Novel data point to a broader mechanism of action of oxidized ATP: the P2X₇ receptor is not the only target

Abstract

Oxidized ATP (oATP) is a Schiff-base-forming reagent that has been used for some years as an antagonist at the P2X₇ receptor (P2X₇R). Preincubation of mononuclear phagocytes with this inhibitor leads to attenuation of several proinflammatory responses triggered by extracellular ATP as well as a few non-nucleotide agonists. Novel data show that oATP reduces NFκB activation and IL-8 release in cells lacking P2X₇R, thus suggesting that some anti-inflammatory effects of oATP may not be due to blockade of the P2X₇R. This effect of oATP resembles the action of other natural or synthetic Schiff-base-forming reagents with immunomodulatory activity.

Receptors for extracellular nucleotides (P2R) are attracting increasing attention in drug development for treatment of pain, inflammation and cancer (Burnstock, 1996; Di Virgilio et al., 2001). An intriguing, and appealing, member of the P2R family is the P2X₇ receptor (P2X₇R). This receptor is a nondesensitizing plasma membrane ion channel that assembles from three or more 595 AA-long subunits (Surprenant et al., 1996; Kim et al., 2001). The amino-acid sequence predicts a membrane topology with two transmembrane stretches, a bulky extracellular domain and cytoplasmic N- and C-termini. In the extracellular region, there are 10 cysteines, three N-linked glycosylation sites and 18–21 lysines, depending on the species of origin. Upon transient stimulation, the P2X₇R behaves like many other channels selective for mono and divalent cations. However, upon sustained stimulation the aqueous pore dilates to admit molecules with a molecular mass up to 900 Da, irrespective of the charge.

The physiological functions of the P2X₇R are largely unknown, but researchers have been long intrigued by the potent cytotoxic effect caused by its sustained stimulation and by the ability of this receptor to drive the release of large amounts of IL-1β (and other inflammatory cytokines) from immune and inflammatory cells (Di Virgilio et al., 2001). Mice with the P2x₇ gene deleted (p2x₇^−/−) have been generated and are currently being exploited to investigate P2X₇R function. These animals are fertile and do not show dramatic phenotypic alterations, however, they appear to have impaired bone formation and remodelling, and reduced ability to develop acute inflammation of the joints in a typical model of experimental arthritis induced by anticollagen antibodies (Labasi et al., 2002; Ke et al., 2003). These in vivo observations strengthen previous in vitro data hinting at an important modulatory role of the P2X₇R in immune cell physiology. However, acceptance of P2X₇ as an important immunomodulatory receptor, and, as a result, identification of this receptor as a promising target for the development of novel anti-inflammatory drugs, has been so far precluded by the frustrating lack of potent and selective blockers.

Two compounds bearing aldehyde functions, pyridoxal-phosphate-6-azophenyl-2′,4′-disulfonic acid (PPADS) and periodate-oxidized ATP (oATP), are currently in use as P2X₇R blockers. PPADS was initially introduced as a wide-spectrum P2 inhibitor (Lambrecht et al., 1992), and later shown to be 10- to 20-fold selective for P2X over P2Y receptors. oATP was originally introduced as an affinity reagent for nucleotide-binding-proteins, and later proposed as a selective antagonist of macrophage P2X₇R, with no antagonist activity on coexpressed P2Y receptors (Murgia et al., 1993). Inhibition by oxidized ATP is irreversible and requires a prolonged (1–2 h) incubation. Later studies showed that that P2X₁R and P2X₂R are also blocked by this reagent.

