John J. Haddad. doi: /cyto , available online at on

Transcription

1 doi: /cyto , available online at on RECOMBINANT TNF-α MEDIATED REGULATION OF THE IκB-α/NF-κB SIGNALING PATHWAY: EVIDENCE FOR THE ENHANCEMENT OF PRO- AND ANTI-INFLAMMATORY CYTOKINES IN ALVEOLAR EPITHELIAL CELLS John J. Haddad The signaling transduction mechanism mediated by tumor necrosis factor-α (TNF-α) in the alveolar epithelium is not well characterized. It was subsequently hypothesized that recombinant murine TNF-α (rmtnf-α) selectively regulates the inhibitory κb (IκB-α)/nuclear factor-κb (NF-κB) pathway and interferes with the endogenous biosynthesis of pro-inflammatory (stimulatory) and anti-inflammatory (inhibitory) cytokines. The cytokine rmtnf-α induced, in a timeand dose-dependent manner, the degradation of IκB-α within the cytosolic compartment, an effect associated with up-regulating its phosphorylation. This allowed the biphasic regulation of selective NF-κB subunit nuclear translocation, thereby mediating a dual excitatory mechanism on NF-κB activation. The immunoregulatory effect of rmtnf-α was associated with a time-dependent induction of pro-inflammatory [interleukin (IL)-1β, IL-6 and TNF-α] and anti-inflammatory (IL-10) cytokine biosynthesis. These results indicate a novel involvement of an IκB-α/NF-κB-sensitive pathway mediating the effect of TNF-α, which is associated with an autocrine, endogenous mechanism mediating the regulation of cytokine signaling Elsevier Science Ltd. All rights reserved. There is increasing evidence implicating the alveolar epithelium as a dynamic barrier that plays an important role in regulating the inflammatory and metabolic responses to oxidative stress and the accompanying inflammatory signal, 1 5 sepsis, endotoxemia, and other critical illnesses in the lung. 6 9 The respiratory epithelium is a primary target of an inflammatory/ infectious condition at the epithelial-blood interface and is itself capable of amplifying an inflammatory signal by recruiting inflammatory cells and by producing inflammatory mediators. 1 5 Many of the side effects of lipopolysaccharide-endotoxin (LPS), derived form the cell wall of gram negative bacteria, and other inflammatory mediators such as tumor necrosis factor-α (TNF-α), are secondary to the overproduction of pro-inflammatory mediators, such as the pleiotropic cytokines, which exacerbate the pathophysiological From the Neuroscience Research Laboratory, Department of Anesthesia & Perioperative Care, University of California, Medical Center S-261, San Francisco, California , USA Correspondence to: haddadj@anesthesia.ucsf.edu Received 25 July 2001; received in revised form 2 January 2002; accepted for publication 5 January /02/$-see front matter 2002 Elsevier Science Ltd. All rights reserved. KEY WORDS: cytokine/iκb-α/il-10/inflammation/nf-κb/tnf-α condition by activating and recruiting inflammatory cells. 6 9 Therefore, the suppression of a proinflammatory signal, and the down-stream conjugated inflammatory pathways, and augmentation of a counter-inflammatory response has been a major focus of the approach to the treatment of inflammatory diseases. For instance, glucocorticoids, 10 extracellular purines, 11 phosphodiesterase selective inhibitors, 12,13 pyrimidylpiperazine, and adrenoreceptor agonism/ antagonism 17 have been widely used to counteract the effects of inflammatory cytokines and subsequently suppress the protracted pathophysiological conditions in vitro and in vivo. Although the transcription factor NF-κB has been originally recognized in regulating gene expression in B-cell lymphocytes, 18 subsequent studies have demonstrated that it is one member of a ubiquitously expressed family of Rel-related transcription factors that serve as critical regulators of many genes, including those of pro-inflammatory cytokines The translocation and activation of NF-κB in response to various stimuli are sequentially organized at the molecular level. 18,21 In its inactive state, the heterodimeric NF-κB, which is mainly composed of two subunits, p50 (NF-κB 1 ) and p65 (RelA), is present in the cytoplasm associated with its inhibitory protein, IκB. 18,21,22 CYTOKINE, Vol. 17, No. 6 (21 March), 2002: pp

2 302 / Haddad CYTOKINE, Vol. 17, No. 6 (21 March, 2002: ) Upon stimulation, such as with cytokines and LPS, IκB-α, the major cytosolic inhibitor of NF-κB, undergoes phosphorylation on serine/threonine residues, ubiquitination and subsequent proteolytic degradation, thus unmasking the nuclear localization signal on p65 subunit and allowing nuclear translocation of the complex. This sequential propagation of signaling ultimately results in the release of NF-κB subunits from IκB-α, thereby promoting translocation and initiation of gene transcription and regulation. 18,21 TNF-α is a potent inflammatory cytokine, which exerts its pleiotropic activities through ligand-induced cross-linking of specific receptors, virtually present in almost all cell types. 23 During inflammation, TNF-α released in the inflammatory environment transiently activates immune and non-immune cells, thereby causing a rapid but transient release of reactive oxygen species (ROS), a crucial part of the defence mechanism against invading microbial pathogens and tumor cell metastasis. 