OUP user menu

X-ROS signalling is enhanced and graded by cyclic cardiomyocyte stretch

Benjamin L. Prosser, Christopher W. Ward, W. Jonathan Lederer
DOI: http://dx.doi.org/10.1093/cvr/cvt066 307-314 First published online: 21 March 2013


Aims A sustained, single stretch of a cardiomyocyte activates a transient production of reactive oxygen species by membrane-located NADPH oxidase 2 (Nox2). This NoX2-dependent ROS (X-ROS) tunes cardiac Ca2+ signalling by reversibly sensitizing sarcoplasmic reticulum Ca2+ release channels. In contrast to static length changes, working heart cells are cyclically stretched and shortened in the living animal. Additionally, this stretch cycle is constantly varied by changes in the pre-load and heart rate. Thus, the objective of this study was (i) to characterize X-ROS signalling during stretch-shortening cycles and (ii) to evaluate how the amplitude (pre-load) and frequency (heart rate) of cell stretch affects X-ROS and Ca2+ signalling.

Methods and results Single adult rat cardiomyocytes were attached to MyoTak™-coated micro-rods and stretched, while ROS production and Ca2+ signals were monitored optically. Although a sustained stretch led to only a transient burst of ROS, cyclic stretch-shortening cycles led to a steady-state elevation of ROS production. Importantly, this new redox state was graded by both the amplitude of stretch (3–15%) and cycle frequency (1–4 Hz). Elevated ROS production enhanced Ca2+ signalling sensitivity as measured by the Ca2+ spark rate.

Conclusion The steady-state level of ROS production in a cardiomyocyte is graded by the amplitude and frequency of cell stretch. Thus, mechanical changes that depend on the pre-load and heart rate regulate a dynamic redox balance that tunes cellular Ca2+ signalling.

  • Stretch
  • ROS
  • NADPH oxidase
  • Calcium sparks
  • Nox2

1. Introduction

X-ROS is the novel signalling pathway seen in ventricular myocytes1 and in skeletal muscle2 that arises when cell stretch triggers local ROS (reactive oxygen species) production by membrane-located NADPH oxidase 2 (Nox2). A single, sustained (multi-second) stretch produces a transient rise in ROS with a half-time of decay ∼5 s1 (Figure 1). In cardiomyocytes, this ROS burst sensitizes local Ca2+ release channels (ryanodine receptors, RyR2) in the sarcoplasmic reticulum (SR). In normal heart cells, this produces a transient burst of elementary Ca2+ release events known as Ca2+ sparks3,4 and increases the fidelity of excitation–contraction (EC) coupling.5 In cardiomyocytes from hearts with specific pathologies, X-ROS can trigger arrhythmogenic, regenerative Ca2+ release events known as Ca2+ waves.1,6

In reports to date X-ROS signalling has been characterized only in cells that were given a single stretch and held at a constant length.1,2,7 However, in vivo cardiomyocytes are stretched and shorten cyclically with each heartbeat. The amplitude of the cyclic stretch is dependent on the pre-load filling pressure of the heart, which regulates the extent of diastolic stretch. The frequency of cyclic stretch is of course determined by the heart rate. Here, we use repetitive stretch paradigms to mimic the cardiac cycle and we investigate how the amplitude and frequency of cell stretch affect X-ROS signalling by examining ROS production and Ca2+ sparks. Our results reveal that under this physiological length cycling, X-ROS signalling is magnified and has a sustained effect on Ca2+ signalling. Cyclic stretch elevates the steady-state level of ROS production in the cell, and this level is graded by both the amplitude and frequency of stretch. In turn, the elevated ROS proportionately modulates RyR2 Ca2+ release channels, manifest as an increase in the Ca2+ spark rate. Thus, our findings hold the implication that the mechanical changes graded by both pre-load and heart rate affect a dynamic redox state in the heart cell that tunes cardiac Ca2+ signalling.

2. Methods

2.1 Rodent models

Animal care and procedures were approved and performed in accordance with the standards set forth by the University of Maryland, Baltimore. Institutional Animal Care and Use Committee and the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication, 8th Edition, 2011).

2.2 Materials

Diphenyleneiodonium (DPI) and H2 dichlorofluorescein (DCF) diacetate were purchased from Sigma-Aldrich (St Louis, MO, USA). Gp91ds-tat was purchased from Anaspec (Fremont, CA, USA). DPI was used at 3 μmol/L and gp91ds-tat at 1 μmol/L.

