Jump to
While sympathetic stimulation of the heart produces chronotropic, inotropic, and lusitropic effects, increased frequency alone causes a positive force-frequency relationship (FFR) and frequency-dependent acceleration of relaxation (FDAR).1 That is, contraction amplitude and relaxation rate are increased with increasing frequency in most species (including humans). The key mechanism involved in the positive FFR is increased sarcoplasmic reticulum (SR) Ca2+ load, due to increased Ca2+ influx and decreased Ca2+ efflux.1,2 Ca2+ influx increases due to more L-type Ca2+ current (ICa) per unit time, while Ca2+ efflux via Na+-Ca2+ exchange (NCX) decreases because the diastolic time is reduced and [Na+]i increases. Enhanced SR Ca2+-pump function causes FDAR and also augments SR Ca2+ loading. Various signaling pathways are involved (eg, CaMKII).3
In human heart failure, the FFR reverses (ie, from positive to negative) due to an inability of the SR to increase Ca2+ content.4 This negative FFR is a main contributor to the loss of contractile reserve in the failing heart. Of many pathways that can modify FFR, nitric oxide (NO) signaling is the topic addressed by Khan et al in this issue of Circulation Research.5
Nitric Oxide and Cardiac Function
NO synthase (NOS) produces NO from l-arginine, and cardiac myocytes express all three NOS isoforms.6,7 NOS1 (nNOS) and NOS3 (eNOS) are constitutively expressed and produce low amounts of NO (regulated by [Ca2+-calmodulin]i levels). NOS2 (iNOS) is expressed during inflammatory responses (eg, cytokines, sepsis, heart failure) and continuously produces large amounts of NO (compared with NOS1 and NOS3),7 independent of Ca2+.
NO can increase or decrease contractility.8 Many factors modulate NO functional effects: different NO concentrations lead to cGMP-dependent or -independent signaling,9,10 adrenergic state,11 NO species,12 NO effects on targets involved in excitation-contraction coupling (EC coupling) and NOS isoforms.6 These pathways work in concert to determine the net effect of NO. Different NO species (eg, NO· or NO+) activate different pathways. For example, NO· activates cGMP-dependent protein kinase (PKG) and NO+ can have direct effects via nitrosylation, and these can differentially modulate targets (eg, PKG phosphorylation decreases ICa and channel nitrosylation increases ICa).13
Recently, Hare’s group provided provocative evidence that the effects of NO (positive or negative) are due to different isoforms of NOS (1 and 3) and their cellular localization.6 NOS1 is localized to the SR and coimmunoprecipitates (Co-IPs) with ryanodine receptors (RyRs), and NOS3 is localized to caveolae and Co-IPs with caveolin-3. Thus, we should envision NO signaling (Figure) as locally controlled (similar to β-adrenergic [β-AR] signaling).14
NOS signaling in cardiac myocytes. ATP indicates SR Ca2+-ATPase; NOx, different congeners of NO· (see text for other abbreviations).
For NOS2, the situation may differ. Since NOS2 is not localized (cytosolic) and produces large amounts of NO, NOS2 signaling is probably less “compartmentalized” and can affect various EC coupling related proteins/processes (eg, ICa, RyRs, and energetics).7,15,16 These effects can contribute to cardiac dysfunction in various diseases where myocyte NOS2 is expressed (eg, rejection of transplanted hearts).17
NOS3 seems to depress contractility.6,18 Since NOS3 is localized to caveolae and can inhibit ICa,19,20 it was hypothesized that NO produced by NOS3 selectively regulates ICa, but not RyRs.6 However, since most of the Ca2+ channels thought to be important in EC coupling are at dyadic sarcolemmal-SR junctions, the impact of putative caveolar NOS3-ICa signaling on EC coupling is not completely clear.
NOS1 also regulates cardiac myocyte function.5,6,21,22 In NOS1 knockout mice, basal contractility and β-AR responsiveness were depressed.6 Since NOS1 is localized to the SR (and Co-IPs with RyRs), Hare’s group hypothesized that NO produced via NOS1 stimulates RyRs and contractility without altering ICa. Indeed, exogenous NO can directly regulate RyR activity (via nitrosylation) increasing channel open probability,23 although NOS1 was not directly implicated. A complicating aspect here is that RyRs are very closely colocalized with L-type Ca2+ channels, probably 0 to 20 nm apart.1 Given the rapid diffusion of NO, it is difficult to envision how NO can affect RyRs without also altering ICa. Also, using the same NOS1 knockout mice, Casadei’s group21,22 found opposite results (increased basal contractility and β-AR responsiveness) and a flat FFR and increased SR Ca2+ load. Discrepancies between these groups may be technical, such as temperature (37°C versus 22°C), frequency, degree of β-AR stimulation (high versus low),11 or age (or hypertrophic stage) at which animals were studied. Thus, while basal NOS1 activity may be positively inotropic (opposite to NOS3), results are mixed. Even Khan et al5 found little basal change in NOS1 knockout, although negative effects were apparent at higher frequencies.
NOS1 and FFR
Khan et al5 hypothesized that NOS1 (but not NOS3) would modulate FFR due to its localization to the SR (and the crucial role of SR Ca2+ load in FFR). Indeed, knockout of NOS1 (but not NOS3) did depress FFR and FDAR, and SR Ca2+ load failed to rise at high frequency. The failure of SR Ca2+ load to rise could be due to decreased SR Ca2+ uptake or increased release (or both). The authors favored increased release, because NO can directly activate RyRs23 and NOS1 Co-IPs with RyRs.6 They suggest that oxygen radicals could irreversibly activate RyRs, increasing leak in the NOS1 knockout. However, it is unclear how NOS1 would prevent this. Moreover, the main difference occurs at higher frequency, but at reduced diastolic interval (and comparable SR Ca2+ load), such a diastolic leak should be less influential.
Indeed, NO-induced changes in SR Ca2+ load may be equally or more influential than NO effects on RyRs in intact cardiac myocytes,11 as with β-AR stimulation.24 Furthermore, while diastolic RyR leak does increase with frequency, this was secondary to increased SR Ca2+ load25 (which was not observed in NOS1 knockouts).5 We suggest that reduced SR Ca2+ uptake may also be involved in the NOS1 knockout phenotype. This would be consistent with the reported unaltered [Ca2+]i decline despite increase in SERCA2 and decrease in phospholamban (PLB) expression5 (changes that may be compensatory) in the NOS1 knockout mice. That is, the SERCA2a/PLB changes should have accelerated [Ca2+]i decline and increased SR Ca2+ load, but neither was observed. Given that NO can also affect PLB and SERCA2a, the influence of NOS1 on PLB and SERCA2a should be entertained (although NO effects in isolation seem to inhibit function).26,27 Nevertheless, depressed Ca2+-pump function in the NOS1 knockout is a possibility. This could also explain the lack of FDAR in NOS1 knockout mice.
In addition, “compartmentalization” of NO may be more complex than just localization of NOS isoforms and may involve colocalization of other NO/cGMP signaling molecules (superoxide dismutase, cGMP-specific phosphodiesterase [PDE-V], guanylate cyclase)28,29 along with the SR transporters and ICa.
The study by Khan et al5 does implicate NOS1 in modulating cardiac (and SR) function and contributes to our understanding of local NO signaling.
Footnotes
-
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
References
This Issue
Jump to
Article Tools
Share this Article
- The Real Estate of NOS Signaling
