Oxygen homeostasis, thiol equilibrium and redox regulation of signalling transcription factors in the alveolar epithelium

Cellular Signalling 14 (2002) 799 – 810 www.elsevier.com/locate/cellsig

Review article

Oxygen homeostasis, thiol equilibrium and redox regulation of signalling transcription factors in the alveolar epithelium John J. Haddad* Neuroscience Research Laboratory, Department of Anesthesia and Perioperative Care, University of California at San Francisco, Medical Sciences Building S-261, San Francisco, CA 94143-0542, USA Received 30 October 2001; accepted 28 January 2002

Abstract There is growing evidence linking the pathophysiology of lung disease to an imbalance state of reduction – oxidation (redox) equilibrium. The therapeutic potential of glutathione, an ubiquitous sulfhydryl thiol, and its immunopharmacological properties in the airway epithelium bears clinical consequences for the paediatric treatment of respiratory distress (RD). Dynamic variation in alveolar pO2 and its effect on redox state may impose a direct role in modulating the pattern of gene expression in lung tissues and, accordingly, could be pivotal in determining cellular fate under these conditions. Hypoxia-inducible factor-1a (HIF-1a) and nuclear factor-nB (NF-nB) are redox-sensitive transcription factors of particular importance because their differential activation by reducing and oxidizing signals, respectively, regulate the expression/ suppression of O2-responsive genes. The regulation of these transcription factors, therefore, which is redox sensitive, is consistent with their roles in coordinating adaptive homeostatic responses to oxidative stress. Functionally, the relationship between O2, glutathione biosynthesis and transcription factor activity bears typical implications for the pattern of cellular survivorship and alveolarization on exposure to O2-linked stresses. In this review, I discuss (1) the HIF-1a/NF-nB responsiveness to dynamic changes in pO2 characteristic of the transition period from placental to pulmonary-based respiration, (2) the capacity of the alveolar epithelium to engage in glutathione biosynthesis and redox shuttling, effectively forming a feedback mechanism governing gene expression, (3) the restitution of antioxidant/prooxidant equilibrium following oxidative challenge and its dependency on the adaptive coordination of responses between redox-associated signalling pathways controlling apoptosis and genetic regulatory factors and (4) a likely association between oxidative stress and the evolution of an inflammatory signal through the pleiotropic O2-sensitive cytokines. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Apoptosis; Cytokine; Glutathione; HIF-1a; NF-nB; Pathophysiology; Redox equilibrium

Abbreviations: NAC, N-acetyl-L-cysteine; ARE, AU-rich element; Bcl-2, B-cell leukemia/lymphoma-2; BCNU, 1,3-bis-(2-chloroethyl)-1-nitrosourea; BSO, L-buthionine-(S,R)-sulfoximine; CO, carbon monoxide; DHEA, dehydroepiandrosterone; ENO-1, enolase-1; EPO, erythropoietin; fATII, foetal alveolar type II epithelial cells; GSH, L-g-glutamyl-L-cysteinyl-glycine; GSSG, glutathione oxidized disulfide; GSH-PX, glutathione peroxidase; GSSG-RD, glutathione reductase; GS, glutamyl synthase; g-GCS, g-glutamylcysteine synthetase; g-GCE, g-glutamylcysteine ethylester; G-CSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte – macrophage colony-stimulating factor; Hsp, heat shock protein; HO-1, heme oxygenase-1; H2O2, hydrogen peroxide; OH, hydroxyl radical; HIF-1a, hypoxia-inducible factor-1a; InB, inhibitory-nB; IKK, InB kinase; Ign, immunoglobulin n light chain; iNOS, inducible nitric oxide synthase; IFN, interferon; IL, interleukin; IGF-2, insulin-like growth factor-2; IGFBP-1, IGF binding protein-1; ICAM-1, intracellular adhesion molecule-1; Ii, invariant chain; LPS, lipopolysaccharide; LT, lymphotoxin; M-CSF, macrophage colony-stimulating factor; MHC, major histocompatibility complex; MAPK, mitogen-activated protein kinase; MKK, MAPK kinase; MEKK, MKK kinase; MAdCAM-1, mucosal addressin cell adhesion molecule-1; NF-nB, nuclear factor-nB; NIK, NF-nB inducing kinase; LMP, proteasome subunit; PDTC, pyrrolidine dithiocarbamate; ROS, reactive oxygen species; ROOH, reactive peroxide; RK, reactivity kinase; redox, reduction – oxidation; RD, respiratory distress; O2
, superoxide anion; SOD, superoxide dismutase; TCR, T-cell receptor; TBP, TATA binding protein; TRX, thioredoxin; TF-IIB/D, transcription factor-IIB/D; TGF-3, transforming growth factor-3; TAP, transporter associated with antigen processing; TNF-a, tumor necrosis factor-a; VCAM-1, vascular cell adhesion molecule-1; VEGF, vascular endothelial growth factor. * Tel.: +1-415-476-8984; fax: +1-415-476-8841. E-mail address: [email protected] (J.J. Haddad). 0898-6568/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 8 9 8 - 6 5 6 8 ( 0 2 ) 0 0 0 2 2 - 0

800

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

Living organisms from prokaryotes to complex eukaryotes have developed an elaborate sequence of adaptive mechanisms in order to maintain oxygen (O2) homeostasis. In mammals, for instance, the development of sophisticated systems, such as the respiratory and cardiovascular systems, allows the capturing and appropriate distribution of O2 as a substrate for oxidative phosphorylation, the major biochemical reaction for the derivation of ATP [1– 4]. Within the context of an integral physiological system, tightly regulating O2 concentrations is, therefore, a necessary optimum in the face of a continuously changing environment.

[2,4,18,19]. Apparently, any perturbations in maintaining homeostatic mechanisms in response to changes in O2 levels are critical, thereby determining cellular characteristic integrity. The clinical, biochemical and histological responses of the lungs to such variations consequently characterize the efficiency and specificity of the antioxidant system in combating stress [12,17,20,21]. In certain lung pathophysiological conditions, for instance, oxygenation of terminal airways becomes uneven, such that this temporal and spatial variance in O2 abundance essentially determines the survivorship of lung cells via O2-dependent activation of cell regulators and genes critical to defending their structural/ functional characteristics [3,11,18,21,22].

