diabetes-associated sustained activation of the ...george perry,7 ann-marie schmidt,5 david m....

Diabetes-Associated Sustained Activation of theTranscription Factor Nuclear Factor-�BAngelika Bierhaus,

1,2Stephan Schiekofer,

1,2Markus Schwaninger,

1Martin Andrassy,

1,2

Per M. Humpert,2

Jiang Chen,1,2

Mei Hong,2

Thomas Luther,3

Thomas Henle,3

Ingrid Kloting,4

Michael Morcos,1

Marion Hofmann,5

Hans Tritschler,6

Bernd Weigle,3

Michael Kasper,3

Mark Smith,7

George Perry,7

Ann-Marie Schmidt,5

David M. Stern,5

Hans-Ulrich Haring,2

Erwin Schleicher,2

and Peter P. Nawroth1,2

Activation of the transcription factor nuclear factor-�B(NF-�B) has been suggested to participate in chronicdisorders, such as diabetes and its complications. Incontrast to the short and transient activation of NF-�Bin vitro, we observed a long-lasting sustained activationof NF-�B in the absence of decreased I�B� in mononu-clear cells from patients with type 1 diabetes. This wasassociated with increased transcription of NF-�Bp65. Acomparable increase in NF-�Bp65 antigen and mRNAwas also observed in vascular endothelial cells of dia-betic rats. As a mechanism, we propose that binding ofligands such as advanced glycosylation end products(AGEs), members of the S100 family, or amyloid-�peptide (A�) to the transmembrane receptor for AGE(RAGE) results in protein synthesis–dependent sus-tained activation of NF-�B both in vitro and in vivo.Infusion of AGE-albumin into mice bearing a �-globinreporter transgene under control of NF-�B also resultedin prolonged expression of the reporter transgene. Invitro studies showed that RAGE-expressing cells in-duced sustained translocation of NF-�B (p50/p65) fromthe cytoplasm into the nucleus for >1 week. SustainedNF-�B activation by ligands of RAGE was mediated byinitial degradation of I�B proteins followed by new

synthesis of NF-�Bp65 mRNA and protein in the pres-ence of newly synthesized I�B� and I�B�. These datademonstrate that ligands of RAGE can induce sustainedactivation of NF-�B as a result of increased levels of denovo synthesized NF-�Bp65 overriding endogenous neg-ative feedback mechanisms and thus might contributeto the persistent NF-�B activation observed in hyper-glycemia and possibly other chronic diseases. Diabetes50:2792–2808, 2001

Tissue culture models of cellular activation pro-vide easily accessible systems for detailed anal-ysis of mechanisms potentially underlying thepathogenesis of human disease. However, the

time course of such in vitro models is usually significantlyabbreviated, limited to hours to days, compared with thepace of disorders under study in vivo. This indicates theimportance of seeking out mechanisms in cell culture thatmight bridge the gap that accounts for the chronicity ofcellular perturbation observed in the intact organism.

The transcription factor nuclear factor-�B (NF-�B) hasbeen proposed as a critical bridge between oxidant stressand gene expression (1–8). Exposure of cells to inflamma-tory, infectious, or other stressful stimuli results in rapidphosphorylation and degradation of I�B� and the subse-quent release and translocation of NF-�B into the nucleus(1–11). This mechanism ensures quick and finely tunedcellular responses in the absence of de novo proteinsynthesis. Because transcription of I�B� is positivelyautoregulated by NF-�B (9–11), activation of NF-�B isusually self-terminated within minutes to hours (1–11).Such a scenario lends itself to analysis by short-term invitro studies in which stimulus-induced responses aretransient and the system returns to the baseline state overhours. Consequently, induction of NF-�B and enhancedtranscription of its target genes in vitro have been studiedmainly in the setting of acute cellular responses.