2.3 Cardiomyocyte isolation

Adult, male, Sprague–Dawley rats (8–16 weeks) were terminally anaesthetized by the injection of pentobarbital (200 mg/kg), followed by the excision of the heart and enzymatic isolation of ventricular myocytes as previously described8. Cardiomyocytes were stored in a normal Tyrode's solution containing (in mmol/L): NaCl 140, KCl 5, CaCl2 1.8, MgCl2 0.5, HEPES 5, Glucose 5, NaH2PO4 0.33. Experiments were performed at room temperature, 22°C.

2.4 Cardiomyocyte attachment and stretch

All experiments were performed in custom-fabricated cell chambers (Four-hour Day Foundation, Towson, MD, USA) mounted on an LSM 510 inverted confocal microscope (Carl Zeiss, Jena, Germany) with a 40x Oil 1.2NA objective. Cell stretch was performed as previously described1. Briefly, glass micro-rods were coated with a biological adhesive, MyoTak™ [World Precision Instruments (WPI), Sarasota FL, USA; Ionoptix, Milton, MA, USA]. One glass micro-rod was connected to a force transducer (KG7, WPI), and the other to a length controller (WPI). Myocytes were attached at both ends by gently pressing down with the MyoTak-coated micro-rod and then lifting the cell from the chamber bottom. Axial stretch was applied by movement of the length controller in response to variable voltage output. The stretch waveform was rounded by an 8-pole Bessel filter (time to peak stretch was 100 ms for all stretch paradigms) to mimic diastolic filling. For cyclic stretch protocols, the diastolic:systolic ratio was 1.5 at all frequencies of stretch (e.g. at 2 Hz cyclic stretch the cell was stretched for 300 ms and shortened for 200 ms each cycle). The sarcomere length was monitored with a high-speed video camera (Aurora Scientific, Ontario, Canada). An average sarcomere length prior to stretch was 1.8 μm. Supplementary material online, Videos S1 and S2 depict static and cyclic stretch protocols. Myocytes were typically subjected to three to four stretch-release trials, with 30–60 s rest allowed between trials. As each stretch produced a similar change in the sarcomere length, DCF fluorescence, and Ca2+ spark rate, trials were pooled for the analysis. Cells did not commonly display Ca2+ waves, and cells with Ca2+ waves were discarded from the analysis. Similar sarcomere length changes are achieved and maintained with static, 1 Hz, and 4 Hz cyclic stretch (Supplementary material online, Videos S1 and S2). Cells showed no signs of damage or disturbed integrity of cell membranes in response to any of the stretch paradigms in this study (See Supplementary material online, Videos).

2.5 Ca2+ spark measurements

Cells were loaded with Fluo-4 by 10 min incubation with 3 µmol/L of fluo-4-acetoxymethyl (AM) ester (Invitrogen) and 0.01% Pluronic® F127 (a poloxamer made by BASF, Florham Park, NJ, USA) and allowed an additional 10 min for de-esterification. Cells were scanned using a 488 nm argon ion laser in a confocal line-scan mode at 1.92 ms/line. Automated analysis of line-scan images for Ca2+ spark location and properties was performed using custom routines written in Interactive Data Language (IDL version 8.1, ITT Visual Information Solutions, Boulder, CO, USA). Contact C.W.W. for more information.

2.6 DCF measurements

Cells were loaded with the cell permeant ROS indicator DCF by a 20 min exposure to a Tyrode's solution containing 2.5 µmol/L of H2DCFDA. The cells were imaged using a confocal line-scanning microscopy at 3.07 ms/line. DCF has limitations as an indicator9 and can produce artefactual signal amplification upon continuous light exposure. Therefore, cells were imaged at very low laser intensity and a sampling rest interval of 6 ms was introduced between each line scan. Additionally, the scan line was moved 2 μm on each trial to avoid any damage to the cell. The slope of the DCF signal is indicative of ROS production. In Figures 1, 2, and 4, the slope prior to stretch has been subtracted from the entire trace so that the initial slope is zero and changes in slope can be easily visualized (as described in 1,2). In the experiments in Figures 14, the slope is expressed normalized to the slope prior to stretch in that same cell (termed ‘rest’), as this comparison avoids differences in DCF loading and resting signal between cells. As the slope was linear for the duration of cyclic stretch (see Figures 1, 2 and 4), the slope was calculated from the entire stretch duration for all cyclic stretch experiments. For measurements of resting (non-stretched) DCF fluorescence in separate experiments (Supplementary material online, Figure S1), cells were plated in 96-well plates, loaded with DCF as above, and imaged on an Olympus inverted microscope (IX-50; 20× air objective) equipped for wide-field fluorescence and CCD imaging. This technique allowed the imaging of a larger number of cells simultaneously to determine resting DCF fluorescence (as opposed to stretch experiments, where only a single cell can be assayed). A 45 s imaging epoch sampled at 1 Hz was used to assay the rate of ROS production in quiescent, non-stretched cardiomyocytes. A minimum of six trials (multiple cells averaged per trial) from a minimum of two hearts were imaged per condition (ctrl, DPI, gp91ds-TAT, scramble-TAT). All methods have been previously described.10

