Taming the neutrophil: calcium clearance and influx mechanisms as novel targets for pharmacological control

The destructive power of activated neutrophils

The crucial involvement of polymorphonuclear leucocytes, predominantly neutrophils, in host defence is underscored by the relative abundance of these cells, with approximately 50 billion circulating in the bloodstream [1]. Following extra-vasation, neutrophils attracted to sites of tissue inflammation by chemotaxins internalize microbial pathogens. Inside the phagosome these are targeted and destroyed by toxic reactive oxidants and serine proteases released into the phagolysosome. Production and/or release of these toxic oxidants and proteases by neutrophils is stringently controlled to protect surrounding cells and tissues [2], but may be excessive as a result of inappropriate activation and necrosis. Excessive and/or protracted activation of neutrophils during hyperacute and chronic inflammatory disorders predisposes to tissue injury. Important examples of inflammatory conditions in which the neutrophil is the primary offender are shown in Table 1.

Notwithstanding the cardiovascular system, the predominance of inflammatory airways disorders shown in Table 1 demonstrates the susceptibility of the lung to neutrophil-mediated injury. The lung is clearly a primary target for neutrophil recruitment and activation, which in turn contributes to the pathogenesis of conditions such as chronic obstructive pulmonary disease (COPD) [3], asthma [4] and cystic fibrosis [5].

Corticosteroids and neutrophils

Few currently available therapeutic agents, including corticosteroids, effectively down-regulate neutrophil pro-inflammatory activity. Insensitivity to corticosteroids may therefore be a feature of those disorders in which the neutrophil is the predominant inflammatory cell type.

The relative insensitivity of neutrophils to corticosteroids is attributable to a combination of mechanisms. First, many of the proinflammatory activities of these cells [production of reactive oxidants, release of granule polypeptides, generation of prostanoids, eicosanoids and platelet activating factor (PAF)] occur within seconds of activation and are independent of de novo protein synthesis. Secondly, neutrophils which are now recognized to be an important source of newly synthesized cytokines [6,7], particularly interleukin (IL)-8 and tumour necrosis factor (TNF)-α, contain comparatively high levels of the functionally inactive beta isoform of the glucocorticoid receptor (GR), the synthesis of which is further up-regulated on exposure of the cells to IL-8 [8], rendering them even less corticosteroid-sensitive. Moreover, neutrophils, unlike other types of immune and inflammatory cells, have been reported to be relatively insensitive to the apoptosis-inducing actions of corticosteroids [9,10].

Clearly, the design and development of novel, neutrophil-directed anti-inflammatory, chemotherapeutic strategies is a priority.

Mobilization of Ca²⁺

Intracellular Ca²⁺ in neutrophils is reportedly stored in specialized storage vesicles termed calciosomes [11]. This may, however, be somewhat of an oversimplification as there appear to be at least two distinct cellular locations for Ca²⁺ stores in neutrophils that may have differential involvement in activation of proinflammatory functions, and may utilize different molecular/biochemical mechanisms of Ca²⁺ mobilization [12]. One site is located peripherally under the plasma membrane and appears to be involved in the activation of β₂-integrins, while the other is localized in the perinuclear space and is mobilized by chemoattractants such as the synthetic tripeptide, N-formyl-L-methionyl-L-leucyl-L-phenylalanine (FMLP) [12]. Mitochondria may also serve as calcium-storage organelles [13], with neutrophils possessing a more extensive mitochondrial network than previously recognized [14].

The molecular/biochemical mechanisms involved in Ca²⁺ mobilization following chemoattractant receptor-mediated activation of neutrophils are well characterized. Leucocyte membrane receptors for chemoattractants, including FMLP, C5a, leukotriene B₄, PAF and chemokines, belong to the 7-transmembrane, G-protein-coupled family of receptors. Occupation of these receptors, which are controlled by various Gα and Gβγ subunits, results in activation of the β isoforms of phospholipase C which in turn mediate production of inositol-1,4,5-triphosphate (IP₃) by hydrolysis of phosphatidylinositol 4,5 biphosphate [15,16]. IP₃ interacts with Ca²⁺-mobilizing receptors on intracellular storage vesicles, resulting in discharge of the cation into the cytosol. These events are extremely rapid, occurring within a few seconds of ligand-receptor binding, and result in a five to 10-fold increase in the cytosolic free Ca²⁺ concentration above a basal value of about 100 n M [17]. Only modest increases in IP₃, of around 15% of maximal, are required to cause complete mobilization of intracellular Ca²⁺[18,19]. The peak increase in cytosolic Ca²⁺ is followed by a rapid, progressive decline in cytosolic Ca²⁺ with a return to basal values within several minutes. The rate of decline in the concentration of cytosolic Ca²⁺ appears to be governed by two mechanisms. First, the efficiency of the systems which promote clearance of Ca²⁺ from the cytosol [20,21] and secondly, those which regulate the time of onset, rate and magnitude of influx of extracellular cation [22].