Reactive oxygen intermediates are generated by pro-cesses that occur over seconds. However, increasing evi-dence suggests a role for oxidative stress in chronicdegenerative diseases such as atherosclerosis (1,6,12,13),diabetes (14–16), and Alzheimer’s disease (17–19). Thisindicates the relevance of signal transduction systemssuch as NF-�B, which are capable of transforming theappearance and disappearance of short-lived oxygen freeradicals into more sustained signals for cellular activation

From the 1Department of Medicine I and Department of Neurology, Universityof Heidelberg, Heidelberg, Germany; 2Department of Medicine IV, UniversityTubingen, Tubingen, Germany; 3Department of Anatomy, Department ofImmunology, Department of Pathology and Institute of Food Chemistry,Technical University Dresden, Dresden, Germany; 4Department of LaboratoryAnimal Science, Institute of Pathophysiology, University Greifswald, Karls-burg, Germany; 5Columbia University, Department of Physiology, New York,New York; 6ASTA-Medica, Frankfurt am Main, Germany; and 7Institute ofPathology, Case Western Reserve University, Cleveland, Ohio.

Address correspondence and reprint requests to Peter P. Nawroth, MD,Department of Medicine I, University of Heidelberg, Bergheimer Strasse 58,69115 Heidelberg, Germany. E-mail: [email protected].

Received for publication 19 October 2000 and accepted in revised form12 September 2001.

M.S. and G. P. serve as consultants/collaborators for Panacea Pharmaceu-ticals, Prion Development Laboratories, and Voyager Laboratories. A.-M.S.and D.M.S. are consultants for and have received a research grant fromTransTech Pharma. P.P.N. has received a research grant from ASTA-Medica.

A�, amyloid-� peptide; AGE, advanced glycosylation end product; BAEC,bovine aortic endothelial cell; �-Gal, �-galactosidase; CML, carboxymethyl-lysine; DMEM, Dulbecco’s modified Eagle’s medium; ECL, enhanced chemi-luminescence; EMSA, electrophoretic mobility shift assay; ERK, extracellularsignal-regulated kinases; FCS, fetal calf serum; HO-1, heme oxygenase-1;HPLC, high-performance liquid chromatography; HPRT, hypoxanthine gua-nine phosphoribosyl transferase; LPS, lipopolysaccharide; NF-�B, nuclearfactor-�B; ODN, oligodeoxynucleotide; PBMC, peripheral blood mononuclearcell; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; PMSF,phenylmethylsulfonyl fluoride; [PS]ODN, phosphorothioate oligodeoxynucle-otide; RAGE, receptor for AGE; ROS, reactive oxygen species; RT, reversetranscription; rt, room temperature; SMC, smooth muscle cell; sRAGE, solubleRAGE; TBS, Tris-buffered saline; TNF-�, tumor necrosis factor-�.

2792 DIABETES, VOL. 50, DECEMBER 2001

(8,16). Consistent with this concept, a recent study dem-onstrated that hyperglycemia-dependent overproductionof mitochondrial superoxide results in NF-�B activation(16). NF-�B, however, also acts as a protective factoragainst programmed cell death by inducing antiapoptoticgenes such as A1, A20, XIAP, Bcl-2, and BcL-XL and thusensures cell survival in situations of perpetuated cellularactivation (20–23). Therefore, more sustained protein syn-thesis–dependent pathways of NF-�B activation might bepresent in addition to immediate, self-controlled proteinsynthesis–independent pathways of NF-�B activation torescue cells from cell death in the presence of a persistentstimulus.

Pathophysiologically relevant cellular perturbants asso-ciated with oxidant stress and chronic disorders areadvanced glycosylation end products (AGEs) (14,16,24–26), S100 proteins (27), and amyloid-� peptides (A�)(28,29). AGEs are the result of nonenzymatic glycation andglycoxidation that accumulate during normal aging(14,24–26). Their deposition is enhanced with concomi-tant hyperglycemia, oxidant stress, carbonyl stress, anddisorders associated with delayed macromolecular turn-over, such as renal failure and amyloidosis (16,18,24–26,28–40). Carboxymethyllysine (CML), a rapidly formedglycoxidation product that increases in mononuclearblood cells of healthy volunteers undergoing hyperglyce-mic clamps already within 2 h (S.S., P.P.N., unpublishedobservation), is one of the earliest markers of AGE forma-tion (36–40). A� is the primary component of the extra-cellular senile plaques in Alzheimer’s disease (29) and hasbeen strongly linked to the pathogenesis of this devastat-ing neurodegenerative disease (18,29,30). AGEs, S100 pro-teins, and A� are ligands of the receptor for AGEs (RAGE)(24,26–32,41–49). Due to their long-life nature, their ubiq-uitous presence in the plasma, vessel wall, and tissues,and their recognition by RAGE, whose expression levelsseem to increase in the course of chronic disorders(24,29,31,32,41–49), we considered the hypothesis thatligand-receptor interaction may alter cellular homeostasisin a self-perpetuating manner.