23 The signaling mechanism mediating the effect of TNF-α in the respiratory epithelium, however, has not been well characterized and has yet to be ascertained. The present investigation has attempted, therefore, to unravel the role that recombinant TNF-α plays in regulating the IκB-α/NF-κB signaling transduction pathway and the downstream cytokine transduction. In particular, TNF-α mediated regulation of NF-κB selective subunit translocation and subsequent activation was shown to be dependent on IκB-α phosphorylation and degradation within the cytosolic compartment. This regulatory effect was associated with the induction of pro-inflammatory (IL-1β, IL-6 and TNF-α) and anti-inflammatory (IL-10) endogenous cytokine biosynthesis, indicating the involvement of an autocrine, cytokine-sensitive mechanism mediating the effect of TNF-α on IκB-α/NF-κB signaling. RESULTS rmtnf-α mediated regulation of IκB-α signaling: Abundance versus phosphorylation state The modulatory role of rmtnf-α in regulating IκB-α phosphorylation and degradation is not well characterized in the alveolar epithelium. Subsequently, we designed a series of experiments to investigate this potential immunoregulatory role. The effect of rmtnf-α on IκB-α regulation is shown in Figure 1. A time-response analysis of the effect of rmtnf-α on IκB-α abundance and phosphorylation within the cytosolic compartment is shown in Figure 1A. The effect of rmtnf-α on IκB-α abundance was immediate, evident as early as 2 min (1/30 h) post addition of rmtnf-α, and continued to degrade in a timedependent manner with variable abundance between 1 h and 48 h of stimulation (Fig. 1A). This inhibitory effect of rmtnf-α on IκB-α abundance was associated with up-regulating its phosphorylation in a biphasic, time-dependent manner (Fig. 1A). To assess for a relationship between the abundance of IκB-α and its phosphorylation state, we determined the efficient ratio R=[IκB-α]/[pIκB-α], as shown in Figure 1B. The histogram analysis revealed a biphasic effect of rmtnf-α on IκB-α (Fig. 1B). The protein expression of IκB-α within the cytosolic compartment decreased with rmtnf-α immediately after addition of the cytokine, in a time-dependent manner. The level of IκB-α expression and the amount of its phosphorylation were similar at 10 min (1/6 h); however, the level of phosphorylation exceeded its abundance, especially between 15 and 20 min (1/4 1/3 h) (Fig. 1B). Despite the observation that the abundance of IκB-α was relatively higher than its phosphorylation state between 30 min and 2 h, IκB-α abundance in comparison to the time period between 0 and 5 min is evidently lower, indicating an active phosphorylation state (Fig. 1B). Again, the level of IκB-α expression and the amount of its phosphorylation were similarly comparable at 4h 8 h, thereby its phosphorylation takes over at between 16 h and 24 h, thereafter exhibiting comparable ratio at 48 h (Fig. 1B). The dose-response curve at 30 min for the effect of rmtnf-α on the abundance of IκB-α is shown in Figure 1C. At a concentration of 0.1 ng/ml, rmtnf-α induced degradation of IκB-α, which continued to degrade in the cytosolic compartment with ascending rmtnf-α concentration (Fig. 1C). The housekeeping gene protein product, β-actin, was used as an internal reference for semi-quantitative loading per parallel lanes, as shown in Figure 1C. The effect of rmtnf-α on the nuclear translocation of selective NF-κB subunits As shown in Figure 2, rmtnf-α (10 ng/ml) mediated, in a time-dependent manner, the nuclear translocation of selective NF-κB subunits. Gel analysis of the abundance of these subunits within the nuclear compartment revealed a biphasic stimulatory effect (Fig. 2). rmtnf-α induced NF-κB 1 (p50) accumulation in as early as 2 min post addition, then subsiding to re-elevate again after 30 min, when it remained abundant until 16 h, thereafter declining (Fig. 2). The effect of rmtnf-α on NF-κB 2 (p52) is marginal during early exposure, but slightly increased after h to become significantly abundant at around 16 h, thereafter remaining mildly in the nuclei (Fig. 2). In a manner roughly similar to the effect on NF-κB 1 (p50), rmtnf-α allowed nuclear accumulation of RelA (p65) immediately in as early as 2 min post addition, declining thereafter, but re-elevating at 30 min and continuing to be strongly abundant in the nucleus afterwards (Fig. 2). The abundance of RelB (p68) was variable to the effect of rmtnf-α in comparison with either

3 TNF-α mediated regulation of IκB-α/NF-κB/303 Figure 1. The regulatory effect of rmtnf-α on IκB-α abundance/phosphorylation states. (A) Time-response analysis of the effect of rmtnf-α on IκB-α abundance and phosphorylation within the cytosolic compartment. The effect of rmtnf-α on IκB-α abundance was immediate, and continued to degrade with variable abundance. This inhibitory effect of rmtnf-α on IκB-α abundance was associated with up-regulating its phosphorylation in a biphasic, time-dependent manner. (B) A relationship between the abundance of IκB-α and its phosphorylation state, as determined by the efficient ratio R=[IκB-α]/pIκB-α]. (C) The dose-response curve at 30 min for the effect of rmtnf-α on the abundance of IκB-α. rmtnf-α induced degradation of IκB-α, which continued to degrade in the cytoplasm with ascending concentration. The housekeeping gene protein product, β-actin, was used as an internal reference for semi-quantitative loading per parallel lanes. φ P<0.05, [IκB-α] abundance is higher in comparison with [piκb-α]; ψ P<0.05, [piκb-α] phosphorylation is higher in comparison with [IκB-α]. n=3 