eNOS, iNOS or no NOS, that is the question!


Khalid M. Naseem, Centre for Atherothrombosis Research, Medical Biosciences, University of Bradford, Bradford, UK.
Tel.: +44 1274 236215; fax: +44 1274 309742.
E-mail: k.m.naseem@bradford.ac.uk

While it is accepted that endothelial-derived nitric oxide (NO) regulates platelet function, the role of platelet-derived NO has been a widely debated topic in recent years. NO is a ubiquitous free radical gaseous messenger synthesized by nitric oxide synthases (NOS) from the amino acid l-arginine [1]. The endothelial form of the enzyme (eNOS) releases NO into the blood continually in response to the pulsatile flow of blood. NO diffuses into platelets and binds to its intracellular receptor, the hem-containing enzyme soluble guanylyl cyclase (sGC), leading to elevated cGMP levels, which acts as a second messenger coordinating the biological actions of NO [2]. The NO/cGC/cGMP system leads to targeted inhibition of numerous platelet activatory pathways [3]. The presence of an eNOS/sGC/cGMP signaling system endogenous to platelets was first suggested by Radmonski et al., who demonstrated elevated cGMP in response to platelet activation, an effect that was abrogated by NOS inhibitors [4,5]. Subsequently, numerous studies confirmed the ability of platelets to synthesis NO as evidenced by measurement of NO release, NO2/NO3 and l-citrulline (a by-product of the eNOS reaction). Furthermore, the presence of eNOS, and to a lesser extent iNOS, has been demonstrated in different species including bovine, murine, porcine and human platelets (reviewed in Ref. [5]).

Despite confidence of the presence of the platelet NOS/sGC/cGMP system, the physiological role of NO released from platelets has remained controversial [5]. Freedman et al. [6] demonstrated that platelet-derived NO acted in a paracrine fashion to inhibit the recruitment of platelets into aggregates. This was seemingly confirmed by a follow-up study from the same group demonstrating that mice with eNOS−/− platelets showed increased bleeding times and accelerated thrombosis in vivo [7]. More recently the assumption that platelet-derived NO is inhibitory has been challenged. Pharmacological and transgenic animal studies have provided evidence that platelet-derived NO and cGMP have a stimulatory role. The authors propose that low concentrations of NO promote agonist-induced dense granule secretion and aggregation [8]. These data are supported by studies with eNOS-deficient mice, which have diminished secretion and aggregation in response to low doses of thrombin and collagen, as well as a reduced ability to form occlusive arterial thrombus in an in vivo thrombosis model [9]. Other studies with eNOS-deficient animals have confused this issue further, with both Iafarati and Ozuyaman publishing data indicating that the absence of platelet eNOS has little consequence for both platelet function and arterial thrombus formation [10,11]. Although, differences in both experimental conditions and animal models can partly explain the nature of this conflicting data, it is difficult for both expert and non-expert readers to draw any conclusions as to the role of platelet NO. As such, the field requires much more detailed and systematic studies to evaluate the presence of NOs in different species under strict experimental conditions, and detailed evaluation of the NOS/sGC/cGMP pathway using multiple readouts.

In the current issue of the Journal, a detailed study by Gambaryan et al. [12] have attempted to clarify some of the issues related to platelet NOS. The authors report two potentially important observations that may have significant implications for this field of research. Their current data suggests, first, that iNOS and eNOS are not present in either mouse or human platelets, and second, that cGMP can be synthesized in platelets independently of NO. In the first instance, the authors demonstrate that widely used NOS inhibitors do not effect Ca2+ dependent or independent platelet cGMP formation, but abolish cGMP formation in endothelial cells, suggesting a NOS-independent mechanism present in platelets. Having made this initial observation, murine platelets deficient in eNOS (eNOS−/−) and human platelets treated with NOS inhibitors, were used to evaluate the presence and activity of eNOS and iNOS in platelets. NOS inhibitors had no effect on platelet activation (as measured by fibrinogen binding) in humans, and importantly no difference in agonist-induced activation was observed between wild type (WT)and NOS-deficient murine platelets. Taken together these data suggest that NOS in platelet, regardless of isoforms, have no measurable influence on platelet activation and aggregation, questioning the physiological relevance of platelet-derived NO. The authors demonstrate that iNOS and eNOS protein are not expressed in human platelets. Importantly, iNOS and eNOS mRNA is present in human platelets prepared by standard techniques, but absent from a newly developed method for preparation of highly purified platelets. Critically these experiments use appropriate cell controls, endothelial cells for eNOS and activated macrophages for iNOS, to confirm the presence or absence of each isoform. Thus, the initial conclusion from the present work indicates that earlier studies reporting the presence of NOS isoforms may be as a result of potential contamination from other cells types. Certainly erythrocytes and leukocytes contain eNOS and may represent sources of potential contamination during the isolation of platelets [13]. In addition to demonstrating the absence of eNOS from platelets, the authors remind the readers of the potential pitfalls of commercially available antibodies. One eNOS antibody used here clearly gives a positive result for eNOS in eNOS-deficient platelets, with similar data shown using a phospho-specific eNOS antibody. In endothelial cells, eNOS activity is regulated by phosphorylation of serine1177 [14,15], and the present data now places a question mark over studies demonstrating phospho-serine1177 of NOS in platelets using this antibody. Critically, the one eNOS antibody where specificity is clear, demonstrates that eNOS protein is not present in both WT and eNOS−/− platelets.