Clearance of Ca²⁺ from the cytosol of activated neutrophils

Following activation of neutrophils, restoration of Ca²⁺ homeostasis is essential to prevent Ca²⁺ overload and hyperactivity of the cells. This is achieved by rapid clearance of Ca²⁺, primarily through the action of two adenosine triphosphate (ATP)-driven pumps operating in unison. These are the plasma membrane Ca²⁺-ATPase, which is a Ca²⁺-efflux pump, and the endo-membrane Ca²+-ATPase which resequesters the cation into intracellular stores [23]. In activated neutrophils, these two Ca²⁺ pumps appear to contribute more or less equally to the clearance of cytosolic Ca²⁺[24,25].

The plasma membrane Ca²⁺-ATPase of neutrophils is modulated by calmodulin, acidic phospholipids and polyunsaturated fatty acids which shift the pump to a higher-affinity state for Ca²⁺, resulting in enhanced maximal velocity [26]. In the case of FMLP-activated neutrophils, a dramatic and transient efflux of Ca²⁺ immediately follows release of the cation from stores, proceeds over a 30-s time course and results in discharge of about 50% of Ca²⁺ mobilized from stores [24,25].

Activation of neutrophils with FMLP also results in immediate, transient activation of adenylate cyclase [27,28]. This results from the interaction of adenosine, generated by dephosphorylation of adenylates, with G-protein/adenylyl cyclase-coupled adenosine receptors (AR) of the A_2A receptor subtype on the neutrophil membrane [27,28], leading to activation of adenosine 3′,5′-cyclic monophosphate (cAMP)-dependent protein kinase A (PKA). PKA in turn up-regulates the Ca²⁺ sequestering/resequestering activity of the endomembrane Ca²⁺-ATPase by phosphorylation of the regulatory polypeptide, phospholamban [29]. The endomembrane Ca²⁺-ATPase contributes to Ca²⁺ clearance from the cytosol of activated neutrophils by promoting resequestration of cation mobilized from stores, as well as sequestration of incoming Ca²⁺ during store-operated influx [30–32].

Operating in harmony, these two Ca²⁺ pumps (plasma membrane and endomembrane) appear to be the primary contributors to clearance of cytosolic Ca²⁺ in activated neutrophils.

The role of membrane depolarization

Efficient clearance of Ca²⁺ by the plasma membrane and endomembrane Ca²⁺-ATPases is facilitated by the membrane depolarizing action of NADPH oxidase which limits the influx of extracellular Ca²⁺. NADPH oxidase is the membrane-associated, electron-transporting, superoxide-generating system of phagocytes, which undergoes Ca²⁺-dependent activation during exposure of the cells to chemoattractants, cytokines and opsonized antigens. The dramatic decrease in membrane potential that accompanies activation of NADPH oxidase, and which is coincident with release of Ca²⁺ from neutrophil intracellular stores, efflux of the cation and activation of superoxide production, has been attributed to the electrogenic activity of the oxidase [22], as well as to the action of a rapidly activated H⁺ conductance with resultant influx of H⁺[33]. This type of abruptly occurring membrane depolarization has been shown to limit the influx of extracellular Ca²⁺[34,35]. When the cells are depolarized, the driving force for entry of Ca²⁺ is abolished because the electrical component of the electrochemical gradient promoting Ca²⁺ entry is markedly reduced.

Membrane repolarization and store-operated influx of Ca²⁺

The uptake of extracellular Ca²⁺ into FMLP-activated neutrophils is a delayed event, with net influx becoming detectable only at around 1 min after addition of the chemoattractant, and proceeding gradually over a 5-min time-course. This type of Ca²⁺ influx is characteristic of store-operated mechanisms, being secondary to emptying of intracellular Ca²⁺ stores and necessary for store refilling.

Interestingly, the time-course of influx of extracellular Ca²⁺ is superimposable on that of membrane repolarization, suggestive of a mechanistic interrelationship between these two events. It has been proposed that membrane repolarization in activated eosinophils is mediated by an NADPH oxidase-associated H⁺ extrusion channel which, being electrogenic, results in repolarization of the cell membrane [33,36].