independent experiments performed with separate cell cultures. abundance of p50 or p65 subunit. In the absence of rmtnf-α we observed constitutive abundance of p68 within the nucleus, variably affected, but not enhanced, by the addition of rmtnf-α (Fig. 2). The abundance of p68 roughly stood there at different time points, with particular availability at 20 min and at between 2 h and 24 h, as shown in Figure 2. Finally, the effect of rmtnf-α on c-rel (p75) nuclear abundance (Fig. 2) revealed an immediate induction in as early as 2 min, thereafter subsiding to re-escalate at 30 min, subsiding again and re-appearing at 4 16 h. The housekeeping gene protein product, β-actin, was used as an internal reference for semi-quantitative loading per parallel lanes, as shown in Figure 2. The effect of rmtnf-α on the nuclear activation of NF-κB: Biphasic DNA-binding analysis In association with the differential, biphasic effect of rmtnf-α on NF-κB selective subunit nuclear translocation, there was a dose-dependent up-regulation of the DNA-binding activity of this transcription factor, as shown in Figure 3A. The activation of NF-κB increased in the presence of ascending concentrations of rmtnf-α, with effective maximal activation at 10 ng/ml (Fig. 3A). Histogram analysis of the gelshifted bands revealed the 10 ng/ml concentration as the most effective and optimum concentration that induced maximum activation (Fig. 3B). The timeresponse analysis of the effect of rmtnf-α on NF-κB activation is shown in Figure 4A. In association with a dual regulation of the nuclear translocation of selective NF-κB subunits, rmtnf-α up-regulated the activation of this transcription factor in a biphasic manner (Fig. 4A). The activation in response to rmtnf-α was immediate in as early as 2 5 min post addition, thereafter subsiding, but remained active between 10 and 20 min. Maximal activation with rmtnf-α was observed at 30 min post addition and remained thereafter steadily active until 24 h culturing in the continuous presence of rmtnf-α, to become inactive beyond

4 304 / Haddad CYTOKINE, Vol. 17, No. 6 (21 March, 2002: ) Figure 2. The effect of rmtnf-α on the nuclear translocation of selective NF-κB subunits. rmtnf-α (10 ng/ml) mediated, in a time-dependent manner, the nuclear translocation of selective NF-κB subunits. Gel analysis of the abundance of these subunits revealed a biphasic stimulatory effect. rmtnf-α induced p50 accumulation in as early as 2 min post addition, then subsiding to re-elevate again after 30 min, when it remained abundant until 16 h, thereafter declining. The effect of rmtnf-α on p52 is marginal during early exposure, but slightly increased after h to become significantly abundant at around 16 h, thereafter remaining mildly in the nucleus. In addition, rmtnf-α allowed nuclear accumulation of p65 immediately in as early as 2 min post addition, declining thereafter, but re-elevating at 30 min and continuing to be strongly abundant in the nucleus afterwards. The abundance of p68 was variable to the effect of rmtnf-α in comparison to either abundance of p50 or p65. The abundance of p68 prevailed at different time points, with particular availability at 20 min and at between 2 h and 24 h. The effect of rmtnf-α on p75 nuclear abundance revealed an immediate induction in as early as 2 min, thereafter subsiding to re-elevate at 30 min, subsiding again and re-appearing at 4 16 h. The housekeeping gene protein product, β-actin, was used as an internal reference for semi-quantitative loading per parallel lanes. n=3 independent experiments performed with separate cell cultures. this time point at 48 h (Fig. 4A). Histogram analysis of the gel-shifted bands exhibiting the time-response curve is displayed in Figure 4B. Analysis of the effect of rmtnf-α on the pro- and anti-inflammatory cytokine biosynthesis The novel interference of rmtnf-α in regulating IκB-α/NF-κB signaling pathway was associated with a dual effect on extracellularly secreted cytokines. As shown in Figure 5A, rmtnf-α (10 ng/ml) up-regulated the production of IL-1β, in a time-dependent manner. There was an initial increase in IL-1β ( 2 4h) after addition of rmtnf-α. This was followed by an intermediate decrease ( 8 16 h) and a maximum increase at about h (Fig. 5A). Similarly, rmtnf-α induced the biosynthesis of IL-6 in a time-dependent manner, with maximum enhancement at around h, as shown in Figure 5B. Of interest, the observation reported that rmtnf-α induced the endogenous production of TNF-α, effective at 16 h, which continued to increase at 24 h and maximize at 48 h, as shown in Figure 5C. The most intriguing observation of all is the ability of rmtnf-α to augment the secretion of a potent anti-inflammatory cytokine (IL-10) in a manner similar to its ability to up-regulate the production of pro-inflammatory cytokines. The concentration of IL-10 in the supernatant appeared at 8 h post addition of rmtnf-α, continued to increase exponentially to maximize at around 48 h (Fig. 5D). DISCUSSION To accommodate an ever changing microenvironment, cells adjust the pattern of gene expression by adaptive regulation of a host of transcription factors, which bind their respective cognate sites in the regulatory elements of targeted genes. 21,24 26 NF-κB comprises the Rel family of inducible transcription factors that are key mediators in regulating the progression of the inflammatory process. 26 Therefore, activation/ regulation of the NF-κB/Rel transcription family, via nuclear translocation of cytoplasmic entities and complexes, plays a central role in the evolution of inflammation through regulation of genes essentially involved

