under all experimental conditions tested, including K⁺ _in/K⁺ _out, Cs⁺ _in/K⁺ _out, K⁺ _in/Cs⁺ _out, and Cs⁺ _in/Cs⁺ _out solutions. Furthermore, during the transition of cell perfusion from K⁺ _in/K⁺ _out to K⁺ _in/Cs⁺ _out solutions (Fig. 7C), I_Ca,L values did not change (n = 6). The I_Ca,L values were determined by calculating the maximum calcium peak current at a given potential at the point on the late current curve that corresponds to that potential. For Cs⁺ _in/Cs⁺ _out solutions, two variants of I_Ca,L values were presented (Fig. 7B): (1) up to zero and (2) up to a characteristic point on the late current curve. Typically, only the value (1) is used for calculations in the literature.

It has been shown that, on the background of cell stretch, Gd³⁺ at a concentration of 5 µM blocks I_SAC [6]. The addition of Gd³⁺ abolished the dependence of I_SAC on all local stretch values. Our experiments revealed that Gd³⁺ also had a minor inhibitory effect on the I_L current at − 45 and − 80 mV, suggesting that I_SAC may have contributed to the net currents of the non-stretched cells. This is depicted in Fig. 7, where the blue triangles indicate the control and the red triangles indicate the application of the blocker (n = 6). Additionally, Gd³⁺ did not affect the background current, I_K1 (Fig. 7A).

To investigate potential changes in L-type Ca²⁺ current through Ca_V1.2 channels, we applied Gd³⁺, which is known to be a nonspecific blocker of the mechanically gated channel’s current I_SAC. Our results indicated that the addition of 5 µM of Gd³⁺ eliminated L-type Ca²⁺ current in K⁺ _in/K⁺ _out (n = 6), Cs⁺ _in/K⁺ _out (n = 6), K⁺ _in/Cs⁺ _out (n = 6), and Cs⁺ _in/Cs⁺ _out (n = 6) solutions (blue circles in the control compared to red circles after application of the blocker).

Discussion

Channel transcripts and their mechanosensitivity

In ventricular cardiomyocytes, mechanosensitive channels are believed to play a role in the regulation of the contractile properties of the heart [5]. In our experiments, we found significant amounts of transcripts for the mechanosensitive channels TRPM7, TRPC1, and TRPM4 (Table 1).

TRPM7 is a nonselective cation channel that is permeable to both Ca²⁺ and Mg²⁺ [102]. It has been shown to contribute to the stretch-induced current in ventricular cardiomyocytes by mediating a Ca²⁺ influx in response to mechanical stretch [103, 104]. This influx of Ca²⁺ is believed to activate downstream signaling pathways that regulate contractility [103]. TRPC1 was shown to contribute to the stretch-induced current in ventricular cardiomyocytes by mediating a non-selective cation influx in response to mechanical stretch, which influx is believed to depolarize the cell membrane and contribute to the regulation of contractility [104]. TRPM4 is a Ca²⁺-activated non-selective cation channel that has been shown to contribute to the stretch-induced current in ventricular cardiomyocytes [45]. It is believed to be activated by Ca²⁺ influx through other channels such as TRPC1 and TRPM7 [37, 42, 105]. TRPM4 activation is believed to contribute to contractility regulation by modulating the duration of AP [106].

Studies on mechanosensitivity in murine ventricular myocytes have shown a link between stretch-activated TRPC6 channels as well as stretch-deactivated GK1 and especially Kir2.3 channels [22]. GK1 channels are formed by Kir2.1, Kir2.2, and Kir2.3 proteins [107, 108]. In our experiments, we found multiple transcripts of Kir2.1 and Kir2.2 in isolated ventricular myocytes from rats, while Kir2.3 was absent. Therefore, in K⁺ _in/K⁺ _out solutions, the mechanosensitivity of ventricular myocytes is probably determined by the stretch-activated TRPM7, TRPC1, and TRPM4 channels and the stretch-deactivated GK1 (Kir2.1 and Kir2.2) channels. Our findings are consistent with other studies [109], which have shown that ventricular I_K1 is based primarily on the heteromeric assembly of the Kir2.1 and Kir2.2 channels, while the Kir2.3 channels are more relevant in the atrium.