Notwithstanding the crucial role of proton efflux in the regulation of intracellular pH, additional mechanisms may contribute to membrane repolarization in activated neutrophils. Importantly, charge compensatory mechanisms may vary according to the nature of the signal transduction pathways utilized. Membrane repolarization in chemoattractant-activated neutrophils is associated with up-regulation of the activity of the electrogenic plasma membrane Na⁺/Ca²⁺ exchanger operating in reverse mode [37].

Treatment of activated neutrophils with KB-R7943, a selective inhibitor of the Na⁺/Ca²⁺ exchanger operating in reverse mode [38], significantly inhibited the rate and magnitude of recovery of the membrane potential towards resting levels (Fig. 1). Importantly, inhibition of membrane repolarization by KB-R7943 is associated with marked attenuation of store-operated Ca²⁺ uptake by these cells, suggesting that these processes are interdependent [37].

The role of membrane repolarization in regulating the rate of calcium influx is further supported by the observation that addition of the NADPH oxidase inhibitor, diphenyleneiodonium chloride (DPI) to neutrophils 1 min after FMLP potentiated the rate and magnitude of repolarization (Fig. 1), with a corresponding increased rate of store-operated Ca²⁺ influx [37].

Recently, high conductance Ca²⁺-activated K⁺ channels have been elegantly demonstrated to participate in charge compensation following activation of neutrophils with phorbol esters and opsonized microorganisms. The resultant electrogenic K⁺ efflux (influx in the case of phagocytic vacuoles) is a prerequisite for intravacuolar activation of antimicrobial proteases [39,40].

Store-operated Ca²⁺ channels and neutrophils

Although the precise molecular identity of the store-operated Ca²⁺ channels (SOCs) operative in human neutrophils and other cell types has not been conclusively established, one particular family of non-voltage-activated Ca²⁺ channels, the family of transient receptor potential (TRP) channels has attracted considerable interest. These channels have been the subject of several recent reviews [41,42], and their salient features with respect to neutrophils can be summarized as follows:

• There are three major subgroups within the TRP gene family, the proposed nomenclature for these being TRPC, TRPV and TRPM, each of which currently contains seven (TRPC 1, 2, 3, 4, 5, 6, 7) five (TRPV 1, 2, 4, 5, 6) and four (TRPM 1, 2, 5, 7) members, respectively [42].

• The presence of members of all three TRP channels has been demonstrated in leucocytes and leucocytic cell lines, with mRNA for TRPC6 having been demonstrated in human neutrophils, eosinophils and lymphocytes [42]. In addition, the ADP-ribose-activated long TRP channel 2 (LTRP 2) is expressed on neutrophil membranes and patch-clamp electrophysiological studies have demonstrated Na⁺ and Ca²⁺ ion conductivity across these channels [43].

• Overexpression of TRPs in mammalian cells has been reported to result in enhancement of store-operated Ca²⁺ entry in many, but not all studies [41,44], while reduction of TRP expression using antisense strategies has been shown to decrease store-operated uptake of the cation [41].

• As described below, filling of intracellular Ca²⁺ stores through TRPs may involve physical interaction of the channel proteins with IP₃ receptors on storage vesicles, compatible with a conformational coupling mechanism of depletion-activated Ca²⁺ entry [41,45].

Although store-depletion activated mechanisms predominate during the refilling of intracellular Ca²⁺ stores, other pathways for Ca²⁺ entry may exist as chemokines utilizing CXC4, CCR1 and CCR5 on neutrophils are able to stimulate Ca²⁺ influx without Ca²⁺ release from storage vesicles [46].

Interestingly, in HEK 293 cells, a physical association has been reported to exist between TRPC3 channels and the Na⁺/Ca²⁺ exchanger, with the TRPC3-mediated Ca²⁺ entry being dependent on the exchanger operating in reverse mode [47], as has been reported for neutrophils [37].

How do calcium storage vesicles ‘communicate’ with plasma membrane store-operated calcium channels?

Several mechanisms have been proposed to explain the activation of Ca²⁺ entry by store depletion and these have been the topic of several recent reviews [41,48].

Briefly, the conformational coupling model proposes that intracellular Ca²⁺ stores are located in close proximity to the plasma membrane, enabling physical coupling between the IP₃ receptors of the stores and the proteins of the Ca²⁺ entry channels. When the IP₃ receptors open during intracellular Ca²⁺ mobilization, the resultant conformational change in the receptor protein activates the Ca²⁺ entry channels, leading to Ca²⁺ influx [41,45]. In human neutrophils, the magnitude of influx of Ca²⁺ appears to be related directly to the intracellular IP₃ concentration [49].

The soluble intracellular messenger, calcium influx factor (CIF), may diffuse from the cytosol to activate plasma membrane calcium channels, triggering Ca²⁺ entry. The bioactive sphingolipid, sphingosine-1-phosphate, has been reported to mimic closely the actions of CIF [50], but this remains to be established conclusively [51].