5 TNF-α mediated regulation of IκB-α/NF-κB/305 Figure 3. The effect of rmtnf-α on NF-κB nuclear activation. (A) The activation of NF-κB increased in the presence of ascending concentrations of rmtnf-α, with effective maximal activation at 10 ng/ml. (B) Histogram analysis of the gel-shifted bands revealed the 10 ng/ml concentration as the most effective concentration that induced maximum activation. P<0.05, P<0.01, P<0.001, as compared with control (rmtnf-α=0 ng/ml). n=3 independent experiments performed with separate cell cultures. in encoding pro-inflammatory cytokines and other inflammatory mediators. 21,24 26 The NF-κB/Rel family includes five members: NF-κB 1 [p50/p105 (p50 precursor)], NF-κB 2 [p52/p100 (p52 precursor)], RelA (p65), RelB (p68) and c-rel (p75). 25 Despite the ability of most Rel members (with the exception of p68) to homodimerize, as well as form heterodimers, with each other, the most prevalent activated form of NF-κB is the heterodimer p50 p65, which possesses the transactivity domains necessary for gene regulation. 21,24 26 The NF-κB members contain a Rel homology domain (RHD), which is responsible for dimer formation, nuclear translocation, sequence-specific consensus DNA recognition and interaction with inhibitory-αb (IκB), the cytosolic inhibitors of NF-κB In unstimulated cells, NF-κB resides in the cytoplasm as an inactive NF-κB/IκB complex, a mechanism which hinders the recognition of the nuclear localization signal (NLS) by the nuclear import machinery, thereby retaining the NF-κB complex within the cytosol. 21,24 26 Although the inflammatory signals mediated by TNF-α are well recognized in other systems and cell models, this role in the fetal alveolar epithelium is not well characterized. The cytokine rmtnf-α induced, in a time- and dose-dependent manner, the degradation of IκB-α within the cytosolic compartment, an effect associated with up-regulating its phosphorylation. This requirement allowed the biphasic regulation of selective NF-κB subunit nuclear translocation (supershift analysis revealed the involvement of the p50 p65 complex; Data not shown), thereby mediating a dual excitatory mechanism on NF-κB activation. The immunoregulatory effect of rmtnf-α was also associated with the induction of pro-inflammatory (IL-1β, IL-6 and TNF-α) and anti-inflammatory (IL-10) cytokine biosynthesis, indicating a novel involvement of an IκB-α/NF-κB-sensitive pathway mediating the effect of TNF-α in association with an autocrine, endogenous mechanism governing the regulation of cytokine signaling. In particular, the interference of this cytokine in regulating the phosphorylation of IκB-α, the major cytosolic inhibitor of NF-κB, suggested the involvement of an upstream kinase. In this respect, signals emanating from membrane receptors, such as those for IL-1 and TNF-α, activate members of the MEKK-related family, including NF-κB inducing kinase (NIK) and MEKK 1, both of which are involved in the activation of IκB kinases, IKK 1 and IKK 2,

6 306 / Haddad CYTOKINE, Vol. 17, No. 6 (21 March, 2002: ) Figure 4. The time-response analysis of the effect of rmtnf-α on NF-κB. (A) rmtnf-α up-regulated the activation of this transcription factor in a biphasic manner. The activation was immediate in as early as 2 5 min, thereafter subsiding, but remained active between 10 and 20 min, with maximal activation at 30 min, and continued to be steadily active until 24 h, to become inactive beyond this time point at 48 h. (B) Histogram analysis of the gel-shifted bands exhibiting the time-response curve. P<0.05, P<0.01, P<0.001, as compared with control (rmtnf-α=0 ng/ml). n=3 independent experiments performed with separate cell cultures. components of the IKK signalsome. 27,28 IKK 1 and IKK 2 were identified as components of the highmolecular weight complex containing a number of proteins involved in NF-κB regulation. 27,28 These kinases phosphorylate members of the IκB family, including IκB-α, at specific serines within their amino termini, thereby leading to site-specific ubiquitination and degradation by the 26S proteasome. Therefore, the immunoregulatory potential of rmtnf-α in up-regulating the phosphorylation of IκB-α and its subsequent degradation strongly implicated an upstream kinase, probably the IKK complex, as a potential target for the inflammatory action of this cytokine. Moreover, the selective interference mediated by rmtnf-α in regulating the translocation/activation of NF-κB and the expression of its inhibitor (IκB-α) is of particular interest since it suggests that this compound s regulatory role in the alveolar space resides within and/or above the upstream pathway regulating the phosphorylation of IκB-α, thereby modulating the downstream pathway governing NF-κB translocation and activation, and subsequently interfering with the regulation of cytokine signaling. The promoters of genes encoding cytokines, particularly IL-1α, IL-1β, IL-2, IL-3, IL-6, IL-8, IL-12 and TNF-α, contain multiple cis-acting motifs including those that bind such transcription factors as NF-κB. Furthermore, the release of free NF-κB upon extracellular stimulation (TNF-α, for example) due to IκB phosphorylation and degradation, leads to DNA binding to