RAGE is a member of the immunoglobulin superfamilyof cell surface molecules (24,27–29,31,32,41,50) and isexpressed by endothelial cells, monocytes/macrophages,neurons, and a range of other cells (50) whose dysfunctionhas been linked to chronic disorders such as diabetes,Alzheimer’s disease, inflammation, and atherosclerosis.Binding of ligands to RAGE triggers signal transductionmechanisms, including formation of lipid peroxides, acti-vation of NF-�B, and subsequent expression of genesregulated by NF-�B (27–29,31,32,41–45,51–53).

There is increasing evidence that NF-�B activation invivo occurs in a more sustained manner. ConstitutiveNF-�B activation has been observed in different malignantcells (54–58) as a result of a defect or dysregulation ofI�B� and I�B� or aberrant activation of I�B-kinase (58). Apersistent NF-�B activation has also been suggested inatherosclerosis (13,59), Crohn’s disease (60–62), and Lis-teria monocytogenes infection (63). In vitro data havesuggested that hypophosphorylated I�B� (64–67) or denovo synthesis of c-rel can mediate longer-lasting NF-�Bactivation (68,69).

However, experiments performed in nonimmune and

nonmalignant cells were not directed toward understand-ing mechanisms underlying chronic NF-�B activation invivo, and the mechanism by which the normally short-acting response of NF-�B activation is converted to apersistent cell perturbation is not understood. RAGE wasrecently shown to be involved in chronic disease, such asatherosclerosis, amyloidosis, diabetes, inflammation, tu-mor proliferation, and metastasis (27–29,31,43–49,53).Therefore, we studied whether ligand-stimulated RAGEactivation can act like a master switch to turn short-lastingactivation into a prolonged activation of NF-�B.

RESEARCH DESIGN AND METHODS

Reagents. Reagents were obtained as follows: Dulbecco’s modified Eagle’smedium (DMEM; 4,500 mg/l glucose), RPMI 1648, HEPES buffer solution,L-glutamine, penicillin-streptomycin mixture, and phosphate-buffered saline(PBS; pH 7.4) were from Biowhittaker (Walkerville, MD). Fetal calf serum(FCS) was from Gibco/BRL (Dreieich, Germany). [�-32P]ATP (3,000 Ci/mmolat 10 mCi/ml), L-[35S]methionine (�1,000 Ci/mmol at 10 mCi/ml), Hybond-N-nylonfilter, enhanced chemiluminescence (ECL)-nitrocellulose membranes,ECL detection reagents, and Hyperfilm X-ray films were obtained fromAmersham (Freiburg, Germany). A� peptides (1–40 and 25–35), S100B,phenylmethylsulfonyl fluoride (PMSF), actinomycin D, and cycloheximidewere purchased from Sigma (Deisenhofen, Germany). Poly dI/dC and oli-go(dt) was from Pharmacia (Freiburg, Germany). Anti-p65 (#sc-109), anti-I�B� (#sc-847), anti-I�B� (#sc-945), anti–c-Jun KM-1 polyclonal antibodies,the respective second antibodies, and protein A/G PLUS-agarose were ob-tained from Santa Cruz (Heidelberg, Germany). Monoclonal anti-p65 antibod-ies, specific for active NF-�B, were obtained from Boehringer Mannheim(Mannheim, Germany). Soluble RAGE (sRAGE) preparations used throughoutthis study have previously been described in detail (29,31,32,45,49,70). Poly-clonal anti-RAGE antibodies, generated in goat with recombinant RAGEprepared in Escherichia coli as antigen, were a gift of Dr. M.A. Shearman(Merck, Sharpe & Dome, Essex, U.K.). Anti–heme oxygenase-1 (HO-1) anti-bodies were from Dr. M.A. Smith (Case Western Reserve University, Cleve-land, OH). Vectastain ABC kits were purchased from Vector Laboratories(Burlingame, CA). The vector NF-�B-Luc and primers for human HPRT werefrom Clontech (Heidelberg, Germany), and Effectene transfection reagent wasobtained from Qiagen (Hilden, Germany).Preparation of AGE-albumin, CML-albumin and A� peptides. AGE-albumin was prepared by incubating bovine or mouse serum albumin with 200mmol/l glucose-6-phosphate at 37°C for 4 to 8 weeks in 100 mmol/l phosphate(pH 7.4) and 0.5 mmol/l sodium azide. At the end of incubation, AGE-albuminpreparations were dialyzed against 5 l of 100 mmol/l phosphate and 10 mmol/lEDTA for 24 h and 0.9% NaCl for 12 h. Control albumin was also dialyzedagainst these buffers. Alternatively, glycated bovine albumin was purchasedfrom Sigma. Nonglycated bovine albumin, heat-inactivated AGE-bovine albu-min, and bovine albumin incubated with the synthetic substrate sorbitolserved as negative controls. AGE-albumin preparations were characterized bychromatographic means (see below).Quantification of furosine, pentosidine, and other amino acids after

