The amplitude of the calcium transient determines the level of activation of the myofilaments and therefore the intensity of contraction and it is finely modulated. The two main factors that determine the amount of calcium released from the SR and the amplitude of the calcium transient are: the amplitude of the L-type calcium current (Trafford et al., 2001) and the calcium concentration in the SR (Shannon et al., 2000). The SR Ca concentration can be increased by stimulation of SERCA activity and prolongation of the duration of the AP and decreased by stimulation of NCX activity.

In vivo the main modulator of the calcium transient amplitude is β adrenergic stimulation that via cAMP-mediated activation of protein kinase A stimulates both the L-type calcium channel and SERCA. This leads to an increase in L-type calcium current and SR Ca content and therefore a substantial increase in calcium transient amplitude (Hussain and Orchard, 1997).

Release of calcium from the SR can also occur independently from an AP during diastole (Venetucci et al., 2008). This process is called spontaneous calcium release (SCR) and occurs when the SR calcium concentration reaches a crucial threshold level (Diaz et al., 1997). In addition recently Belevych et al. (2012) elegantly demonstrated that soon after a systolic calcium transient the RyRs enter a refractory state and calcium wave occur only once the RyR have recovered from this refractory state even if the SR threshold is reached. Therefore the two conditions necessary for the generation of SCR are: (1) Increased SR Ca content up to the SR threshold for SCR (2) Recovery of RyR from refractory state. SCR activates the NCX that generates an inward current and a DAD. When a DAD reaches the threshold for activation of the Na channels, it initiates an AP which in turn causes TA and arrhythmias. The process of SCR and generation of DADs is shown in Figure 1B. SCR, DAD, and TA are responsible for the onset of arrhythmias in various clinical settings including digitalis-induced arrhythmias, some forms of heart failure and catecholaminergic polymorphic VT (a genetic arrhythmia syndrome caused by mutations of the RyR and CASQ2 genes). The scheme illustrated above suggests that there are two main therapeutic strategies that can be used to prevent the onset of calcium handling related arrhythmias: prevention of the onset of an SCR and prevention of triggering of an AP by a DAD. In view of the fact that the generation of DAD through SCR has been shown to be dependent on threshold SR calcium content (Diaz et al., 1997; Trafford et al., 2000; Jiang et al., 2005), the first strategy can be achieved by preventing the SR calcium content from reaching the threshold for SCR either by decreasing SR calcium content (Venetucci et al., 2007; Llach et al., 2011) or by raising SR threshold for SCR via inhibition of RyR (Venetucci et al., 2006; Maxwell et al., 2012). The second strategy can be achieved via inhibition of sodium channels that reduces the number of sodium channels ready to be activated when a DAD occurs and therefore reduces the likelihood that the DAD will initiate and AP and cause TA. Whilst this strategy can be beneficial in preventing arrhythmias caused by TA such as in heart failure, it could also be pro-arrhythmic during acute ischaemia by facilitating re-entry. In the following section, we will illustrate the experimental evidence that suggests that PUFA prevent SCR, DADs and TA via some of the mechanisms described.

In 2001 Leifert et al. (2001) gave a DHA and EPA enriched diet to adult rats for 3 weeks. This diet increased the levels of incorporation of DHA (from 6 to 20% of total phospholipids) and EPA (from 0 to 3.2%). The incorporated PUFA did not affect calcium transient amplitude but decreased its rate of decay. In addition, incorporation of PUFA did not affect SR Ca content. The authors suggested that the reduction in calcium transient rate of decay was due to reduced NCX-mediated removal in the PUFA group. It is likely that the effects are more complex because isolated reduction of NCX function would have also increased SR calcium content. The Coronel group studied the effects of a PUFA-enriched diet on pig ventricular myocytes. The diets were given for 8 weeks and produced a significant increase in the levels of incorporated DHA (from not detectable to around 7%) and EPA (from not detectable to around 15%). The most striking effect of incorporated PUFA was reduction in AP duration which was more prominent at slow frequencies and was mainly due to increase in two repolarizing currents: the slow component of the delayed rectifier current and inward rectifier K current (Den Ruijter et al., 2006; Verkerk et al., 2006). The analysis of calcium handling showed that the incorporated PUFA did not produce a significant effect on calcium transient amplitude but accelerated its rate of decay. SR calcium content was not affected but both L-type calcium current amplitude and NCX function were significantly reduced. The acceleration of the calcium transient rate of decay is probably explained by the shorter AP duration, but what is surprising is the fact that despite reduction in L-type calcium current amplitude, the calcium transient amplitude itself was not affected. Recently Billman et al. (2010) assessed the effects of DHA and EPA supplements (Omacor 1–2 or 4 tablets a day) on contractility and calcium transient in mixed breed dogs. The tablets were given for 3 months. The supplements produced a significant increase in EPA and DHA incorporation levels in cardiac membranes. However, the increased incorporation levels did not produce any significant effect on cardiac contractility, calcium transient amplitude and on L-type calcium current amplitude.