initiate transcription of related genes, including immunoreceptors, cytokines and, interestingly, its own inhibitor, IκB Two unique features of the NF-κB/IκB complex system are deduced from its feedback regulation. The transcriptional activation of NF-κB triggers the synthesis of IκB, and NF-κB activated transcription is maintained by continuous degradation of IκB, which is sustained by an extracellular stimulus (in this particular case rmtnf-α) Thus, the expression of IκB parallels both NF-κB activity and the duration of the activating extracellular stimulation, suggesting that this temporal parallelism between IκB accumulation/degradation and an effective external stimulation is a mechanism allowing dual, biphasic, regulation of NF-κB within the respiratory epithelium. Of particular interest, unraveling the downstream cytokine pathway that is regulated by NF-κB/IκB remains a major target for the search for novel therapeutic agents that tend to suppress the inflammatory signal mediated, but have the potential though to up-regulate a counter-inflammatory response. One such novel approach is the discovery

7 TNF-α mediated regulation of IκB-α/NF-κB/307 A Baseline B IL-6 (pg/ml) Baseline C Baseline D IL-10 (pg/ml) Baseline Figure 5. The effect of rmtnf-α on pro- and anti-inflammatory cytokine biosynthesis. (A) rmtnf-α up-regulated the secretion of IL-1β, in a time-dependent manner. (B) Similarly, rmtnf-α induced the biosynthesis of IL-6 in a time-dependent manner. (C) rmtnf-α induced the production of TNF-α, effective at 16 h. (D) rmtnf-α augmented the secretion of IL-10 in a manner similar to its ability to up-regulate the production of pro-inflammatory cytokines. P<0.05, P<0.01, P<0.001, as compared with control (rmtnf-α=0 ng/ml). n=3 independent experiments performed with separate cell cultures. reported in this investigation of the potential immunoregulatory role that rmtnf-α exhibits in augmenting an inflammatory signal, yet up-regulate an anti-inflammatory loop; however, whether a direct association between either pathway is established that is likely to regulate IκB-α/NF-κB signaling transduction has yet to be ascertained as part of the mechanism mediating the effect of TNF-α. The so-called pro-inflammatory cytokines, which include TNF-α, are involved in a plethora of cellular actions, particularly inflammatory/infectious conditions. 29 Locally, these cytokines stimulate leukocyte proliferation, cytotoxicity, synthesis and release of proteolytic enzymes, and synthesis of prostaglandins that initiate a cascade of secondary cytokine transcription/biosynthesis, including the amplification of TNF-α and other inflammatory mediators. 29 These cytokines, for instance, have the ability to raise the thermoregulatory set point systemically and, via differential influences on the expression of iron binding proteins, mediate a redistribution of iron from extracellular to intracellular sites and stores. Moreover, inflammatory mediators orchestrate a metabolic reaction that reduces any energy consumption not directed at repelling the microbial pathogen, redirects host resources to the defence effort, and sets up a defence network that protects non-leukocyte cells from collateral damage by anti-microbial effectors. 29 One such secondary cytokine, IL-10, is often referred to as a counter-inflammatory (inhibitory) cytokine because of its temporal association with the aforementioned processes, thereby acting to keep potentially destructive inflammatory responses from overshooting. 29 The ability of rmtnf-α to mediate the biosynthesis of IL-10, a potent anti-inflammatory cytokine, remains of particular interest. IL-10 was originally identified as a cytokine inhibitory factor, which suppressed the biosynthesis of an array of pro-inflammatory mediators, including IL-1β, IL-4, IL-8, TNF-α, TNF-α, interferon-γ (IFN-γ), and granulocyte/macrophage colony stimulating factor (GMCSF). 30 The level of regulation by IL-10 is both transcriptional and translational, raising the possibility of IL-10 acting on transcription factors involved in regulating inflammatory genes. 30 It has also been reported that IL-10 selectively regulated NF-κB activation, a phenomenon well correlated with a dose-dependent inhibition of the release of pro-inflammatory cytokines. 30 Moreover, we have previously shown that rhil-10 down-regulated NF-κB activation, an effect associated with the inhibition of inflammatory cytokine biosynthesis. 16 On the mechanism of action of IL-10, it seems that it involves the blockade of a reaction required for the release of IκB from the complex in intact cells. 16,31 For example, certain experiments using cell-free preparations have suggested that certain protein kinases phosphorylate