acid hydrolysis. AGE-albumin samples were hydrolyzed in the presence of 6N hydrochloric acid for 23 h at 110°C; the hydrolysate was dried in vacuo ona Speedvac concentrator (Savent, Farmingdale, NY) and dissolved in 0.2Nsodium citrate buffer (pH 2.2). Analysis of amino acids, including CML, wasperformed on an Alpha Plus amino acid analyzer (LKB Biochrom, Cambridge,U.K.), using a stainless steel column (150 � 4 mm) filled with ion-exchangeresin, DC4A-spec sodium form (Benson, Reno, NV). Composition of elutionbuffers, ninhydrin reagent, and running conditions are described elsewhere(71,72). Injection volume was 10–80 �l. After separation on the ion-exchangecolumn, amino acids were initially detected using either a photodiode arraydetector PDA 996 (Waters, Eschborn, Germany; for furosine quantification, adetection wavelength of 280 nm was used) or a fluorescence detector SFM 25(Kontron, Eching, Germany; for pentosidine quantification, measuring wasachieved at excitation/emission wavelengths of 335/385 nm), connectedbetween the outlet of the column and the ninhydrin reaction coil. Aftersubsequent ninhydrin derivatization, amino acids were detected at 570 and440 nm. Protein content of the plasma samples was calculated as the sum ofthe amino acids. From values of furosine, fructoselysine was calculated bymultiplying the molar furosine values by 3.1, the known conversion factor forthe degradation of fructoselysine during acid hydrolysis (73,74). For quantifi-cation of pentosidine, reference material was synthesized according to Henleet al. (75). Furosine was obtained from Neosystem (Strasbourg, France).