1.1. The airway epithelium: a competitive barrier

1.3. Oxidative stress and pathophysiology

The process of breathing is the initiating step of respiration, which includes the movement of O2 from the lungs to the tissues and the process of cellular respiration, which produces ATP. The lungs are cone-shaped organs divided into lobes, each of which is subdivided into lobules having bronchioles that serve many alveoli. Each alveolar sac is made up of simple squamous epithelium surrounded by blood capillaries, thereby allowing for efficient and rapid gas exchange across this barrier [5 – 7]. The airway epithelium, however, is not an inert barrier but a major participant in signalling mechanisms in development and pathophysiology [2 – 5,8 – 11]. Any damage caused to the airway epithelium, therefore, can adversely affect its normal physiology and regulatory processes [2,3,8 – 12]. The major functions of the airway epithelium include the following: (1) a physiological barrier to diffusion and osmotic processes, (2) an integral metabolic capacity to synthesize and degrade chemical components endogenously produced or exogenously introduced and (3) a secretory ability evident by the production of mucous, cytokines and chemokines, hormones, growth factors and enzymes. This underlies, therefore, the significance of a physiologically competent epithelium where, in the case of metabolic failure or noxious damage, the situation might lead to abnormalities in the normal development and functioning of the lungs [8 – 12].

Accumulating body of evidence in recent years has linked the pathogenesis of human diseases to increased oxidative stress [15,19,20]. In particular, reactive O2 species (ROS), which are partially reduced metabolites, may contribute to alveolar – capillary membrane perturbations and the development of lung injury [23 – 28]. Oxidative cell injury involves the modification of cellular macromolecules by ROS, often leading to cell death and lysis of sensitive cells, resulting in the microvascular and alveolar perturbations [13,17,22]. Oxidative stress appears to increase lung antioxidants in some experimental models, and hypoxia and hyperoxia modulate foetal lung growth [11,21 – 24]. Furthermore, there is growing evidence supporting the concept of a cross-talk between oxidative stress and the up-regulation of a proinflammatory signal via the participation of cytokines [24 – 29]. These peptide hormones are major participants in the pathophysiology of respiratory distress (RD) and have been recognized as signalling molecules responsive to dynamic variation in pO2. This integral association between oxidative stress and an inflammatory state may affect cellular reduction – oxidation (redox) potential, thereby imposing a direct role in modulating the pattern of gene expression in lung tissues and determining cell fate and survival.

1. Introduction and background

1.4. Redox equilibrium and oxygen-sensitive transcription factors

1.2. Oxygen homeostasis and adaptation mechanisms All forms of aerobic life are thus faced with the threat of oxidation from molecular O2 and have developed elaborate mechanisms of antioxidant defenses to cope with this potential problem [13 – 16]. In particular, key developmental changes in the late gestational (pre-term) lung have evolved to allow production of surfactants and enzymatic and nonenzymatic antioxidants in preparation for the first breaths at birth [2,4,17]. Moreover, the chronology of maturation of the lung antioxidant system seems to be paralleling that pattern of prenatal maturation of the surfactant system, thereby pointing out to the developing stages fetuses undergo in preparation to be born into an O2-rich environment

Oxygen- and redox-sensitive transcription factors are likely to be regulated by O2 availability, bind specific DNA consensus sequences and activate the expression of several genes, particularly those controlling adaptive homeostasis [1,2,8 – 10,22,30 – 32]. Amongst such factors, the hypoxia-inducible factor-1a (HIF-1a) and nuclear factornB (NF-nB), whose activation states are differentially regulated in oxidative stress, remain of particular importance [1,2,9,10,16]. HIF-1a, first identified in vitro as a DNA binding activity expressed under hypoxic conditions [33], has its concentration and activity increased exponentially on lowering O2 tensions over physiologically relevant ranges [1,2,32]. The ubiquitous activation of HIF-1a is thus con-

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

sistent with the significant role that this transcription factor plays in coordinating adaptive responses to hypoxia. NFnB, on the other hand, first identified as a transcriptional factor that regulates antibody release in B-cells [34], is central to the regulation and expression of stress response genes in the face of inflammatory and oxidative challenges [2,9,10,22,31]. This review discusses the regulation of redox-sensitive transcription factors, HIF-1a and NF-nB, and elaborates on their responsiveness to dynamic changes in DpO2. Moreover, a relationship is established between antioxidant defense mechanisms involving glutathione homeostasis and the up-regulation of HIF-1a and NF-nB in the perinatal lung. Of note, adaptive apoptosis signalling mechanisms during the transition from placental to pulmonary-based respiration are investigated with the potential regulatory role of reactive thiols. Finally, evidence is provided for an integral and intimate relationship between oxidative stress and the augmentation of an inflammatory signal, mediated by the O2- and redox-responsive pleiotropic cytokines.

2. Oxygen responsiveness and the regulation of transcription factors: molecular aspects

801

by stimulating erythropoiesis. VEGF is a transcriptional regulator of vascularization. Glycolytic transporters and enzymes increase the efficiency of anaerobic generation of ATP. It is expected that any reduction of tissue oxygenation in vivo and in vitro would, therefore, provide a mechanistic stimulus for a graded and adaptive response mediated by HIF-1a [1,16,22]. The oxygen- and redox-sensitive pathways mediating the regulation of HIF-1a are schematized in Fig. 1. The array of genes directly controlled by HIF-1a is given in Table 1. The activation of HIF-1a is consistent with its role in coordinating adaptive homeostatic responses under hypoxic conditions [1,8,14,16,22,30]. The alveolar epithelium, in particular, is genetically responsive to changes in O2 availability, as it functionally acts as more than just an inert barrier for gas exchange [2 – 7,9,11,27,28]. The O2 history of the perinatal epithelium is continuously laying at c 23– 30 Torr, which represents the O2 transfer capacity of the umbilical vein in utero [1 – 7]. The reasoning is that a relative hyperoxic shift during birth transition from placental to pulmonary-based respiration would rather constitute such a potential signalling mechanism for the activation of O2-sensitive transcriptional factors, including HIF-1a

2.1. Oxygen homeostasis, redox signalling and the regulation of HIF-1a The process of capturing O2 from the environment and its distribution as a substrate for biochemical conversion, either by oxidative phosphorylation or by glycolysis, has been conserved through evolution by the development of advanced systems. In order to maintain O2 homeostasis, a process that is essential for survival, O2 concentration must be tightly maintained within a narrow physiological range [15,16,30]. However, this system may fail thereby causing the subsequent induction of hypoxia and resulting in a failure to generate sufficient ATP to sustain metabolic activities or promoting hyperoxia, which crucially contributes to the generation of ROS. ROS at low concentrations might participate in signalling pathways, but their excessive generation could be cytotoxic and often cytocidal [1 –3,15 – 17,19,22,30]. Adaptive responses to hypoxia, therefore, involve the regulation of gene expression, at least in part by HIF-1a, whose expression, stability and transcriptional activity were reported to increase exponentially on lowering pO2 [1 – 3,9,16,22,30,34,35]. In hypoxia, multiple systemic responses are induced, including angiogenesis, erythropoiesis and glycolysis. HIF-1a is a crucial mediator for increasing the efficiency of O2 delivery through erythropoietin (EPO) and vascular endothelial growth factor (VEGF) [36 – 40]. A wellcontrolled process of adaptation parallels this mechanism to decreased O2 availability through expression and activation of glucose transporters and glycolytic enzymes [1,8,16,25]. EPO is responsible for increasing blood O2-carrying capacity

Fig. 1. Hypoxia signal transduction. Reduction of cellular O2 concentration is associated with redox changes that lead to altered phosphorylation of HIF-1a, which increases its stability and transcriptional activity, resulting in the induction of downstream gene expression. Putative inducers (horizontal arrows) and inhibitors (blocked arrows) of different stages in the proposed pathway are indicated.