8 308 / Haddad CYTOKINE, Vol. 17, No. 6 (21 March, 2002: ) IκB causing its release and allowing activation of the NF-κB complex. 31 Despite the unequivocal observation therein reported which gave an evidence for the amplification of an anti-inflammatory pathway alongside the IκB-α/NF-κB/inflammatory cytokine pathway mediated by rmtnf-α, the molecular basis for this biphasic mechanism remains to be elucidated. However, the possibility of TNF-α mediating an endogenous regulatory pathway via IL-10 which requires a dual regulation of IκB-α/NF-κB signaling is likely to implicate a novel feedback and/or feedforward mechanism working within the respiratory epithelium. In summary, the results of the present investigation could be highlighted as follows: (i) The cytokine rmtnf-α induced, in a time- and dose-dependent manner, the cytosolic degradation of IκB-α, aneffect associated with up-regulating its phosphorylation; (ii) this allowed the biphasic translocation of selective NF-κB subunits onto the nuclear compartment, thereby mediating a dual excitatory mechanism on NF-κB activation; and (iii) the immunoregulatory effect of rmtnf-α was associated with the induction of pro-inflammatory (IL-1β, IL-6 and TNF-α) and antiinflammatory (IL-10) cytokine biosynthesis. These results indicated the involvement of an IκB-α/NF-κBsensitive pathway, which is associated with an autocrine, endogenous mechanism mediating the regulation of cytokine signaling. MATERIAL AND METHODS Chemicals and reagents Unless indicated otherwise, chemicals of the highest analytical grade were purchased from Sigma-Aldrich (Dorset, England, UK). All experimental procedures involving the use of live animals were approved under the Animals (Scientific Procedures) Act, 1986 (UK). Primary cultures of alveolar epithelia Fetal alveolar type II (fatii) epithelial cells were isolated from lungs of fetuses, essentially as reported elsewhere. 2 5 Briefly, fetal rats were removed from pregnant Sprague-Dawley rats by caesarian section at day 19 of gestation (term=22 days), the lungs excised, teased free from heart and upper airway tissue, and were finely minced then washed free of erythrocytes using sterile, chilled Mg 2+ - and Ca 2+ -free Hanks balanced salt solution (HBSS; 0.5 ml/ fetus). The cleaned lung tissue was re-suspended in 1 ml/fetus HBSS containing trypsin (0.1 mg/ml), collagenase (0.06 mg/ ml) and DNase I (0.012% w/v), and was agitated at 37 C for 20 min. The solution was then centrifuged at 100 g for 2 min to remove undispersed tissue, the supernatant was saved to a fresh sterile tube and an equal volume of Dulbecco s modified Eagle medium (DMEM) with 10% (v/v) fetal calf serum (FCS) was added to the supernatant. After passing the supernatant through a 120 μm pore sterile mesh, the filtrate was centrifuged at 420 g for 5 min, the pellet re-suspended in 20 ml DMEM/FCS and the cells were placed into a T-150 culture flask for 1 h at 37 C to enable fibroblasts and non-epithelial cells to adhere. Unattached cells were washed three times by centrifugation at 420 g for 5 min each and then seeded onto 24 mm diameter Transwell-clear permeable supports (Costar; 0.4 μm pore size) at a density of cells per filter and were allowed to adhere overnight at 152 Torr ( 21% O 2 /5% CO 2 ). DMEM/FCS was exchanged for 4 ml of serum free PC-1 media (Biowhittaker, MD, USA) pre-equilibrated to po 2 =152 Torr and 37 C 24h later and cells were maintained at this po 2 until the experiment. In each case, and under conditions of independent treatments, the adenylate energy charge, an index of cell viability and competence, remained 0.7 and transepithelial monolayer resistance was monitored constantly at Ωcm 2 or more. 2 5 rmtnf-α treatment, western analysis and electrophoretic mobility shift assay Time-dependent analysis (0 48 h) of the effect of recombinant murine TNF-α (rmtnf-α; 10 ng/ml) on IκB-α/NF-κB signaling was performed. For the dose-response curve, cells were treated with rmtnf-α (0 30 ng/ml) for 0.5 h and filters were twice washed with pre-equilibrated PC-1 medium. Following treatments, sub-cellular extracts (cytosolic and nuclear) were prepared essentially as previously described. 2,3 Briefly, filters were washed twice in 5 ml ice-cold, preequilibrated phosphate buffered saline (PBS; ph ) and cells ( ) were collected and centrifuged at 420 g for 5 min at 4 C. Nuclei were released by suspending the centrifuged pellet in 250 μl buffer A (lysis buffer) containing (in mm): 10 Tris-HCl (ph 7.8), 10 KCl, 2.5 NaH 2 PO 4, 1.5 MgCl 2, 1 Na 3 VO 4, 0.5 dithiothreitol (DTT), aminoethyl-benzene sulfonyl fluoride-hcl (AEBSF), and 2 μg/ml each of leupeptin, pepstatin A and aprotinin. The suspension was left in ice for 10 min followed by a 45 s homogenization at a moderate speed. Nuclei were collected by centrifuging the slurry at 4500 g for 5 min at 4 C and re-suspending in 100 μl buffer B (buffer A adjusted to (in mm): 20 Tris-HCl (ph 7.8), 420 KCl, 20% (v/v) Glycerol). The supernatant thus formed is termed the cytosolic extract. The nuclei were then lysed at 4 C for 30 min with gentle agitation, the debris cleared by centrifugation at g for an additional 30 min at 4 C and the supernatant frozen in liquid nitrogen and stored at 70 C until used. In all cases, protein contents were determined by the Bradford method using BSA as a standard. 2,3 Cytosolic and nuclear proteins (20 25 mg) were resolved over SDS-PAGE (7.5%) gels at RT, blotted onto nitrocellulose membrane, and non-specific binding sites were subsequently blocked. Mouse monoclonal IgG 1 anti-iκb-α (H-4), IgG 2b anti-(phosphorylated) piκb-α (B-9), rabbit polyclonal IgG anti-p50 (NLS), anti-p52 (K-27), antip65 (RelA; A), anti-p68 (RelB; C-19), and anti-p75 (c-rel; N) (Santa Cruz Biotechnology, UK) antibodies were used for primary detection. Anti-rabbit Ig-biotinylated antibody (Amersham Life Science, UK) was employed for secondary detection followed by the addition of streptavidin-hrp conjugate and visualized on film by chemiluminescence. β-actin standard was used as an internal reference for semi-quantitative loading in parallel lanes for each variable.