The rationale for using oATP as a P2X₇ receptor antagonist was that the nucleotide structure of this Schiff-base-forming reagent might allow selective modification of lysine residues in the vicinity of the ATP-binding site of the P2X₇R, and thus block the receptor. Over the last 10 years, oATP has proven to be a valuable P2X₇R reagent when used to antagonize ATP-dependent stimulation of this receptor, less valuable when used to infer the participation of P2X₇R in cellular responses due to stimulation with non-nucleotide ligands. In this issue of the British Journal of Pharmacology, George Dubyak and co-workers (Beigi et al., 2003) have examined oATP effects in three human cell lines lacking P2X₇R expression, and provide sound evidence in support of an inhibitory effect of this reagent independent of P2X₇R. This observation raises two important questions: (1) under which conditions can oATP be used as a reliable P2X₇ receptor blocker? (2) What are the site(s) and mechanism(s) of the P2X₇R-independent anti-inflammatory effect of oATP? As regards the conditions under which oATP can be used as an unambiguous probe of P2X₇R function, Beigi et al. (2003) confirmed that blockade of ATP-induced Ca²⁺ influx, inward currents or membrane permeabilization by oATP in the vast majority of cases are due to P2X₇R blockade. With respect to the mechanism of P2X₇R-independent actions of oATP, these authors suggest that it might directly interfere with cytokine or pattern recognition receptors (PRR), such as Toll-like receptors, expressed on the plasma membrane of immune cells. Alternatively, they hypothesize that oATP might gain access to the cell cytoplasm (maybe by fluid-phase pinocytosis), and covalently modify and inhibit intracellular kinases. However, I think that an alternative (or additional) pathways should also be considered.

oATP is a Schiff-base-forming reagent and an immunomodulatory action of Schiff-base-forming compounds is not new. Small Schiff-base-forming molecules have long been known to mimic cell surface ligands expressed by antigen-presenting cells and provide costimulatory signals to CD4⁺ T cells, with a resulting Th1 polarization of the T cells (Rhodes et al., 1995). One such compound, tucaresol, developed at Glaxo Smith Kline, is currently being evaluated as an oral immunopotentiatory drug to be associated with vaccines. In vitro and in vivo studies show that Schiff-base formation by tucaresol increases T-cell receptor-dependent cytokine production and the ability to cope with tumors and with viral, bacterial and protozoal infections (Charo et al., 2002). The mechanism of this effect of tucaresol is unclear and the plasma membrane receptors involved have never been identified. Furthermore, the two compounds tucaresol and oATP have opposite effects on immune cell responses: stimulatory effects for the former, and inhibitory effects for the latter. However, this is not a major problem, as the effect might be dose-dependent, and a proper analysis of the dependency of oATP effects on the dose has not yet been carried out. Quite interestingly, Schiff-base-forming drugs seem to mimic the effect of endogenous Schiff-base-forming compounds, such as p-hydroxy-phenylacetaldehyde, a small highly reactive molecule produced by activated neutrophil myeloperoxidase. This neutrophil metabolite is also a potent amplifier of Th1 responses in vitro and in vivo. As suggested by Rhodes (2002), small Schiff-base-forming products, endogenously generated in inflammatory conditions, might function as danger signals linking the innate and adaptive immune responses. It is intriguing to notice this additional convergence between Schiff-base-forming compounds and P2R, since extracellular nucleotides are currently viewed as early danger signals at inflammatory sites (Gallucci & Matzinger, 2001), and P2R as sensors of danger (La Sala et al., 2003).

In conclusion, according to the new data provided by Beigi et al. (2003), there are few doubts that oATP inhibits more cell receptors and pathways than just the P2X₇ type, but this is not by itself surprising: many so-called ‘selective' receptor antagonists have turned out to be much less selective than initially thought. What emerges from these studies is that oATP can be used as a reliable tool to study P2X₇ receptor functions only under selected experimental conditions, in particular when the stimulant is a P2X₇R agonist such as ATP or benzoylbenzoyl ATP (BzATP), the read-out is a bona fide P2X₇R-dependent response, and an independent control, for example, P2X₇R-less clones of the same lineage as the cells under investigation, is available. Outside these boundaries, oATP inhibition should not be taken as unequivocal indication of the involvement of P2X₇R, and in any case the results should be interpreted with extreme caution. Does this mean the end of oATP as a tool to investigate purinergic modulation of immune cells? Probably not. Recognition of effects of oATP unrelated to P2X₇R blockade might shed light on the long-known immunomodulatory effect of Schiff-base-forming compounds, help to understand better the mechanisms involved in immunostimulation, and suggest new approaches for the development of immunomodulatory drugs.

Abbreviations

BzATP

benzoylbenzoyl ATP

oATP

oxidized ATP

PPADS, pyridoxal-phosphate-6-azophenyl-2′

4′-disulfonic acid

P2R

P2 receptor

P2X₇R

P2X₇ receptor