Increased Expression of the DNA-Binding Cytokine HMGB1 in Human Atherosclerotic Lesions
http://www.100md.com
《动脉硬化血栓血管生物学》
N. Kalinina; A. Agrotis; Y. Antropova; G. DiVitto; P. Kanellakis; G. Kostolias; O. Ilyinskaya; E. Tararak; A. Bobik
From the Baker Heart Research Institute (N.K., A.A., G.D., P.K., G.K., A.B.), Alfred Hospital, Melbourne, Australia; and the Institute of Experimental Cardiology (N.K., Y.A., O.I., E.T.), Cardiology Research Complex, Moscow, Russia.
ABSTRACT
Objective— Atherosclerosis is a chronic inflammatory response of the arterial wall to injury. High-mobility group box 1 (HMGB1) is a DNA-binding protein, which on release from cells exhibits potent inflammatory actions. We examined its expression in atherosclerotic lesions and regulation by cytokines.
Methods and Results— In atherosclerotic lesions, HMGB1 protein is expressed by endothelial cells, some intimal smooth muscle cells, and macrophages. As atherosclerosis develops and progresses from fatty streaks to fibrofatty lesion, the number of HMGB1-producing macrophages increases markedly. Studies using the THP-1 cell line indicated that HMGB1 mRNA expression could be markedly upregulated by inflammatory cytokines, interferon (IFN)-, tumor necrosis factor (TNF)- and also transforming growth factor (TGF)-?. IFN-, TNF-, TWEAK, and TGF-? induced an intracellular redistribution of HMGB1 and stimulated secretion by THP-1 cells and human blood monocytes. Inhibitors of MEK1/MEK2, protein kinase C, and PI-3/Akt, which inhibit lysosomal degranulation and mRNA translation, attenuated cytokine-induced HMGB1 secretion.
Conclusions— Macrophage is the major cell type responsible for HMGB1 production in human atherosclerotic lesions. Inflammatory cytokines and TGF-? increase HMGB1 expression and secretion by monocyte/macrophages. HMGB1 appears to be a common mediator of inflammation induced by inflammatory cytokines and is likely to contribute to lesion progression and chronic inflammation.
The expression and regulation of high-mobility group box 1 (HMGB1) proinflammatory cytokine in human atherosclerosis were examined. HMGB1 is mostly expressed by macrophages and its expression increases during atherogenesis. Inflammatory cytokines upregulate HMGB1 expression and secretion. HMGB1 is a common mediator of inflammation and may contribute to atherosclerotic lesion progression.
Key Words: high-mobility group box 1 ? macrophages ? cytokines ? inflammation ? atherosclerosis
Introduction
High-mobility group box chromosomal protein 1 (HMGB1; previously called HMG1 or amphoterin) is a ubiquitous nuclear protein present in many eukaryotic cells.1 As a nuclear protein it stabilizes nucleosomes and enables bending of DNA, which facilitates gene transcription.2 HMGB1 possesses 2 separate and characteristic DNA-binding domains, a 3-hydroxy-3-methylglutaryl (HMG) A box and B box. Recent studies unexpectedly revealed that extracellular released HMGB1 exerts distinctly different cellular actions from its intranuclear effects.3 When released from cells, HMGB1 is a potent stimulator of macrophages4 and a proinflammatory mediator of inflammation.5 Its effects on macrophages include elevations in the secretion of tumor necrosis factor (TNF)-, interleukin (IL)-1, IL-1?, IL-6, and macrophage inflammatory proteins (MIP-1 and MIP-1?). It also elicits proinflammatory responses in endothelial cells (ECs)6 and induces chemotaxis in vascular smooth muscle cells (SMCs).7 HMGB1 plays a critical role in several inflammatory diseases such as septic shock,8 acute lung inflammation,9 and rheumatoid arthritis.10
See cover
The mechanisms that govern HMGB1 production, secretion, and action are only partially understood. Inflammatory stimuli, such as lysophosphatidylcholine or interferon (IFN)- lead to the release of preformed HMGB1.11,12 HMGB1 can also be released from necrotic cells to trigger inflammation.13 Its actions appear dependent on interactions with several membrane receptors, including RAGE.14 Ligation of HMGB1 to receptors results in the activation of multiple kinases, including ERK1/ERK2, p38MAP kinase, and JNK kinase,15 the rapid phosphorylation, and nuclear localization of the cAMP response element-binding protein.16 In ECs this leads to increased expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1, RAGE, and secretion of proinflammatory cytokines, TNF-, IL-8, and monocyte chemotactic protein-1.6
The pathogenesis of atherosclerosis is characterized by a chronic inflammatory fibroproliferative response of the arterial wall to injury.17 Because HMGB1 is a mediator of inflammatory processes8,9 and human diseases,10 and because its signaling receptor RAGE is expressed in human atherosclerotic lesions,18,19 we studied the expression and distribution of HMGB1 immunoreactivity in different stages of human atherosclerosis. We also sought to examine the contribution of cytokines commonly present in human atherosclerotic lesions to its expression and release by macrophages and potential signaling mechanisms regulating these processes.
Materials and Methods
Aortic Tissues
Thoracic and abdominal aortas were collected during autopsy, not later than 6 hours after death, from 20 males and females, at the Russian Cardiology Research Complex, Moscow (please see http://atvb.ahajournals.org for specific details and Table I). Specimen collection was approved by the Ethics Committee of the Cardiology Research Industrial Complex, Moscow. Aortic segments were frozen in OCT (Miles Inc, Elkhart, Ind) and stored at –80°C. Atherosclerotic lesions were classified according to American Heart Association guidelines after staining with Oil Red O, hematoxylin and eosin, and/or trichrome.20,21
Antibodies
Please see online Methods, available at http://atvb.ahajournals.org, for specific details. Anti-RAGE antibodies are described elsewhere.22
Immunohistochemical Procedures
The expression of HMGB1 was examined in aortic cross-sections, as previously described.21 Please see http://atvb.ahajournals.org for specific details.
Immunofluorescence and Confocal Imaging
Immunofluorescent staining of cultured monocytes/macrophages was performed as previously described11 and HMGB1 distribution was examined using a Zeiss META Channel confocal microscope. Please see http://atvb.ahajournals.org for specific details.
Monocytes and HMGB1 Secretion
Human blood monocytes were isolated from venous blood of healthy donors using Histopaque (Sigma) as described previously.11 Monocytes were cultured in RPMI-1640 medium (Gibco BRL) containing heat-inactivated 15% fetal calf serum (JRH BioSciences). The promonocyte (THP-1) human cell line (American Cell Type Collection) was cultured in 10% fetal calf serum/RPMI-1640 medium and 1 μmol/L ?-mercaptoethanol (Sigma). Please see http://atvb.ahajournals.org for specific details of stimulation of HMGB1 secretion by monocytes and THP-1 cells. To evaluate cytotoxity in THP-1 cell cultures, lactate dehydrogenase (LDH) activity was assessed using a CytoTox96 colorimetric kit (Promega).
Western Blotting
Cells were lysed in lysis buffer (1% Nonidet P-40, 2.5 mmol/L Na2VO4, 10 mmol/L Tris-HCl, 150 mmol/L NaCl, 1 mmol/L EDTA, 100 mmol/L NaF, 50 mg/mL aprotinin, 50 mg/mL leupeptin, and 1 μmol/L of PMSF). Proteins for Western blotting were quantitated using the "Coomassie plus" protein assay kit; Pierce) with bovine serum albumin as standard. Volumes of culture media taken for analysis were normalized to cell counts. Western blots using 40 μg of protein or normalized volumes of culture media were performed using the ECL Western Blotting System (Amersham) as previously described.21 To ensure equal protein loading, polyvinylidene fluoride were stained by Ponceau S (Sigma) before probing with antibodies.
Reverse-Transcription Polymerase Chain Reaction
Messenger RNA encoding HMGB1 was assessed using RT-PCR23 and DNAase-treated RNA isolated from THP-1 cells using RNAeasy kit (Qiagen). Please see http://atvb.ahajournals.org.
SMCs Migration Assay
Migration of SMCs in response to HMGB1-containing medium was examined using Boyden chambers as described previously.24 Please see http://atvb.ahajournals.org.
Statistical Analysis
Ratios of HMGB1-expressing cells in human atherosclerotic lesions were analyzed using Kruskal-Wallis 1-way test on ranks (please see online Methods, available at http://atvb.ahajournals.org, for cell count criteria). Numbers of migrated SMCs were analyzed using Student t test. Differences with P<0.05 were considered as statistically significant. Data are expressed as mean±SEM.
Results
HMGB1 Expression in Normal Intima and Atherosclerotic Lesions
In normal intima, SMCs expressing HMGB1 were mostly localized close to the endothelium (Figure 1B). HMGB1 was also present in endothelium of normal intima, as well as in microvessels within adventitia (Figure 1B, 1H, and 1J). Most of solitary CD68-positive macrophages observed within normal intima expressed HMGB1. Approximately 50% of those cells contained HMGB1 in their cytoplasm. Medial SMC did not express HMGB1 (Figure 1C).