9 TNF-α mediated regulation of IκB-α/NF-κB/309 Western blots were scanned by NIH MagiScanII and subsequently quantitated by UN-Scan-IT automated digitizing system (version 5.1; 32-bit), and the ratio of the density of the band to that of β-actin was subsequently performed. 2,3,22 Custom deoxy-oligonucleotide probe sequences were purchased from Genosys, UK: NF-κB, 5 -AGTTGAGG GGACTTTCCCAGGC-3 (binding sequence underlined). Gel-purified double-stranded DNA was end-labeled with [γ 32 P]-ATP (NEN Life Sciences Products, UK). Identical amounts of radioactive probe ( counts min 1 ) were added to binding reactions containing 1 5 μg fatii nuclear extracts in a final volume of 40 μl in DNA binding buffer (20 mm HEPES (ph 7.9); 1 mm MgCl 2 ; 4% Ficoll). Reaction mixtures were incubated for 30 min at 25 C before separating on non-denaturing 4% polyacrylamide gels at RT and subjected to electrophoresis with 1:10 5X Tris-Borate- EDTA buffer. A non-specific competitive polydeoxyinosinicdeoxycytidylic acid [poly(di-dc)] (Amersham Pharmacia Biotech, UK) was added to reaction mixtures after addition of labeled probe. Gels were transferred to ion-exchange chromatography paper, vacuum dried and then electronically visualized on a Packard Instant phosphorimager. Specific quantitation of the corresponding DNA gel shift bands was performed with phosphorimaging. 2,3,22 Assessment of pro- and anti-inflammatory cytokines by enzyme-linked immunosorbent assay The bioactivity of extracellularly released cytokines was measured by a two-site, solid-phase, developed sandwich enzyme-linked immunosorbent assay (ELISA). 4,5 Immunoaffinity purified polyclonal rabbit anti-rat IL-1β, IL-6, IL-10 and TNF-α (2 μg/ml) primary antibodies were used to coat high-binding microtiter plates (MaxiSorp, Nunc, UK). Recombinant rat and biotinylated immunoaffinity purified sheep anti-rat cytokine (R&D Systems) were used as standard and recognition antibodies, respectively. The color was developed incorporating streptavidin-poly-horseradish peroxidase coupled reaction with the chromagen 3,3,5,5 tetramethyl-benzidine dihydrochloride (TMB), and the optical density (O.D.) was measured at 450 nm. Inter- and intra-assay coefficients of variations (CV) were <10%, and the minimum detectable sensitivity for each cytokine is 2 pg/ml. Results interpolated from the linear regression of the standard curve are expressed as pg/ml. 4,5 Statistical analysis and data presentation Data are the means and the error bars the SEM. Statistical evaluation of the difference in mean separation was performed by one-way analysis of variance (ANOVA), followed by post hoc Tukey s test, and the a priori level of significance at 95% confidence level was considered at P Acknowledgements This work was supported by grants from the Medical Research Council (MRC, UK), Anonymous Trust and Tenovus-Scotland (Dr Stephen C. Land). J.J.H. is a recipient of the George John Livanos prize (London). Parts of this work were presented at FASEB Experimental Biology-2001 in Orlando, Florida, USA (Dr John J. Haddad). REFERENCES 1. Freeman BA, Panus PC, Matalon S, Buckley BJ, Baker RR (1993) Oxidant injury to the alveolar epithelium: Biochemical and pharmacologic studies. Res Rep Health Eff Inst 54: Haddad JJ, Land SC (2000) O 2 -evoked regulation of HIF-1α and NF-κB in perinatal lung epithelium requires glutathione biosynthesis. Am J Physiol Lung Cell Mol Physiol 278:L492 L Haddad JJ, Olver RE, Land SC (2000) Antioxidant/prooxidant equilibrium regulates HIF-1α and NF-κB redox sensitivity: Evidence for inhibition by glutathione oxidation in alveolar epithelial cells. J Biol Chem 275: Haddad JJ, Safieh-Garabedian B, Saadé NE, Kanaan SA, Land SC (2001) Chemioxyexcitation (ΔpO 2 /ROS)-dependent release of IL-1β, IL-6 and TNF-α: Evidence of cytokines as oxygen sensitive mediators in the alveolar epithelium. Cytokine 13: Haddad JJ, Safieh-Garabedian B, Saadé NE, Land SC (2001) Thiol regulation of pro-inflammatory cytokines reveals a novel immunopharmacological potential of glutathione in the alveolar epithelium. J Pharmacol Exp Therap 296: Pittet JF, Wiener-Kronish JP, Serikov V, Matthay MA (1995) Resistance