A. BIERHAUS AND ASSOCIATES

DIABETES, VOL. 50, DECEMBER 2001 2793

Pyrraline analysis after enzymatic hydrolysis. Enzymatic hydrolysis wasperformed using four enzymes (pepsin, pronase, aminopeptidase, and proli-dase) as described previously (75). For pyrraline detection, amino acidanalysis was performed as described above with the combination of the PDA(set at a wavelength of 297 nm) and ninhydrin derivatization (75). Alterna-tively, pyrraline analysis was performed via isocratic ion-pair reversed-phasehigh-performance liquid chromatography (HPLC) with UV detection (76).Pyrraline reference material was synthesized according to Henle and Bach-mann (77).Preparation of CML-albumin and A� peptides. In vitro synthesis ofCML-albumin was performed as described by Schleicher et al. (36). Assays forendotoxin showed AGE-albumin and CML preparations to contain virtuallyundetectable levels of lipopolysaccharide (LPS; �10 pg/ml at a proteinconcentration of 5 mg/ml according to the Limulus assay [Sigma]). SyntheticA� peptides (1–40) were purchased from Sigma and made up freshly everytime. To allow fibril formation, we kept preparations at room temperature (rt)for at least 4 h before use.Isolation of CML-modified proteins from erythrocytes. Erythrocyteswere lysed in 0.9% NaCl, 1.5 mmol/l PMSF, 0.1 mmol/l leupeptin, 20 mg/mlsoybean inhibitor, and 2 mmol/l benzamidine by three freeze-thaw cyclesfollowed by pulsed ultrasonication for 3 min and a final freeze-thaw cycle.Insoluble material was removed by a 5-min centrifugation at 6,000g at 4°C, andthe supernatant was used for extraction of CML-modified proteins as previ-ously described (70). Ten micrograms of total erythrocyte lysate was loadedonto a 1-ml anti–CML-Sepharose column (36,70) and incubated for 2 h at rt(70). After extensive washing, CML-modified proteins were eluted with 1 mol/lglycine (pH 3.0), neutralized, and dialyzed in DMEM. Protein concentrationwas determined using the BCA assay system (Pierce, Rockford, IL).Plasmids. The SV-40 driven luciferase control plasmid pGL2-control, thepromoterless plasmid pGL2-basic, and the �-galactosidase (�-Gal) controlplasmid pSV-�-Gal were obtained from Promega (Heidelberg, Germany). Theplasmid NF-�B-Luc, which contains four tandem copies of the NF-�B consen-sus sequence fused to a TATA-like promoter region from the herpes simplexvirus thymidine kinase promoter, was from Clontech (Heidelberg, Germany).Transgenic mouse model. Mice transgenic for an NF-�B–driven �-globinreporter gene (tg14) were provided by Dr. Thomas Wirth (Wurzburg, Ger-many) and have been previously characterized in detail (78). Eight-week-oldfemale mice were left untreated or treated once with 1,000 �g of mouseAGE-albumin (500 �g i.p. and 500 �g i.v.) at time point 0 and kept for anadditional 6 days. Where indicated, mice received sRAGE (25 �g/mouse) oranti-RAGE antibodies (40 �g/mouse) as described (29,31,32). At the end of theexperiments, mice were killed and organs were removed, immediately snap-frozen, and analyzed by reverse transcription–polymerase chain reaction(RT-PCR) and electrophoretic mobility shift assays (EMSAs).Diabetic rat model. The kidneys of diabetic rats and of nondiabeticSprague-Dawley rats were provided by Dr. I. Koting (Karlsburg, Germany).Diabetic BB/O(ttawa)K(arlsburg) rats were bred as previously described indetail (79), developed diabetes at the age of 104 days (16 days), and werekept for 57 days (9 days) before harvest. During this time, rats were treatedwith sustained-release insulin implants (Linplant; Mollegaard, Skensved, Den-mark) with an insulin rate of 2 units/24 h. With this treatment, the meanplasma glucose amounted to 136 24 mg/dl.Patient characteristics. Blood from six patients with newly manifesteddiabetes type 1 (two men, four women; age 35.1 11.7 years; HbA1c 12.4 2.3%) was collected on the day the diagnosis was made. Four healthyvolunteers (three men, one woman; age 31.5 5.5 years) served as controlsubjects. From three of these patients, repetitive blood sampling was per-formed over 3 consecutive days. HbA1c levels were routinely determined byHPLC (Diamat; Biorad, Munich, Germany; normal range 4.5–6.1%) in thecentral laboratory unit of the hospital. The study was approved by the ethicalcommittee of the Department of Medicine, University of Tubingen, andinformed consent had been obtained from all patients studied.Preparation of peripheral blood mononuclear cells and immunocyto-