Figure 1. Immunohistochemical identification of HMGB1 expression in human aorta. A to C, In normal aorta, HMGB1 (black stain) is located in ECs, some intimal SMCs (double yellow arrows), and CD68-positive monocytes (red–brown stained cells), but not in the medial SMCs. In fatty streak (D and E) and fibrofatty lesion (F and G), HMGB1 (black) is present within the cytoplasm and nuclei of almost all CD68-positive macrophages (cells stained red–brown, yellow arrows). H to J, In adventitia, HMGB1 is present in ECs of microvessels (mv) (double yellow arrows) and macrophages (yellow arrows). Green arrowhead indicates CD68-positive macrophage not expressing HMGB1. I, Region parallel to (E) incubated with nonspecific rabbit IgGs instead of the primary antibody. Original magnification in A, D, F, and H was x80; B, C, E, G, J, and I magnification x400.
In fatty streaks and fibrofatty lesions, 13.6±1.9% of the SMC population expressed HMGB1, a frequency similar to that observed in normal intima (12.8±1.9%; for difference P>0.05), suggesting that only a limited number of intimal SMCs are capable of expressing this peptide in amounts detected by immunohistochemistry (Figure 1D, 1E, 1F, and 1G). Although the number of macrophages increased markedly in fatty streaks and fibrofatty lesions, the proportion that expressed HMGB1 did not alter significantly (95.3±3.1% in fatty streaks and 95.1±2.9% in fibrofatty lesions, P>0.05). However, the proportion of macrophages containing HMGB1 in both cytoplasm and nuclei increased markedly (93.8±1.1% in fibrofatty lesions compared with 45.2±8.8% in normal intima; P<0.01), suggesting secretion of HMGB1 in atherosclerotic lesions. Intense HMGB1 immunostaining was also observed in regions adjacent to necrotic core of lesions.
Most of the HMGB1-positive cells within fatty streaks and fibrofatty lesions also expressed RAGE (please see Figure I, available online at http://atvb.ahajournals.org).
Cytokines Elevate HMGB1 mRNA Expression
Because HMGB1 is upregulated in macrophages of atherosclerotic lesions, we examined the dependency of upregulation of its mRNA on cytokines. IFN- induced a time-dependent increase in HMGB1 mRNA, which was maximal after 24 hours (Figure 2A). TNF- induced a similar time-dependent increase in mRNA. TGF-? also elevated HMGB1 mRNA levels (Figure 2B). HMGB1 mRNA did not increase in unstimulated monocytes after 24 hours in culture (control in Figure 2A) or after 48, 72, and 96 hours (not shown). Because IFN-, TNF-, and TGF-? can signal via extracellular signal kinase (ERK1/ERK2),25–27 we examined the extent to which ERK1/ERK2 might contribute to the elevations in HMGB1 mRNA. Pre-exposure to PD98059 did not affect the ability of IFN-, TNF-, or TGF-? to elevate HMGB1 mRNA (Figure 2B). Similarly, protein kinase C, which is also implicated in signaling by these cytokines,28–30 does not contribute to the elevations in HMGB1 mRNA, because bisindolylmaleimide did not attenuate the mRNA elevations (not shown). In contrast, the phosphatidylinositol-3 (PI-3) kinase inhibitor wortmannin attenuated the elevations in HMGB1 by all 3 cytokines (Figure 2); PI-3 kinase has previously been implicated in cell signaling by all 3 cytokines.31–33 Thus, PI-3 kinase appears essential for IFN-, TNF-, and TGF-? induced elevations in HMGB1 mRNA in monocytes.
Figure 2. Reverse-transcription polymerase chain reaction (RT-PCR) analysis of HMGB1 mRNA expression in THP-1 monocytes. A, Level of HMGB1 mRNA increases after 24 hours of treatment with either IFN- or TNF-. Level of HMGB1 mRNA in control cultures did not change after 48, 72, or 96 hours (as indicated in Results and Cytokines Elevate HMGB1 mRNA Expression sections), and the control result shown reflects a 24-hour culture. B, Pretreatment with wortmannin but not PD098059 decreases levels of HMGB1 mRNA in THP-1 monocytes induced by IFN-, TNF-, or TGF-?. Equal RNA loading for RT-PCR was verified using expression of GAPDH mRNA.
Regulation of HMGB1 Protein Expression and Secretion by Cytokines
Lysophosphatidylcholine and IFN- increase HMGB1 secretion by monocytes.11,12 Consequently, we examined whether other cytokines that are expressed in human atherosclerotic lesions might also elevate HMGB1 secretion by monocyte/macrophages. TNF-, IFN-, and TGF-?1 induced secretion of HMGB1 in freshly isolated blood monocytes (please see Figure II, available online at http://atvb.ahajournals.org).
Control unstimulated THP-1 monocytes secreted very low amounts of HMGB1 into the medium (Figure 3). Proinflammatory members of the TNF superfamily, TNF-, TWEAK, CD40L (not shown), and IFN- markedly increased the secretion of HMGB1 into the medium (Figure 3A). Secretion increased in a time-dependent manner over 96 hours. TGF-?1 also stimulated HMGB1 secretion, which was apparent after 72 hours and continued through to at least 96 hours. In all instances, intracellular levels of HMGB1 remained relatively constant, suggesting that the increased release of HMGB1 was also accompanied by significant increases in HMGB1 protein synthesis. HMGB1-containing media concentrated from THP-1 cells markedly increased ERK1/2 phosphorylation in rat aortic SMCs and induced their migration (44.2±1.6 cells/field in HMGB1-containing medium versus 10.2±0.5 cells/field in control medium; P<0.05) (please see Figure III, available online at http://atvb.ahajournals.org). Inhibition of SAPK/p38 MAPK and NF-B by pre-incubating the monocytes with SB203580 and isohelenin did not affect the IFN-, TNF-, and TGF-?–stimulated increases in HMGB1 secretion (Figure 3B). In contrast inhibitors of ERK1/ERK2 (PD098059), protein kinase C (bisindoylmaleimide) and PI3-kinase (wortmannin) all attenuated secretion stimulated by these cytokines (Figure 3B), suggesting that ERK1/ERK2, protein kinase C, and PI3-kinase activities are necessary for cytokine-stimulated HMGB1 secretion.
Figure 3. Secretion of HMGB1 protein by THP-1 cells analyzed by Western blotting. A, Level of HMGB1 protein expression and secretion after treating THP-1 monocytes with IFN-, TNF-, TGF-?, or TWEAK for 24, 48, 72, or 96 hours. B, Pretreatment of cells with PD098059, bisindoylmaleimide, and wortmannin attenuates HMGB1 secretion induced by 48-hour incubation with cytokines but not the secretion of endolysosomal marker cathepsin D.
HMGB1 lacks a secretory peptide and is thought to be secreted via endolysosomes in a manner analogous to IL-1?.11 Because endolysosomes contain cathepsin D, we sought to determine whether IFN-, TFN-, and TGF-? affect cathepsin D secretion in a manner similar to HMGB1. In contrast to HMGB1, only TNF- and TGF-? elevated the secretion of cathepsin D into the medium (Figure 3B); inhibitors of ERK1/ERK2, protein kinase C, and PI-3 kinase did not uniformly attenuate its secretion. It would appear that HMGB1 secretion occurs independently of endolysosomes containing cathepsin D.
Because cell necrosis can also be responsible for HMGB1 release, we examined cytotoxity in cultures by measuring LDH activity. LDH activity in monocytes treated with cytokines averaged 103.4±7.9% compared with control (100%) over 3 cytokines (IFN-, TFN-, and TGF-?1), whereas HMGB1 release was 5-times greater compared to control (Figure 3). Therefore, the increase in HMGB1 secretion is unlikely to be increased because of spontaneous release from necrotic cells.
Regulation of HMGB1 Cellular Localization by Cytokines
To determine whether the cytokines that induce HMGB1 secretion also affect its cellular distribution, THP-1 monocyte/macrophage cultures were immunostained with anti-HMGB1 antibodies and nuclei identified using propidium iodide (PI). Control monocyte/macrophages constitutively expressed HMGB1 and maintained an intracellular pool mostly localized in the nucleus (Figure 4). Exposure to TGF-?1 induced a redistribution of HMGB1 from the nucleus and into cytoplasm (Figure 4). IFN- also induced a similar redistribution, which was most apparent after 2 days when HMGB1 was being actively secreted into the medium. In contrast, TNF- did not induce substantial redistribution of HMGB1 during this time despite inducing active secretion of HMGB1 (Figure 4). Thus, it would appear that although a number of cytokines induce translocation of HMGB1 from the nucleus to the cytoplasm, when substantial amounts of HMGB1 are present in the cytoplasm, this redistribution is apparently not necessary for active secretion of HMGB1. Pretreatment of cells with inhibitors of SAPK/p38 MAP kinase, NF-B, MEK1/2, protein kinase C, or PI-3 kinase did not affect cytokine-induced redistribution of HMGB1 in the cells (not shown).