Figure 1

ROS production during sustained vs. cyclic stretch. (A) Length change during a sustained (red) and cyclic stretch (blue). (B) Average DCF traces from the protocol in (A). The sustained stretch (red) is fit by a polynomial function (r2 = 0.96, n = 8 cells), while the cylic stretch (blue) is fit by a linear function (r2 = 0.98, n = 5 cells). (C) Derivative traces calculated from the fits to average data in (B) show the rates of ROS production during static and cyclic stretch.

Figure 2

Cyclic stretch sustainably elevates Nox2-ROS production in a frequency-dependent manner. (A) Top—prolonged cyclic stretch protocol. Bottom—resulting DCF traces in cells cyclically stretched (n = 12), not stretched (n = 12), treated with DPI (n = 7) or gp91ds (n = 8). (B) The quantification of DCF slopes from experiment in (A), normalized to the pre-stretch slope. Linear fits were applied to the four different time intervals over the protocol, and the slope was calculated for each interval. (C) Top—frequency-dependent cyclic stretch protocol. Cells were cyclically stretched at 1 Hz (black inset) or 4 Hz (red inset). Bottom—DCF traces for cells held at resting length (n = 13), cyclically stretched at 1 Hz (n = 13) 2 Hz (not shown, n = 14) or 4 Hz (n = 13). (D) Quantification of DCF slopes from experiment depicted in (C). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 3

Contraction vs. stretch-induced ROS production. (A) Representative steady-state contractions in an isolated, unloaded cardiomyocyte stimulated at 1 Hz. The change in the sarcomere length depicted is ∼15% (average ΔL = 14.1%). (B) The quantification of DCF slopes in cells at rest and during contraction from 1 Hz stimulation (n = 14 cells). Gp91ds (1μm) blocks the contraction-induced ROS (n = 12 cells). (C) The quantification of DCF slopes in cells at rest and during 1 Hz cyclic stretch to 3% (n = 9), 6% (n = 12), 10% (n = 12), and 15% (n = 15) of the cell length. *P < 0.05, **P < 0.01.

Figure 4

X-ROS is graded by the amplitude of cell stretch. (A) Top—length-dependent cyclic stretch protocol. Bottom—average DCF traces and linear fits from cells stretched to 3, 6, and 10% of the cell length. (B) The quantification of DCF slopes from 2 Hz cyclic stretch to 0% (n = 14), 3% (n = 9), 6% (n = 16), 10% (n = 14), and 15% (n = 16) of the cell length. *P < 0.05, **P < 0.01.

2.7 Statistics

All the data are presented as mean value ± standard error of the means (SEM) unless otherwise noted. Normality of data sets was ensured for each statistical calculation. All significance tests were done using Student's t-test, and significance was set at P < 0.05.

3. Results

Figure 1 depicts the fundamental difference in ROS production that occurs when a heart cell is cyclically stretched when compared with being stretched and held at a constant length (i.e. ‘static’ stretch, see Supplementary material online, Video S1). Static stretch (red) produces a transient increase in the DCF signal slope (Figure 1B), corresponding to a transient increase in ROS production (Figure 1C). ROS production peaks within seconds of stretch and then declines with a half-time of the decay of 4–5 s, consistent with earlier observations1 (Figure 1C). In contrast, a 1 Hz cyclic stretch (blue) produces a similar increase in the DCF slope, but the slope remains elevated for the duration of cyclic stretch (Figure 1B). This corresponds to an elevated level of ROS production that is sustained during cyclic cardiomyocyte stretch (Figure 1C).