logical detection of activated NF-�Bp65. Immediately after venipuncture,peripheral blood mononuclear cells (PBMCs) were separated as previouslydescribed in detail (80–82). Isolated PBMC were washed three times in coldPBS (pH 7.4, 4°C), seeded on coverslips, and fixed in 4% paraformaldehydeaccording to standard methods. Staining for NF-�Bp65 was performed asdescribed below using a monoclonal antibody (Boehringer Mannheim), whichexclusively recognizes the activated form of NF-�B (13).Immunocytochemistry. Immunocytochemistry was performed by indirectimmunoperoxidase techniques using PBMC fixed onto poly-L-lysine–coatedglass slides with 4% paraformaldehyde; slides were washed twice in PBS (pH7.4) for 5 min. After the last wash, slides were treated with 0.3% hydrogenperoxide, dissolved in methanol, for 30 min at rt. Blocking was performed with0.6% normal goat or horse serum for 30 min (rt) before the affinity-purified

polyclonal rabbit antibody against I�B� (concentration: 0.01 �g/�l; 60 min; rt)or the monoclonal mouse antibody for activated NF-�Bp65 (concentration:0.01 �g/�l; 60 min; rt) were used as primary antibodies. Slides were washed 3times for 10 min in PBS (pH 7.4), and detection of signals was performed usingVectastain ABC kits (Vector Laboratories) according to the manufacturer’sinstructions, as previously described in detail (61,80,81). Peroxidase activitywas visualized with 0.05% 3,3-diaminobenzidine-tetrahydrochloride (VectorLaboratories) before the slides were counterstained with Mayer’s hematoxylin(Vector Laboratories). Controls for immunospecificity were included in allexperiments by omission of the primary antibody and its replacement by PBSand matching concentrations of normal rabbit serum (data not shown).RT-PCR. RT and PCR for �-globin, NF-�Bp65, hypoxanthine guanine phos-phoribosyl transferase (HPRT), and �-actin were performed as described byLernbecher et al. (78). cDNA was reverse-transcribed from 2 �g of totalRNA with oligo-dT primers (Pharmacia), AMV-reverse transcriptase, andTaq-polymerase (Promega). When RT-PCR was performed from PBMC, Taq-polymerase was replaced by PFU-polymerase (Stratagene, Heidelberg, Ger-many). Amplification was performed using the following primer pairs: �-globintransgene forward 5-TTTCTGATAGGCAGCCTGCACTGGT-3, �-globin trans-gene reverse 5-CATAGTTGTGTTCAGATCGATCTGG-3; p65-forward-1 5-GCTACAAGTGCGAGGGGC-3, p65-reverse-1 5-GGGGTCTGCGTAGGGAGGG-3;p65-reverse-2 5-GGCCTGCCTGATGGGTCCC-3; actin-forward 5-AGAGGTATCCTGACCCTGAAGTACC-3, actin-reverse 5-CCACCAGACAACACTGTGTTGGCAT-3; HPRT-forward-1 5-GGCGTCGTGATTAGTGATGATGAACC-3,HPRT-reverse 5-CTTGCGACCTTGACCATCTTTGGA-3 The following condi-tions were used: 1� [95°C, 360 s]; 40� [94°C, 40 s; 65°C, 60 s.; 72°C, 120 s]; 1�[72°C, 600 s] for �-globin and actin, 1� [95°C, 360 s]; 35� [94°C, 40 s; 68°C,60 s; 72°C, 120 s]; 1� [72°C, 600 s] for NF-�Bp65, 1� [95°C, 360 s]; 1� [94°C,60 s; 52°C, 120 s; 72°C, 120 s]; 1� [94°C, 60 s; 58°C, 60 s; 72°C, 120 s]; 32�[94°C, 40 s; 62°C, 60 s; 72°C, 60 s]; 1� [72°C, 600 s] for HPRT and separatedonto 2% agarose gels. Verification of PCR products was performed byreamplification with different primers recognizing sequences internal to theamplification products and by Southern blot analysis. Reactions that lackedtemplate RNA or AMV-reverse transcriptase served as internal controls.EMSA. Nuclear proteins were harvested as described elsewhere (51,52,61,80–84) and assayed for transcription factor binding activity using the NF-�Bconsensus sequence 5-AGTTGAGGGGACTTTCCCAGGC-3. Specificity ofbinding was ascertained by competition with a 160-fold molar excess ofunlabeled consensus oligonucleotides and by supershift experiments (datanot shown).Immunoblot (Western blot) analysis. Cytoplasmic and nuclear fractionswere prepared as previously described in detail (51,52,61,84). Twenty micro-grams of cytoplasmic extracts or 10 �g of nuclear extracts was separated onto10–12% SDS-PAGE, followed by electroblotting to ECL-nitrocellulose mem-branes. Membranes were incubated with primary antibodies for NF-�Bp65,I�B�, and I�B� (1:100) for 60 min at rt. After washing (2 � 7 min inTris-buffered saline [TBS], 0.05% Tween), the secondary antibody (horseradishperoxidase–coupled rabbit IgG, 1:2000) was added, and incubation wascontinued for 30 min at rt. Membranes were washed 3� 5 min as abovefollowed by a last 5-min wash in TBS. Immunoreactive proteins were detectedwith the ECL-Western blot system and subsequent autoradiography for 2 min.In vitro phosphorylation of I�B�. For in vitro phosphorylation, 5 � 106