Figure 4. Confocal analysis of intracellular distribution of HMGB1 protein in THP-1 monocytes. In control cells, HMGB1 (green) is mostly localized in the nuclei (counterstained by propidium iodide , red) as shown by yellow representing colocalization. Monocytes exposed to either TGF-? or IFN- for 2 days contained high amounts of HMGB1 in their cytoplasm with less protein localized in the nuclei. In contrast, THP-1 cells treated with TNF- did not exhibit as great a shift in HMGB1 distribution. Bottom panels show the lack of staining with a control IgG.
Discussion
This study demonstrates that HMGB1 is highly expressed in human atherosclerotic lesions and may contribute to the sustained inflammation of fatty rich lesions. HMGB1 expression is highest in fibrofatty lesions and mostly associated with macrophages. Cytokines TNF- and other members of the TNF superfamily, as well as TGF-?1 and IFN-, appear responsible for its upregulation in monocyte/macrophages and its secretion. Upregulation of its mRNA was dependent on PI-3kinase, whereas secretion stimulated by the cytokines appears dependent on MEK1/2, protein kinase C, and PI-3 kinase.
HMGB1, a 30-kDa member of the high-mobility group nonhistone chromosomal protein family,2,34 is also a mediator of delayed endotoxin lethality and systemic inflammation.3 Recent studies have associated local expression of HMGB1 with local sites of tissue inflammation.8–10 For example, in rheumatoid arthritis, HMGB1 was localized to the cytoplasm of CD68-positive cells infiltrating the sublining layer, suggesting that it is secreted by synovial macrophages.35 In normal aortic intima, only a small number of cells, intimal SMCs, some ECs, and occasionally present macrophages expressed HMGB1, whereas in fibrofatty lesions macrophages were the major cell type expressing HMGB1, with expression nearly always in the cytoplasm. This distribution is similar to that reported in rheumatoid arthritis, where HMGB1 levels are also increased in synovial fluid, indicative of secretion by macrophages.35 Secreted HMGB1 can profoundly affect the function of cells associated with atherosclerotic lesions, particularly macrophages, ECs, and vascular SMCs, promoting local inflammation, the accumulation of monocyte/macrophages, and even remodeling of the lesion. HMGB1 induces human monocytes to release a large array of inflammatory cytokines, including TNF-, IL-1, IL-1?, IL-1RA, IL-6, IL-8, and MIP-1 and MIP-1?.4 Its actions on ECs include increasing the expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and RAGE, as well as increasing the secretion of TNF-, IL-8, monocyte chemotactic protein-1, and tissue plasminogen activator.6 HMGB1 also induces cytoskeleton reorganization and chemotaxis in vascular SMCs.7
Our findings that the proinflammatory cytokines IFN- and TNF-, as well as TGF-?1, elevate HMGB1 mRNA in monocyte/macrophages suggest that HMGB1 exerts its inflammatory action downstream of proinflammatory cytokines. The finding that TNF- induces HMGB1 secretion is particularly important because the secreted HMGB1 may in turn induce further secretion of TNF-,4 raising the possibility that a proinflammatory loop exists between TNF- and HMGB1, which increases the severity of inflammation and prolongs its duration. Whether such a proinflammatory loop also exists for other inflammatory cytokines of the TNF superfamily and HMGB1 remains to be determined. Our findings also suggest that PI-3 kinase is essential for the upregulation of HMGB1 mRNA by these cytokines. PI-3 kinase is known to regulate the activity of Sp1-responsive promoters,36 which are abundant in the 5'-region of the human HMGB1 gene.37 TGF-?, TNF-, and IFN- are all known to potently stimulate phosphorylation of Akt.33,38,39
HMGB1 is released from necrotic cells, inducing local inflammation.13 In macrophages, release of HMGB1 can also be induced by lysophosphatidylcholine and IFN-.11,12 IFN-–dependent release of HMGB1 could be inhibited by a specific inhibitor of Janus kinase 2.12 In both instances, release appeared to be associated with its translocation from the nucleus to the cytoplasm.11,12 We have demonstrated that a number of proinflammatory cytokines of the TNF superfamily, TNF-, TWEAK, and CD40L (not shown), as well as TGF-?1 and IFN-, are also capable of stimulating HMGB1 secretion from human monocyte/macrophages. These cytokines stimulated secretion in a somewhat delayed manner, being most apparent after 48 hours. Secretion was accompanied by increased HMGB1 protein synthesis, because intracellular HMGB1 levels were unaltered during this time. Our findings indicate that HMGB1 secretion stimulated by cytokines is in part dependent on transient elevations in its mRNA and elevations in its biosynthesis. Secretion can be also accompanied by its translocation from the nucleus to the cytoplasm. This was most apparent with TGF-?, IFN-, and CD40L. TNF- was least effective in inducing such translocation, although it is a potent stimulant of HMGB1 secretion. Because HMGB1 lacks a hydrophobic signal sequence, it appears to be secreted via a yet to be fully defined nonclassical, vesicle-mediated secretory pathway analogous to that used by IL-1?.11 However, in contrast to IL-1?–containing secretory lysosomes,40 those responsible for HMGB1 secretion do not apparently contain cathepsin D. We found no correlation between HMGB1 and cathepsin D secretion under a variety of different conditions. Cytokine-induced HMGB1 secretion could be attenuated by inhibitors of MEK1/MEK2, protein kinase C, and PI-3 kinase/Akt, suggesting that such agents might be useful in attenuating HMGB1-mediated inflammation. Although the mechanisms by which these agents attenuate HMGB1 secretion needs to be clarified, it is possible that they inhibit lysosomal degranulation and/or HMGB1 protein synthesis via effects on mRNA translation. Protein kinase C, MEK1/MEK2, and PI-3 kinase appear essential for lysosomal degranulation41–43 and also regulate mRNA translational processes.43–46
In summary, this is the first study to our knowledge to identify that macrophages are a major source of HMGB1 in human atherosclerotic lesions. Its expression and secretion by macrophages is highly regulated by cytokines. Upregulation and secretion of HMGB1 has the potential to amplify inflammatory responses and may also contribute to macrophage accumulation, thereby promoting atherogenesis. Definition of its precise roles in the development and progression of atherosclerosis will require additional in vivo investigations.
References
Mosevitsky MI, Novitskaya VA, Iogannsen MG, Zabezhinsky MA. Tissue specificity of nucleo-cytoplasmic distribution of HMG1 and HMG2 proteins and their probable functions. Eur J Biochem. 1989; 185: 303–310.
Bustin M. Regulation of DNA-dependent activities by the functional motifs of the high-mobility-group chromosomal proteins. Mol Cell Biol. 1999; 19: 5237–5246.
Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J, Frazier A, Yang H, Ivanova S, Borovikova L, Manogue KR, Faist E, Abraham E, Andersson J, Andersson U, Molina PE, Abumrad NN, Sama A, Tracey KJ. HMG1 as a late mediator of endotoxin lethality in mice. Science. 1999; 285: 248–251.
Andersson U, Wang H, Palmblad K, Aveberger AC, Bloom O, Erlandsson-Harris H, Janson A, Kokkola R, Zhang M, Yang H, Tracey KJ. High mobility group 1 protein (HMG-1) stimulates proinflammatory cytokine synthesis in human monocytes. J Exp Med. 2000; 192: 565–570.
Andersson U, Erlandsson-Harris H, Yang H, Tracey KJ. HMGB1 as a DNA-binding cytokine. J Leukoc Biol. 2002; 72: 1084–1091.
Fiuza C, Bustin M, Talwar S, Tropea M, Gerstenberger E, Shelhamer JH, Suffredini AF. Inflammation-promoting activity of HMGB1 on human microvascular endothelial cells. Blood. 2003; 101: 2652–2660.
Degryse B, Bonaldi T, Scaffidi P, Muller S, Resnati M, Sanvito F, Arrigoni G, Bianchi ME. The high mobility group (HMG) boxes of the nuclear protein HMG1 induce chemotaxis and cytoskeletal reorganization in rat SMCs. J Cell Biol. 2001; 152: 1197–1206.
Fang WH, Yao YM, Shi ZG, Wu Y, Lu LR, Sheng ZY. The significance of changes in high mobility group-1 protein mRNA expression in rats after thermal injury. Shock. 2002; 17: 329–333.
Abraham E, Arcaroli J, Carmody A, Wang H, Tracey KJ. HMG-1 as a mediator of acute lung inflammation. J Immunol. 2000; 2950–2954.
Kokkola R, Sundberg E, Ulfgren AK, Palmblad K, Li J, Wang H, Ulloa L, Yang H, Yan XJ, Furie R, Chiorazzi N, Tracey KJ, Andersson U, Harris HE. High mobility group box chromosomal protein-1: a novel proinflammatory mediator in synovitis. Arthritis Rheum. 2002; 46: 2598–2603.
Gardella S, Andrei C, Ferrera D, Lotti LV, Torrisi MR, Bianchi ME, Rubartelli A. The nuclear protein HMGB1 is secreted by monocytes via a non-classical, vesicle-mediated secretory pathway. EMBO Reports. 2002; 3: 995–1001.
Rendon-Mitchell B, Ochani M, Li J, Wang H, Yang H, Susarla S, Czura C, Mitchell RA, Chen G, Sama AE, Tracey KJ, Wang H. IFN- induces high mobility box 1 protein release partially through a TNF-dependent mechanism. J Immunol. 2003; 170: 3890–3897.