If cyclic stretch indeed causes a steady-state change in ROS production then the increase in the DCF slope should be maintained for minutes, not just seconds. Thus, we cyclically stretched cells for 3 min and monitored the resulting DCF signals before, during, and after stretch (Figure 2A). As quantified in Figure 2B, 1 Hz cyclic stretch 8% of the cell length causes a ∼1.5-fold increase in the DCF slope over the first 90 s of cyclic stretch (orange bars), and this ∼1.5-fold increase is maintained over the last 90 s of cyclic stretch (red bars). Approximately 20 s after the release of stretch, the DCF slope returns to the pre-stretch level for the final 40 s of imaging (blue bars). Importantly, the stretch-induced ROS production is blocked by the Nox inhibitor DPI and the highly specific Nox2 inhibitory peptide gp91ds-tat11 (Figure 2A and B), verifying that Nox2 is the source of ROS elicited by cyclic stretch1,2. Neither DPI nor gp91ds-tat had any a significant effect on basal ROS production in non-stretched cells in control experiments (Supplementary material online, Figure S1). In time control experiments, there are no significant changes in the DCF slope in non-stretched cells over the same duration (Figure 2A and B, non-stretched).

The effect of the stretch frequency (i.e. heart rate) on ROS generation is shown in Figure 2C and D (see also Supplementary material online, Video S2). From resting length, cells were cyclically stretched to 10% of cell length at 1, 2, or 4 Hz for 60 s and the resulting DCF signals were recorded. Increasing the frequency of stretch to 2 or 4 Hz causes a significant increase in ROS production when compared with 1 Hz stretch. There was no significant difference found between 2 and 4 Hz. Thus, there is a steep graded increase in the rate of ROS production with the heart rate.

In the experiments described in Figures 1 and 2, quiescent (non-stimulated) cells were stretched in order to isolate the effects of diastolic stretch on ROS production. Next, we investigated the effect of contraction-induced shortening and relaxation on ROS production by field stimulating isolated, unloaded cardiomyocytes. DCF fluorescence was imaged for 1 min in the quiescent cell to establish the resting rate of ROS production, and then the cell was electrically paced at 1 Hz for 2 min while DCF fluorescence was monitored. For the last 5 s of the imaging protocol, the sarcomere length change was quantified (Figure 3A). The average resting sarcomere length was 1.79 ± 0.01 μm, and peak shortening occurred at 1.54 ± 0.02 μm (n = 14 cells). The average sarcomere length change of 0.25 ± 0.01 μm was equivalent to a 14.1% change in the sarcomere length. Cell stimulation alone (in unloaded, contracting myocytes) caused a small, but a significant 1.3-fold increase in ROS production in control cells (treated with scramble-TAT peptide) (Figure 3B). Treatment with the Nox2 inhibitor gp91ds blocked this small contraction-induced ROS increase (Figure 3B), while causing no significant change in resting ROS production (Supplementary material online, Figure S1). This finding suggests that contraction-induced shortening and relaxation alone is sufficient to generate some Nox2-ROS production. The result is also consistent with a recent report in the skeletal muscle that demonstrates contraction-induced ROS in the cytosol arises from Nox2 (as opposed to mitochondrial sources).12

To compare contraction-induced vs. stretch-induced ROS, we next examined the effect of an increase in the diastolic length on ROS production in quiescent, cyclically stretched myocytes. Following 1 min of imaging to establish resting ROS production, cells were stretched at 1 Hz to 3, 6, 10, and 15% of the cell length while monitoring ROS production (Figure 3C). A 3% change in the cell length produced an insignificant change in ROS production, whereas a 6% length change produced a significant increase in ROS. ROS production further increased at 10 and 15% cell stretch. By comparison to the 1.3-fold change in ROS production caused by a ∼15% contraction-induced shortening and relaxation (Figure 3B), a 15% stretch caused a 2.3-fold increase in ROS production (Figure 3C). These results show how the steady-state rate of ROS production in the heart cell is graded by the amplitude of diastolic stretch, or pre-load.

As demonstrated in Figure 3C, at 1 Hz cyclic stretch ROS production continued to increase with the amplitude of stretch up to a 15% cell stretch, which is near the end of the physiological sarcomere length range.13 Twenty per cent cell stretch was not pursued as such a large stretch can cause membrane damage to the myocyte and confound results (Prosser, unpublished observation,14). To see if ROS production could be further increased, we examined the length-dependence of ROS production at 2 Hz cyclic stretch. Cells were stretched at 2 Hz to 3, 6, 10, and 15% of the cell length (or held at resting length, i.e. 0% stretch) while monitoring ROS production (protocol and average traces are shown in Figure 4A). Consistent with Figure 2D, ROS production was increased at all cell lengths at 2 Hz compared with 1 Hz cyclic stretch. At 2 Hz cyclic stretch, a 3% change in the cell length produces a small, but a significant increase in ROS production, whereas a 6% length change produces a significantly larger increase in ROS. ROS production peaks at 10% cell stretch, and is not further increased when a 15% stretch is imposed (Figure 4B).