Leifert et al. (2001) reported that incorporation of DHA and EPA in membranes of rat ventricular myocytes reduced the incidence of calcium waves and DADs during challenge with the β agonist isoproterenol. From the data illustrated by the authors, it is difficult to understand the mechanisms responsible for these protective effects. The authors also showed that the incorporation of DHA and EPA does not affect the increase in calcium transient amplitude produced by isoproterenol. Similarly to what described at baseline there was a reduction in the rate of decay of the calcium transient that was significant only at 0.2 Hz. From these data is difficult to establish how incorporated PUFA reduced the incidence of calcium waves. It is not clear whether PUFA affected SR threshold for calcium waves and/or SR calcium content. Similar to what was previously described by Leifert et al. in rat cardiac myocytes, the Coronel group demonstrated that incorporation of DHA and EPA in the membrane of pig ventricular myocytes reduced the incidence of calcium waves and DADs during challenge with norepinephrine (Berecki et al., 2007). They investigated in detail the mechanism responsible for these effects and demonstrated that membrane incorporation of PUFA reduces the response to norepinephrine. In particular, incorporated PUFA blunted the increase in SR calcium content produced by norepinephrine and therefore attenuated the increase in calcium transient produced by norepinephrine. Incorporated PUFA also prevented the prolongation of AP potential produced by norepinephrine and it is unclear whether the blunting in the increase of SR calcium content is simply due to the effects on AP duration or also involves blunting of the stimulation of SERCA activity. In a recent paper, Billman et al. showed that incorporation of DHA and EPA in cardiac membranes of dogs did not affect the response to isoproterenol. The authors however did not investigate whether incorporated PUFA prevented the onset of calcium waves and DADs. However the finding that they do not attenuate response to isoproterenol would suggest that in these experiments they would not prevent DADs. An important issue that was not investigated by all these studies is whether incorporated PUFA have any effect on RyR and affect the threshold for SCR. Some studies inferred that because calcium transient amplitude was not affected there was no effect on RyR.

In summary, the studies that have investigated the effects of incorporated PUFA on Ca handling have reached conflicting conclusions. The cause of these differences remains unclear. Billman et al. have suggested that they may be related to the different species utilized in the studies. More specifically pigs (utilized by the Coronel Group) have a calcium-dependent transient outward current that is absent in dog and human myocytes. This could cause different effects of incorporated PUFA on AP duration and therefore SR calcium content before and after adrenergic stimulation. However this does not explain the differences in the effects of incorporated PUFA on the L-type Ca current. In addition, the pig study that documented attenuated adrenergic response used norepinephrine (β1 and α1 agonist) to produce adrenergic stimulation while the other two studies that documented limited effects utilized isoproterenol (β1 and β2 agonist). This raises the possibility that stimulation with norepinephrine is more susceptible to modulation by incorporated PUFA. This is a possibility that needs to be investigated.

The large amount of experimental evidence gained with in vitro experiments has not been supported by in vivo experiments. Several studies have demonstrated that in vivo both infusion of PUFA (free PUFA) (Billman et al., 1999) and chronic dietary supplementation (that leads to membrane incorporation and increased circulating free PUFA) protect from ischaemia-reperfusion related arrhythmias (McLennan et al., 1988; London et al., 2007). However a recent study evaluating the effects of dietary n-3 PUFA on susceptibility to post-myocardial infarction ventricular fibrillation in dogs, showed that despite significant increases in circulating as well as left ventricular PUFA levels, PUFA not only failed to prevent ischaemia-induced VF. Contrary to expectations dietary PUFA exerted pro-arrhythmic effects facilitating the onset of VF both in non-infarcted animals and in low-risk post-MI dogs that did not have VF prior to initiation of PUFA diet (Billman et al., 2012). These conflicting results highlight the fact that the pathogenesis of ischaemia reperfusion related arrhythmias is complex and certainly does not involve just alterations in calcium handling (Janse and Wit, 1989). Therefore this experimental evidence cannot be used as proof that PUFA modulate calcium handling in vivo. The evidence that PUFA in vivo modulate myocardial calcium handling is limited. This question could be answered by studies on cardiac contractility and on calcium handling related arrhythmias. To date no study has assessed the effects of acute PUFA administration on cardiac function and the two studies that have assessed the effects of PUFA diet supplementation on cardiac function have not confirmed the in vitro findings. In 1992, McLennan et al. (1992) demonstrated that dietary supplementation with PUFA in marmosets increased ejection fraction by enhancing LV filling. Recently Billman et al. (2010) studied the effects of PUFA supplementation on LV function (assessed by echo) and demonstrated that despite significant increases in the incorporation of DHA and EPA in the cardiac membrane (and probably in circulating free PUFA) there was no change in LV function. Interestingly these results are very different from what one would expect on the basis of in vitro studies that point more towards a reduction in calcium transient amplitude, cardiac contractility and LV function. These data suggest that in vivo there are complex modulating factors that mitigate or abolish the effects of PUFA observed in vitro. The evidence that PUFA prevent arrhythmias which are exclusively due to abnormalities in calcium handling, calcium waves and DADs (such as catecholaminergic and digoxin-related arrhythmias) is also limited. Only one study (Gudbjarnason et al., 1989) demonstrated that in rats a PUFA diet prevented the onset of VF following isoproterenol infusion. There is clearly a need for more targeted studies that specifically assess the anti-arrhythmic potential of PUFA in arrhythmia syndromes caused exclusively by abnormalities in calcium handling. To this purpose, a study in animal models and or patients with CPVT would address this point.

Dr. Sankaranarayanan's research is funded by a grant from the British Heart Foundation (Grant Reference FS/11/15/28693). Dr. Venetucci's research is funded by a grant from the British Heart Foundation (Grant Reference FS/10/63/28374).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.