Scaffidi P, Misteli T, Bianchi ME. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature. 2002; 418: 191–195.
Hori O, Brett J, Slattery T, Cao R, Zhang J, Chen JX, Nagashima M, Lundh ER, Vijay S, Nitecki D, Morser J, Stern D, Schmidt AM. The receptor for advanced glycation end products (RAGE) is a cellular binding site for amphoterin. Mediation of neurite outgrowth and co-expression of range and amphoterin in the developing nervous system. J Biol Chem. 1995; 270: 25752–25761.
Taguchi A, Blood D, del Toro G, Canet A, Lee DC, Qu W, Tanji N, Lu Y, Lalla E, Fu C, Hofmann MA, Kislinger T, Ingram M, Lu A, Tanaka H, Hori O, Ogawa S, Stern DM, Schmidt AM. Blockade of RAGE-amphoterin signalling suppresses tumour growth and metastases. Nature. 2000; 405: 354–358.
Huttunen HJ, Kuja-Panula J, Rauvala H. Receptor for advanced glycation end products (RAGE) signaling induces CREB-dependent chromogranin expression during neuronal differentiation. J Biol Chem. 2002; 277: 38635–38646.
Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Cipollone F, Iezzi A, Fazia M, Zucchelli M, Pini B, Cuccurullo C, De Cesare D, De Blasis G, Muraro R, Bei R, Chiarelli F, Schmidt AM, Cuccurullo F, Mezzetti A. The receptor RAGE as a progression factor amplifying arachidonate-dependent inflammatory and proteolytic response in human atherosclerotic plaques. Circ. 2003; 108: 1070–1077.
Bucciarelli LG, Wendt T, Qu W, Lu Y, Lalla E, Rong LL, Goova MT, Moser B, Kislinger T, Lee DC, Kashyap Y, Stern DM, Schmidt AM. RAGE blockade stabilizes established atherosclerotic lesions in diabetic apolipoprotein E-null mice. Circ. 2002; 106: 2827–2835.
Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insul W, Rosenfeld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis. Circ. 1995; 92: 1355–1374.
Kalinina N, Agrotis A, Tararak E, Antropova Y, Kanellakis P, Ilyinskaya O, Quinn MT, Smirnov V, Bobik A. Cytochrome b558-dependent NAD(P)H oxidase-phox units in smooth muscle and macrophages of atherosclerotic lesions. Arterioscler Thromb Vasc Biol. 2002; 22: 2037–2043.
Candido R, Forbes JM, Thomas MC, Thallas V, Dean RG, Burns WC, Tikellis C, Ritchie RH, Twigg SM, Cooper ME, Burrell LM. A breaker of advanced glycation end products attenuates diabetes-induced myocardial structural changes. Circ Res. 2003; 92: 785–792.
Ward MR, Agrotis A, Kanellakis P, Dilley R, Jennings G, Bobik A. Inhibition of protein tyrosine kinases attenuates increases in expression of transforming growth factor-? isoforms and their receptors following arterial injury. Arterioscler Thromb Vasc Biol. 1997; 17: 2461–2470.
Palumbo R, Sampaolesi M, De Marchis F, Tonlorenzi R, Colombetti S, Mondino A, Cossu G, Bianchi ME. Extracellular HMGB1, a signal of tissue damage, induces mesoangioblast migration and proliferation. J Cell Biol. 2004; 164: 441–449.
Balnchette J, Jarmillo M, Olivier M. Signalling events involved in interferon-–inducible macrophage nitric oxide generation. Immunology. 2003; 108: 513–522.
Goetze S, Kineshior K, Meehan WP, Collins A, Fleek E, Hsueh WA, Law RE. TNF- induces expression of transcription factors c-fos, Egr-1 and Ets-1 in vascular lesions through extracellular signal-regulated kinases 1/2. Atherosclerosis. 2001; 159: 93–101.
Tokuda H, Hatakeyama D, Akamatsu S, Tanable K, Yoshida M, Shibata T, Kozawa O. Involvement of MAP kinases in TGF-?–stimulated vascular endothelial growth factor synthesis in osteoblasts. Arch Biochem Biophys. 2003; 415: 117–125.
Deb DK, Sassano A, Lekmine F, Majchrzak B, Bverma A, Kambhampati S, Uddin S, Rahman A, Fish EN, Platanias LC. Activation of protein kinase C by IFN-. J Immunol. 2003; 171: 267–273.
Rahman JK, Anwar KN, Frey RS, Minshall RD, Malik AB. Tumor necrosis factor- induces early-onset endothelial adhesivity by protein kinase Czeta-dependent activation of intercellular adhesion molecule-1. Circ Res. 2003; 92: 1089–1097.
Runyan CE, Schnaper HW, Poncelet AC. Smad3 and PKC mediate TGF-?–mediated collagen-1 expression in human mesangial cells. Am J Physiol. 2003; 285: F413–F422.
Kawanaka H, Jones MK, Szabo IL, Baatar D, Pai R, Tsugawa K, Sugimachi K, Sarfeh IJ, Tarnawski AS. Activation of eNOS in rat portal hypertensive gastric mucosa is mediated by TNF- via the PI-3 kinase-Akt signaling pathway. Hepatology. 2002; 35: 393–402.
Kin G, Jun JB, Elkon KB. Necessary role of phosphatidylinositol 3-kinase in transforming growth factor-? mediated activation of Akt in normal and rheumatoid arthritis synovial fibroblasts. Arthritis Rheum. 2002; 46: 12504–11511.
Navarro A, Anad-Apte B, Tanabe Y, Feldman G, Larner AC. A PI-3 kinase-dependent, STAT1-independent signaling pathway in interferon-stimulated monocyte adhesion. J Leukoc Biol. 2003; 73: 540–545.
Bianchi ME, Beltrame M, Paonessa G. Specific recognition of cruciform DNA by nuclear HMG1. Science. 1989; 243: 1056–1059.
Taniguchi N, Kawahara K, Yone K, Hashiguchi T, Yamakuchi M, Goto M, Inoue K, Yamada S, Ijiri K, Matsunaga S, Nakajima S, Maruyama I. High mobility group box chromosomal protein 1 plays a role in the pathogenesis of rheumatoid arthritis as a novel cytokine. Arthritis Rheum. 2003; 48: 971–981.
Garcia A, Cereghini S, Sontag E. Protein phosphatase 2A and phosphatidylinositol 3-kinase regulate the activity of Sp1-responsive promoters. J Biol Chem. 2000; 275: 9385–9389.
Lum HK, Lee K-L, D. the human HMGB1 promoter is modulated by a silencer and an enhancer-containing intron. Biochimica et Biophysica Acta. 2000; 1520: 79–84.
Kim G, Jun JB, Elkon KB. Necessary role of phosphatidylinositol 3-kinase in transforming growth factor ?–mediated activation of Akt in normal and rheumatoid arthritis synovial fibroblasts. Arthritis Rheum. 2002; 46: 1504–1511.
Barsacchi R, Perrotta C, Bulotta S, Moncada S, Borgese N, Clementi E. Activation of endothelial nitric-oxide synthase by tumor necrosis factor-: a novel pathway involving sequential activation of neutral sphingomyelinase, phosphatidylinositol-3'-kinase and Akt. Mol Pharmacol. 2003; 63: 886–895.
Gardella S, Andrei C, Lotti LV, Poggi A, Torrisi MR, Zocchi MR, Rubartelli A. CD8+ T lymphocytes induce polarized exocytosis of secretory lysosomes by dendritic cells with release of interleukin-1? and cathepsin D. Immunobiology. 2001; 98: 2152–2159.
Tapper H, Sundler R. Protein kinase C and intracellular pH regulate zymosan-induced lysosomal enzyme secretion in macrophages. J Leukoc Biol. 1995; 58: 485–494.
Badewa AP, Heiman As. Inhibition of CCL11, CCL24 and CCL26-induced degranulation in HL-60 eosinophilic cells by specific inhibitors of MEK1/MEK2, p38 MAP kinase and PI 3-kinase. Immunopharmacol Immunotoxicol. 2003; 25: 145–157.
Huber M, Hughes MR, Krystal G. Thapsigargin-induced degranulation of mast cells is dependent on transient activation of phosphatidylinositol-3 kinase. J Immunol. 2000; 165: 124–133.
Stolovich M, Tang H, Hornstein E, Levy G, Cohen R, Bae SS, Birnbaum MJ, Meyuhas O. Transduction of growth or mitogen signals into translational activation of TOP mRNAs is fully reliant on the phosphatidylinositol 3-kinase-mediated pathway but requires neither 1 nor rpS6 phosphorylation. Mol Cell Biol. 2002; 22: 8101–8113.
Kumar V, Pandey P, Sabatini D, Kumar M, Majumder PK, Bharti A, Carmichael G, Kufe D, Kharbanda S. Functional interaction between RAFT1/FRAP/mTOR and protein kinase C in the regulation of cap-dependent initiation of translation. EMBO J. 2000; 19: 1087–1097.
Petegnief V, Friguls B, Sanfeliu C, Sunol C, Planas AM. Transforming growth factor- attenuates N-methyl-D-aspartic acid toxicity in cortical cultures by preventing protein synthesis inhibition through an Erk1/2-dependent mechanism. J Biol Chem. 2003; 278: 29552–29559.(Role of Activated Macroph)
From the Baker Heart Research Institute (N.K., A.A., G.D., P.K., G.K., A.B.), Alfred Hospital, Melbourne, Australia; and the Institute of Experimental Cardiology (N.K., Y.A., O.I., E.T.), Cardiology Research Complex, Moscow, Russia.
ABSTRACT
Objective— Atherosclerosis is a chronic inflammatory response of the arterial wall to injury. High-mobility group box 1 (HMGB1) is a DNA-binding protein, which on release from cells exhibits potent inflammatory actions. We examined its expression in atherosclerotic lesions and regulation by cytokines.
Methods and Results— In atherosclerotic lesions, HMGB1 protein is expressed by endothelial cells, some intimal smooth muscle cells, and macrophages. As atherosclerosis develops and progresses from fatty streaks to fibrofatty lesion, the number of HMGB1-producing macrophages increases markedly. Studies using the THP-1 cell line indicated that HMGB1 mRNA expression could be markedly upregulated by inflammatory cytokines, interferon (IFN)-, tumor necrosis factor (TNF)- and also transforming growth factor (TGF)-?. IFN-, TNF-, TWEAK, and TGF-? induced an intracellular redistribution of HMGB1 and stimulated secretion by THP-1 cells and human blood monocytes. Inhibitors of MEK1/MEK2, protein kinase C, and PI-3/Akt, which inhibit lysosomal degranulation and mRNA translation, attenuated cytokine-induced HMGB1 secretion.
Conclusions— Macrophage is the major cell type responsible for HMGB1 production in human atherosclerotic lesions. Inflammatory cytokines and TGF-? increase HMGB1 expression and secretion by monocyte/macrophages. HMGB1 appears to be a common mediator of inflammation induced by inflammatory cytokines and is likely to contribute to lesion progression and chronic inflammation.
The expression and regulation of high-mobility group box 1 (HMGB1) proinflammatory cytokine in human atherosclerosis were examined. HMGB1 is mostly expressed by macrophages and its expression increases during atherogenesis. Inflammatory cytokines upregulate HMGB1 expression and secretion. HMGB1 is a common mediator of inflammation and may contribute to atherosclerotic lesion progression.
Key Words: high-mobility group box 1 ? macrophages ? cytokines ? inflammation ? atherosclerosis
Introduction
High-mobility group box chromosomal protein 1 (HMGB1; previously called HMG1 or amphoterin) is a ubiquitous nuclear protein present in many eukaryotic cells.1 As a nuclear protein it stabilizes nucleosomes and enables bending of DNA, which facilitates gene transcription.2 HMGB1 possesses 2 separate and characteristic DNA-binding domains, a 3-hydroxy-3-methylglutaryl (HMG) A box and B box. Recent studies unexpectedly revealed that extracellular released HMGB1 exerts distinctly different cellular actions from its intranuclear effects.3 When released from cells, HMGB1 is a potent stimulator of macrophages4 and a proinflammatory mediator of inflammation.5 Its effects on macrophages include elevations in the secretion of tumor necrosis factor (TNF)-, interleukin (IL)-1, IL-1?, IL-6, and macrophage inflammatory proteins (MIP-1 and MIP-1?). It also elicits proinflammatory responses in endothelial cells (ECs)6 and induces chemotaxis in vascular smooth muscle cells (SMCs).7 HMGB1 plays a critical role in several inflammatory diseases such as septic shock,8 acute lung inflammation,9 and rheumatoid arthritis.10
See cover
The mechanisms that govern HMGB1 production, secretion, and action are only partially understood. Inflammatory stimuli, such as lysophosphatidylcholine or interferon (IFN)- lead to the release of preformed HMGB1.11,12 HMGB1 can also be released from necrotic cells to trigger inflammation.13 Its actions appear dependent on interactions with several membrane receptors, including RAGE.14 Ligation of HMGB1 to receptors results in the activation of multiple kinases, including ERK1/ERK2, p38MAP kinase, and JNK kinase,15 the rapid phosphorylation, and nuclear localization of the cAMP response element-binding protein.16 In ECs this leads to increased expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1, RAGE, and secretion of proinflammatory cytokines, TNF-, IL-8, and monocyte chemotactic protein-1.6
The pathogenesis of atherosclerosis is characterized by a chronic inflammatory fibroproliferative response of the arterial wall to injury.17 Because HMGB1 is a mediator of inflammatory processes8,9 and human diseases,10 and because its signaling receptor RAGE is expressed in human atherosclerotic lesions,18,19 we studied the expression and distribution of HMGB1 immunoreactivity in different stages of human atherosclerosis. We also sought to examine the contribution of cytokines commonly present in human atherosclerotic lesions to its expression and release by macrophages and potential signaling mechanisms regulating these processes.
Materials and Methods
Aortic Tissues
Thoracic and abdominal aortas were collected during autopsy, not later than 6 hours after death, from 20 males and females, at the Russian Cardiology Research Complex, Moscow (please see http://atvb.ahajournals.org for specific details and Table I). Specimen collection was approved by the Ethics Committee of the Cardiology Research Industrial Complex, Moscow. Aortic segments were frozen in OCT (Miles Inc, Elkhart, Ind) and stored at –80°C. Atherosclerotic lesions were classified according to American Heart Association guidelines after staining with Oil Red O, hematoxylin and eosin, and/or trichrome.20,21
Antibodies
Please see online Methods, available at http://atvb.ahajournals.org, for specific details. Anti-RAGE antibodies are described elsewhere.22
Immunohistochemical Procedures
The expression of HMGB1 was examined in aortic cross-sections, as previously described.21 Please see http://atvb.ahajournals.org for specific details.
Immunofluorescence and Confocal Imaging
Immunofluorescent staining of cultured monocytes/macrophages was performed as previously described11 and HMGB1 distribution was examined using a Zeiss META Channel confocal microscope. Please see http://atvb.ahajournals.org for specific details.
Monocytes and HMGB1 Secretion
Human blood monocytes were isolated from venous blood of healthy donors using Histopaque (Sigma) as described previously.11 Monocytes were cultured in RPMI-1640 medium (Gibco BRL) containing heat-inactivated 15% fetal calf serum (JRH BioSciences). The promonocyte (THP-1) human cell line (American Cell Type Collection) was cultured in 10% fetal calf serum/RPMI-1640 medium and 1 μmol/L ?-mercaptoethanol (Sigma). Please see http://atvb.ahajournals.org for specific details of stimulation of HMGB1 secretion by monocytes and THP-1 cells. To evaluate cytotoxity in THP-1 cell cultures, lactate dehydrogenase (LDH) activity was assessed using a CytoTox96 colorimetric kit (Promega).
Western Blotting
Cells were lysed in lysis buffer (1% Nonidet P-40, 2.5 mmol/L Na2VO4, 10 mmol/L Tris-HCl, 150 mmol/L NaCl, 1 mmol/L EDTA, 100 mmol/L NaF, 50 mg/mL aprotinin, 50 mg/mL leupeptin, and 1 μmol/L of PMSF). Proteins for Western blotting were quantitated using the "Coomassie plus" protein assay kit; Pierce) with bovine serum albumin as standard. Volumes of culture media taken for analysis were normalized to cell counts. Western blots using 40 μg of protein or normalized volumes of culture media were performed using the ECL Western Blotting System (Amersham) as previously described.21 To ensure equal protein loading, polyvinylidene fluoride were stained by Ponceau S (Sigma) before probing with antibodies.
Reverse-Transcription Polymerase Chain Reaction
Messenger RNA encoding HMGB1 was assessed using RT-PCR23 and DNAase-treated RNA isolated from THP-1 cells using RNAeasy kit (Qiagen). Please see http://atvb.ahajournals.org.
SMCs Migration Assay
Migration of SMCs in response to HMGB1-containing medium was examined using Boyden chambers as described previously.24 Please see http://atvb.ahajournals.org.
Statistical Analysis
Ratios of HMGB1-expressing cells in human atherosclerotic lesions were analyzed using Kruskal-Wallis 1-way test on ranks (please see online Methods, available at http://atvb.ahajournals.org, for cell count criteria). Numbers of migrated SMCs were analyzed using Student t test. Differences with P<0.05 were considered as statistically significant. Data are expressed as mean±SEM.
Results
HMGB1 Expression in Normal Intima and Atherosclerotic Lesions
In normal intima, SMCs expressing HMGB1 were mostly localized close to the endothelium (Figure 1B). HMGB1 was also present in endothelium of normal intima, as well as in microvessels within adventitia (Figure 1B, 1H, and 1J). Most of solitary CD68-positive macrophages observed within normal intima expressed HMGB1. Approximately 50% of those cells contained HMGB1 in their cytoplasm. Medial SMC did not express HMGB1 (Figure 1C).
Figure 1. Immunohistochemical identification of HMGB1 expression in human aorta. A to C, In normal aorta, HMGB1 (black stain) is located in ECs, some intimal SMCs (double yellow arrows), and CD68-positive monocytes (red–brown stained cells), but not in the medial SMCs. In fatty streak (D and E) and fibrofatty lesion (F and G), HMGB1 (black) is present within the cytoplasm and nuclei of almost all CD68-positive macrophages (cells stained red–brown, yellow arrows). H to J, In adventitia, HMGB1 is present in ECs of microvessels (mv) (double yellow arrows) and macrophages (yellow arrows). Green arrowhead indicates CD68-positive macrophage not expressing HMGB1. I, Region parallel to (E) incubated with nonspecific rabbit IgGs instead of the primary antibody. Original magnification in A, D, F, and H was x80; B, C, E, G, J, and I magnification x400.
In fatty streaks and fibrofatty lesions, 13.6±1.9% of the SMC population expressed HMGB1, a frequency similar to that observed in normal intima (12.8±1.9%; for difference P>0.05), suggesting that only a limited number of intimal SMCs are capable of expressing this peptide in amounts detected by immunohistochemistry (Figure 1D, 1E, 1F, and 1G). Although the number of macrophages increased markedly in fatty streaks and fibrofatty lesions, the proportion that expressed HMGB1 did not alter significantly (95.3±3.1% in fatty streaks and 95.1±2.9% in fibrofatty lesions, P>0.05). However, the proportion of macrophages containing HMGB1 in both cytoplasm and nuclei increased markedly (93.8±1.1% in fibrofatty lesions compared with 45.2±8.8% in normal intima; P<0.01), suggesting secretion of HMGB1 in atherosclerotic lesions. Intense HMGB1 immunostaining was also observed in regions adjacent to necrotic core of lesions.
Most of the HMGB1-positive cells within fatty streaks and fibrofatty lesions also expressed RAGE (please see Figure I, available online at http://atvb.ahajournals.org).
Cytokines Elevate HMGB1 mRNA Expression
Because HMGB1 is upregulated in macrophages of atherosclerotic lesions, we examined the dependency of upregulation of its mRNA on cytokines. IFN- induced a time-dependent increase in HMGB1 mRNA, which was maximal after 24 hours (Figure 2A). TNF- induced a similar time-dependent increase in mRNA. TGF-? also elevated HMGB1 mRNA levels (Figure 2B). HMGB1 mRNA did not increase in unstimulated monocytes after 24 hours in culture (control in Figure 2A) or after 48, 72, and 96 hours (not shown). Because IFN-, TNF-, and TGF-? can signal via extracellular signal kinase (ERK1/ERK2),25–27 we examined the extent to which ERK1/ERK2 might contribute to the elevations in HMGB1 mRNA. Pre-exposure to PD98059 did not affect the ability of IFN-, TNF-, or TGF-? to elevate HMGB1 mRNA (Figure 2B). Similarly, protein kinase C, which is also implicated in signaling by these cytokines,28–30 does not contribute to the elevations in HMGB1 mRNA, because bisindolylmaleimide did not attenuate the mRNA elevations (not shown). In contrast, the phosphatidylinositol-3 (PI-3) kinase inhibitor wortmannin attenuated the elevations in HMGB1 by all 3 cytokines (Figure 2); PI-3 kinase has previously been implicated in cell signaling by all 3 cytokines.31–33 Thus, PI-3 kinase appears essential for IFN-, TNF-, and TGF-? induced elevations in HMGB1 mRNA in monocytes.
Figure 2. Reverse-transcription polymerase chain reaction (RT-PCR) analysis of HMGB1 mRNA expression in THP-1 monocytes. A, Level of HMGB1 mRNA increases after 24 hours of treatment with either IFN- or TNF-. Level of HMGB1 mRNA in control cultures did not change after 48, 72, or 96 hours (as indicated in Results and Cytokines Elevate HMGB1 mRNA Expression sections), and the control result shown reflects a 24-hour culture. B, Pretreatment with wortmannin but not PD098059 decreases levels of HMGB1 mRNA in THP-1 monocytes induced by IFN-, TNF-, or TGF-?. Equal RNA loading for RT-PCR was verified using expression of GAPDH mRNA.
Regulation of HMGB1 Protein Expression and Secretion by Cytokines
Lysophosphatidylcholine and IFN- increase HMGB1 secretion by monocytes.11,12 Consequently, we examined whether other cytokines that are expressed in human atherosclerotic lesions might also elevate HMGB1 secretion by monocyte/macrophages. TNF-, IFN-, and TGF-?1 induced secretion of HMGB1 in freshly isolated blood monocytes (please see Figure II, available online at http://atvb.ahajournals.org).
Control unstimulated THP-1 monocytes secreted very low amounts of HMGB1 into the medium (Figure 3). Proinflammatory members of the TNF superfamily, TNF-, TWEAK, CD40L (not shown), and IFN- markedly increased the secretion of HMGB1 into the medium (Figure 3A). Secretion increased in a time-dependent manner over 96 hours. TGF-?1 also stimulated HMGB1 secretion, which was apparent after 72 hours and continued through to at least 96 hours. In all instances, intracellular levels of HMGB1 remained relatively constant, suggesting that the increased release of HMGB1 was also accompanied by significant increases in HMGB1 protein synthesis. HMGB1-containing media concentrated from THP-1 cells markedly increased ERK1/2 phosphorylation in rat aortic SMCs and induced their migration (44.2±1.6 cells/field in HMGB1-containing medium versus 10.2±0.5 cells/field in control medium; P<0.05) (please see Figure III, available online at http://atvb.ahajournals.org). Inhibition of SAPK/p38 MAPK and NF-B by pre-incubating the monocytes with SB203580 and isohelenin did not affect the IFN-, TNF-, and TGF-?–stimulated increases in HMGB1 secretion (Figure 3B). In contrast inhibitors of ERK1/ERK2 (PD098059), protein kinase C (bisindoylmaleimide) and PI3-kinase (wortmannin) all attenuated secretion stimulated by these cytokines (Figure 3B), suggesting that ERK1/ERK2, protein kinase C, and PI3-kinase activities are necessary for cytokine-stimulated HMGB1 secretion.
Figure 3. Secretion of HMGB1 protein by THP-1 cells analyzed by Western blotting. A, Level of HMGB1 protein expression and secretion after treating THP-1 monocytes with IFN-, TNF-, TGF-?, or TWEAK for 24, 48, 72, or 96 hours. B, Pretreatment of cells with PD098059, bisindoylmaleimide, and wortmannin attenuates HMGB1 secretion induced by 48-hour incubation with cytokines but not the secretion of endolysosomal marker cathepsin D.
HMGB1 lacks a secretory peptide and is thought to be secreted via endolysosomes in a manner analogous to IL-1?.11 Because endolysosomes contain cathepsin D, we sought to determine whether IFN-, TFN-, and TGF-? affect cathepsin D secretion in a manner similar to HMGB1. In contrast to HMGB1, only TNF- and TGF-? elevated the secretion of cathepsin D into the medium (Figure 3B); inhibitors of ERK1/ERK2, protein kinase C, and PI-3 kinase did not uniformly attenuate its secretion. It would appear that HMGB1 secretion occurs independently of endolysosomes containing cathepsin D.
Because cell necrosis can also be responsible for HMGB1 release, we examined cytotoxity in cultures by measuring LDH activity. LDH activity in monocytes treated with cytokines averaged 103.4±7.9% compared with control (100%) over 3 cytokines (IFN-, TFN-, and TGF-?1), whereas HMGB1 release was 5-times greater compared to control (Figure 3). Therefore, the increase in HMGB1 secretion is unlikely to be increased because of spontaneous release from necrotic cells.
Regulation of HMGB1 Cellular Localization by Cytokines
To determine whether the cytokines that induce HMGB1 secretion also affect its cellular distribution, THP-1 monocyte/macrophage cultures were immunostained with anti-HMGB1 antibodies and nuclei identified using propidium iodide (PI). Control monocyte/macrophages constitutively expressed HMGB1 and maintained an intracellular pool mostly localized in the nucleus (Figure 4). Exposure to TGF-?1 induced a redistribution of HMGB1 from the nucleus and into cytoplasm (Figure 4). IFN- also induced a similar redistribution, which was most apparent after 2 days when HMGB1 was being actively secreted into the medium. In contrast, TNF- did not induce substantial redistribution of HMGB1 during this time despite inducing active secretion of HMGB1 (Figure 4). Thus, it would appear that although a number of cytokines induce translocation of HMGB1 from the nucleus to the cytoplasm, when substantial amounts of HMGB1 are present in the cytoplasm, this redistribution is apparently not necessary for active secretion of HMGB1. Pretreatment of cells with inhibitors of SAPK/p38 MAP kinase, NF-B, MEK1/2, protein kinase C, or PI-3 kinase did not affect cytokine-induced redistribution of HMGB1 in the cells (not shown).
Figure 4. Confocal analysis of intracellular distribution of HMGB1 protein in THP-1 monocytes. In control cells, HMGB1 (green) is mostly localized in the nuclei (counterstained by propidium iodide , red) as shown by yellow representing colocalization. Monocytes exposed to either TGF-? or IFN- for 2 days contained high amounts of HMGB1 in their cytoplasm with less protein localized in the nuclei. In contrast, THP-1 cells treated with TNF- did not exhibit as great a shift in HMGB1 distribution. Bottom panels show the lack of staining with a control IgG.
Discussion
This study demonstrates that HMGB1 is highly expressed in human atherosclerotic lesions and may contribute to the sustained inflammation of fatty rich lesions. HMGB1 expression is highest in fibrofatty lesions and mostly associated with macrophages. Cytokines TNF- and other members of the TNF superfamily, as well as TGF-?1 and IFN-, appear responsible for its upregulation in monocyte/macrophages and its secretion. Upregulation of its mRNA was dependent on PI-3kinase, whereas secretion stimulated by the cytokines appears dependent on MEK1/2, protein kinase C, and PI-3 kinase.
HMGB1, a 30-kDa member of the high-mobility group nonhistone chromosomal protein family,2,34 is also a mediator of delayed endotoxin lethality and systemic inflammation.3 Recent studies have associated local expression of HMGB1 with local sites of tissue inflammation.8–10 For example, in rheumatoid arthritis, HMGB1 was localized to the cytoplasm of CD68-positive cells infiltrating the sublining layer, suggesting that it is secreted by synovial macrophages.35 In normal aortic intima, only a small number of cells, intimal SMCs, some ECs, and occasionally present macrophages expressed HMGB1, whereas in fibrofatty lesions macrophages were the major cell type expressing HMGB1, with expression nearly always in the cytoplasm. This distribution is similar to that reported in rheumatoid arthritis, where HMGB1 levels are also increased in synovial fluid, indicative of secretion by macrophages.35 Secreted HMGB1 can profoundly affect the function of cells associated with atherosclerotic lesions, particularly macrophages, ECs, and vascular SMCs, promoting local inflammation, the accumulation of monocyte/macrophages, and even remodeling of the lesion. HMGB1 induces human monocytes to release a large array of inflammatory cytokines, including TNF-, IL-1, IL-1?, IL-1RA, IL-6, IL-8, and MIP-1 and MIP-1?.4 Its actions on ECs include increasing the expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and RAGE, as well as increasing the secretion of TNF-, IL-8, monocyte chemotactic protein-1, and tissue plasminogen activator.6 HMGB1 also induces cytoskeleton reorganization and chemotaxis in vascular SMCs.7
Our findings that the proinflammatory cytokines IFN- and TNF-, as well as TGF-?1, elevate HMGB1 mRNA in monocyte/macrophages suggest that HMGB1 exerts its inflammatory action downstream of proinflammatory cytokines. The finding that TNF- induces HMGB1 secretion is particularly important because the secreted HMGB1 may in turn induce further secretion of TNF-,4 raising the possibility that a proinflammatory loop exists between TNF- and HMGB1, which increases the severity of inflammation and prolongs its duration. Whether such a proinflammatory loop also exists for other inflammatory cytokines of the TNF superfamily and HMGB1 remains to be determined. Our findings also suggest that PI-3 kinase is essential for the upregulation of HMGB1 mRNA by these cytokines. PI-3 kinase is known to regulate the activity of Sp1-responsive promoters,36 which are abundant in the 5'-region of the human HMGB1 gene.37 TGF-?, TNF-, and IFN- are all known to potently stimulate phosphorylation of Akt.33,38,39
HMGB1 is released from necrotic cells, inducing local inflammation.13 In macrophages, release of HMGB1 can also be induced by lysophosphatidylcholine and IFN-.11,12 IFN-–dependent release of HMGB1 could be inhibited by a specific inhibitor of Janus kinase 2.12 In both instances, release appeared to be associated with its translocation from the nucleus to the cytoplasm.11,12 We have demonstrated that a number of proinflammatory cytokines of the TNF superfamily, TNF-, TWEAK, and CD40L (not shown), as well as TGF-?1 and IFN-, are also capable of stimulating HMGB1 secretion from human monocyte/macrophages. These cytokines stimulated secretion in a somewhat delayed manner, being most apparent after 48 hours. Secretion was accompanied by increased HMGB1 protein synthesis, because intracellular HMGB1 levels were unaltered during this time. Our findings indicate that HMGB1 secretion stimulated by cytokines is in part dependent on transient elevations in its mRNA and elevations in its biosynthesis. Secretion can be also accompanied by its translocation from the nucleus to the cytoplasm. This was most apparent with TGF-?, IFN-, and CD40L. TNF- was least effective in inducing such translocation, although it is a potent stimulant of HMGB1 secretion. Because HMGB1 lacks a hydrophobic signal sequence, it appears to be secreted via a yet to be fully defined nonclassical, vesicle-mediated secretory pathway analogous to that used by IL-1?.11 However, in contrast to IL-1?–containing secretory lysosomes,40 those responsible for HMGB1 secretion do not apparently contain cathepsin D. We found no correlation between HMGB1 and cathepsin D secretion under a variety of different conditions. Cytokine-induced HMGB1 secretion could be attenuated by inhibitors of MEK1/MEK2, protein kinase C, and PI-3 kinase/Akt, suggesting that such agents might be useful in attenuating HMGB1-mediated inflammation. Although the mechanisms by which these agents attenuate HMGB1 secretion needs to be clarified, it is possible that they inhibit lysosomal degranulation and/or HMGB1 protein synthesis via effects on mRNA translation. Protein kinase C, MEK1/MEK2, and PI-3 kinase appear essential for lysosomal degranulation41–43 and also regulate mRNA translational processes.43–46
In summary, this is the first study to our knowledge to identify that macrophages are a major source of HMGB1 in human atherosclerotic lesions. Its expression and secretion by macrophages is highly regulated by cytokines. Upregulation and secretion of HMGB1 has the potential to amplify inflammatory responses and may also contribute to macrophage accumulation, thereby promoting atherogenesis. Definition of its precise roles in the development and progression of atherosclerosis will require additional in vivo investigations.
References
Mosevitsky MI, Novitskaya VA, Iogannsen MG, Zabezhinsky MA. Tissue specificity of nucleo-cytoplasmic distribution of HMG1 and HMG2 proteins and their probable functions. Eur J Biochem. 1989; 185: 303–310.
Bustin M. Regulation of DNA-dependent activities by the functional motifs of the high-mobility-group chromosomal proteins. Mol Cell Biol. 1999; 19: 5237–5246.
Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J, Frazier A, Yang H, Ivanova S, Borovikova L, Manogue KR, Faist E, Abraham E, Andersson J, Andersson U, Molina PE, Abumrad NN, Sama A, Tracey KJ. HMG1 as a late mediator of endotoxin lethality in mice. Science. 1999; 285: 248–251.
Andersson U, Wang H, Palmblad K, Aveberger AC, Bloom O, Erlandsson-Harris H, Janson A, Kokkola R, Zhang M, Yang H, Tracey KJ. High mobility group 1 protein (HMG-1) stimulates proinflammatory cytokine synthesis in human monocytes. J Exp Med. 2000; 192: 565–570.
Andersson U, Erlandsson-Harris H, Yang H, Tracey KJ. HMGB1 as a DNA-binding cytokine. J Leukoc Biol. 2002; 72: 1084–1091.
Fiuza C, Bustin M, Talwar S, Tropea M, Gerstenberger E, Shelhamer JH, Suffredini AF. Inflammation-promoting activity of HMGB1 on human microvascular endothelial cells. Blood. 2003; 101: 2652–2660.
Degryse B, Bonaldi T, Scaffidi P, Muller S, Resnati M, Sanvito F, Arrigoni G, Bianchi ME. The high mobility group (HMG) boxes of the nuclear protein HMG1 induce chemotaxis and cytoskeletal reorganization in rat SMCs. J Cell Biol. 2001; 152: 1197–1206.
Fang WH, Yao YM, Shi ZG, Wu Y, Lu LR, Sheng ZY. The significance of changes in high mobility group-1 protein mRNA expression in rats after thermal injury. Shock. 2002; 17: 329–333.
Abraham E, Arcaroli J, Carmody A, Wang H, Tracey KJ. HMG-1 as a mediator of acute lung inflammation. J Immunol. 2000; 2950–2954.
Kokkola R, Sundberg E, Ulfgren AK, Palmblad K, Li J, Wang H, Ulloa L, Yang H, Yan XJ, Furie R, Chiorazzi N, Tracey KJ, Andersson U, Harris HE. High mobility group box chromosomal protein-1: a novel proinflammatory mediator in synovitis. Arthritis Rheum. 2002; 46: 2598–2603.
Gardella S, Andrei C, Ferrera D, Lotti LV, Torrisi MR, Bianchi ME, Rubartelli A. The nuclear protein HMGB1 is secreted by monocytes via a non-classical, vesicle-mediated secretory pathway. EMBO Reports. 2002; 3: 995–1001.
Rendon-Mitchell B, Ochani M, Li J, Wang H, Yang H, Susarla S, Czura C, Mitchell RA, Chen G, Sama AE, Tracey KJ, Wang H. IFN- induces high mobility box 1 protein release partially through a TNF-dependent mechanism. J Immunol. 2003; 170: 3890–3897.
Scaffidi P, Misteli T, Bianchi ME. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature. 2002; 418: 191–195.
Hori O, Brett J, Slattery T, Cao R, Zhang J, Chen JX, Nagashima M, Lundh ER, Vijay S, Nitecki D, Morser J, Stern D, Schmidt AM. The receptor for advanced glycation end products (RAGE) is a cellular binding site for amphoterin. Mediation of neurite outgrowth and co-expression of range and amphoterin in the developing nervous system. J Biol Chem. 1995; 270: 25752–25761.
Taguchi A, Blood D, del Toro G, Canet A, Lee DC, Qu W, Tanji N, Lu Y, Lalla E, Fu C, Hofmann MA, Kislinger T, Ingram M, Lu A, Tanaka H, Hori O, Ogawa S, Stern DM, Schmidt AM. Blockade of RAGE-amphoterin signalling suppresses tumour growth and metastases. Nature. 2000; 405: 354–358.
Huttunen HJ, Kuja-Panula J, Rauvala H. Receptor for advanced glycation end products (RAGE) signaling induces CREB-dependent chromogranin expression during neuronal differentiation. J Biol Chem. 2002; 277: 38635–38646.
Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Cipollone F, Iezzi A, Fazia M, Zucchelli M, Pini B, Cuccurullo C, De Cesare D, De Blasis G, Muraro R, Bei R, Chiarelli F, Schmidt AM, Cuccurullo F, Mezzetti A. The receptor RAGE as a progression factor amplifying arachidonate-dependent inflammatory and proteolytic response in human atherosclerotic plaques. Circ. 2003; 108: 1070–1077.
Bucciarelli LG, Wendt T, Qu W, Lu Y, Lalla E, Rong LL, Goova MT, Moser B, Kislinger T, Lee DC, Kashyap Y, Stern DM, Schmidt AM. RAGE blockade stabilizes established atherosclerotic lesions in diabetic apolipoprotein E-null mice. Circ. 2002; 106: 2827–2835.
Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insul W, Rosenfeld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis. Circ. 1995; 92: 1355–1374.
Kalinina N, Agrotis A, Tararak E, Antropova Y, Kanellakis P, Ilyinskaya O, Quinn MT, Smirnov V, Bobik A. Cytochrome b558-dependent NAD(P)H oxidase-phox units in smooth muscle and macrophages of atherosclerotic lesions. Arterioscler Thromb Vasc Biol. 2002; 22: 2037–2043.
Candido R, Forbes JM, Thomas MC, Thallas V, Dean RG, Burns WC, Tikellis C, Ritchie RH, Twigg SM, Cooper ME, Burrell LM. A breaker of advanced glycation end products attenuates diabetes-induced myocardial structural changes. Circ Res. 2003; 92: 785–792.
Ward MR, Agrotis A, Kanellakis P, Dilley R, Jennings G, Bobik A. Inhibition of protein tyrosine kinases attenuates increases in expression of transforming growth factor-? isoforms and their receptors following arterial injury. Arterioscler Thromb Vasc Biol. 1997; 17: 2461–2470.
Palumbo R, Sampaolesi M, De Marchis F, Tonlorenzi R, Colombetti S, Mondino A, Cossu G, Bianchi ME. Extracellular HMGB1, a signal of tissue damage, induces mesoangioblast migration and proliferation. J Cell Biol. 2004; 164: 441–449.
Balnchette J, Jarmillo M, Olivier M. Signalling events involved in interferon-–inducible macrophage nitric oxide generation. Immunology. 2003; 108: 513–522.
Goetze S, Kineshior K, Meehan WP, Collins A, Fleek E, Hsueh WA, Law RE. TNF- induces expression of transcription factors c-fos, Egr-1 and Ets-1 in vascular lesions through extracellular signal-regulated kinases 1/2. Atherosclerosis. 2001; 159: 93–101.
Tokuda H, Hatakeyama D, Akamatsu S, Tanable K, Yoshida M, Shibata T, Kozawa O. Involvement of MAP kinases in TGF-?–stimulated vascular endothelial growth factor synthesis in osteoblasts. Arch Biochem Biophys. 2003; 415: 117–125.
Deb DK, Sassano A, Lekmine F, Majchrzak B, Bverma A, Kambhampati S, Uddin S, Rahman A, Fish EN, Platanias LC. Activation of protein kinase C by IFN-. J Immunol. 2003; 171: 267–273.
Rahman JK, Anwar KN, Frey RS, Minshall RD, Malik AB. Tumor necrosis factor- induces early-onset endothelial adhesivity by protein kinase Czeta-dependent activation of intercellular adhesion molecule-1. Circ Res. 2003; 92: 1089–1097.
Runyan CE, Schnaper HW, Poncelet AC. Smad3 and PKC mediate TGF-?–mediated collagen-1 expression in human mesangial cells. Am J Physiol. 2003; 285: F413–F422.
Kawanaka H, Jones MK, Szabo IL, Baatar D, Pai R, Tsugawa K, Sugimachi K, Sarfeh IJ, Tarnawski AS. Activation of eNOS in rat portal hypertensive gastric mucosa is mediated by TNF- via the PI-3 kinase-Akt signaling pathway. Hepatology. 2002; 35: 393–402.
Kin G, Jun JB, Elkon KB. Necessary role of phosphatidylinositol 3-kinase in transforming growth factor-? mediated activation of Akt in normal and rheumatoid arthritis synovial fibroblasts. Arthritis Rheum. 2002; 46: 12504–11511.
Navarro A, Anad-Apte B, Tanabe Y, Feldman G, Larner AC. A PI-3 kinase-dependent, STAT1-independent signaling pathway in interferon-stimulated monocyte adhesion. J Leukoc Biol. 2003; 73: 540–545.
Bianchi ME, Beltrame M, Paonessa G. Specific recognition of cruciform DNA by nuclear HMG1. Science. 1989; 243: 1056–1059.
Taniguchi N, Kawahara K, Yone K, Hashiguchi T, Yamakuchi M, Goto M, Inoue K, Yamada S, Ijiri K, Matsunaga S, Nakajima S, Maruyama I. High mobility group box chromosomal protein 1 plays a role in the pathogenesis of rheumatoid arthritis as a novel cytokine. Arthritis Rheum. 2003; 48: 971–981.
Garcia A, Cereghini S, Sontag E. Protein phosphatase 2A and phosphatidylinositol 3-kinase regulate the activity of Sp1-responsive promoters. J Biol Chem. 2000; 275: 9385–9389.
Lum HK, Lee K-L, D. the human HMGB1 promoter is modulated by a silencer and an enhancer-containing intron. Biochimica et Biophysica Acta. 2000; 1520: 79–84.
Kim G, Jun JB, Elkon KB. Necessary role of phosphatidylinositol 3-kinase in transforming growth factor ?–mediated activation of Akt in normal and rheumatoid arthritis synovial fibroblasts. Arthritis Rheum. 2002; 46: 1504–1511.
Barsacchi R, Perrotta C, Bulotta S, Moncada S, Borgese N, Clementi E. Activation of endothelial nitric-oxide synthase by tumor necrosis factor-: a novel pathway involving sequential activation of neutral sphingomyelinase, phosphatidylinositol-3'-kinase and Akt. Mol Pharmacol. 2003; 63: 886–895.
Gardella S, Andrei C, Lotti LV, Poggi A, Torrisi MR, Zocchi MR, Rubartelli A. CD8+ T lymphocytes induce polarized exocytosis of secretory lysosomes by dendritic cells with release of interleukin-1? and cathepsin D. Immunobiology. 2001; 98: 2152–2159.
Tapper H, Sundler R. Protein kinase C and intracellular pH regulate zymosan-induced lysosomal enzyme secretion in macrophages. J Leukoc Biol. 1995; 58: 485–494.
Badewa AP, Heiman As. Inhibition of CCL11, CCL24 and CCL26-induced degranulation in HL-60 eosinophilic cells by specific inhibitors of MEK1/MEK2, p38 MAP kinase and PI 3-kinase. Immunopharmacol Immunotoxicol. 2003; 25: 145–157.
Huber M, Hughes MR, Krystal G. Thapsigargin-induced degranulation of mast cells is dependent on transient activation of phosphatidylinositol-3 kinase. J Immunol. 2000; 165: 124–133.
Stolovich M, Tang H, Hornstein E, Levy G, Cohen R, Bae SS, Birnbaum MJ, Meyuhas O. Transduction of growth or mitogen signals into translational activation of TOP mRNAs is fully reliant on the phosphatidylinositol 3-kinase-mediated pathway but requires neither 1 nor rpS6 phosphorylation. Mol Cell Biol. 2002; 22: 8101–8113.
Kumar V, Pandey P, Sabatini D, Kumar M, Majumder PK, Bharti A, Carmichael G, Kufe D, Kharbanda S. Functional interaction between RAFT1/FRAP/mTOR and protein kinase C in the regulation of cap-dependent initiation of translation. EMBO J. 2000; 19: 1087–1097.
Petegnief V, Friguls B, Sanfeliu C, Sunol C, Planas AM. Transforming growth factor- attenuates N-methyl-D-aspartic acid toxicity in cortical cultures by preventing protein synthesis inhibition through an Erk1/2-dependent mechanism. J Biol Chem. 2003; 278: 29552–29559.(Role of Activated Macroph)