bovine aortic endothelial cells (BAECs) were incubated in the presence of 500nmol/l AGE-albumin for the times indicated in the figure legends. At the endof incubation, cells were washed 3 times in 0.9% NaCl, harvested by scraping,collected by a 5-min centrifugation at 500g, 4°C, and resuspended in 1 ml of0.9% NaCl. Preparations were counted in a Neugebauer chamber and adjustedfor the same number of cells, collected by centrifugation as above, andresuspended in 500 �l of 0.9% NaCl. Whole-cell lysate was obtained bymechanical disruption with an Ultrathurrax (Wheaton, Milleville, NJ), clearedby centrifugation at 10,000g at 4°C for 5 min, and immediately supplementedwith 5 �l of 100 mmol/l PMSF (80). In vitro phosphorylation of a 70-kDaI�B�-GST fusion protein (Santa Cruz) was performed for 30 min at 37°C in 35mmol/l Tris (pH 7.5), 15 mmol/l MgCl2, 1 mmol/l MnCl2, 2 �l of 100 mmol/lrATP, and 1 �l of �-32P-ATP (10 �Ci) with 15 �l of the above whole-cell extract(85). The reaction was completed by the addition of 180 �l of IP buffer (20mmol/l Tris [pH 7.5], 150 mmol/l NaCl, 5 mmol/l EDTA, 0.2% SDS, 1%Triton-X100, 0.1 mmol/l PMSF, 1 unit/ml aprotinin) supplemented with 1mmol/l sodium orthovanadate (85). After addition of 2 �l of I�B� (FL)antibody and 10 �l of protein A/G Sepharose, incubation was continued for16 h at 4°C. At the end of incubation, precipitates were collected bycentrifugation at 10,000g for 10 min at 4°C; washed twice in IP buffer followedby a final wash in 150 mmol/l NaCl, 50 mmol/l Tris-HCl (pH 7.5), and 5 mmol/lEDTA; and then resuspended in 40 �l of denaturing buffer (125 mmol/lTris-HCl [pH 6.8], 4% SDS, 20% glycerol, 10% 2-mercaptoethanol) (86). Theimmunoprecipitate was analyzed by SDS-PAGE and autoradiography (85,86).

