Glucose associated NETosis in patients with ST-elevation myocardial infarction: an observational study

Neutrophil extracellular traps (NETs), comprising fragments of neutrophil chromatin covered with neutrophil granule proteins extruding into the extracellular space upon neutrophil activation [1], have lately been identified as novel players in coronary artery disease (CAD). In stable CAD, levels of circulating NETs markers have been associated with CAD severity and prognosis [2, 3], while in acute ST elevation myocardial infarction (STEMI), levels of circulating NETs markers and the amount of NETs in aspirated coronary thrombi have both been associated with myocardial infarct size [4‐6]. NETs probably enhance endothelial dysfunction, activate plasmacytoid dendritic cells and stimulate proinflammatory cytokine secretion from macrophages, which all drive atherosclerosis progression [7]. Substantial data also indicate that NETs are involved in thrombosis per se by serving as a scaffold for platelet entrapment and activation, coagulation activation and inbihition of fibrinolysis [7‐11]. During NETs formation, NETosis, the enzyme peptidylarginine deiminase 4 (PAD4) catalyzes conversion of arginine residues of histones to citrulline in the neutrophil chromatin. This citrullination process results in histone charge change and chromatin decondensation, preceding neutrophil cell membrane rupture and NETs release, and is regarded a prerequisite for NETosis [12].

Multiple inducers of NETosis have been suggested, including gram positive and negative bacteria, fungi, lipopolysaccharide (LPS), proinflammatory interleukins, calcium channel opening agents and glucose [13]. NETosis has been reported to be glucose dependent in general and circulating NETs markers are shown to be elevated in patients with type 1 and 2 diabetes mellitus [14‐19]. The metabolic requirements for NETosis in CAD have, however, not been addressed. In the present investigation we aimed to explore whether the circulating NETs markers double stranded deoksyribonucleic acid (dsNDA) and myeloperoxidase-DNA (MPO-DNA) are associated with plasma glucose and the glucometabolic status in a cohort of STEMI patients. We further explored whether an acute glucose load by oral glucose tolerance test (OGTT) 3 months after the STEMI had any impact on NETosis, assessed by PAD4 messenger ribonucleic acid (mRNA) levels in circulating leukocytes as well as serum levels of dsDNA and MPO-DNA. Our hypothesis was that hyperglycemia would act as a NETosis stimulating agent in STEMI patients, thus contributing to increased understanding of NETs in CAD. If NETs have a negative impact on myocardial infarct size, identification of NETosis stimulating agents may participate in paving the way for directed anti-NETs therapy.

While the glucometabolic status overall only modestly influenced serum levels of NETs markers in this cohort of STEMI patients, plasma glucose per se was associated with levels of dsDNA both in the acute phase and in the stable condition after 3 months. An acute glucose load by OGTT performed 3 months after the STEMI resulted in increased gene expression of PAD4, suggestive of enhanced NETosis, although not accompanied by an increase in dsDNA or MPO-DNA levels. The latter might be caused by a delayed time course of NETs release.

The observed positive correlation between serum levels of dsDNA and plasma glucose is in line with existing literature suggesting that NETosis might be glucose dependent [14, 15, 17, 18]. Nevertheless, such association has not previously been specifically addressed in patients with acute MI, in which acute hyperglycemia has been associated with poorer outcome [25, 26]. It is interesting to note that the mechanisms linking acute hyperglycemia to poorer outcome in MI, eg. generation of reactive oxygen species (ROS), enhanced platelet and coagulation cascade activation, endothelial dysfunction and upregulated inflammation [26] overlap with suggested NETs mediated effects. Thus, the association between circulating dsDNA and plasma glucose in this cohort of patients with STEMI could shed light on how hyperglycemia is linked to adverse effects in acute MI. The absence of any association between the NETs markers and the glucometabolic status in the acute setting in our study is not easily discussed along with existing literature. However, higher levels of dsDNA were observed in patients with IFG and T2DM compared to patients with NGR after 3 months, which reflect a more stady state. The somewhat inconsistency could be due to the fact that patients with known diabetes and therefore more longlasting hyperglycemia were excluded from the study. Thus, the patients diagnosed with glucometabolic disturbances were probably in an “early” stage of their tendency towards hyperglycemia. It has, nevertheless, also been reported that neutrophils may become insensitive to NETs inducing stimuli during chronic hyperglycemia [27], which probably were the case for many of the patients with glucometabolic disturbances in this cohort. Based on our observations, however, NETs might be associated with the glucometabolic status in the more stable CAD setting. The complex pathophysiology during the acute course of a MI does not make it unlikely that the regulation of NETs and NETosis differ along the time axis from the acute event. Underpinning the latter, dsDNA and MPO-DNA levels were both significantly higher in the acute phase than after 3 months.

Gene expression levels of the NETosis related enzyme PAD4 increased significantly during the acute glucose load by OGTT after 3 months. The increase in PAD4 was not accompanied by increased levels of dsDNA or MPO-DNA. As the time course of NETs release upon different inducers of NETosis in CAD is unknown, this observation may imply that an acute glucose load in a stable state after STEMI is associated with NETosis. Although not reported in CAD, different metabolic intracellular signaling pathways have been suggested to be involved in the link between glucose and NETosis (Fig. 4). First, inhibition of glycolysis has been reported to inhibit NETosis [15]. Second, NETs formation depends on generation of reactive oxygen species (ROS) and intracellular hyperglycemia increases ROS production in neutrophils [28‐30]. Third, the pentose phosphate pathway generates NADPH that might be used as a cofactor in ROS production [31]. Fourth, inhibition of the polyl pathway has been reported to abbrogate NETs formation completely [17]. Exessive ROS per se have also been reported to increase transcription and activation of PAD4 directly by increasing intracellular Ca⁺⁺ concentration [29]. It has moreover been reported that acute glucose fluctuations, like during OGTT, rather than chronic hyperglycemia trigger ROS production [32]. The latter could be discussed in line with our observation of modest association between the NETs markers and the glucometabolic status, representing a more chronic situation. Taken together, the observed upregulation of PAD4 as a marker of NETosis during an acute oral glucose load is well in line with current literature and indicate that NETosis in the stable setting after a STEMI might be upregulated by acute fluctuations in plasma glucose levels. Further work to verify our observations and to explore details of the intracellular pathways linking hyperglycemia to NETosis in patients with CAD and acute MI is warranted. It also remains to be determined whether the observed associations between plasma glucose and markers of NETosis potentially can be translated into meaningful clinical applications.

In this cohort of STEMI patients, plasma glucose correlated to the NETs marker dsDNA both in the acute phase and after 3 months. Although the glucometabolic status overall was only modestly associated with the selected NETs markers, an acute oral glucose load after 3 months resulted in upregulated PAD4 gene expression, suggestive of enhanced NETosis. Potential modulators of NETosis, like plasma glucose, might become clinically relevant in the quest to influence adverse effects of NETosis in CAD.