Given the fundamental differences in ROS production during static vs. cyclic cell stretch, we explored how this may differentially affect RyR2 Ca2+ signals. Cells were held at resting length for 10 s, and then either statically or cyclically stretched for 60 s before returning the cells to resting length (Figure 5). Ca2+ sparks were imaged during the first 10 s at resting length (pre-stretch) and for the first 10 s of stretch (initial stretch); imaging was then halted for 40 s of sustained stretch to avoid photodamage to the cell. Imaging was then resumed for the last 10 s of stretch (sustained stretch) and a further 10 s when the cell was returned to resting length (post-stretch).

Figure 5

Static cell stretch transiently elevates the Ca2+ spark rate. (A) The surface plot profile of Ca2+ sparks in a cell statically stretched. In yellow is the stretch protocol and time intervals. (B) Ca2+ spark histogram from experiment depicted in (A) (n = 16 cells). (C) The quantification of the Ca2+ spark rate from different intervals in (A). *P < 0.05 compared with the pre-stretch value.

Static stretch initially causes a rapid increase in the spark rate (Figure 5A top, Figure 5B and C red bars), which is dependent upon Nox2-ROS production as previously described.1 Consistent with the transient nature of this ROS production (Figure 1C), the spark rate returns back to the pre-stretch value during sustained static stretch, and is not further altered following the release of stretch (Figure 5A bottom, Figure 5B and C blue and purple bars).

Cyclic stretch has a dramatically different effect on the spark rate. It initially causes a similar rapid increase in the spark rate as static stretch (Figure 6A top, Figure 6B and C red bars)—however, upon sustained cyclic stretch, the spark rate remains elevated (or even slightly increased) and stays elevated (or slightly increased) upon the release of stretch (Figure 6A bottom, Figure 6B and C blue and purple bars). One minute following the release of stretch, the spark rate has returned to the pre-stretch level (Figure 6C green bar), suggesting that this process is reversible and is not indicative of cell damage. Importantly, treatment with the Nox2 inhibitor gp91ds abolished both the acute and sustained changes in the spark rate seen with cyclic stretch (Figure 6D and E). These findings are consistent with elevated steady-state Nox2-ROS production increasing RyR2 Ca2+ signalling sensitivity during cyclic stretch.

Figure 6

Cyclic cell stretch sustainably elevates the Ca2+ spark rate. (A) The surface plot profile of Ca2+ sparks in a cell cyclically stretched at 2 Hz. In yellow is the stretch protocol and time intervals. (B) Ca2+ spark histogram from experiment in (A) (n = 16 cells). (C) The quantification of the Ca2+ spark rate at different imaging intervals in (B and D) Ca2+ spark histogram from same experimental paradigm depicted in A in cells pre-treated with gp91ds (1 μM, n = 9). (E) The quantification of the Ca2+ spark rate at different imaging intervals in (D). *P < 0.05 compared with the pre-stretch value.

4. Discussion

Here, we show that ROS generation and Ca2+ signalling sensitivity in the heart cell are tuned by the amplitude and frequency of diastolic stretch. These findings suggest that the redox-state of the heart cell may be graded through X-ROS signalling during normal physiological cycling. Importantly, cellular Ca2+ signalling tracks these acute or chronic mechanical changes in the myocyte that occur during physiological or pathophysiological cardiac stress. ROS are produced by Nox2 in the transverse-tubule and sarcolemmal membranes1,15,16 and act locally to reversibly increase the Ca2+ signalling sensitivity of RyR2s. The stretch-induced sensitization of RyR2s helps to ensure their reliable triggering by Ca2+ influx (ICa),1 thus promoting the fidelity of EC coupling when heart rate and pre-load are elevated. This may prove particularly important when ICa is reduced1 or EC coupling is impaired, such as during T-tubule remodelling in cardiac hypertrophy or failure.17

Our work is consistent with the findings of Sanchez et al.,15 who demonstrated that exercise and tachycardia (both of which increase the heart rate) increased Nox activity and SR calcium release rates through oxidative modification of RyR2s. Frequency-dependent X-ROS provides a mechanism for elevated Nox2 activity during exercise, which importantly provides a powerful preconditioning effect against ischaemia-reperfusion damage.