of the alveolar epithelium to injury from septic shock in sheep. Am J Respir Crit Care Med 151: Matthay MA, Fukuda N, Frank J, Kallet R, Daniel B, Sakuma T (2000) Alveolar epithelial barrier: Role in lung fluid balance in clinical lung injury. Clin Chest Med 21: Matuschak GM, Lechner AJ (1996) Targeting the alveolar epithelium in acute lung injury: Keratinocyte growth factor and regulation of the alveolar epithelial barrier. Crit Care Med 24: m 9. Laffon M, Pittet JF, Modelska K, Matthay MA, Young DM (1999) Interleukin-8 mediates injury from smoke inhalation to both the lung endothelial and the alveolar epithelial barriers in rabbits. Am J Respir Crit Care Med 160: Visser J, Van Boxel-Dezaire A, Methorst D, Brunt T, Ronald De Kloet E, Nagelkerken L (1998) Differential regulation of interleukin-10 (IL-10) and IL-12 by glucocorticoids in vitro. Blood 91: Haskó G, Kuhel DG, Németh ZH, Mabley JG, Stachlewitz RF, Virág L, Lohinai Z, Southan GJ, Salzman AL, Szabó C (2000) Inosine inhibits inflammatory cytokine production by a posttranscriptional mechanism and protects against endotoxin-induced shock. J Immunol 164: Haskó G, Szabó C, Németh ZH, Salzman AL, Sylvester Vizi E (1998) Suppression of IL-12 production by phosphordiesterase inhibition in murine endotoxemia is IL-10 independent. Eur J Immunol 28: Doherty AM (1999) Phosphodiesterase 4 inhibitors as novel anti-inflammatory agents. Curr Opin Chem Biol 3: Fukuda T, Sumichika H, Murata M, Hanano T, Adachi K, Hisadome M (2000) A novel dual regulator of tumor necrosis factor-α and interleukin-10 protects mice from endotoxin-induced shock. Eur J Pharmacol 391: Hisadome M, Fukuda T, Sumichika H, Hanano T, Adachi K (2000) A novel anti-rheumatic drug suppresses tumor necrosis factor-α and augments interleukin-10 in adjuvant arthritic rats. Eur J Pharmacol 409: Haddad JJ, Safieh-Garabedian B, Saadé NE, Land SC (2001) The biphasic immunoregulation of pyrimidylpiperazine (Y-40138) is IL-10 sensitive and requires NF-κB targeting in the alveolar epithelium. Br J Pharmacol 133: Zhang H, Kim YK, Govindarajan A, Baba A, Binnie M, Ranieri VM, Liu M, Slutsky AS (1999) Effect of adrenoreceptors on endotoxin-induced cytokines and lipid peroxidation in lung explants. Am J Respir Crit Care Med 160:

10 310 / Haddad CYTOKINE, Vol. 17, No. 6 (21 March, 2002: ) 18. Sen R, Baltimore D (1986) Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Cell 46: Baldwin AS (1996) The NF-κB and IκB proteins: New discoveries and insights. Annu Rev Immunol 14: Siebenlist U, Franzoso G, Brown K (1994) Structure, regulation and function of NF-κB. Annu Rev Cell Biol 10: Baldwin AS (2001) The transcription factor NF-κB and human disease. J Clin Invest 107: Haddad JJ, Lauterbach R, Saadé NE, Safieh-Garabedian B, Land SC (2001) α-melanocyte-related tripeptide, Lys- D -Pro-Val, ameliorates endotoxin-induced nuclear factor κb translocation and activation: Evidence for involvement of an interleukin-1β receptor antagonism in the alveolar epithelium. Biochem J 355: Vandenabeele P, Declercq W, Beyaert R, Fiers W (1995) Two tumor necrosis factor receptors: Structure and function. Trends Cell Biol 5: Makarov SS (2000) NF-κB as a therapeutic target in chronic inflammation: Recent advances. Mol Med Today 6: Tak PP, Firestein GS (2001) NF-κB: A key role in inflammatory disease. J Clin Invest 107: Yamamoto Y, Gaynor RB (2001) Therapeutic potential of inhibition of the NF-κB pathway in the treatment of inflammation and cancer. J Clin Invest 107: Mercurio F, Manning AM (1999) Multiple signals converging on NF-κB. Curr Opin Cell Biol 11: Zandi E, Rothwarf DM, Delhasse M, Hayakawa M, Karin M (1997) The IκB kinase complex (IKK) contains two kinase subunits, IKKα and IKKβ, necessary for IκB phopshorylation and NF-κB activation. Cell 91: Kelley J (1990) Cytokines of the lung. Am Rev Respir Dis 141: Wang P, Wu P, Siegel MI, Egan RW, Billah MM (1995) Interleukin (IL)-10 inhibits nuclear factor κb (NF-κB) activation in human monocytes. J Biol Chem 270: Shirakawa F, Mizel S (1989) In vitro activation and nuclear translocation of NF-κB catalyzed by cyclic AMP-dependent protein kinase and protein kinase C. Mol Cell Biol 9: