- Email: [email protected]
α-Tocopherol decreases tumor necrosis factor-α mRNA and protein from activated human monocytes by inhibition of 5-lipoxygenase
Free Radical Biology & Medicine 38 (2005) 1212 – 1220 www.elsevier.com/locate/freeradbiomed
Original Contribution
a-Tocopherol decreases tumor necrosis factor-a mRNA and protein from activated human monocytes by inhibition of 5-lipoxygenase Sridevi DevarajT, Ishwarlal Jialal Laboratory for Atherosclerosis and Metabolic Research, Department of Pathology and Laboratory Medicine, University of California, Davis Medical Center, Sacramento, CA 95817, USA Received 8 October 2004; revised 4 January 2005; accepted 11 January 2005
Abstract Cardiovascular disease is the leading cause of morbidity in Westernized populations. Low levels of a-tocopherol (AT) are associated with increased incidence of atherosclerosis and increased intakes appear to be protective. AT supplementation decreases interleukin 1 and 6 release from human monocytes. Thus, the aim of this study was to examine the effect of AT on an important proinflammatory cytokine, tumor necrosis factor-a (TNF) release from human monocytes. AT supplementation (1200 IU/day for 3 months) significantly decreased TNF release from activated human monocytes. Mechanisms that were examined included its effect as a general antioxidant, its inhibitory effect on protein kinase C (PKC), and the cycloxygenase-lipoxygenase pathway. While AT decreased TNF release from activated monocytes, other antioxidants had no effect on TNF release. Specific PKC inhibitors had no effect on TNF release from activated monocytes. The inhibition of TNF release by AT in activated monocytes was reversed by leukotriene B4 (LTB4), a major product of the 5-lipoxygenase (5-LO) pathway. Similar observations were seen with inhibitors of 5-lipoxygenase. Indomethacin, a COX inhibitor, in the presence and absence of AT failed to affect TNF activity. These findings suggest that AT decreases TNF release from activated human monocytes via inhibition of 5-lipoxygenase. Also, AT as well as a 5-LO inhibitor significantly decreased TNF mRNA. Furthermore, AT and the 5-LO inhibitor decreased NFnb-binding activity. Thus, in activated human monocytes, AT appears to inhibit TNF mRNA and protein by inhibition of 5-LO. D 2005 Elsevier Inc. All rights reserved. Keywords: a tocopherol; Monocytes; Cytokine; Lipoxygenase; Free radicals
Introduction Cardiovascular disease is the leading cause of morbidity and mortality in the United states. Several lines of evidence point to the pivotal role of inflammation in atherosclerosis [1,2]. Evidence from knockout models shows that the monocyte-macrophage is a crucial cell in atherosclerosis [3–5]. Monocytes secrete several proinflammatory cytokines, the most notable being interleukin-1, tumor necrosis factor-a, and interleukin-6. Several lines of evidence Abbreviations: AT, a-tocopherol; BIM, bisindoleylmaleimide; BT, h-tocopherol; COX, cycloxygenase; LTB4, leukotriene B4; 5-LO, 5lipoxygenase; PKC, protein kinase C; ROS, reactive oxygen species; RRR-AT, RRR-a-tocopherol; SOD, superoxide dismutase; TNF, tumor necrosis factor. T Corresponding author. Fax: (916) 734 6593. E-mail address: [email protected] (S. Devaraj). 0891-5849/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.freeradbiomed.2005.01.009
suggest that the proinflammatory cytokine, tumor necrosis factor-a (TNF) is proatherogenic [6–8]. TNF-a is a multifunctional cytokine that exerts pleiotropic biological actions [9–13]. It activates endothelial cells, promotes monocyteendothelial cell adhesion, stimulates angiogenesis, and induces smooth muscle cell proliferation. Increased mRNA for TNF-a has been documented in aortas of Watanabe heritable hyperlipidemic rabbits and carotid atherosclerotic plaques. TNF-a processing via its receptor can promote apoptosis and thus could contribute to the necrotic core of the atherosclerotic lesion. TNF-a also plays a role in the pathogenesis of obesity-linked insulin resistance [9–13]. Hence, in this study, we examined if TNF release from LPSactivated human monocytes could be modulated. We have previously shown that, RRR-a-tocopherol (RRR-AT), at high doses, exerts anti-inflammatory effects [14]. We showed that in vivo supplementation of human
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
volunteers with RRR-AT (1200 IU/day) significantly decreases monocyte superoxide anion release, lipid oxidation, interleukin-6 and interleukin-1h release, and adhesion to endothelial cells [14]. The inhibition of superoxide anion release and lipid oxidation was mediated by inhibition of protein kinase C-a by AT. Subsequently, we showed that AT inhibits IL-1h release via inhibition of 5-lipoxygenase (5LO) [15] and decreased monocyte-endothelial cell adhesion by decreasing expression of CD11b and VLA-4 and decreased DNA-binding activity of the transcription factor, NFnb [16]. In this study, we examined if monocyte TNF release can be modulated with RRR-AT and the mechanisms involved. The potential mechanisms that were examined included its effect as a general antioxidant, its inhibitory effect on protein kinase C activity, and its effect on the cycloxygenase-lipoxygenase pathway, since it has previously been shown that AT can inhibit both cycloxygenase and lipoxygenase in certain systems [15,17–19]. Also, the effect of AT on mRNA for TNF was studied.
1213
Determination of protein kinase C (PKC) activity PKC activity in human monocytes in the presence and absence of 50 and 100 AM AT was performed by a radioimmunoassay technique using reagents from Amersham Corp. as described previously [14]. The PKC activity assay is based on the PKC-catalyzed transfer of the g -phosphate group of ATP to a PKC-specific peptide. PKC activity is expressed as nanomoles of phosphate transferred per million cells. Quantitation of leukotriene B4 (LTB4) and prostaglandin E2 (PGE2) levels LTB4 levels were measured in the cell culture supernatants from monocytes activated with LPS in the absence and presence of 5-LO inhibitors and AT and assayed by an enzyme immunoassay as described previously (Amersham Corp.) [15]. PGE2 levels were determined using an identical enzyme immunoassay as described for LTB4 [15]. The inter assay CVs for these assays was b10%.
Materials and methods Dose response of AT on TNF release Isolation of human peripheral blood monocytes For the in vivo study, subjects (n = 21) were supplemented with RRR-AT (1200 IU/d) for a period of 8 weeks followed by a 6-week washout phase as described previously [14]. For the in vitro studies, fasting blood was obtained from healthy volunteers who were not taking any antioxidant supplements. All subjects gave informed consent and this study was approved by the Institutional Review Board. Human mononuclear cells were separated from 120 ml of blood from fasting, healthy volunteers by Ficoll Hypaque gradient as described previously [14]. Cells were plated and incubation was carried out at 378C for 2 h in 5% CO2/95% air, after which nonadherent cells were removed after washing three times with phenol red-free RPMI 1640 medium. Nonspecific esterase staining revealed that 88% of cells isolated in this manner were monocytes [14]. All incubations were carried out immediately following isolation of the monocytes. All reagents used were tested for endotoxin contamination by the Limulus endotoxin assay [14], which was b12.5 pg/ml. The viability of the monocytes was found to be 94% by trypan blue exclusion [14]. LPS (1 `ıg/ml) was used to activate monocytes. TNF release was measured in the supernatants following an overnight incubation. Release of TNF The release of TNF was measured in the culture supernatants by ELISA using the human immunoassay kit (Biotrak High Sensitivity Kit, Amersham Corp., Arlington Heights, IL). Interassay CV was b10%.
Mononuclear cells were incubated with AT (25, 50, and 100 AM) during the 2-h adherence incubation and for 30 min following the three washes with RPMI 1640. Thereafter the monocytes were activated with LPS and incubated at 378C for 18 h. The supernatants were then collected and TNF secretion was assayed by ELISA. The cells were dissolved with 0.1 N NaOH and the protein content was measured by the method of Lowry et al. and TNF activity was expressed as nanograms of TNF per milligram protein. Potential mechanisms by which AT could inhibit TNF release include: (i) inhibition of protein kinase C; (ii) inhibition of reactive oxygen species; (iii) by acting as a chain-breaking antioxidant; (iv) inhibition of 5-lipoxygenase and thereby leukotriene B4; (v) inhibition of the cycloxygenase pathway; and (vi) inhibition of TNF synthesis. Role of protein kinase C Mononuclear cells were incubated with PKC inhibitors, Calphostin C (250 and 500 nM) or with bisindoleylmaleimide (0.1, 1 AM) during the 2-h incubation for adherence of monocytes. Thereafter, the cells were washed three times with RPMI 1640 medium and monocytes were activated with LPS. Following an 18-h incubation, TNF activity was measured in the supernatants. The concentrations of the inhibitors used did not show any antioxidant activity as determined by assaying the lag phase of copper-catalyzed LDL oxidation (data not shown).
1214
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Role of reactive oxygen species The effect of the antioxidant enzymes, superoxide dismutase (PEG-SOD, 100 Ag/ml) and Tiron (1 AM), was tested on human monocytes by adding them during the 2-h adherence incubation of the mononuclear cells prior to addition of LPS. TNF activity was measured after an 18-h incubation. To test if inhibition of TNF release by AT was purely an antioxidant property, human monocytes were also incubated in presence of other antioxidants, h-tocopherol (50 and 100 AM), which is not an inhibitor of PKC [20] and ascorbate (0.5 and 1 mM). Role of 5-lipoxygenase We have earlier shown that AT inhibits 5-LO activity as evidenced by a decrease in the product, LTB4 [15]. To test the hypothesis that AT could inhibit TNF release via inhibition of 5-LO, we tested the effect of two inhibitors of 5-LO, MK886 and REV5901, on TNF release and LTB4 levels. Human monocytes were incubated with 1 AM MK886 and 10 AM REV5901 for 30 min before addition of LPS and incubated for 18 h at 378C. Both TNF and LTB4 levels were measured in the supernates. Role of cycloxygenase Monocytes were incubated in presence of the cycloxygenase (COX) inhibitor, indomethacin (10 AM) and in presence of 100 AM AT for 2 h before the addition of LPS. PGE2 release and TNF levels were measured in the cell culture supernatants following an 18-h incubation. Effect of AT on TNF mRNA Monocytes were incubated with AT (100 AM) or MK886 (1 AM) during the 2-h adherence incubation and for 30 min following three washes with RPM1 1640 medium. Thereafter, the monocytes were incubated with LPS for 4 h at 378C. RNA was isolated using Trizol reagents from Gibco BRL [27]. TNF mRNA was quantitated using RNase protection assay, employing reagents from Pharmingen using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as control [28].
assessed in nuclear extracts by EMSA as described previously [16]. Statistical analysis Data were expressed as the mean F SE of at least three experiments. Statistical analysis was performed by Student’s t test to determine differences. Significance was defined at the 5% level.
Results Effect of in vivo AT supplementation on TNF release from activated human monocytes AT supplementation resulted in 2-fold enrichment in monocytes as shown previously [14]. Also, AT supplementation resulted in a significant inhibition of TNF release from activated human monocytes compared to both baseline and washout phases (Fig. 1). Effect of in vitro AT enrichment on TNF release from human monocytes The effect of AT on TNF release from LPS-activated monocytes is shown in Fig. 2. There was a dose-dependent inhibition of TNF release from LPS-activated monocytes ( P b 0.001). Since AT has been shown to inhibit PKC activity in smooth muscle cells and platelets [21,22] and we have demonstrated this effect in human monocytes [14,15], we tested the effect of specific inhibitors of PKC on TNF release and the results are shown in Fig. 3; both the catalytic and the regulatory subunit inhibitors of PKC had no effect on TNF release from LPS-activated monocytes. Since AT is a known chain-breaking antioxidant which preserves membrane integrity [23], it could prevent induction of TNF release by decreasing reactive oxygen species (ROS). Hence, we tested the effect of scavengers of ROS,
Effect of AT on NFjB DNA-binding activity Monocytes were incubated in presence and absence of inhibitors to NF-nB, SN-50 (10 AM) and Bay 11-7085 (10 AM), and TNF release was assayed. Also, monocytes were incubated in the presence and absence of AT (100 AM) or MK886 (1 AM) during the 2-h adherence incubation and for 30 min following three washes with RPM1 1640 medium. Thereafter, the monocytes were incubated with LPS for 4 h at 378C. NF-nB DNA-binding activity was
Fig. 1. Effect of RRR-AT supplementation on monocyte TNF release. Healthy human volunteers were supplemented with RRR-AT (1200 IU/day) for 8 weeks followed by a 2-week washout. Monocytes were isolated and activated with LPS and TNF release was measured in the supernates and described under Materials and methods.
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
1215
Fig. 2. Dose response effect of AT on TNF release. Human peripheral blood mononuclear cells were isolated by Ficoll Hypaque gradient and incubated for 2 h at 378C in 5% C02/95% air in the presence and absence of AT (25–100 AM), after which nonadherent cells were removed by three washes with RPMI 1640 medium. Thereafter the cells were incubated for an additional 30 min in the presence and absence of AT. Monocytes were then activated with LPS and TNF release was measured following an 18-h incubation at 378C by sandwich ELISA as described under Materials and methods. Data are mean F SE of four separate experiments performed in duplicate.
namely, superoxide dismutase (PEG-SOD) and Tiron, on TNF release from LPS-activated human monocytes. Incubation of monocytes with PEG-SOD or Tiron did not produce any significant reduction in TNF release from LPSactivated cells (Fig. 4). To test if the inhibition of TNF release by AT was purely an antioxidant function, we tested the effect of other antioxidants, ascorbate and h-tocopherol, on TNF release from human monocytes. Incubation with ascorbate (0.5 and 1 mM) or h-tocopherol (BT, 50 and 100 AM) did not affect TNF release from activated monocytes (Fig. 5). Leukotriene B4 , a product of 5-LO activity, has been shown to increase TNF release from monocytes [24,25] and AT has been shown to inhibit 5-LO [15,17–19]. As shown in Fig. 6, AT (100 AM) also decreased TNF release from
LPS-activated monocytes. AT also decreased LTB4 levels from activated monocytes (LPS: 99 F 34 pmol/mg protein; LPS + AT 100 AM: 42 F 21 pmol/mg protein). Since AT decreased LTB4 levels and TNF release from activated monocytes, it could possibly act via inhibition of 5-LO. To test this hypothesis, we added LTB4 (10 7 M) to activated monocytes incubated with AT. When LTB4 was added along with AT, there was a reversal of the inhibition of TNF release from activated monocytes incubated with AT (Fig. 6). To further determine the role of LTB4 on TNF release from activated monocytes, we tested the effect of two inhibitors of 5-LO, MK886 and REV5901, on TNF release and LTB4 levels from activated monocytes. The 5-LO inhibitors, MK886 and REV5901, significantly inhibited
Fig. 3. Effect of AT on TNF release in human monocytes: Role of PKC. Mononuclear cells were incubated with bisindoleylmaleimide (0.1 and 1.0 AM) or AT during the 2-h adherence incubation as described in the legend to Fig. 2. Thereafter the cells were stimulated with LPS.TNF was measured as described under Materials and methods. Data are mean F SE of three experiments.
1216
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Fig. 4. Effect of superoxide dismutase (SOD) and Tiron on TNF release from LPS-activated monocytes. Monocytes were incubated with either PEG-SOD (100 Ag/ml) or Tiron (1 mM) during the 2-h adherence incubation as described in the legend to Fig. 1. Thereafter the cells were activated with LPS. TNF release was measured following an 18-h incubation at 378C. Data are mean F SE of four experiments.
LTB4 levels in LPS-activated monocytes as shown previously [15]. Both inhibitors were also able to inhibit TNF release from activated human monocytes (42 and 51% inhibition, respectively). When LTB4 was added to either system (i.e., LPS + MK886 or LPS + REV5901), TNF release was restored (Fig. 7). Furthermore, the combination of AT + MK886 was not additive when compared to either alone with regard to inhibition of TNF release (Fig. 7). To examine the possibility that AT could act via the cycloxygenase pathway, we measured the release of a major product of COX, prostaglandin E2, from cells incubated in the presence of AT and in the presence of indomethacin, a known inhibitor of COX. While indomethacin decreased PGE2 release from activated cells by 94%, AT only produced a nonsignificant (12%) decrease in PGE2 release
as shown previously [15]. However, indomethacin did not have any effect on TNF release from activated monocytes (data not shown). Furthermore, when MK886 was incubated alone or with indomethacin, there was a significant reduction in TNF release from activated monocytes (Fig. 8). Also, in the presence of indomethacin, AT significantly decreased TNF release from activated cells (data not shown). As shown in Fig. 9, AT as well as the 5-LO inhibitor, MK886, significantly decreased mRNA synthesis for TNF from LPS-activated monocytes. Since the TNF promoter has nb-response elements, we tested the effect of NF-nb inhibitors on TNF. Both Bay 11 and SN 50 significantly reduced TNF-a release from activated monocytes (LPS, 4.5 F 1.4 nmol/mg protein; LPS + Bay 11, 3.2 F
Fig. 5. Effect of antioxidants on TNF release. Human monocytes were incubated along with AT or BT (50 and 100 AM, respectively) or with ascorbate (1 mM) during the 2-h adherence incubation and for an additional 30 min as described in the legend to Fig. 1. Following activation with LPS, monocytes were incubated overnight and TNF release was measured in the supernatants by ELISA as described under Materials and methods. Data are mean F SE of three experiments.
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Fig. 6. Effect of AT on TNF release from human monocytes: Role of 5-LO. Mononuclear cells were incubated with AT (100 AM) during the 2-h adherence incubation and then for an additional 30 min as described in the legend to Fig. 1. LTB4 (10 7 M) was added to one set of monocytes that had been preincubated with AT prior to activation with LPS. TNF was measured in supernates by ELISA. Data are mean F SE of four experiments performed in duplicate.
1.2T nmol/mg protein; LPS+SN-50, 2.4 F 1.1T nmol/mg protein, TP b 0.05 compared to LPS, n = 3). We then tested the effect AT as well as the 5-LO inhibitor MK886 on NFnb-binding activity and both significantly decreased the activity of the transcription factor (Fig. 10).
Discussion Much data support the concept that atherosclerosis is an inflammatory process [1,2]. IL-1h, IL-6, and TNF are a major proinflammatory cytokines released by activated monocytes. We have shown that AT supplementation (1200 IU/day) to human volunteers resulted in a significant decrease in monocyte IL-1h, IL-6 release [14]. In this study, we show
1217
that in vivo supplementation of healthy human volunteers with RRR-AT (1200 IU/day) significantly decreased monocyte TNF release. However, the exact mechanism by which TNF release from monocytes is modulated is not well understood. In this study, we show that in vitro enrichment of human monocytes with AT decreases TNF release from activated monocytes as shown following in vivo supplementation. Thus, we investigated potential mechanisms via which AT could effect this decrease in TNF. To gain some insight into the inhibitory effect of AT on TNF release, we first investigated the effect of PKC inhibition by AT on TNF release. PMA and other phorbol esters are thought to induce TNF activity through activation of cAMP and PKC [26]. AT has been shown to inhibit PKC activity in monocytes, vascular smooth muscle cells, and human platelets [15,21,22]. However, no appreciable decrease in TNF release from LPS-activated monocytes was observed in our studies using the regulatory or catalytic subunit inhibitor of PKC, Calphostin C and bisindoleylmaleimide (BIM), respectively. While in PMA-activated monocytes, PKC and MAPK appear to drive TNF-a release [26], Shames et al. [27] have also shown previously that LPS-induced TNF release from human monocytes is independent of PKC and was not decreased in the presence of staurosporine or BIM. Thus, AT does not appear to inhibit TNF release from activated human monocytes via inhibition of PKC. Since AT is a known chain-breaking antioxidant which provides membrane integrity, it could prevent induction of TNF release by decreasing ROS. Hence, we studied the effect of PEG-SOD on TNF release from activated monocytes. Both encapsulated SOD and Tiron had no effect on TNF release from human monocytes. To differentiate its general antioxidant effect from other intracellular effects, we
Fig. 7. Effect of 5-Lipoxygenase inhibitors on TNF release from LPS-activated monocytes. Adherent monocytes were incubated with AT or the 5-LO inhibitors, MK886 (1 AM) and REV5901 (10 AM), for 30 min before the addition of LPS. LTB4 (10 7 M) was added to one set of monocytes that had been preincubated with the 5-LO inhibitors. Following an 18-h incubation, LTB4 and TNF release was measured in the supernatants as described under Materials and methods. Data are mean F SE of four experiments performed in duplicate.
1218
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Fig. 8. Effect of AT on TNF release from monocytes: Role of COX. Monocytes were incubated with AT (100 AM) or Indomethacin or MK886 or MK886 + Indomethacin during the 2-h adherence incubation and for an additional 30 min at 378C following the washes. Thereafter, the cells were activated with LPS and TNF levels were measured following an overnight incubation at 378C. Data are mean F SE of three experiments performed in duplicate.
studied the effect of BT, another antioxidant, which does not inhibit PKC [20]. However, BT had no significant effect on TNF release from activated monocytes. Another potent water-soluble antioxidant, ascorbate, also failed to significantly affect TNF release from activated monocytes. Thus in activated human monocytes, AT does not appear to act through a classical chain-breaking antioxidant mechanism to inhibit TNF release. Leukotrienes are a family of potent proinflammatory lipid compounds derived from the metabolism of arachidonic acid via the 5-lipoxygenase pathway. The product of 5-LO, LTB4, has been shown to enhance TNF production from resting and LPS-activated monocytes [24,25]. AT has been shown to inhibit 5-LO activity in monocytes [15]. AT (50 and 100 A) significantly inhibited LTB4 and TNF release from activated monocytes. When LTB4 was added to the system containing AT and LPS, there was a reversal of the inhibition of TNF release from activated monocytes. Thus, AT appears to inhibit TNF release from LPS-activated monocytes via inhibition of LTB4. To test this hypothesis further we also studied specific inhibitors of 5-LO, such as MK886 and REV5901, on TNF release. In the present
study, both inhibitors decreased TNF release from activated monocytes to the same extent as that seen with AT. Also, in both systems, addition of LTB4 reversed the inhibition of TNF release from activated monocytes. In the murine macrophage cell line (RAW 264), 5-LO inhibitors significantly decreased LPS-induced TNF-a release and this effects appeared to be via decreasing TNF synthesis and decreasing message stability [28]. Cycloxygenase-derived arachidonate metabolites, such as PGE2, have been shown to decrease LPS-induced cytokine release [29]. AT had no significant effect on PGE2 levels from LPS-activated monocytes [15]. Also, we show that indomethacin, a COX inhibitor, while significantly decreasing PGE2 levels in LPS-activated monocytes, did not produce any significant difference in TNF release from activated monocytes. Furthermore, when AT was added along with indomethacin, there was inhibition of TNF release from activated monocytes. This suggests that inhibition of the cycloxygenase pathway seems to be overridden by an inhibition of 5-lipoxygenase in the presence of AT.
Fig. 9. Effect of AT on TNFmRNA from activated human monocytes. Monocytes were incubated with AT (100 AM) during the 2-h adherence incubation and for an additional 30 min at 378C following the washes. Thereafter, the cells were activated with LPS for 4 h at 378C. RNA was isolated and TNF mRNA was quantitated by RNAse protection assay as described under Materials and methods.
Fig. 10. Effect of AT on NF-nb DNA-binding activity in activated human monocytes. Monocytes were incubated with AT (100 AM) or MK886 during the 2-h adherence incubation and for an additional 30 min at 378C following the washes. Thereafter, the cells were activated with LPS for 4 h at 378C. Nuclear extracts were prepared and EMSA was run as described under Materials and methods. 10 unlabeled probe was used as competitor.
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
To examine if AT had an effect on TNF synthesis, TNF mRNA was measured. AT significantly decreased mRNA expression for TNF in presence of AT as well as in presence of MK886. Stability of mRNA also did not seem to be altered in presence of AT (data not shown). Since the TNF promoter has nb-response elements, we tested the effect of AT as well as MK 886 on NF-nB binding since this transcription factor activation results in increased synthesis for TNF [30]. We have previously shown that AT significantly decreases NF-nB binding [16]. MK886 also resulted in similar reduction in NF-nB DNA-binding activity. Several investigators [31–33] have shown that LPS induction of human monocytes results in release of TNF via activation of NF-nB. Leukotrienes activate NF-nb [34,35] and leukotriene inhibitors have been shown to decrease NF-nB activity [36]. In this study, we show inhibition of 5-LO by AT results in decreased NF-nB binding and in turn decreased synthesis and secretion of TNF from LPS-activated human monocytes. Thus, we conclude that AT modulation of TNF release in LPS-activated human monocytes appears to be via inhibition of 5-LO, decreased NF-nB activity, resulting in decreased mRNA and protein for TNF. These data provide additional evidence for an anti-inflammatory effect of high-dose RRR-a-tocopherol. While a recent statistically questionable and selective meta-analyses has reported increased mortality with high dose a-tocopherol [37], they omitted 2 studies that clearly showed the benefit of combined RRR-a-tocopherol (natural AT) and vitamin C supplementation on the primary endpoint . Also, in the 11 studies in which they suggested harm from high-dose AT supplementation, it should be pointed out that in 5, AT was used along with other antioxidants including betacarotene, which has previously been shown to be harmful. Furthermore, in 3 of the studies, which used AT alone (CHAOS, SPACE, ADCS), there was a significant benefit on the primary endpoint without a significant increase in mortality. The validity of the meta-analyses with respect to the heterogeneity of the different studies including diverse populations, sample sizes, dose and duration of AT, antioxidant cocktails, form of AT (RRR- vs all rac), omission of use of biomarkers of oxidative stress, and inflammation is thus circumspect. Thus, we believe that while the benefits of high-dose RRR-AT remains to be proven, it should be pointed out that the American Heart Association, in a recent Advisory [38], reviewed in detail antioxidant vitamin supplements and cardiovascular disease and did not conclude that a-tocopherol increased mortality.
Acknowledgments This work was supported in part by research grants from the American Diabetes Association, National Institutes of Health K24 AT 00596.