The local stretch of cells modulates I _Ca,L, probably through Ca _V 1.2 channels

One of the main questions of the study was to determine the cause of changes in I_Ca,L. Our observations showed a decrease in ^SI_Ca,L with discrete cell stretching, while there was a corresponding increase in Δ^SI^tc_Ca,L and Δ^SI_Ca,L during cell stretching at 6, 8, and 10 μm.

In general, membrane channels (including voltage-gated) are expected to respond to mechanical stimuli because mechanical energy can affect the energy barriers between different conformational states of a channel protein [110]. For example, longitudinal stretching has been shown to modify the gating mechanism of; Na_V channels [111], such as Na_V1.5 [81, 82] and Na_V1.6 [83], Ca_V channels such as Ca_V1.2 [84], Ca_V1.3 [85], and Ca_V2.2 channels [86], as well as K_V channels such as K_V1.2 [67], K_V1.5, and K_V3.2 [87, 112]. In addition, mechanical energy can affect the conformation of proteins associated with cytoskeletal elements [113, 114], such as F-actin [114] and integrins [115]. This can also activate stretch-sensitive kinases (Src, MAP) or phosphatases, which can, in turn, modulate the channel protein by phosphorylation and dephosphorylation [109]. Finally, mechanical stretch can also impact the production rate of reactive oxygen species, or NO, which can modify channel gating through oxidative or nitrosative mechanisms [116], which in turn can play a role in stretch-induced effects on Ca²⁺ release from the sarcoplasmic reticulum (SR) [116, 117].

The L-type Ca²⁺ current is responsible for generating the negative current wave observed at the start of the depolarizing clamp step (Fig. 3A.1, B.1, and C.1), as well as the long-lasting plateau of the AP. Interestingly, the local mechanical stretch has been shown to decrease I_Ca,L, with the reduction being observed across the entire range of clamp potentials. In particular, the voltage dependence of I_Ca,L appears to remain unchanged under these conditions (see Fig. 3A.2, B.3, and C.4).

In ventricular myocytes, some nonselective channels are known to be activated by an increase in cytosolic calcium concentration ([Ca²⁺]_C) [118]. To investigate whether stretch-induced changes in [Ca²⁺]_C might play a role in these effects, Gannier et al. (1996) conducted experiments in which they prevented possible stretch-induced increments in [Ca²⁺]_C [119]. Interestingly, they have found that chelation of the [Ca²⁺]_C does not have a significant effect on I_SAC. However, we observed significant effects of BAPTA on the stretching effects on I_Ca,L. Specifically, while I_SAC in Cs⁺ _in/Cs⁺ _out solutions was found to be insensitive to chelation of [Ca²⁺]_C, the stretch-induced reduction of I_Ca,L disappeared after dialyzing the cell with 5 mM BAPTA in the patch pipette [6].

Since stretch has been shown to increase [Ca²⁺]_C and BAPTA can chelate cytosolic Ca²⁺, it is likely that the stretch-induced reduction of ^SI_Ca,L results from the stretch-induced increase in [Ca²⁺]_C followed by Ca²⁺-mediated inactivation of the Ca²⁺ channel (Ca²⁺-calmodulin interaction with the Ca²⁺ channel α subunit) [12]. We propose that the primary effect of stretch is to increase [Ca²⁺]_C via an increase in Ca²⁺ influx through SACs or via Ca²⁺ release from the SR in the vicinity of the L-type Ca²⁺ channel. Previous research has suggested that I_Ca,L can be reduced by an increase in [Ca²⁺]_C and by stimulation of I_P (current due to electrogenic sodium pumping) by an increase in [Na⁺]_C. The concept of indirect activation by stretch should also be applied to the current generated by Na⁺/Ca²⁺-exchange if [Na⁺] is elevated by Na⁺ influx due to I_SAC [120] and to K⁺ channels activated by Ca²⁺ or Na⁺ ions [121]. Alternatively, the stretch may increase [Ca²⁺]_C by Ca²⁺ release from the SR [117].

Cytoskeletal F-actin fibers were found to be involved specifically in I_SAC activation and regulation of the Ca_V1.2 channel gating mechanism. In the context of cell stretch, which has been shown to increase I_SAC and decrease I_Ca,L, treatment with cytochalasin D, a toxin known to depolymerize F-actin, blocks the effects of stretch on late currents and leads to further reduction of I_Ca,L [8], which also corroborates the effect of cell stretch, probably on the Ca_V1.2 channel.

We make such an assumption based on the analysis of the transcript quantities for Ca²⁺ ion channel genes. As we have demonstrated, the transcript counts for the Ca_V1.2 channel gene are exceptionally high, measuring 1336.66 ± 71.8. In contrary, for Ca_V1.3, this count is quite low at 17.33 ± 2.03, and for Ca_V1.1, it is minimal at 0.33 ± 0.33. We did not detect other channels involved in the whole L-type Ca²⁺ current. Based on these numerical values, we can infer the primary role of the Ca_V1.2 channel in the formation of the L-type Ca²⁺ current that we observed.

Voltage-gated current modulation or mechanosensitivity of the (Ca_V1.2) channel

It is now well established that Gd³⁺ blocks a range of ion channels, including SACs (mechanically gated non-selective cation channels, MGC_ns, and mechanically gated potassium channels, MGC_K), which was previously assumed in the study by Yang et al. (1989) [122]. In addition, Gd³⁺ has been found to block several other ion channels, such as BK_Ca channels cloned from chick ventricular myocytes [123] and delayed rectifier current (I_K) in guinea pig single ventricular myocytes, including both I_KR (rapidly activated) and I_KS (slowly activated), while the background current I_K1, was not affected by Gd³⁺ [124]. Moreover, Gd³⁺ was a potent I_Na blocker near the threshold potential for Na⁺ channels in rabbit ventricular myocytes [122]. It has been demonstrated that Gd³⁺ is also a potent blocker of L-type Ca²⁺ channels in isolated guinea pig ventricular myocytes [125].

In our experiments, treatment with the nonspecific inhibitor of I_SAC, Gd³⁺, on the background of cell stretching eliminated both I_SAC and I_Ca,L at all stretching magnitudes and voltage steps relative to the V_hp. This response to Gd³⁺ was maintained in Cs⁺ _in/Cs⁺ _out solutions, indicating that voltage-gated channels, apparently Ca_V1.2 in adult rat ventricular myocytes, have additional mechanosensitive properties.

Conclusion

Our analysis revealed the presence of transcripts for the TRPM7, TRPC1, and TRPM4 channels that are known to exhibit mechanosensitivity. Furthermore, we detected mechanosensitive transcripts of the Kir6.2 and Kir6.1 channels, as well as transcripts of GK1 channels formed by the Kir2.1 and Kir2.2 proteins. Although the detection of the TREK-1/K_2P2.1 transcript was limited, indirect data from other studies supports its presence [77]. The highest number of RNA reads among all Ca_V channels was detected for Ca_V1.2 channels, which themselves exhibit mechanosensitivity.

Cell stretching at various magnitudes discretely increased I_SAC and decreased I_Ca,L probably through Ca_V1.2 channels in K⁺ _in/K⁺ _out, Cs⁺ _in/K⁺ _out, K⁺ _in/Cs⁺ _out, and Cs⁺ _in/Cs⁺ _out solutions. However, treatment with the nonspecific blocker of I_SAC, Gd³⁺, on the background of cell stretching eliminated both I_SAC and I_Ca,L at all stretching magnitudes and voltage steps relative to V_hp. The study suggests that voltage-gated Ca_V1.2 channels in adult rat ventricular myocytes have additional mechanosensitive properties, as evidenced by the maintained response to Gd³⁺ in the Cs⁺ _in/Cs⁺ _out solution.