Background
Gamma-glutamyltransferase [GGT; (5-l-glutamyl)-peptide⁄amino acid 5-glutamyl transferase; EC 2.3.2.2], a member of the structural superfamily of the N-terminal nucleophilic hydrolases expressed by a wide number of cell types [
1] including circulating monocytes [
2,
3], hydrolyzes extracellular glutathione (GSH) to provide cysteine for its intracellular re-synthesis. Along with its pivotal role in that antioxidant biological process [
1], however, GGT also generates reactive oxygen species (ROS) [
4] and activates NFkB [
5,
6], a redox-sensitive transcription factor [
7] key in Tissue Factor (TF) gene expression [
8], a major regulator of haemostasis and thrombosis [
8,
9]. Therefore, deductive reasoning makes it plausible to hypothesize a cross-talk between GGT and TF, a possibility consistent with the highly consistent epidemiological association of circulating GGT levels with acute coronary events (see [
10] for a review), a pathological process favoured by TF (see [
11] for a review). The assumption, corroborated by recent reports of enzymatically active protein in human atheromatous plaques [
12], requires, however, to document a mechanistic link between GGT and TF for which at the moment no evidence is available.
Discussion
Discussion of the effects of GGT on TF expression needs some preliminary comments about endotoxin contamination, an important concern in the light of the frequent pollution by that bacterial product of cell cultures even if grown in carefully controlled experimental conditions [
22]. However, three orders of considerations make that possibility unlikely. First, GGT-induced PCA was abrogated by heat, a procedure generally ineffective on heat-resistant LPS [
21]. Secondly, LPS levels in our preparation were less than <0.1 ng/mL, a concentration with a minor, if any, procoagulant action and certainly far less than that of GGT, a behaviour excluding an important role for heat-sensitive LPS strains [
22]. Thirdly, inhibition of TF activity by an anti-GGT antibody provided additional proof of the specificity of the procoagulant response to hrGGT.
Having this background in mind, the main and original outcome of this study was the demonstration of the procoagulant properties of GGT and their independence from the enzymatic action of the molecule. That latter, quite intriguing conclusion is based upon the insensitivity of natural GGT-stimulated PCA to acivicin, a highly specific GGT inhibitor [
20] and, more importantly, by the maintained procoagulant effect of hrGGT, a wheat germ-derived protein devoid of enzymatic activity because of a missing post-translational glycosylation apparatus [
17,
18], an assumption fully verified by our GSH hydrolysis experiments reported in Fig.
1. Thus, hrGGT induced a concentration-dependent procoagulant effect, increased TFag and upregulated mRNA, a behaviour this latter showing that GGT-induced TF gene transcription is an early event since our mRNA assays were obtained after only 2 h of exposure to the molecule, quite similar in this regard to the response induced by unrelated cytokines and inflammatory agonists (e.g. [
27]). However, additional studies are needed to evaluate the pattern of GGT-induced gene expression over a longer time course and the reader should also be aware that our PCR procedure based upon the use of GAPDH as a single and quite variable reference gene may present some technical limitations [
28,
29]. Moreover, GGT-induced TF gene transcription was likely located at the level of NFkB activation given the inhibitory effect of BAY-10-772, a pharmacological antagonist [
24] of that crucial controller of redox stimuli converging upon TF gene [
7,
8]. The conclusion was strengthened by the negative modulation of PCA exerted by NAC, a sulfhydryl-group donor that, by increasing the antioxidant thiol pool and scavenging the excess of intracellular ROS, downregulates NFkB [
25]. Therefore, our data, besides constituting, to the best of our knowledge, the first demonstration of a direct TF procoagulant effect of GGT, also provide consistent, albeit admittedly indirect, evidence for increased ROS production and NFkB activation as the pathophysiological mechanism linking GGT to TF expression in PBMCs.
That GGT might induce ROS generation is a concept dating back to two decades ago or so when Glass and Stark showed oxidative damage of cell surface proteins and membrane lipids as a consequence of GSH cleavage by GGT inducing auto-oxidation of the sulfur via a Fenton reaction resulting in the iron–dependent production of oxygen radicals [
30]. However, that mechanism, which requires an enzymatically functional GGT molecule, cannot evidently apply to our data. Rather, direct, non enzymatic TF stimulation is closely consonant with similar results obtained in evaluating the role of the GGT as a bone resorbing factor [
31]. The data open obvious questions about the GGT-operated signal transduction pathways upstream NFkB activation including the mechanisms allowing membrane permeation of the exogenous protein and the interaction with its intracellular targets. Although our results cannot answer this specific point, one might conjecture of a specific but insofar not identified GGT-receptor or lipid-driven endocytosis [
32]. Perhaps, some role may play a chemokine-like CX3C motif contained in the GGT molecule [
33] or exogenous GGT might mimick cytokine-like activities of GGT-related proteins produced by genes different from the GGT1 [
34]. For example, the GGT2 gene located close to the GGT1 chromosomal region seems to encode for a full length, enzymatically inactive protein, not localized to the plasma membrane [
35] and, according to some Authors, involved in redox modulation [
36]. It might also be possible to envisage protein-protein interactions leading to the expression of other procoagulant cytokines. Additional unresolved questions to be addressed in future studies regard an understanding of how the enzymatic and non-enzymatic activity of GGT may complement each other as well as the role of circulating GGT levels in the activation of the TF procoagulant pathway.
The data reported in this study may have pathophysiological and clinical relevance. In fact, previous experiments have shown upregulated GGT transcription in response to NADPH oxidase-mediated ROS generation [
37] as well as to agents endowed with TF-stimulating properties [
8] such as Tumor Necrosis Factor(TNF)-alpha [
38] and phorbol esters [
39]. Moreover, monocytes, a cell line harbouring GGT [
2,
3], have recently been shown to release a GGT fraction when exposed to LPS [
40], a product of Gram-negative bacteria that initiates the pathogen-induced inflammatory response [
41] of which activation of coagulation is a prominent component [
9]. In that framework, thus, it is conceivable to hypothesize a vicious circle by which circulating or within-plaque [
12,
40] GGT stimulates TF expression, and plaque-derived cytokines induce procoagulant GGT expression [
38]. Both arms of this self-reverberating mechanism hinging around NFkB activation [
42] may amplify the pro-thrombotic potential of vulnerable atheromatous plaques possibly in synergism with TF expressed by activated B- and T-lymphocytes [
14,
43], a tempting but at the moment only speculative hypothesis worth being pursued in future studies.