1219
References [1] Libby, P. Inflammation in atherosclerosis. Nature 420 (6917): 868 – 874; 2002. [2] Lusis, A. J. Atherosclerosis. Nature 407 (6801):233 – 241; 2000. [3] Boring, L.; Gosling, J.; Cleary, M.; Charo, I. F. Decreased lesion formation in CCR2-/- mice reveals a role for chemokines in the initiation of atherosclerosis. Nature 394 (6696):894 – 897; 1998. [4] Gu, L.; Okada, Y.; Clinton, S. K.; Gerard, C.; Sukhova, G. K.; Libby, P.; Rollins, B. J. Absence of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice. Mol. Cell. 2 (2):275 – 281; 1998. [5] Qiao, J. H.; Tripathi, J.; Mishra, N. K.; Cai, Y.; Tripathi, S.; Wang, X. P.; Imes, S.; Fishbein, M. C.; Clinton, S. K.; Libby, P.; Lusis, A. J.; Rajavashisth, T. B. Role of macrophage colony-stimulating factor in atherosclerosis: studies of osteopetrotic mice. Am. J. Pathol. 150 (5):1687 – 1699; 1997. [6] Sack, M. Tumor necrosis factor-a in cardiovascular biology and the potential role for anti-tumor necrosis factor-a therapy in heart disease. Pharmacol. Ther. 94 (1–2):123 – 135; 2002. [7] Rodriguez-Maldonado, E.; Hernandez-Rebollar, A. E.; Lopez-Marure, R.; Masso, F.; Montano, L. F. Role of tumor necrosis factor alpha in atherosclerosis. Arch. Cardiol. Mex. 71 (3):241 – 249; 2001. [8] Halle, M.; Berg, A.; Northoff, H.; Keul, J. Importance of TNF-alpha and leptin in obesity and insulin resistance: a hypothesis on the impact of physical exercise. Exerc. Immunol. Rev. 4:77 – 94; 1998. [9] Dandona, P.; Weinstock, R.; Thusu, K., et al. Tumor necrosis factoralpha in sera of obese patients: fall with weight loss. J. Clin. Endocrinol. Metab. 83 (8):2907 – 2910; 1998. [10] Hotamisligil, G.; Arner, P.; Caro, J., et al. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J. Clin. Invest. 95 (5):2409 – 2415; 1995. [11] Kern, P.; Saghizadeh, M.; Ong, J., et al. The expression of tumor necrosis factor in human adipose tissue. regulation by obesity, weight loss, and relationship to lipoprotein lipase. J. Clin. Invest. 95 (5):2111 – 2119; 1995. [12] Yamakawa, T.; Tanaka, S.; Yamakawa, Y., et al. Augmented production of tumor necrosis factor-alpha in obese mice. Clin. Immunol. Immunopathol. 75 (1):51 – 56; 1995. [13] Hotamisligil, G.; Shargill, N.; Spiegelman, M. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science 259:87 – 91; 1993. [14] Devaraj, S.; Li, D.; Jialal, I. The effects of alpha tocopherol supplementation on monocyte function. Decreased lipid oxidation, interleukin 1 beta secretion, and monocyte adhesion to endothelium. J. Clin. Invest. 98 (3):756 – 763; 1996. [15] Devaraj, S.; Jialal, I. Alpha-tocopherol decreases interleukin-1 beta release from activated human monocytes by inhibition of 5-lipoxygenase. Arterioscler. Thromb. Vasc. Biol. 19 (4):1125 – 1133; 1999. [16] Islam, K. N.; Devaraj, S.; Jialal, I. Alpha-tocopherol enrichment of monocytes decreases agonist-induced adhesion to human endothelial cells. Circulation 98 (21):2255 – 2261; 1998. [17] Goetzl, E. J. Vitamin E modulates the lipoxygenation of arachidonic acid in leukocytes. Nature 288:183 – 185; 1980. [18] Reddanna, P.; Rao, M. K.; Reddy, C. C. Inhibition of 5-lipoxygenase by vitamin E. FEBS Lett. 193 (1):39 – 43; 1985. [19] Reddanna, P.; Whelan, J.; Burgess, J. R.; Eskew, M. L.; Hildenbrandt, G.; Zarkower, A.; Scholz, R. W.; Reddy, C. The role of vitamin E and selenium on arachidonic acid oxidation by way of the 5-lipoxygenase pathway. Ann. N.Y. Acad. Sci. 570:136 – 145; 1989. [20] Fazzio, A.; Marilley, D.; Azzi, A. The effect of alpha-tocopherol and beta-tocopherol on proliferation, protein kinase C activity and gene expression in different cell lines. Biochem. Mol. Biol. Int. 41 (1):93 – 101; 1997. [21] Mahoney, C. W.; Azzi, A. Vitamin E inhibits protein kinase C activity. Biochem. Biophys. Res. Commun. 154 (2):694 – 697; 1998.
1220
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
[22] Freedman, J. E.; Farhat, J. H.; Loscalzo, J.; Keaney, J. F. Alphatocopherol inhibits aggregation of human platelets by a protein kinase C-dependent mechanism. Circulation 94 (10):2434 – 2440; 1996. [23] Burton, G. W.; Ingold, K. U. Vitamin E as an in vitro and in vivo antioxidant. Ann. N.Y. Acad. Sci. 570:7 – 22; 1989. [24] Schade, F. U.; Engel, R.; Jakobs, D. Lipoxygenase inhibitors but not site specific 5-lipoxygenase blockers protect against endotoxic shock and inhibit production of tumor necrosis factor. Eicosanoids 5 Suppl:S45 – S47; 1992. [25] Stankova, J.; Dupuis, G.; Gagnon, N.; Thivierge, M.; Turcotte, S.; Rola-Pleszczynski, M. Priming of human monocytes with leukotriene B4 enhances their sensitivity in IL-2-driven tumor necrosis factor-alpha production. Transcriptional and post-transcriptional upregulation of IL-2 receptors. J. Immunol. 150 (9):4041 – 4051; 1993. [26] Kontny, E.; Ziolkowska, M.; Ryzewska, A.; Maslinski, W. Protein kinase c-dependent pathway is critical for the production of proinflammatory cytokines (TNF-alpha, IL-1beta, IL-6). Cytokine 11:839 – 848; 1999. [27] Shames, B. D.; Selzman, C. H.; Pulido, E. J.; Meng, X.; Meldrum, D. R.; McIntyre, R. C. Jr.; Harken, A. H.; Banerjee, A. LPS-Induced NFkappaB activation and TNF-alpha release in human monocytes are protein tyrosine kinase dependent and protein kinase C independent. J. Surg. Res. 83 (1):69 – 74; 1999. [28] Lin, H. I.; Chu, S. J.; Wang, D.; Feng, N. H. Pharmacological modulation of TNF production in macrophages. J. Microbiol Immunol. Infect. 37 (1):8 – 15; 2004. [29] Rouzer, C. A.; Kingsley, P. J.; Wang, H.; Zhang, H.; Morrow, J. D.; Dey, S. K.; Marnett, L. J. Cyclooxygenase-1-dependent prostaglandin synthesis modulates tumor necrosis factor-alpha secretion in lipopolysaccharide-challenged murine resident peritoneal macrophages. J. Biol Chem. 279 (33):34256 – 34268; 2004. [30] Kwon, H. J.; Lee, K. W.; Yu, S. H.; Han, J. H.; Kim, D. S. NFkappaB-dependent regulation of tumor necrosis factor-alpha gene
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
expression by CpG-oligodeoxynucleotides. Biochem. Biophys. Res. Commun. 311 (1):129 – 138; 2003. Zhou, Z.; Wang, L.; Song, Z.; Saari, J. T.; McClain, C. J.; Kang, Y. J. Abrogation of nuclear factor-kappaB activation is involved in zinc inhibition of lipopolysaccharide-induced tumor necrosis factor-alpha production and liver injury. Am. J. Pathol. 164 (5):1547 – 1556; 2004. Araujo, F.; Slifer, T.; Li, S.; Kuver, A.; Fong, L.; Remington, J. Gemifloxacin inhibits cytokine secretion by lipopolysaccharide stimulated human monocytes at the post-transcriptional level. Clin. Microbiol. Infect. 10 (3):213 – 219; 2004. Haversen, L.; Ohlsson, B. G.; Hahn-Zoric, M.; Hanson, L. A.; Mattsby-Baltzer, I. Lactoferrin down-regulates the LPS-induced cytokine production in monocytic cells via NF-kappa B. Cell. Immunol. 220 (2):83 – 95; 2002. Kawano, T.; Matsuse, H.; Kondo, Y.; Machida, I.; Saeki, S.; Tomari, S.; Mitsuta, K.; Obase, Y.; Fukushima, C.; Shimoda, T.; Kohno, S. Cysteinyl leukotrienes induce nuclear factor kappa b activation and RANTES production in a murine model of asthma. J. Allergy Clin. Immunol. 112 (2):369 – 374; 2003. Aoki, Y.; Qiu, D.; Zhao, G. H.; Kao, P. N. Leukotriene B4 mediates histamine induction of NF-kappaB and IL-8 in human bronchial epithelial cells. Am. J. Physiol. 274 (6 Pt 1):L1030 – L1039; 1998. van Puijenbroek, A. A.; Wissink, S.; van der Saag, P. T.; Peppelenbosch, M. P. Phospholipase A2 inhibitors and leukotriene synthesis inhibitors block TNF-induced NF-kappaB activation. Cytokine 11 (2):104 – 110; 1999. Miller, E. R. 3rd; Pastor-Barriuso, R.; Dalal, D.; Riemersma, R. A.; Appel, L. J.; Gualla, E. Meta-analysis: high-dosage vitamin E supplementation may increase all-cause mortality. Ann Intern Med. 142 (1):37–46; 2004. Kris-Etherton, P. M.; Lichtenstein, A. H.; Howard, B. V.; Steinberg, D.; Witztum, J. L. Nutrition Committee of the American Heart Association Council on Nutrition, Physical Activity, and Metabolism. Antioxidant vitamin supplements and cardiovascular disease. Circulation 110 (5):637 – 641; 2004.
Original Contribution
a-Tocopherol decreases tumor necrosis factor-a mRNA and protein from activated human monocytes by inhibition of 5-lipoxygenase Sridevi DevarajT, Ishwarlal Jialal Laboratory for Atherosclerosis and Metabolic Research, Department of Pathology and Laboratory Medicine, University of California, Davis Medical Center, Sacramento, CA 95817, USA Received 8 October 2004; revised 4 January 2005; accepted 11 January 2005
Abstract Cardiovascular disease is the leading cause of morbidity in Westernized populations. Low levels of a-tocopherol (AT) are associated with increased incidence of atherosclerosis and increased intakes appear to be protective. AT supplementation decreases interleukin 1 and 6 release from human monocytes. Thus, the aim of this study was to examine the effect of AT on an important proinflammatory cytokine, tumor necrosis factor-a (TNF) release from human monocytes. AT supplementation (1200 IU/day for 3 months) significantly decreased TNF release from activated human monocytes. Mechanisms that were examined included its effect as a general antioxidant, its inhibitory effect on protein kinase C (PKC), and the cycloxygenase-lipoxygenase pathway. While AT decreased TNF release from activated monocytes, other antioxidants had no effect on TNF release. Specific PKC inhibitors had no effect on TNF release from activated monocytes. The inhibition of TNF release by AT in activated monocytes was reversed by leukotriene B4 (LTB4), a major product of the 5-lipoxygenase (5-LO) pathway. Similar observations were seen with inhibitors of 5-lipoxygenase. Indomethacin, a COX inhibitor, in the presence and absence of AT failed to affect TNF activity. These findings suggest that AT decreases TNF release from activated human monocytes via inhibition of 5-lipoxygenase. Also, AT as well as a 5-LO inhibitor significantly decreased TNF mRNA. Furthermore, AT and the 5-LO inhibitor decreased NFnb-binding activity. Thus, in activated human monocytes, AT appears to inhibit TNF mRNA and protein by inhibition of 5-LO. D 2005 Elsevier Inc. All rights reserved. Keywords: a tocopherol; Monocytes; Cytokine; Lipoxygenase; Free radicals
Introduction Cardiovascular disease is the leading cause of morbidity and mortality in the United states. Several lines of evidence point to the pivotal role of inflammation in atherosclerosis [1,2]. Evidence from knockout models shows that the monocyte-macrophage is a crucial cell in atherosclerosis [3–5]. Monocytes secrete several proinflammatory cytokines, the most notable being interleukin-1, tumor necrosis factor-a, and interleukin-6. Several lines of evidence Abbreviations: AT, a-tocopherol; BIM, bisindoleylmaleimide; BT, h-tocopherol; COX, cycloxygenase; LTB4, leukotriene B4; 5-LO, 5lipoxygenase; PKC, protein kinase C; ROS, reactive oxygen species; RRR-AT, RRR-a-tocopherol; SOD, superoxide dismutase; TNF, tumor necrosis factor. T Corresponding author. Fax: (916) 734 6593. E-mail address: [email protected] (S. Devaraj). 0891-5849/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.freeradbiomed.2005.01.009
suggest that the proinflammatory cytokine, tumor necrosis factor-a (TNF) is proatherogenic [6–8]. TNF-a is a multifunctional cytokine that exerts pleiotropic biological actions [9–13]. It activates endothelial cells, promotes monocyteendothelial cell adhesion, stimulates angiogenesis, and induces smooth muscle cell proliferation. Increased mRNA for TNF-a has been documented in aortas of Watanabe heritable hyperlipidemic rabbits and carotid atherosclerotic plaques. TNF-a processing via its receptor can promote apoptosis and thus could contribute to the necrotic core of the atherosclerotic lesion. TNF-a also plays a role in the pathogenesis of obesity-linked insulin resistance [9–13]. Hence, in this study, we examined if TNF release from LPSactivated human monocytes could be modulated. We have previously shown that, RRR-a-tocopherol (RRR-AT), at high doses, exerts anti-inflammatory effects [14]. We showed that in vivo supplementation of human
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
volunteers with RRR-AT (1200 IU/day) significantly decreases monocyte superoxide anion release, lipid oxidation, interleukin-6 and interleukin-1h release, and adhesion to endothelial cells [14]. The inhibition of superoxide anion release and lipid oxidation was mediated by inhibition of protein kinase C-a by AT. Subsequently, we showed that AT inhibits IL-1h release via inhibition of 5-lipoxygenase (5LO) [15] and decreased monocyte-endothelial cell adhesion by decreasing expression of CD11b and VLA-4 and decreased DNA-binding activity of the transcription factor, NFnb [16]. In this study, we examined if monocyte TNF release can be modulated with RRR-AT and the mechanisms involved. The potential mechanisms that were examined included its effect as a general antioxidant, its inhibitory effect on protein kinase C activity, and its effect on the cycloxygenase-lipoxygenase pathway, since it has previously been shown that AT can inhibit both cycloxygenase and lipoxygenase in certain systems [15,17–19]. Also, the effect of AT on mRNA for TNF was studied.
1213
Determination of protein kinase C (PKC) activity PKC activity in human monocytes in the presence and absence of 50 and 100 AM AT was performed by a radioimmunoassay technique using reagents from Amersham Corp. as described previously [14]. The PKC activity assay is based on the PKC-catalyzed transfer of the g -phosphate group of ATP to a PKC-specific peptide. PKC activity is expressed as nanomoles of phosphate transferred per million cells. Quantitation of leukotriene B4 (LTB4) and prostaglandin E2 (PGE2) levels LTB4 levels were measured in the cell culture supernatants from monocytes activated with LPS in the absence and presence of 5-LO inhibitors and AT and assayed by an enzyme immunoassay as described previously (Amersham Corp.) [15]. PGE2 levels were determined using an identical enzyme immunoassay as described for LTB4 [15]. The inter assay CVs for these assays was b10%.
Materials and methods Dose response of AT on TNF release Isolation of human peripheral blood monocytes For the in vivo study, subjects (n = 21) were supplemented with RRR-AT (1200 IU/d) for a period of 8 weeks followed by a 6-week washout phase as described previously [14]. For the in vitro studies, fasting blood was obtained from healthy volunteers who were not taking any antioxidant supplements. All subjects gave informed consent and this study was approved by the Institutional Review Board. Human mononuclear cells were separated from 120 ml of blood from fasting, healthy volunteers by Ficoll Hypaque gradient as described previously [14]. Cells were plated and incubation was carried out at 378C for 2 h in 5% CO2/95% air, after which nonadherent cells were removed after washing three times with phenol red-free RPMI 1640 medium. Nonspecific esterase staining revealed that 88% of cells isolated in this manner were monocytes [14]. All incubations were carried out immediately following isolation of the monocytes. All reagents used were tested for endotoxin contamination by the Limulus endotoxin assay [14], which was b12.5 pg/ml. The viability of the monocytes was found to be 94% by trypan blue exclusion [14]. LPS (1 `ıg/ml) was used to activate monocytes. TNF release was measured in the supernatants following an overnight incubation. Release of TNF The release of TNF was measured in the culture supernatants by ELISA using the human immunoassay kit (Biotrak High Sensitivity Kit, Amersham Corp., Arlington Heights, IL). Interassay CV was b10%.
Mononuclear cells were incubated with AT (25, 50, and 100 AM) during the 2-h adherence incubation and for 30 min following the three washes with RPMI 1640. Thereafter the monocytes were activated with LPS and incubated at 378C for 18 h. The supernatants were then collected and TNF secretion was assayed by ELISA. The cells were dissolved with 0.1 N NaOH and the protein content was measured by the method of Lowry et al. and TNF activity was expressed as nanograms of TNF per milligram protein. Potential mechanisms by which AT could inhibit TNF release include: (i) inhibition of protein kinase C; (ii) inhibition of reactive oxygen species; (iii) by acting as a chain-breaking antioxidant; (iv) inhibition of 5-lipoxygenase and thereby leukotriene B4; (v) inhibition of the cycloxygenase pathway; and (vi) inhibition of TNF synthesis. Role of protein kinase C Mononuclear cells were incubated with PKC inhibitors, Calphostin C (250 and 500 nM) or with bisindoleylmaleimide (0.1, 1 AM) during the 2-h incubation for adherence of monocytes. Thereafter, the cells were washed three times with RPMI 1640 medium and monocytes were activated with LPS. Following an 18-h incubation, TNF activity was measured in the supernatants. The concentrations of the inhibitors used did not show any antioxidant activity as determined by assaying the lag phase of copper-catalyzed LDL oxidation (data not shown).
1214
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Role of reactive oxygen species The effect of the antioxidant enzymes, superoxide dismutase (PEG-SOD, 100 Ag/ml) and Tiron (1 AM), was tested on human monocytes by adding them during the 2-h adherence incubation of the mononuclear cells prior to addition of LPS. TNF activity was measured after an 18-h incubation. To test if inhibition of TNF release by AT was purely an antioxidant property, human monocytes were also incubated in presence of other antioxidants, h-tocopherol (50 and 100 AM), which is not an inhibitor of PKC [20] and ascorbate (0.5 and 1 mM). Role of 5-lipoxygenase We have earlier shown that AT inhibits 5-LO activity as evidenced by a decrease in the product, LTB4 [15]. To test the hypothesis that AT could inhibit TNF release via inhibition of 5-LO, we tested the effect of two inhibitors of 5-LO, MK886 and REV5901, on TNF release and LTB4 levels. Human monocytes were incubated with 1 AM MK886 and 10 AM REV5901 for 30 min before addition of LPS and incubated for 18 h at 378C. Both TNF and LTB4 levels were measured in the supernates. Role of cycloxygenase Monocytes were incubated in presence of the cycloxygenase (COX) inhibitor, indomethacin (10 AM) and in presence of 100 AM AT for 2 h before the addition of LPS. PGE2 release and TNF levels were measured in the cell culture supernatants following an 18-h incubation. Effect of AT on TNF mRNA Monocytes were incubated with AT (100 AM) or MK886 (1 AM) during the 2-h adherence incubation and for 30 min following three washes with RPM1 1640 medium. Thereafter, the monocytes were incubated with LPS for 4 h at 378C. RNA was isolated using Trizol reagents from Gibco BRL [27]. TNF mRNA was quantitated using RNase protection assay, employing reagents from Pharmingen using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as control [28].
assessed in nuclear extracts by EMSA as described previously [16]. Statistical analysis Data were expressed as the mean F SE of at least three experiments. Statistical analysis was performed by Student’s t test to determine differences. Significance was defined at the 5% level.
Results Effect of in vivo AT supplementation on TNF release from activated human monocytes AT supplementation resulted in 2-fold enrichment in monocytes as shown previously [14]. Also, AT supplementation resulted in a significant inhibition of TNF release from activated human monocytes compared to both baseline and washout phases (Fig. 1). Effect of in vitro AT enrichment on TNF release from human monocytes The effect of AT on TNF release from LPS-activated monocytes is shown in Fig. 2. There was a dose-dependent inhibition of TNF release from LPS-activated monocytes ( P b 0.001). Since AT has been shown to inhibit PKC activity in smooth muscle cells and platelets [21,22] and we have demonstrated this effect in human monocytes [14,15], we tested the effect of specific inhibitors of PKC on TNF release and the results are shown in Fig. 3; both the catalytic and the regulatory subunit inhibitors of PKC had no effect on TNF release from LPS-activated monocytes. Since AT is a known chain-breaking antioxidant which preserves membrane integrity [23], it could prevent induction of TNF release by decreasing reactive oxygen species (ROS). Hence, we tested the effect of scavengers of ROS,
Effect of AT on NFjB DNA-binding activity Monocytes were incubated in presence and absence of inhibitors to NF-nB, SN-50 (10 AM) and Bay 11-7085 (10 AM), and TNF release was assayed. Also, monocytes were incubated in the presence and absence of AT (100 AM) or MK886 (1 AM) during the 2-h adherence incubation and for 30 min following three washes with RPM1 1640 medium. Thereafter, the monocytes were incubated with LPS for 4 h at 378C. NF-nB DNA-binding activity was
Fig. 1. Effect of RRR-AT supplementation on monocyte TNF release. Healthy human volunteers were supplemented with RRR-AT (1200 IU/day) for 8 weeks followed by a 2-week washout. Monocytes were isolated and activated with LPS and TNF release was measured in the supernates and described under Materials and methods.
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
1215
Fig. 2. Dose response effect of AT on TNF release. Human peripheral blood mononuclear cells were isolated by Ficoll Hypaque gradient and incubated for 2 h at 378C in 5% C02/95% air in the presence and absence of AT (25–100 AM), after which nonadherent cells were removed by three washes with RPMI 1640 medium. Thereafter the cells were incubated for an additional 30 min in the presence and absence of AT. Monocytes were then activated with LPS and TNF release was measured following an 18-h incubation at 378C by sandwich ELISA as described under Materials and methods. Data are mean F SE of four separate experiments performed in duplicate.
namely, superoxide dismutase (PEG-SOD) and Tiron, on TNF release from LPS-activated human monocytes. Incubation of monocytes with PEG-SOD or Tiron did not produce any significant reduction in TNF release from LPSactivated cells (Fig. 4). To test if the inhibition of TNF release by AT was purely an antioxidant function, we tested the effect of other antioxidants, ascorbate and h-tocopherol, on TNF release from human monocytes. Incubation with ascorbate (0.5 and 1 mM) or h-tocopherol (BT, 50 and 100 AM) did not affect TNF release from activated monocytes (Fig. 5). Leukotriene B4 , a product of 5-LO activity, has been shown to increase TNF release from monocytes [24,25] and AT has been shown to inhibit 5-LO [15,17–19]. As shown in Fig. 6, AT (100 AM) also decreased TNF release from
LPS-activated monocytes. AT also decreased LTB4 levels from activated monocytes (LPS: 99 F 34 pmol/mg protein; LPS + AT 100 AM: 42 F 21 pmol/mg protein). Since AT decreased LTB4 levels and TNF release from activated monocytes, it could possibly act via inhibition of 5-LO. To test this hypothesis, we added LTB4 (10 7 M) to activated monocytes incubated with AT. When LTB4 was added along with AT, there was a reversal of the inhibition of TNF release from activated monocytes incubated with AT (Fig. 6). To further determine the role of LTB4 on TNF release from activated monocytes, we tested the effect of two inhibitors of 5-LO, MK886 and REV5901, on TNF release and LTB4 levels from activated monocytes. The 5-LO inhibitors, MK886 and REV5901, significantly inhibited
Fig. 3. Effect of AT on TNF release in human monocytes: Role of PKC. Mononuclear cells were incubated with bisindoleylmaleimide (0.1 and 1.0 AM) or AT during the 2-h adherence incubation as described in the legend to Fig. 2. Thereafter the cells were stimulated with LPS.TNF was measured as described under Materials and methods. Data are mean F SE of three experiments.
1216
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Fig. 4. Effect of superoxide dismutase (SOD) and Tiron on TNF release from LPS-activated monocytes. Monocytes were incubated with either PEG-SOD (100 Ag/ml) or Tiron (1 mM) during the 2-h adherence incubation as described in the legend to Fig. 1. Thereafter the cells were activated with LPS. TNF release was measured following an 18-h incubation at 378C. Data are mean F SE of four experiments.
LTB4 levels in LPS-activated monocytes as shown previously [15]. Both inhibitors were also able to inhibit TNF release from activated human monocytes (42 and 51% inhibition, respectively). When LTB4 was added to either system (i.e., LPS + MK886 or LPS + REV5901), TNF release was restored (Fig. 7). Furthermore, the combination of AT + MK886 was not additive when compared to either alone with regard to inhibition of TNF release (Fig. 7). To examine the possibility that AT could act via the cycloxygenase pathway, we measured the release of a major product of COX, prostaglandin E2, from cells incubated in the presence of AT and in the presence of indomethacin, a known inhibitor of COX. While indomethacin decreased PGE2 release from activated cells by 94%, AT only produced a nonsignificant (12%) decrease in PGE2 release
as shown previously [15]. However, indomethacin did not have any effect on TNF release from activated monocytes (data not shown). Furthermore, when MK886 was incubated alone or with indomethacin, there was a significant reduction in TNF release from activated monocytes (Fig. 8). Also, in the presence of indomethacin, AT significantly decreased TNF release from activated cells (data not shown). As shown in Fig. 9, AT as well as the 5-LO inhibitor, MK886, significantly decreased mRNA synthesis for TNF from LPS-activated monocytes. Since the TNF promoter has nb-response elements, we tested the effect of NF-nb inhibitors on TNF. Both Bay 11 and SN 50 significantly reduced TNF-a release from activated monocytes (LPS, 4.5 F 1.4 nmol/mg protein; LPS + Bay 11, 3.2 F
Fig. 5. Effect of antioxidants on TNF release. Human monocytes were incubated along with AT or BT (50 and 100 AM, respectively) or with ascorbate (1 mM) during the 2-h adherence incubation and for an additional 30 min as described in the legend to Fig. 1. Following activation with LPS, monocytes were incubated overnight and TNF release was measured in the supernatants by ELISA as described under Materials and methods. Data are mean F SE of three experiments.
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Fig. 6. Effect of AT on TNF release from human monocytes: Role of 5-LO. Mononuclear cells were incubated with AT (100 AM) during the 2-h adherence incubation and then for an additional 30 min as described in the legend to Fig. 1. LTB4 (10 7 M) was added to one set of monocytes that had been preincubated with AT prior to activation with LPS. TNF was measured in supernates by ELISA. Data are mean F SE of four experiments performed in duplicate.
1.2T nmol/mg protein; LPS+SN-50, 2.4 F 1.1T nmol/mg protein, TP b 0.05 compared to LPS, n = 3). We then tested the effect AT as well as the 5-LO inhibitor MK886 on NFnb-binding activity and both significantly decreased the activity of the transcription factor (Fig. 10).
Discussion Much data support the concept that atherosclerosis is an inflammatory process [1,2]. IL-1h, IL-6, and TNF are a major proinflammatory cytokines released by activated monocytes. We have shown that AT supplementation (1200 IU/day) to human volunteers resulted in a significant decrease in monocyte IL-1h, IL-6 release [14]. In this study, we show
1217
that in vivo supplementation of healthy human volunteers with RRR-AT (1200 IU/day) significantly decreased monocyte TNF release. However, the exact mechanism by which TNF release from monocytes is modulated is not well understood. In this study, we show that in vitro enrichment of human monocytes with AT decreases TNF release from activated monocytes as shown following in vivo supplementation. Thus, we investigated potential mechanisms via which AT could effect this decrease in TNF. To gain some insight into the inhibitory effect of AT on TNF release, we first investigated the effect of PKC inhibition by AT on TNF release. PMA and other phorbol esters are thought to induce TNF activity through activation of cAMP and PKC [26]. AT has been shown to inhibit PKC activity in monocytes, vascular smooth muscle cells, and human platelets [15,21,22]. However, no appreciable decrease in TNF release from LPS-activated monocytes was observed in our studies using the regulatory or catalytic subunit inhibitor of PKC, Calphostin C and bisindoleylmaleimide (BIM), respectively. While in PMA-activated monocytes, PKC and MAPK appear to drive TNF-a release [26], Shames et al. [27] have also shown previously that LPS-induced TNF release from human monocytes is independent of PKC and was not decreased in the presence of staurosporine or BIM. Thus, AT does not appear to inhibit TNF release from activated human monocytes via inhibition of PKC. Since AT is a known chain-breaking antioxidant which provides membrane integrity, it could prevent induction of TNF release by decreasing ROS. Hence, we studied the effect of PEG-SOD on TNF release from activated monocytes. Both encapsulated SOD and Tiron had no effect on TNF release from human monocytes. To differentiate its general antioxidant effect from other intracellular effects, we
Fig. 7. Effect of 5-Lipoxygenase inhibitors on TNF release from LPS-activated monocytes. Adherent monocytes were incubated with AT or the 5-LO inhibitors, MK886 (1 AM) and REV5901 (10 AM), for 30 min before the addition of LPS. LTB4 (10 7 M) was added to one set of monocytes that had been preincubated with the 5-LO inhibitors. Following an 18-h incubation, LTB4 and TNF release was measured in the supernatants as described under Materials and methods. Data are mean F SE of four experiments performed in duplicate.
1218
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
Fig. 8. Effect of AT on TNF release from monocytes: Role of COX. Monocytes were incubated with AT (100 AM) or Indomethacin or MK886 or MK886 + Indomethacin during the 2-h adherence incubation and for an additional 30 min at 378C following the washes. Thereafter, the cells were activated with LPS and TNF levels were measured following an overnight incubation at 378C. Data are mean F SE of three experiments performed in duplicate.
studied the effect of BT, another antioxidant, which does not inhibit PKC [20]. However, BT had no significant effect on TNF release from activated monocytes. Another potent water-soluble antioxidant, ascorbate, also failed to significantly affect TNF release from activated monocytes. Thus in activated human monocytes, AT does not appear to act through a classical chain-breaking antioxidant mechanism to inhibit TNF release. Leukotrienes are a family of potent proinflammatory lipid compounds derived from the metabolism of arachidonic acid via the 5-lipoxygenase pathway. The product of 5-LO, LTB4, has been shown to enhance TNF production from resting and LPS-activated monocytes [24,25]. AT has been shown to inhibit 5-LO activity in monocytes [15]. AT (50 and 100 A) significantly inhibited LTB4 and TNF release from activated monocytes. When LTB4 was added to the system containing AT and LPS, there was a reversal of the inhibition of TNF release from activated monocytes. Thus, AT appears to inhibit TNF release from LPS-activated monocytes via inhibition of LTB4. To test this hypothesis further we also studied specific inhibitors of 5-LO, such as MK886 and REV5901, on TNF release. In the present
study, both inhibitors decreased TNF release from activated monocytes to the same extent as that seen with AT. Also, in both systems, addition of LTB4 reversed the inhibition of TNF release from activated monocytes. In the murine macrophage cell line (RAW 264), 5-LO inhibitors significantly decreased LPS-induced TNF-a release and this effects appeared to be via decreasing TNF synthesis and decreasing message stability [28]. Cycloxygenase-derived arachidonate metabolites, such as PGE2, have been shown to decrease LPS-induced cytokine release [29]. AT had no significant effect on PGE2 levels from LPS-activated monocytes [15]. Also, we show that indomethacin, a COX inhibitor, while significantly decreasing PGE2 levels in LPS-activated monocytes, did not produce any significant difference in TNF release from activated monocytes. Furthermore, when AT was added along with indomethacin, there was inhibition of TNF release from activated monocytes. This suggests that inhibition of the cycloxygenase pathway seems to be overridden by an inhibition of 5-lipoxygenase in the presence of AT.
Fig. 9. Effect of AT on TNFmRNA from activated human monocytes. Monocytes were incubated with AT (100 AM) during the 2-h adherence incubation and for an additional 30 min at 378C following the washes. Thereafter, the cells were activated with LPS for 4 h at 378C. RNA was isolated and TNF mRNA was quantitated by RNAse protection assay as described under Materials and methods.
Fig. 10. Effect of AT on NF-nb DNA-binding activity in activated human monocytes. Monocytes were incubated with AT (100 AM) or MK886 during the 2-h adherence incubation and for an additional 30 min at 378C following the washes. Thereafter, the cells were activated with LPS for 4 h at 378C. Nuclear extracts were prepared and EMSA was run as described under Materials and methods. 10 unlabeled probe was used as competitor.
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
To examine if AT had an effect on TNF synthesis, TNF mRNA was measured. AT significantly decreased mRNA expression for TNF in presence of AT as well as in presence of MK886. Stability of mRNA also did not seem to be altered in presence of AT (data not shown). Since the TNF promoter has nb-response elements, we tested the effect of AT as well as MK 886 on NF-nB binding since this transcription factor activation results in increased synthesis for TNF [30]. We have previously shown that AT significantly decreases NF-nB binding [16]. MK886 also resulted in similar reduction in NF-nB DNA-binding activity. Several investigators [31–33] have shown that LPS induction of human monocytes results in release of TNF via activation of NF-nB. Leukotrienes activate NF-nb [34,35] and leukotriene inhibitors have been shown to decrease NF-nB activity [36]. In this study, we show inhibition of 5-LO by AT results in decreased NF-nB binding and in turn decreased synthesis and secretion of TNF from LPS-activated human monocytes. Thus, we conclude that AT modulation of TNF release in LPS-activated human monocytes appears to be via inhibition of 5-LO, decreased NF-nB activity, resulting in decreased mRNA and protein for TNF. These data provide additional evidence for an anti-inflammatory effect of high-dose RRR-a-tocopherol. While a recent statistically questionable and selective meta-analyses has reported increased mortality with high dose a-tocopherol [37], they omitted 2 studies that clearly showed the benefit of combined RRR-a-tocopherol (natural AT) and vitamin C supplementation on the primary endpoint . Also, in the 11 studies in which they suggested harm from high-dose AT supplementation, it should be pointed out that in 5, AT was used along with other antioxidants including betacarotene, which has previously been shown to be harmful. Furthermore, in 3 of the studies, which used AT alone (CHAOS, SPACE, ADCS), there was a significant benefit on the primary endpoint without a significant increase in mortality. The validity of the meta-analyses with respect to the heterogeneity of the different studies including diverse populations, sample sizes, dose and duration of AT, antioxidant cocktails, form of AT (RRR- vs all rac), omission of use of biomarkers of oxidative stress, and inflammation is thus circumspect. Thus, we believe that while the benefits of high-dose RRR-AT remains to be proven, it should be pointed out that the American Heart Association, in a recent Advisory [38], reviewed in detail antioxidant vitamin supplements and cardiovascular disease and did not conclude that a-tocopherol increased mortality.
Acknowledgments This work was supported in part by research grants from the American Diabetes Association, National Institutes of Health K24 AT 00596.