More recently, additional mechanisms have been proposed. These include (1) a role for mitochondria in Ca²⁺ signalling [52], (2) participation of the cytoskeleton during calcium influx [53], (3) a direct mass-action mechanism [54] and (4) calcium-sensing proteins (CALPs) that may modulate the activity of calcium channels [55].

The various mechanisms utilized by chemoattractant-activated neutrophils to mobilize intracellular Ca²⁺ and to restore Ca²⁺ homeostasis are summarized in Fig. 2.

Cyclic AMP-elevating agents

Cyclic AMP-elevating agents have been reported to interact directly with immune and inflammatory cells, including neutrophils, resulting in cAMP-mediated attenuation of the responses of these cells to various proinflammatory stimuli [56]. Importantly, these agents have been demonstrated to inhibit the activity of phospholipase A₂, as well as the production of a range of proinflammatory mediators including cytokines (especially tumour necrosis factor), prostaglandins, leukotrienes, PAF, reactive oxidants and release of granule-derived enzymes. Cyclic AMP-elevating agents also attenuate β₂-integrin activation/expression by neutrophils, eosinophils, monocytes, macrophages and lymphocytes [56,57].

Antagonists of depletion-activated store-refilling in neutrophils

Interference with the refilling of depleted Ca²⁺ stores in activated neutrophils represents an attractive target for anti-inflammatory chemotherapy, because this would be expected to prevent activation of the proinflammatory activities of the cells on re-exposure to the same or a different receptor-dependent mediator of Ca²⁺ mobilization. Calcium reuptake by stimulated neutrophils increases the cytosolic Ca²⁺ concentration near the plasma membrane which facilitates degranulation and calpain activation [52], as well as β₂-integrin-mediated adhesion to vascular endothelium [58]. Antagonists of calcium influx may therefore attenuate these neutrophil responses. Agents which fall into this category include, first, those which inhibit the membrane repolarization responses of activated neutrophils, particularly inhibitors of the Na⁺, Ca²⁺-exchanger, and secondly, antagonists of store-operated Ca²⁺ channels such as itraconazole [59].

Beta-adrenergic agonists

Beta-adrenergic agonists (β-agonists) bind to β-receptors on inflammatory cells, including neutrophils, with resultant elevation of intracellular cAMP and inhibition of the proinflammatory activities of these cells in vitro[60,61]. In the clinical setting, numerous trials have supported the use of long-acting β-agonists (formoterol and salmeterol) in conjunction with inhaled corticosteroids, for patients with asthma and COPD [62–64].

Currently, the therapeutic potential of β-agonists, at clinically relevant concentrations, in experimental models of acute lung injury (ALI) [65] and in critically ill patients with the acute respiratory distress syndrome (ARDS) has generated interest [66,67].

Second-generation type 4 phosphodiesterase (PDE) inhibitors

PDE is a generic term that encompasses at least 11 distinct enzyme families (isotypes) that hydrolyse cAMP and/or cGMP, representing the only cellular pathway for the degradation of these cyclic nucleotides. Cyclic AMP-specific PDE4 is present in most immune and inflammatory cells, with the 4B2 subtype being the predominant PDE species in human neutrophils and monocytes [68,69]. Consequently, there has been considerable interest in PDE4-selective inhibitors as a potential therapy for acute and chronic inflammatory diseases, including ALI, bronchial asthma and COPD, and these agents have indeed been shown to exert significant anti-inflammatory activity in various animal models [70–72] and in clinical trials in humans [73], several of which are currently in progress. These involve compounds designed to have an improved therapeutic window such as cilomilast [c-4-cyano-4-(3-cyclopentoxyl-4-methoxyphenyl-r-1-cyclohexane carboxylic acid)], also known as Ariflo™ or SB 207499, which has been shown to improve pulmonary functions and symptoms in patients with COPD [74]. Importantly, the anti-inflammatory effects of cilomilast have been confirmed in patients with COPD in whom this agent significantly decreased the numbers of inflammatory cells determined by means of serial bronchial biopsies [75].

Another selective PDE4 inhibitor, roflumilast (3-cyclo-propylmethoxy-4-difluoromethoxy-N-[3,5-di-chloropyrid-4-yl]benzamide), inhibits neutrophil proinflammatory responses in vitro[76] and has also demonstrated efficacy in the management of exercise-induced asthma [77] and COPD [73].

Adenosine receptor agonists

The well-recognized broad-spectrum anti-inflammatory actions of adenosine and its analogues are for the most part mediated through activation of the A_2AR subtype. This is certainly the case for neutrophils, monocytes, mast cells and T lymphocytes [30,78]. The A_2AR subtype is a G-protein (G_s)-coupled receptor linked to activation of adenylyl cyclase, resulting in cAMP-mediated anti-inflammatory activity [79,80].

The prototype A_2AR agonist, CGS 21680 (2(4-[(2-carboxyethyl)phenyl] ethylamino)-5′-N-ethylcarboxamido adenosine), although lacking the required receptor specificity for therapeutic application, together with ZM 241385 (4-(2-[7-amino-2-(2-fury 1)[1,2,4)triazolo[2,3-a][1,3,5]triazin-5-yl aminoethyl phenol), a highly selective antagonist of A_2ARs, have been useful in probing the presence and anti-inflammatory properties of A_2ARs in immune and inflammatory cells [81,82]. CGS 21680 has also proved useful in the development of novel, experimental A_2AR agonists with improved receptor specificity and anti-inflammatory properties [79,83].

The anti-inflammatory potential of adenosine, adenosine A_2A receptor agonists, adenosine precursors and enzyme inhibitors which elevate adenosine levels holds promise in the management of a range of inflammatory disorders including asthma, sepsis, auto-immune diseases, myocardial ischaemia and reperfusion injury [82,84–86]. The adenosine A_2AR agonists, CGS 21680 and 4-[3-[6-amino-9-(5-ethylcarbamoyl-3,4-dihydroxy-tetrahydro-furan-2-yl)-9H-purin-2-yl]-prop-2-ynyl]-cyclohexanecarboxylic acid methyl ester (ATL146e), exhibited significant anti-inflammatory activity in animal models of allergic asthma [87] and sepsis [88], respectively. During clinical trials in patients with acute myocardial infarction, adenosine significantly reduced infarct size [89]. Potent A_2A receptor agonists are currently being developed and the efficacy of these agents in the treatment of COPD remains to be established [90]. The widespread anatomical distribution of A_2ARs increases the potential for side effects, but this may be overcome by selective topical application in the lung, or by combining A_2AR agonists with selective PDE4 inhibitors, or corticosteroids, which may enable reductions in the dosages of both types of agent [30,91].

Inhibitors of the Na⁺/Ca²⁺-exchanger

KB-R7943 has proved to be a useful experimental agent to probe the involvement of the Na⁺, Ca²⁺-exchanger in regulating the intracellular Ca²⁺ concentration in a variety of cell types such as ventricular myocytes, as well as the validity of the exchanger as a potential target for the prevention and therapy of ischaemia-related cardiovascular damage and dysfunction [92]. SEA0400 (2-[4,2,5-difluorophenyl)methoxy]phenoxy]-5-ethoxyaniline), is a recently developed compound that appears to be a more potent and selective inhibitor of the Na⁺/Ca²⁺-exchanger than KB-R7943, and has been reported to be efficacious in animal models of cerebral ischaemic injury [92,93].

The anti-inflammatory potential of inhibitors of the Na⁺/Ca²⁺-exchanger, while extremely promising is, however, a very new area of immunopharmacological research. Meaningful progress will be dependent on identifying the isoform(s) of the exchanger which are operative not only in human neutrophils, but also in other types of immune and inflammatory cells [94,95], as well as on the design and development of isoform-specific inhibitors of the exchanger which selectively target these cells.

Antagonists of store-operated Ca²⁺ channels

As mentioned earlier, there is considerable interest in store-operated Ca²⁺ channels as potential therapeutic targets [42]. The imidazole antimycotic, itraconazole, antagonizes store-operated Ca²⁺ influx mechanisms in human neutrophils in vitro[59], while the structurally related triazole, fluconazole, improved survival in a limited number of critically ill patients with septic shock, an effect not attributed to its antifungal activity [96]. Clearly, selective and potent inhibitors of the TRP channel family have therapeutic potential for a range of inflammatory disorders. Although this field of immunopharmacology is exciting and holds considerable therapeutic promise, meaningful progress is, however, dependent on target validation and on the design and synthesis of selective pharmacological agents.

Phospholipase C (PLC) and NADPH oxidase

U73122, an inhibitor of PLC, has been shown in vivo to block carrageenan-induced paw oedema and leucocyte accumulation, as well as lipopolysaccharide-induced cellular infiltration and prostaglandin production in animal models [97]. Clofazimine, an antileprosy agent, potentiates NADPH oxidase activity [98], but also possesses well-documented anti-inflammatory actions [99], possibly reflecting the recognized role of the oxidase in restraining Ca²⁺ influx [22,35,37,100].