802

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

Table 1 Direct HIF-1 target genes Glucose/energy metabolism and cell proliferation/viability Adenylate kinase-3, aldolase A, aldolase C, enolase-1 (ENO-1), glucose transporter-1, glucose transporter-3, glyceraldehyde-3-phosphate dehydrogenase, hexokinase-1, hexokinase-2, insulin-like growth factor-2 (IGF-2), IGF binding protein-1 (IGFBP-1), IGFBP-3, lactate dehydrogenase A, phosphoglycerate kinase-1, pyruvate kinase M, p21, transforming growth factor-3 (TGF-3) Erythropoiesis and iron metabolism Ceruloplasmin, EPO, transferrin, transferrin receptor Vascular development/remodeling and vasomotor tone Adrenergic receptor, adrenomedullin, endothelin-1, heme oxygenase-1 (HO-1), nitric oxide synthase-2, plasminogen activator inhibitor 1, VEGF, VEGF receptor FLT-1

[1,2,9,19,16,22]. Maintenance of epithelial cells in vitro at static pO2 (23 Torr) allowed maximum induction of HIF-1a protein expression and subsequent nuclear localization and activation, a phenomenon sustained with a shift towards the early postnatal distal lung pO2 at 23 ! 70– 100 Torr [2,9]. Of note is the observation that HIF-1a DNA consensus binding activity ex vivo was sustained with the first respiratory movements but lost thereafter [2,9]. This provided functional evidence that the sensitivity of HIF-1a to subambient pO2 is responsively heightened during birth, which

would support the notion that this transcription factor might participate in signalling mechanisms existing beyond a strictly hypoxic environment. The restitution of redox equilibrium in the face of an oxidative challenge requires an adaptive cross-talk between signalling pathways sensing dynamic variations in pO2 and genetically regulated transcription factors. Glutathione-associated metabolism is crucial for providing an equilibrium interface between oxidative stress and adaptive responses of cytoprotection. L-g-glutamyl-L-cysteinyl-glycine (GSH) is an ubiquitous tripeptide thiol that serves essential functions in aerobic species [2,9,41 –43]. Synthesized by the action of the rate-limiting enzyme g-glutamylcysteine synthetase (gGCS), GSH uniquely provides a functional cysteinyl moiety that is responsible for much of the diverse properties of GSH. Although GSH biosynthesis and recycling occurs intracellularly, its catabolism is extracellular. Endogenously produced radicals such as hydrogen peroxide (H2O2) are effectively reduced by the selenium-dependent GSH peroxidase in the presence of GSH as a substrate. During this reaction, GSH is converted into glutathione oxidized disulfide (GSSG), which is recycled back to 2GSH by GSSG reductase at the expense of NADPH/H + , thus forming what is known as a redox cycle (Fig. 2) [2,9,41 –43]. Furthermore, GSH and GSSG do not exist in equilibrium ratio under normal conditions, because the majority ( > 95%) of glutathione is in the reduced form. Shifting redox equili-

Fig. 2. The schematic of redox cycle shows the relationship between antioxidant enzymes and glutathione. Glutathione (GSH) is synthesized from amino acids by the action of g-GCS, the rate-limiting enzyme, and glutamyl synthase (GS). This reaction requires energy, is ATP limited and is specifically inhibited at the level of g-GCS by BSO. GSH undergoes the glutathione peroxidase (GSH-PX) coupled reaction, thereby detoxifying ROS such as H2O2. A major source of H2O2 is the biochemical conversion of O
 2 by the action of superoxide dismutase (SOD). During this reaction, GSH is oxidized to generate GSSG, which is recycled back to GSH by the action of glutathione reductase (GSSG-RD) at the expense of reduced nicotinamide (NADPH/H + ), thus forming the redox cycle. The reduction of the glutathione pathway is blocked by the action of BCNU. The major source of NADPH/H + comes from the conversion of glucose, a reaction blocked by dehydroepiandrosterone (DHEA).

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

brium, therefore, in favour of a reducing or oxidizing state is potentially of particular significance, as this would dictate the pattern of gene expression of HIF-1a over relevant shifts in pO2 [1,2,9,16,22,30]. Antioxidant/prooxidant equilibrium differentially regulates HIF-1a redox sensitivity [1,2,8,9,30]. For example, selective and irreversible inhibition of g-GCS by L-buthionine-(S,R)-sulfoximine (BSO) [42,43] abrogated hypoxiainduced nuclear localization and stabilization of HIF-1a [2,9]. In parallel, BSO reduced HIF-1a DNA binding activity regardless of the direction of O2 shift [1,2,9,16]. Moreover, there is substantial evidence supporting the notion that g-GCS inhibition is accompanied by intracellular accumulation of superoxide anion (O2
) and H2O2 [16,26,41 –43], believed to play a key role in destabilizing HIF-1a. Although it cannot be inferred that the effect of BSO is exclusively ROS dependent, it appears that maintaining GSH equilibrium and, by inference, the shuttling between reduction and oxidation states, is a prerequisite for HIF-1a activation in hypoxia and under a relatively mild shift in pO2 [1,2,9,16,30]. This assumption is reinforced with the observation that N-acetyl-L-cysteine (NAC), an antioxidant and a precursor for L-cysteine, the rate-limiting amino acid in the biosynthesis of GSH, imposed a reducing environment, thereby protracting HIF-1a stability in the cytosol and subsequently favouring its translocation and activation with a relatively hyperoxic shift [1,2,9,16]. Thus, shifting the equilibrium ratio of GSH/GSSG in favour of a reduction state substantiality allowed HIF-1a stabilization and activation independent of pO 2 , a phenomenon uncoupled from the classical hypoxic induction of HIF-1a [1,2,9,16]. This raised the issue of whether reversing GSH/GSSG equilibrium would interfere in the hypoxic induction of HIF-1a. To meet this end, dithiocarbamates, nonthiol antioxidants, including pyrrolidine dithiocarbamate (PDTC), were enrolled in experiments to span the activation of HIF-1a with ascending DpO2 regimen [1,2,9]. PDTC affects redox potential by its ability to scavenge radical species (a reduction property) and to oxidize directly GSH and related thiols (an oxidation property) [2,9,44]. Although PDTC favoured a GSSG/GSH equilibrium and the stabilization of HIF-1a protein in the cytosolic compartment, it failed to induce its activation under nonhypoxic conditions, ostensibly due to the generation of thiol radicals and thiuram disulfides that are produced during PDTC antioxidant reaction, thus leading to oxidation of GSH [1,2,9,16,30]. Imposing an oxidizing environment by the rapid accumulation of GSSG in the nucleus adversely affects HIF-1a activation, because a reducing equilibrium is necessary for its stabilization and activation [1,9,22,33]. This notion is further supported by rather unequivocal evidence suggesting that HIF-1a consensus binding could be facilitated by increasing GSH/GSSG equilibrium in isolated nuclei [1,2,9,16]. In keeping with this notion, selective inhibition of GSSG reductase by 1,3-bis-(2-chloroethyl)-1-nitrosourea (BCNU)

803

was shown to induce intracellular accumulation of GSSG and subsequently abrogate hypoxia-induced HIF-1a activation [1,2,9,16,33]. It is very likely, therefore, that HIF-1a O2 responsiveness resides over a permissive range of antioxidant buffering capacities, with the demonstration that antioxidant/prooxidant equilibrium effectively uncoupled HIF1a activity from the normal pattern observed in response to variations in pO2. Correlation analysis of the relationship between HIF-1a activation and up-regulation of redoxsensitive enzymes [2,9] is shown in Fig. 3. 2.2. Oxygen homeostasis, redox signalling and the regulation of NF-nB One important and widely investigated transcription factor, which is a major participant in signalling pathways governing cellular responses to environmental (oxidative) stresses, is NF-nB. First identified as a factor that regulates the expression of the immunoglobulin n light chains (Ign) in B-cell lymphocytes [34], NF-nB is recognized as central to regulating the expression of stress-response genes in oxidative challenge [2,9,10,22,31,44,45]. In resting conditions (inactive state), NF-nB exists as homo- or heterodimers of a family of structurally related proteins. The dimers of NF-nB are sequestered in the cytosol through noncovalent interactions with an inhibitory protein termed inhibitory-nB (InB), of which the most important may be InB-a [46 – 49]. The translocation and activation of NF-nB in response to various stimuli, such as cytokines [interleukin (IL)-1 and tumor necrosis factor-a (TNF-a)], microbial agents [lipopolysaccharide (LPS)], oxidative challenge (ROS) and irradiation (UV and g-rays), are sequentially organized at the molecular level. NF-nB activation occurs through the signal-induced phosphorylation of an upstream kinase, termed InB kinase (IKK), by NF-nB inducing kinase (NIK). Phosphorylation of InB marks its ubiquitination and subsequent proteolytic degradation, thereby unmasking the nuclear localization signal and allowing nuclear translocation of NF-nB complex [46 – 49]. A schematic summarizing the potential pathways regulating NF-nB is given in Fig. 4. The array of genes directly controlled by NF-nB is given in Table 2. Independent factors, however, have been recently recognized as alternative pathways regulating the activation of NF-nB. For instance, direct phosphorylation of RelA (p65), the major transactivating member of the nB family, in one of two of its transactivation domains, has been shown to regulate NF-nB activation [45,48]. A further mechanism was subsequently proposed for NF-nB regulation, with the discovery of transcription factor-IIB/D (TF-IIB/D) and TATA binding protein (TBP) as two important regulators of NF-nB transcriptional activity [45]. The dominant-negative form of mitogen-activated protein kinase (MAPK) (p38) expression vector abrogated the interaction of TFIID/TBP with a cotransfected His-p65 fusion protein and selective inhibition of p38 by SB-203580 reduced TF-IID/

804

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

Fig. 3. Correlation regression analysis on HIF-1a percent binding activation and the induced activity of redox-sensitive enzymes. Foetal alveolar type II epithelial cells (fATII) epithelia: (A) A significant correlation exists between GSH-PX and HIF-1a activity (r = .963, P < .01). (B) A significant correlation exists between GSSG-RD and HIF-1a activity (r = .964; P < .01). (C) No correlation has been observed between HIF-1a activity and the activity of g-GCS (r = .020; P >.05). (D) A significant correlation exists between glutathione synthase (GS) and HIF-1a activity (r = .984; P < .01). Neonatal lungs: (E) A significant correlation exists between GSH-PX and HIF-1a activity (r = .538; P < .05). (F) Negative correlation has been observed between HIF-1a activity and the activity of GSSG-RD (r =
.349; P < .05). (G) A significant correlation exists between HIF-1a activity and the activity of g-GCS (r = .649; P < .01). (H) A significant correlation exists between HIF-1a activity and the activity of GS (r = .665; P < .01).

TBP in vitro [45,48]. Finally, modulating intracellular redox equilibrium constitutes a potential mechanism that can manipulate the localization and activation of NF-nB [2,9,10,45], the subject that will be the focus of subsequent discussion. NF-nB nuclear abundance and DNA consensus binding activity were marginal under static pO2 tensions in alveolar epithelial cells [2,9]. The relative hyperoxic shift, however, augmented its translocation and subsequent activation. While spanning the activation profile of HIF-1a and NFnB at various ascending DpO2 regimens, an observation of interest could be particularly noted. The protein expression and DNA binding activity of either factor are prominent under mild hypoxic conditions with approximately identical kinetics [2,9]. This observation suggested a possible crosstalk between these two factors at this particular DpO2, although the molecular basis of this mechanism has yet to be ascertained. In a manner similar to its activation state in vitro, NF-nB protein expression and consensus binding

maximized with oxygenation, especially during the first few days after birth [2,9]. Taken together, these observations indicate that HIF-1a and NF-nB are sufficiently tuned to mediate the changeover from hypoxic to hyperoxic induction of regulatory genetic factors at different pO2 tensions resembling those falling within the range expected during the birth transition from placental to pulmonarybased respiration. Experimental depletion of glutathione by dose-dependent inhibition of g-GCS using BSO resulted in a step-wise inactivation of NF-nB under each activating DpO2 regimen [2,9]. Intriguingly, while HIF-1a appeared maximally active in hypoxia in a low total glutathione environment [1,2,9,16], its activity was substantially more responsive to glutathione depletion compared to all active ranges of NF-nB explored where the control glutathione content was two- to threefold greater [2,9]. Although it cannot be inferred that the effect is specifically ROS dependent, it appears clear that maintenance of the glutathione pool and the shuttling of electrons

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

805

Fig. 4. Hyperoxia and other stress conditions signal transduction. Environmental stress, including hyperoxia, is associated with the activation of upstream kinases that diverge at activating either the p38 mitogen-activated protein kinase pathway or the InB/NF-nB pathway. Translocation of NF-nB into the nucleus triggers the activation of stress-response genes. Abbreviations: ARE, AU-rich element; Hsp27, heat shock protein 27; MEKK, MAPK kinase kinase; MKK, MAPK kinase; RAC/PAC, small G-coupled proteins; RK, reactivity kinase (see discussion for further details).

between reductant and oxidant components of this pathway are prerequisites for transcription factor activation under any given O2 profile. Pretreatment with NAC, for example, has shown no stimulatory effect on the abundance or activation kinetics of NF-nB but rather suppressed NF-nB nuclear localization abundance and activation [2,9,10,31]. As oxidants and ROS play a key role in the biochemical pathways leading to NF-nB activation, it is evident that NAC is acting as a free radical scavenger antioxidant, a role that supersedes, or is distinct from, its ability to increase intracellular glutathione stores. Recent evidence suggested that regulatory transcription factors, such as NF-nB, play critical roles in the early events controlling the molecular response to ROS [44 – 48]. It has also been suggested that the activation of NF-nB by a wide variety of agents can be blocked by NAC, suggesting that the production of reactive metabolites may act as a common pathway for a diverse range of stimuli [2,9,44,45]. This suggests that the ability of NAC to suppress the activation of NF-nB is likely to involve its antioxidant potential, where the circuits involved in NFnB activation are previously schematized [2,9]. Apparently, the mode of action of NAC may not be the same in different experimental models and clinical situations. Consequently, extensive research is warranted to investigate the mechanisms of action of NAC that may offer a clue for elucidating its differential effects [2,9,10,49]. PDTC has been proven a potent inhibitor of NF-nB in the alveolar epithelium [2,9] and in other tissues [44,51]. The inhibitory effects are shown at both the levels of nuclear

abundance and activation, suggesting posttranslational instability and interference with the capacity of NF-nB to bind DNA and activate oxidant-induced stress genes. The distinguishable properties of PDTC mediating its antioxidant and/or prooxidant effects on NF-nB could be highTable 2 Direct NF-nB target genes Cytokines/growth factor IL-1a, IL-1h, IL-2, IL-3, IL-6, IL-8, IL-12, tumor necrosis factor (TNF)-a, lymphotoxin (LT)-a, interferon (IFN)-h, granulocyte colony-stimulating factor (G-CSF), macrophage colony-stimulating factor (M-CSF), granulocyte – macrophage colony-stimulating factor (GM-CSF) Cytokine receptors IL-2 receptor a-chain (IL-2Ra) Stress proteins Serum amyloid A protein (SAA), complement factors B, C3 and C4, a1-acid glycoprotein Adhesion molecules Intracellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), mucosal addressin cell adhesion molecule-1 (MAdCAM-1), E-selectin Immunoregulatory molecules Immunoglobulin n light chain (Ign), major histocompatibility complex (MHC class I and II), T-cell receptor (TCRa and h), h2-microglobulin, invariant chain (Ii), transporter associated with antigen processing (TAP-1), proteasome subunit (LMP-2), inducible nitric oxide synthase (iNOS), inhibitory nB (InB), p53

806

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

lighted as follows: (1) PDTC is much more a potent inhibitor of NF-nB than NAC given that its inhibitory working concentrations (10 –100 AM) are much lower than that of NAC (1 –50 mM). In this respect, PDTC is 100- to 500-fold more potent than NAC. (2) PDTC decreases the GSH/GSSG ratio by elevating GSSG due to oxidation of GSH. Thus, it is likely that PDTC is inhibiting NF-nB activation by acting as a prooxidant, although the possibility of acting as an antioxidant while mediating this inhibition cannot be excluded, as ROS are key mediators of NF-nB activation. This is further supported by the observation that phosphorylation of InB at specific serine residues can be inhibited by dithiocarbamates, pointing to the possibility that NF-nB translocation and subsequent activation is mediated by ROS, which might induce a cytosolic kinase activity [21,33,24]. The assumption of GSSG-mediated inhibition of NF-nB has been verified by the observations that the oxidation of GSH to GSSG induces the formation of NF-

nB/disulfide complex, thereby inhibiting DNA binding [2,9,10,21,22,45]. It is speculated, however, that a derivative and/or a metabolite of PDTC metabolism is generated under these conditions that is responsible for the outcome of this effect, as it is very unlikely that PDTC would directly interact with NF-nB to inhibit its activation, because addition of PDTC (50 AM) to nuclear extracts of cell cultures exposed to various DpO2 failed to inhibit NF-nB activation in mild hyperoxia [2,9,44]. (3) GSSG formed by the prooxidant effect of PDTC drives the formation of an oxidation equilibrium that renders NF-nB inactive [21,22,45]. This is supported by the observation that BCNU, which induced the accumulation of intracellular GSSG at the expense of GSH, reduced oxidant-dependent NF-nB activation [2,9,21,45]. Although oxidizing conditions are necessary for the activation of NF-nB in the cytosol, thereby allowing optimum translocation due to dissociation from InB, NF-nB must be maintained in a reduced state in the

Fig. 5. Correlation regression analysis on NF-nB percent binding activation and the induced activity of redox-sensitive enzymes. fATII epithelia: (A) A significant negative correlation exists between GSH-PX and NF-nB activity (r =
.624; P < .01). (B) Negative correlation exists between GSSG-RD and NFnB activity (r =
.845; P < .01). (C) Positive correlation has been observed between NF-nB activity and the activity of g-GCS (r = .445; P < .05). (D) A significant negative correlation exists between GS and NF-nB activity (r =
.624; P < .05). Neonatal lungs: (E) A significant positive correlation exists between GSH-PX and NF-nB activity (r = .614; P < .05). (F) Negative correlation has been observed between NF-nB activity and the activity of GSSG-RD (r =
.570; P > .05). (G) A significant positive correlation exists between NF-nB activity and the activity of g-GCS (r = .567; P < .05). (H) A significant positive correlation exists between NF-nB activity and the activity of GS (r = .486; P < .05).

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

nucleus for DNA binding to occur [2,9,10,21,22,30,45 – 48]. This assumption is based on the ability of an increased ratio of GSSG/GSH to inhibit NF-nB activation in vitro [2,9,10]. GSSG elevation, therefore, promotes oxidation of protein cysteinyl thiols, shifting the equilibrium of thiol-disulfide exchange significantly in the direction of mixed disulfide formation and ultimately changing protein conformation. Consequently, the binding capacity of regulatory transcription factors is dramatically altered [21,22]. The possibility that PDTC is interfering with the translocation of NF-nB to the nucleus cannot be excluded. Rather, what is evident is that the inhibitory effects on NF-nB support the hypothesis that dithiocarbamates act as prooxidants by elevating the concentration of GSSG, which is capable of direct (by forming a mixed NF-nB-thiol inactive complexes) or indirect (creating an oxidized environment in the nucleus) prevention of NF-nB activation [2,9,21,22,45,47]. Correlation analysis of the relationship between NF-nB activation and up-regulation of redox-sensitive enzymes [2,9] is shown in Fig. 5.

3. Oxygen-sensitive and redox-mediated pathways regulating apoptosis in the lungs The regulation of apoptosis or programmed cell death during the transition from placental- to pulmonary-based respiration is not well characterized in the perinatal epithelium. The kinetics of this process centres on the activation of redox-sensitive transcription factors and cofactors, which modulate the pattern of the molecular response to oxidative stress [2,9,10,21,22,30]. Oxyexcitation (DpO2)-dependent suppression of B-cell leukemia/lymphoma-2 (Bcl-2), for example, thereby favouring the promotion of Bax, suggested the involvement of an O2-responsive pathway [11,50]. The likely pathway implicated could involve cell cycle arrest through the activation of p53, consistent with the hypothesis that hyperoxic injury is characterized by a complex pattern during which the alveolar surface is damaged, denuded and repopulated [50,51,52,53]. It is postulated that chemioxyexcitation (DpO 2/ROS) triggers a process eventually leading to suppression of Bcl-2 and upregulation of Bax via a p53-linked pathway [11,50– 55]. The regulation of the process of apoptosis is ROS and redox sensitive [57 – 63]. Selective inhibition of GSH biosynthesis, for example, up-regulated Bax and p53 regardless of the direction of pO2 [11,50]. Because the depletion of GSH, a major antioxidant, could trigger apoptosis, it is likely that ROS are involved in the induction of Bax and p53 [55,60 – 62]. Of note, GSH depletion also induced the intracellular accumulation of Bcl-2, although redox disequilibrium seems to be in favour of apoptosis [64]. These data indicate that the loss of intracellular GSH might induce oxidative stress, possibly through the formation of ROS, thereby altering redox potential in favour of oxidation equilibrium [55 – 63]. This subsequently leads to up-regu-

807

lation of the downstream agonists of apoptosis and cell cycle arrest regulators. In this respect, effective pharmacological intervention by NAC may indicate modulation of redox equilibrium in association with oxidative stress [2,9,11,50,55 –62]. NAC is an antioxidant thiol, which, after uptake, deacylation and biochemical conversion to GSH, may function as a redox buffer and/or effective ROS scavenger [2,9,49,50]. The inhibitory effect of NAC is evident through suppression of apoptosis in response to oxyexcitation by down-regulating Bax/p53 and inducing Bcl-2, an effect mimicked by g-glutamylcysteine ethylester (g-GCE), a cell-permeant GSH precursor [11,50]. The antioxidant potential of NAC/g-GCE, therefore, may contribute to inhibition of oxidative stress-mediated apoptosis and expression of selective agonists. The potential involvement of the redox-sensitive transcription factor, NF-nB, in apoptosis pathways is well documented [2,9,10,44,45,65,66]. For example, overexpression of the dominant-negative form of InB-a, a major cytosolic inhibitor of NF-nB, promoted apoptosis in vitro [67,68]. Moreover, mice bearing the knockout gene RelA
/
(p65
/
) show more susceptibility to agents stimulating apoptosis than the wild type [69], suggesting a potential role for NF-nB in regulating this process. Nonselective inhibition of NF-nB by PDTC, an antioxidant/prooxidant molecule that decreases the GSH/GSSG ratio [2,9], led to upregulation of Bax independent of p53 [11,50]. However, selective inhibition of NF-nB by sulfasalazine (SSA) induced up-regulation of Bax in a p53-dependent mechanism [11,50]. These observations suggested that NF-nB exhibits an antiapoptotic potential, presumably through regulating responsive genes particularly involved in controlling apoptosis and cell cycle events. Schematic pathways governing O2 and redox sensitivity of apoptosis is shown in Fig. 6. The maturation of the developing lung at different stages of gestation and postnatally occurs within widely different pO2 [2,9,11,50]. The successful and smooth transition from placental to pulmonary-based respiration incurs a relative hyperoxic shift in the lung, an event that constitutes a potential signalling mechanism for determining the potential for apoptosis. The antiapoptotic potential postnatally was dramatically suppressed, especially during the first early hours after birth, in accord with the decline in Bcl-2 expression with ascending DpO2 in vitro [11]. Of particular interest, the proapoptotic potential, resembled by Bax, was not prominently sufficient to exceed the antiapoptotic potential, given that the equilibrium ratio of Bcl-2/Bax remained steadily constant at birth [11,50]. This challenged the assumption that Bax is the major cofactor involved in regulating apoptosis ex vivo. Accordingly, the focus was to determine whether there is a particular involvement of p53, a cell cycle regulator, under which Bcl-2 and Bax are transcriptionally controlled [51]. The birth transition period predominantly up-regulated the expression of p53 in a timedependent manner, indicating that a p53-dependent pathway is involved in apoptosis in the developing lung [11,50].

808

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

Fig. 6. Potential redox and ROS convergent pathways reflecting an intimate relationship in mediating gene activation/repression events, which regulate cell cycle events and the propagation of apoptosis.

Whereas the depression of apoptosis antagonists is concomitant with up-regulating the agonist expression through a p53-dependent pathway in vitro [11], apoptosis in the perinatal developing lung is p53 dependent but possibly Bax insensitive [50].

4. Oxygen-sensitive and redox-mediated pathways regulating cytokines Cytokines act as major participants in mediating molecular responses in physiology and pathophysiology [25,26,70 – 74]. There is accumulating evidence suggesting that the conventionally known ‘‘proinflammatory’’ cytokines are observed as O2-sensitive mediators, indicating that the alveolar epithelium integrates O2-linked pathways mediated by cytokines via a ROS-dependent mechanism [18,25,26,70 – 74]. A direct contact of the epithelium with fluctuations in pO2 at the alveolar space blood barrier during pulmonary-based ventilation is likely, therefore, to up-regulate cytokines. This is strengthened by the evidence implicating the epithelium in a front-line defense strategy activated in preparation for birth into an O2-rich environment [2,9,10,11,50,70 –74]. The generation of ROS might induce pulmonary damage. However, moderate oxidative stress can induce a signalling mechanism mediated, at least in part, by cytokines [70 – 75]. ROS induced proinflammatory cytokine biosynthesis and this response was abrogated by selective antioxidants, suggesting an integral role of

endogenous ROS [25,26,70– 75]. As such, cytokines could form a pivotal link in ROS-dependent pathways, leading to the activation of redox-sensitive transcription factors whose up-regulation determines the specificity of the cellular response to oxidative stress [9,10,21,22,30]. Thus, dynamic variation in pO2 regulates the release of cytokines through a ROS-dependent pathway, thereby bearing consequences for the paediatric treatment of RD in clinical O2 therapy where cytokines are potential participants in pathophysiology. The observation that intracellular ROS participate in cytokine signalling suggested that they might be redox sensitive and responsive to dynamic variations in pO2. There is growing evidence implicating an association between oxidative stress and the development of a proinflammatory state, thereby placing more demand on the utilization of intracellular GSH [2 –4,9 – 12,23 – 29,31,73 –75]. As such, the respiratory epithelium becomes more engaged in regulating enzymes involved in maintaining redox equilibrium [2,9,10,11,25,26,50]. While the glutathione biosynthetic machinery is overwhelmed in disease, the up-regulation of cytokines may contribute to acute exacerbation of the clinical symptoms [70,75,76]. Thiol regulation of proinflammatory cytokines, therefore, bears clinical relevance to the paediatric treatment of RDs, where cytokines are crucial elements in their pathophysiology. Glutathione biosynthesis is selectively blocked via a specific and irreversible inhibition of g-GCS [42,43]. This inhibition might lead to ROS accumulation, and as expected, their inappropriate disposition and intracellular localization can augment a proinflammatory

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

809

Fig. 7. Schematic model of thiol regulation of oxidative stress-induced cytokine secretion in the alveolar epithelium. The predominant form of intracellular glutathione is GSH, which is synthesized by g-GCS. 2-Oxothiazolidine-4-carboxylate (OTC), S-adenosyl-L-methionine (SAM) and NAC are major precursors of cysteine, the rate-limiting substrate in the biosynthesis of GSH, a pathway that is selectively blocked by BSO. BSO up-regulates the formation of intracellular ROS, which are major inducers of cytokine secretion. GSH is either rapidly exported, where the membrane-bound g-glutamyl transpeptidase (gGT) degrades it into its subcellular components to be used for resynthesis, or is converted by oxidation to GSSG by GSH-PX and its mimetic ebselen, at the expense of ROS, which are promptly detoxified into reactive peroxide (ROOH). The formation of GSSG and/or reduction of ROS down-regulate the chemioxyexcitation-dependent cytokine release. GSSG is rapidly recycled back into GSH by GSSG-RD, a pathway that is selectively blocked by BCNU, leading to accumulation of GSSG, which, along with PDTC, a precursor of GSSG, down-regulate the formation of ROS and abrogate the cytokine-dependent sequelae. Cytokines act as major participants in the pathophysiology and aggravation of the clinical symptoms of RDs.

signal [21,22,25,26,70,76]. Replenishing intracellular GSH stores, for example, favoured the reduction of ROS, subsequently suppressing the downstream cytokine-dependent pathway [70,76]. Therefore, selective inhibition of GSH biosynthesis can up-regulate cytokines in a ROS-dependent manner. Furthermore, blockading glutathione redox cycle may uncouple the ROS/cytokine pathway, and GSH precursors have the capacity to suppress intracellular ROS formation and the downstream cytokine-dependent pathway [22,24 – 26,29,70 – 76]. Thus, shifting redox potential in favour of reduction equilibrium negatively interferes with the capacity to up-regulate a proinflammatory signal, thereby bearing consequences for determining the survival of epithelial cells under conditions mimicking clinical O2 therapy [70 – 76]. Thiol-mediated pathways simulating functional mechanisms controlling inflammatory cytokines are schematized in Fig. 7.

potential. Antioxidant/prooxidant equilibrium interferes with the regulatory pathways controlling HIF-1a and NF-nB. A novel equilibrium among signalling agonists/antagonists of apoptosis is exhibited in response to DpO2/ROS, an event associated with the evolution of an inflammatory signal through up-regulating pleiotropic cytokines, recognized as O2-responsive and redox-sensitive molecules that participate in the physiology and pathophysiology of the perinatal lung.

Acknowledgements The author’s own publications are financially supported by the Anonymous Trust (Scotland), the National Institute for Biological Standards and Control (England), the National Institute of Health (NIH; USA), the Tenovus Trust (Scotland), the UK Medical Research Council (MRC, London) and the Wellcome Trust (London). Dr. John J. Haddad holds the George John Livanos prize (London).

5. Conclusions The perinatal epithelium responds to dynamic variations in pO2 by regulating the expression/activation of redoxsensitive transcription factors [2– 5,9,10,25,26,70 – 76]. This responsiveness is coupled to up-regulating glutathione biosynthesis as a major intracellular thiol with antioxidant

References [1] Semenza GL. Curr Opin Cell Biol 2001;13:167 – 71. [2] Haddad JJ, Land SC. Am J Physiol: Lung Cell Mol Physiol 2000; 278:L492 – 503. [3] MacNee W, Rahman I. Trends Mol Med 2001;7:55 – 62.

810

J.J. Haddad / Cellular Signalling 14 (2002) 799–810

[4] Thompson AB, Robbins RA, Romberger DJ. Eur Respir J 1985;8: 127 – 49. [5] Evans MJ, Van Winkle LS, Fanucchi MV, Plopper CG. Exp Lung Res 2001;27:401 – 15. [6] Campiche MA, Gautier A, Hernandez EI, Reymond A. Pediatrics 1963;36:976 – 94. [7] Scrapelli EM. Anat Rec 1998;251:491 – 527. [8] Semenza GL. Respir Res 2000;1:159 – 62. [9] Haddad JJ, Olver RE, Land SC. J Biol Chem 2000;275:21130 – 9. [10] Rahman I, Mulier B, Gilmour PS, Watchorn T, Donaldson K, Jeffrey PK, MacNee W. Biochem Pharmacol 2001;62:787 – 94. [11] Haddad JJ, Land SC. Biochem Biophys Res Commun 2000;271:257 – 67. [12] MacNee W, Rahman I. Am J Respir Crit Care Med 1999;160:S58 – 65. [13] Chabot F, Mitchell JA, Gutteridge JMC, Evans TW. Eur Respir J 1998;11:745 – 57. [14] Putman E, van Golde LMG, Haagsman HP. Lung 1997;175:75 – 103. [15] Fridovich I. J Exp Biol 1998;201:1203 – 9. [16] Gassmann M, Wenger RH. News Physiol Sci 1997;12:214 – 8. [17] Sen CK, Packer L. FASEB J 1996;10:709 – 20. [18] Connelly KG, Repine JE. Annu Rev Med 1997;48:429 – 45. [19] O’Brodovich HM, Mellins RB. Am Rev Respir Dis 1985;132:694 – 709. [20] Saugstad OD. Pediatr Pulmonol 1985;1:167 – 75. [21] Acarregui MJ, Snyder JM, Mendelson CR. Am J Physiol 1993; 264:L465 – 74. [22] Alder V, Yin Z, Tew KD, Ronai Z. Oncogene 1999;18:6104 – 11. [23] Rahman I, Swarska E, Henry M, Stolk J, MacNee W. Thorax 2000; 55:189 – 93. [24] Chen CY, Huang YL, Lin TH. Arch Biochem Biophys 1998; 356:127 – 32. [25] Haddad JJ, Safieh-Garabedian B, Saade´ NE, Kanaan SA, Land SC. Cytokine 2001;13:138 – 47. [26] Haddad JJ, Safieh-Garabedian B, Saade´ NE, Land SC. J Pharmacol Exp Ther 2001;296:996 – 1005. [27] Rahman I, MacNee W. Am J Physiol 1999;277:L1067 – 88. [28] Rahman I, Swarska E, Henry M, Davis M, O’Connor CM, FitzGerald MX, Greening A, MacNee W. Free Radical Biol Med 1999;27:60 – 8. [29] Nicod LP. Thorax 1993;48:660 – 7. [30] Bunn HF, Poyton OR. Physiol Rev 1996;76:839 – 85. [31] Rahman I. Biochem Pharmacol 2000;60:1041 – 9. [32] Semenza GL. Trends Mol Med 2001;7:345 – 50. [33] Wang GL, Semenza GL. J Biol Chem 1995;270:1230 – 7. [34] Sen R, Baltimore D. Cell 1986;46:705 – 16. [35] Jiang BH, Semenza GL, Bauer C, Marti HH. Am J Physiol 1996;271: C1172 – 80. [36] Mahon PC, Hirota K, Semenza GL. Genes Dev 2001;15:2675 – 86. [37] Semenza GL. J Clin Invest 2001;108:39 – 40. [38] Wykoff CC, Pugh CW, Harris AL, Maxwell PH, Ratcliffe PJ. Novartis Found Symp 2001;240:212 – 25. [39] Maxwell PH, Pugh CW, Ratcliffe PJ. Curr Opin Genet Dev 2001;11: 293 – 9. [40] Wykoff CC, Pugh CW, Maxwell PH, Harris AL, Ratcliffe PJ. Oncogene 2000;19:6297 – 305.

[41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76]

Haddad JJ, Land SC. FEBS Lett 2001;505:269 – 74. Meister A. J Biol Chem 1994;269:9397 – 400. Griffith OW, Meister A. J Biol Chem 1979;254:7558 – 60. Schreck R, Albermann K, Baeuerle PA. Free Radical Res Commun 1992;17:221 – 37. Carter AB, Knudston KL, Monick MM, Hunninghake GW. J Biol Chem 1999;274:30858 – 63. Haddad JJ, Lauterbach R, Saade´ NE, Safieh-Garabedian B, Land SC. Biochem J 2001;355:29 – 38. Hagemann C, Blank JL. Cell Signalling 2001;13:863 – 75. Ye RD. J Leukocyte Biol 2001;70:839 – 48. Bernard GR. Am J Med 1991;91:55S – 9S. Haddad JJ, Choudhary KK, Land SC. Biochem Biophys Res Commun 2001;281:311 – 6. Tryka AF, Witschi H, Gosslee DG, McArthur AH, Clapp NK. Am Rev Respir Dis 1986;133:1055 – 9. O’Reilly MA, Staversky RJ, Watkins RH, Maniscalco WM. Am J Respir Cell Mol Biol 1998;19:777 – 85. Bedi A, Mookerjee B. Apoptosis 1998;3:237 – 44. Me´plan C, Richard MJ, Hainaut P. Biochem Pharmacol 2000;59:25 – 33. Cohen GM. Biochem J 1997;326:1 – 16. Hall AG. Eur J Clin Invest 1999;29:238 – 45. Thannickal VJ, Fanburg BL. Am J Physiol: Lung Cell Mol Physiol 2000;279:L1005 – 28. Kazzaz JA, Xu J, Palaia TA, Mantell L, Fein AM, Horowitz S. J Biol Chem 1996;271:15182 – 6. Simon HU, Haj-Yehia A, Levi-Schaffer F. Apoptosis 2000;5:415 – 8. Tan S, Sagara Y, Liu Y, Maher P, Schubert D. J Cell Biol 1998;141: 1423 – 32. Kannan K, Jain SK. Pathophysiology 2000;7:153 – 63. Buttke TM, Sandstorm PA. Immunol Today 1994;15:7 – 10. Kehrer JP. Teratology 2000;62:235 – 6. Celli A, Que FG, Gores GJ, LaRusso NF. Am J Physiol 1998;275: G749 – 57. Bellas RE, FitzGerald MJ, Fausto N, Sonenshein GE. Am J Pathol 1997;151:891 – 6. Hettmann T, DiDonato J, Karin M, Leiden JM. J Exp Med 1999; 189:145 – 57. Beg AA, Baltimore D. Science 1996;274:782 – 6. Liu ZG, Hsu H, Goeddel DV, Karin M. Cell 1996;87:565 – 76. Boothby MR, Mora AL, Scherer DC, Brockman JA, Ballard DW. J Exp Med 1997;185:1897 – 907. Haddad JJ. Cytokines, Cell Mol Ther 2000;6:177 – 87. Haddad JJ, Land SC. Biochem Biophys Res Commun 2001;285: 267 – 72. Haddad JJ, Saade´ NE, Safieh-Garabedian B. Cell Signalling 2002;14: 211 – 8. Rahman I. Antioxid Redox Signal 1999;1:425 – 47. Parmentier M, Hirani N, Rahman I, Donaldson K, MacNee W, Antonicelli F. Eur Respir J 2000;16:933 – 9. Sen CK. Biochem Pharmacol 1998;55:1747 – 58. Haddad JJ. Eur Cytokine Network 2001;12:614 – 24.