1219
References [1] Libby, P. Inflammation in atherosclerosis. Nature 420 (6917): 868 – 874; 2002. [2] Lusis, A. J. Atherosclerosis. Nature 407 (6801):233 – 241; 2000. [3] Boring, L.; Gosling, J.; Cleary, M.; Charo, I. F. Decreased lesion formation in CCR2-/- mice reveals a role for chemokines in the initiation of atherosclerosis. Nature 394 (6696):894 – 897; 1998. [4] Gu, L.; Okada, Y.; Clinton, S. K.; Gerard, C.; Sukhova, G. K.; Libby, P.; Rollins, B. J. Absence of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice. Mol. Cell. 2 (2):275 – 281; 1998. [5] Qiao, J. H.; Tripathi, J.; Mishra, N. K.; Cai, Y.; Tripathi, S.; Wang, X. P.; Imes, S.; Fishbein, M. C.; Clinton, S. K.; Libby, P.; Lusis, A. J.; Rajavashisth, T. B. Role of macrophage colony-stimulating factor in atherosclerosis: studies of osteopetrotic mice. Am. J. Pathol. 150 (5):1687 – 1699; 1997. [6] Sack, M. Tumor necrosis factor-a in cardiovascular biology and the potential role for anti-tumor necrosis factor-a therapy in heart disease. Pharmacol. Ther. 94 (1–2):123 – 135; 2002. [7] Rodriguez-Maldonado, E.; Hernandez-Rebollar, A. E.; Lopez-Marure, R.; Masso, F.; Montano, L. F. Role of tumor necrosis factor alpha in atherosclerosis. Arch. Cardiol. Mex. 71 (3):241 – 249; 2001. [8] Halle, M.; Berg, A.; Northoff, H.; Keul, J. Importance of TNF-alpha and leptin in obesity and insulin resistance: a hypothesis on the impact of physical exercise. Exerc. Immunol. Rev. 4:77 – 94; 1998. [9] Dandona, P.; Weinstock, R.; Thusu, K., et al. Tumor necrosis factoralpha in sera of obese patients: fall with weight loss. J. Clin. Endocrinol. Metab. 83 (8):2907 – 2910; 1998. [10] Hotamisligil, G.; Arner, P.; Caro, J., et al. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J. Clin. Invest. 95 (5):2409 – 2415; 1995. [11] Kern, P.; Saghizadeh, M.; Ong, J., et al. The expression of tumor necrosis factor in human adipose tissue. regulation by obesity, weight loss, and relationship to lipoprotein lipase. J. Clin. Invest. 95 (5):2111 – 2119; 1995. [12] Yamakawa, T.; Tanaka, S.; Yamakawa, Y., et al. Augmented production of tumor necrosis factor-alpha in obese mice. Clin. Immunol. Immunopathol. 75 (1):51 – 56; 1995. [13] Hotamisligil, G.; Shargill, N.; Spiegelman, M. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science 259:87 – 91; 1993. [14] Devaraj, S.; Li, D.; Jialal, I. The effects of alpha tocopherol supplementation on monocyte function. Decreased lipid oxidation, interleukin 1 beta secretion, and monocyte adhesion to endothelium. J. Clin. Invest. 98 (3):756 – 763; 1996. [15] Devaraj, S.; Jialal, I. Alpha-tocopherol decreases interleukin-1 beta release from activated human monocytes by inhibition of 5-lipoxygenase. Arterioscler. Thromb. Vasc. Biol. 19 (4):1125 – 1133; 1999. [16] Islam, K. N.; Devaraj, S.; Jialal, I. Alpha-tocopherol enrichment of monocytes decreases agonist-induced adhesion to human endothelial cells. Circulation 98 (21):2255 – 2261; 1998. [17] Goetzl, E. J. Vitamin E modulates the lipoxygenation of arachidonic acid in leukocytes. Nature 288:183 – 185; 1980. [18] Reddanna, P.; Rao, M. K.; Reddy, C. C. Inhibition of 5-lipoxygenase by vitamin E. FEBS Lett. 193 (1):39 – 43; 1985. [19] Reddanna, P.; Whelan, J.; Burgess, J. R.; Eskew, M. L.; Hildenbrandt, G.; Zarkower, A.; Scholz, R. W.; Reddy, C. The role of vitamin E and selenium on arachidonic acid oxidation by way of the 5-lipoxygenase pathway. Ann. N.Y. Acad. Sci. 570:136 – 145; 1989. [20] Fazzio, A.; Marilley, D.; Azzi, A. The effect of alpha-tocopherol and beta-tocopherol on proliferation, protein kinase C activity and gene expression in different cell lines. Biochem. Mol. Biol. Int. 41 (1):93 – 101; 1997. [21] Mahoney, C. W.; Azzi, A. Vitamin E inhibits protein kinase C activity. Biochem. Biophys. Res. Commun. 154 (2):694 – 697; 1998.
1220
S. Devaraj, I. Jialal / Free Radical Biology & Medicine 38 (2005) 1212–1220
[22] Freedman, J. E.; Farhat, J. H.; Loscalzo, J.; Keaney, J. F. Alphatocopherol inhibits aggregation of human platelets by a protein kinase C-dependent mechanism. Circulation 94 (10):2434 – 2440; 1996. [23] Burton, G. W.; Ingold, K. U. Vitamin E as an in vitro and in vivo antioxidant. Ann. N.Y. Acad. Sci. 570:7 – 22; 1989. [24] Schade, F. U.; Engel, R.; Jakobs, D. Lipoxygenase inhibitors but not site specific 5-lipoxygenase blockers protect against endotoxic shock and inhibit production of tumor necrosis factor. Eicosanoids 5 Suppl:S45 – S47; 1992. [25] Stankova, J.; Dupuis, G.; Gagnon, N.; Thivierge, M.; Turcotte, S.; Rola-Pleszczynski, M. Priming of human monocytes with leukotriene B4 enhances their sensitivity in IL-2-driven tumor necrosis factor-alpha production. Transcriptional and post-transcriptional upregulation of IL-2 receptors. J. Immunol. 150 (9):4041 – 4051; 1993. [26] Kontny, E.; Ziolkowska, M.; Ryzewska, A.; Maslinski, W. Protein kinase c-dependent pathway is critical for the production of proinflammatory cytokines (TNF-alpha, IL-1beta, IL-6). Cytokine 11:839 – 848; 1999. [27] Shames, B. D.; Selzman, C. H.; Pulido, E. J.; Meng, X.; Meldrum, D. R.; McIntyre, R. C. Jr.; Harken, A. H.; Banerjee, A. LPS-Induced NFkappaB activation and TNF-alpha release in human monocytes are protein tyrosine kinase dependent and protein kinase C independent. J. Surg. Res. 83 (1):69 – 74; 1999. [28] Lin, H. I.; Chu, S. J.; Wang, D.; Feng, N. H. Pharmacological modulation of TNF production in macrophages. J. Microbiol Immunol. Infect. 37 (1):8 – 15; 2004. [29] Rouzer, C. A.; Kingsley, P. J.; Wang, H.; Zhang, H.; Morrow, J. D.; Dey, S. K.; Marnett, L. J. Cyclooxygenase-1-dependent prostaglandin synthesis modulates tumor necrosis factor-alpha secretion in lipopolysaccharide-challenged murine resident peritoneal macrophages. J. Biol Chem. 279 (33):34256 – 34268; 2004. [30] Kwon, H. J.; Lee, K. W.; Yu, S. H.; Han, J. H.; Kim, D. S. NFkappaB-dependent regulation of tumor necrosis factor-alpha gene
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
expression by CpG-oligodeoxynucleotides. Biochem. Biophys. Res. Commun. 311 (1):129 – 138; 2003. Zhou, Z.; Wang, L.; Song, Z.; Saari, J. T.; McClain, C. J.; Kang, Y. J. Abrogation of nuclear factor-kappaB activation is involved in zinc inhibition of lipopolysaccharide-induced tumor necrosis factor-alpha production and liver injury. Am. J. Pathol. 164 (5):1547 – 1556; 2004. Araujo, F.; Slifer, T.; Li, S.; Kuver, A.; Fong, L.; Remington, J. Gemifloxacin inhibits cytokine secretion by lipopolysaccharide stimulated human monocytes at the post-transcriptional level. Clin. Microbiol. Infect. 10 (3):213 – 219; 2004. Haversen, L.; Ohlsson, B. G.; Hahn-Zoric, M.; Hanson, L. A.; Mattsby-Baltzer, I. Lactoferrin down-regulates the LPS-induced cytokine production in monocytic cells via NF-kappa B. Cell. Immunol. 220 (2):83 – 95; 2002. Kawano, T.; Matsuse, H.; Kondo, Y.; Machida, I.; Saeki, S.; Tomari, S.; Mitsuta, K.; Obase, Y.; Fukushima, C.; Shimoda, T.; Kohno, S. Cysteinyl leukotrienes induce nuclear factor kappa b activation and RANTES production in a murine model of asthma. J. Allergy Clin. Immunol. 112 (2):369 – 374; 2003. Aoki, Y.; Qiu, D.; Zhao, G. H.; Kao, P. N. Leukotriene B4 mediates histamine induction of NF-kappaB and IL-8 in human bronchial epithelial cells. Am. J. Physiol. 274 (6 Pt 1):L1030 – L1039; 1998. van Puijenbroek, A. A.; Wissink, S.; van der Saag, P. T.; Peppelenbosch, M. P. Phospholipase A2 inhibitors and leukotriene synthesis inhibitors block TNF-induced NF-kappaB activation. Cytokine 11 (2):104 – 110; 1999. Miller, E. R. 3rd; Pastor-Barriuso, R.; Dalal, D.; Riemersma, R. A.; Appel, L. J.; Gualla, E. Meta-analysis: high-dosage vitamin E supplementation may increase all-cause mortality. Ann Intern Med. 142 (1):37–46; 2004. Kris-Etherton, P. M.; Lichtenstein, A. H.; Howard, B. V.; Steinberg, D.; Witztum, J. L. Nutrition Committee of the American Heart Association Council on Nutrition, Physical Activity, and Metabolism. Antioxidant vitamin supplements and cardiovascular disease. Circulation 110 (5):637 – 641; 2004.