Regardless of whether NO acts as a platelet inhibitor or activator, there is a consensus that it’s biological effects occur through formation of cGMP. Thus, sGC, the only known intracellular receptor for NO, is a critical node within NO-mediated signaling events. Here, Gambaryan confirms previous findings that platelet activation by von Willebrand factor (VWF) results in accumulation of cGMP [8,16,17]. However, they demonstrate that VWF-induced cGMP accrual is independent of NO, as it occurs in the presence of NOS inhibitors. Furthermore, NO-independent activation of sGC is associated with tyrosine phosphorylation of the β1 subunit. One consequence of this agonist-induced cGMP is a reduction in platelet activation, as inhibition of VWF-mediated cGMP accural with a specific sGC inhibitor led to an increase in fibrinogen binding. These data present the exciting possibility of a new pathway for the regulation of sGC and cGMP formation independently of NO. Clearly further studies are required to clarify the potential role of this pathway in platelet function.

Taken as a whole, the authors present compelling evidence that iNOS and eNOS may in fact be absent from both human and mouse platelets, and therefore represents an important addition to this area of research. Certainly the evidence to support the presence of iNOS is limited and hence this finding is not surprising. However, the other results of this study must be viewed with caution as the critical issue of eNOS activity has not been addressed here. Gambaryan et al. have relied on markers downstream of NOS, that is, cGMP formation and phosphorylation of vasodilator-activated phosphoprotein (VASP) as a target for cGMP-dependent protein kinase, to evaluate the NOS signaling pathway. In contrast, numerous studies in the past 17 years have demonstrated the release of NO from platelets, and evidence of NOS activity by the measurement of l-citrulline [5]. Thus while the present study, provides strong evidence to demonstrate that eNOS may not be present, as detected by immunoblotting, and places a potential question mark against studies purporting to demonstrate the presence of NOS in platelets, it does not address the issue of a NOS-type activity in platelets. It is important that these two issues be reconciled. Contamination from other cells is unlikely to account for platelet NOS activity described by other studies, as in most cases NO is synthesized in response to physiological platelets agonists. Erythrocytes and leukocytes, the most likely source of contamination, do not respond to these agents. Thus, other possibilities must be considered. For example, can pathways other than the NOS reaction be responsible for the production of l-citrulline, which has been used as gold standard marker of NOS activity. The only other major source of l-citrulline is the urea cycle, which to date has not been described in platelets. If NOS is absent, are there alternate pathways for NO production in response to platelet agonists? Furthermore, could a possible explanation be that NOS is expressed at very low levels that is not detectable by immunoblotting, but is still sufficient to synthesize measurable NO? Certainly further studies by the authors using their new method for high purification of platelet to evaluate NOS activity would help clarify some the unresolved issues, as would the mass spectrometric analysis of platelet eNOS described by other groups.

In summary, the study by Gambaryan et al. present several new possibilities that may underlie the conflicting reports related to platelet NO formation. However, the controversy surrounding the role of agonist-induced cGMP in platelet function continues. Watch this space for further developments….


The author would like to thank R. Riba for critical discussion of this commentary. Research in the author’s laboratory is funded by the British Heart Foundation and Heart Research UK.

Disclosure of Conflict of Interests

The author states that he has no conflict of interest.