In our experiments, unloaded contraction and relaxation from a slack sarcomere length was sufficient to induce a low level of Nox2-ROS production (Figure 3B), consistent with a recent report in the skeletal muscle.12 However, stretching a myocyte from the slack length at a comparable frequency produced a larger increase in ROS production that was graded by the magnitude of stretch (Figure 3C). Taken together, these findings suggest that in the beating heart cell there is likely some basal Nox2-ROS production arising during shortening and relaxation alone, which can then be dynamically modulated by the amplitude and frequency of applied stretch. The total Nox2-ROS production may arise from a single mechanism dependent on restoring forces and cell stretch, or multiple mechanisms that act in concert in a beating myocyte.

Unlike other ROS sources, Nox2 produces ROS in a highly regulated manner, ideal for a role in cell signalling.18,19 During a single static stretch, the rapid enhancement of SR Ca2+ release is attributed solely to local Nox2-ROS production, and not to other potentially confounding sources such as mechanosensitive channels, mitochondrial ROS production or nitric oxide signalling.1,7 However, during sustained cyclic stretch, when ROS production is elevated for minutes or more, there is likely combinatorial activation of additional redox-sensitive signalling cascades. Nitric oxide may also be generated and can interact with Nox2 ROS to produce reactive nitrogen species that may directly or indirectly increase RyR2 activity.2022 Oxidation of Ca2+/calmodulin-dependent protein Kinase II (CaMKII) can increase its activity and stimulate RyR2 through phosphorylation,23,24 and oxidation of calmodulin itself can remove its tonic inhibitory action on RyR2.25 Furthermore, key signalling (e.g. ERK and JNK) and EC coupling proteins (e.g. Na+/Ca2+ exchanger, the L-type Ca2+ channel, and SERCA pump) are all sensitive to oxidation (reviewed in 19,26). In general, a small controlled amount of ROS increases the activity of exchangers, channels and transporters, which would enhance cardiac function during acute stress when heart rate and pre-load are elevated. Our work does not rule out potential influences of these pathways on the sustained increase in Ca2+ signalling sensitivity with cyclic stretch. On the contrary, it is likely that a sustained increase in ROS will affect one or more of these targets. However, the finding that gp91ds completely blocks cyclic stretch-induced ROS production and calcium sparks implies that X-ROS is a necessary upstream driver of these redox-sensitive pathways. While it is difficult to predict which pathways may be affected by X-ROS (particularly given its local signalling nature and the complex time and concentration-dependent effects of ROS27), it will be important to consider X-ROS as a potential initiating stimulus of redox-sensitive pathways activated under physiological and pathological cardiac stress. Potential downstream targets and their patho/physiological ramifications are discussed in detail in a recent review.9

A limitation of our study is that the experiments were conducted at room temperature and at stretch/stimulation rates typically used and compatible with isolated myocyte experiments. These rates are slowed compared with the rodent heart rate in vivo. Therefore, effects of changes in heart rate and pre-load can be evaluated qualitatively, but extrapolation to the in vivo setting will require further the evaluation of X-ROS signalling in the working heart in rodents and larger animal models.

In conclusion, we demonstrate for the first time that rhythmic stretch-shortening cycles elevate steady-state Nox2-dependent ROS production in cardiomyocytes. This sustained level of ROS is graded by the amplitude (pre-load) and frequency of stretch (heart rate). X-ROS reversibly increases the Ca2+ spark rate of RyR2s, thus coupling ROS production and Ca2+ signalling to the mechanical changes set by heart rate and pre-load. During acute physiologic stress, ROS production may be elevated to enhance cardiac function, while under chronic stress, this may contribute to oxidative damage and its pathological consequences.


B.L.P. is supported by a National Institutes of Health (NIH) training grant (T32 HL072751-07) in Cardiovascular Cell Biology and by (5K99HL114879-02) from National Heart Lung and Blood Institute (NHLBI) at NIH. Additional support from NIH (R01 HL106059, R01 HL36974); and Leducq North American-European Atrial Fibrillation Research Alliance; European Union Seventh Framework Program (FP7), Georg August University, ‘Identification and therapeutic targeting of common arrhythmia trigger mechanisms’.


We thank WPI and Ionoptix for supporting instrumentation and equipment development.

Conflict of interest: B.L.P., C.W.W., and W.J.L. have filed a university sponsored US patent for MyoTak.


View Abstract