SUSTAINED NF-�B ACTIVATION


Intrinsic labeling and immunoprecipitation of NF-�Bp65. BAEC in35-mm2 dishes that had been confluent for 3 days were stimulated withAGE-albumin (500 nmol/l) for the times given in the figure legends. Four hoursbefore harvest, the medium was changed to methionine-free medium for 1 h.Thereafter, the medium was changed again to DMEM supplemented with7 �Ci/ml L-[35S]methionine. Where indicated, BAECs were cultivated in thepresence of 0.1 �mol/l antisense-p65 oligonucleotides, added 24 h beforeharvest. At the end of incubation, BAECs were lysed in 400 �l of IP buffer (20mmol/l Tris [pH 7.5], 150 mmol/l NaCl, 5 mmol/l EDTA, 0.2% SDS, 1%Triton-X100, 0.1 mmol/l PMSF, and 1 unit/ml aprotinin) for 20 min at 4°C,before lysates were cleared by centrifugation (10,000g, 10 min, 4°C) (86). Theprotein concentration in the supernatant was determined according to Brad-ford. Total protein lysate (200 �g) was precleared by adding 1 �g of anirrelevant monoclonal mouse antibody (c-jun KM; Santa Cruz) and 20 �l ofprotein A/G PLUS-agarose. The reaction was incubated at 4°C for 30 min andcentrifuged for 5 min at 10,000g, 4°C. The supernatant was incubated with 2�g of a polyclonal anti-p65 antibody (Santa Cruz) for 1 h at 4°C before 20 �lagarose conjugate was added and incubation was continued for 16 h at 4°C. Atthe end of incubation, precipitates were collected by centrifugation at 10,000g

for 10 min at 4°C; washed three times in IP buffer, followed by a last wash in150 mmol/l NaCl, 50 mmol/l Tris-HCl (pH 7.5), and 5 mmol/l EDTA, and thenresuspended in 40 �l of denaturing buffer (125 mmol/l Tris-HCl [pH 6.8], 4%SDS, 20% glycerol, 10% �-mercaptoethanol) (86). Four microliters werecounted in a �-counter to determine incorporation of L-[35S]methionine intoeach sample. Determination was repeated twice. The remaining 32 �l wassubjected to SDS-gel electrophoresis and subsequent autoradiography toconfirm the incorporation data.Tissue culture and transient transfection experiments. BAEC werecultured as previously described (83,87). Bovine arterial smooth muscle cells(SMCs) prepared from bovine aorta were cultured in DMEM containing 10%FCS (88). hT neurons (Stratagene), derived from Ntera/D1(NT2) precursorcells (Stratagene), were cultured according to the manufacturer’s instructions.THP-1 monocytic cells were cultured in RPMI 1648 medium supplementedwith 10% FCS. BAECs and SMCs were cultured to confluence and kept so for3 days in the respective medium containing 10% FCS before the experimentswere started. Experiments with hT-neurons were started 1 day after the cellshad reached confluence. THP-1 cells were seeded 24 h before the test proteinwas added. All experiments were performed without serum deprivationbefore stimulation.

For transfection experiments, BAECs growing in the logarithmic phasewere transfected with an NF-�B–driven luciferase reporter gene (Clontech)using the Effectene Transfection Reagent (Qiagen) according to the manufac-turer’s instructions by adding 0.4 �g of NF-�B DNA for each well at aconcentration of 0.2 �g/ml of culture medium. After applying the DNA to thecells, cells were cultivated for 1 h before stimulation by AGE-albumin (500nmol/l) or A� (1 �mol/l) was performed for 48 h. To correct for efficiency intransfection, we included 0.05 �g of pSV-�-Gal plasmid/ml medium (83). Theratio of luciferase activity to �-gal activity served to normalize luciferaseactivity (83). Each experiment was performed in triplicate.Antisense-phosphorothioate oligodeoxynucleotides. Antisense-RAGE-phosphorothioate oligodeoxynucleotides ([PS]ODNs) (5-G-[PTO]-ACCACTGCCCCTGCTGC-3); sense-RAGE-[PS]ODNs-(5-G-[PTO]-CAGCAGGGGCAGTGGTC-3), antisense-NF-�Bp65-[PS]ODNs (5-G-[PTO]-GGGAACAG-[PTO]-TTCGTCCATGG-[PTO]-C-3), scrambled-antisense-NF-�Bp65-[PS]ODNs (5-A-[PTO]-GTCTGGAG-[PTO]-CCATCGAGGTG-[PTO]-C-3), and sense-NF-�Bp65-[PS]ODNs (5-G-[PTO]-CCATGGACGAA-[PTO]-CTGTTCCC-[PTO]-C-3) derivedfrom the published DNA sequences for bovine RAGE and human NF-�Bp65were synthesized and purified by HPLC (MWG-Biotech, Ebersberg, Germany).[PS]ODNs were chemically modified by introducing phosphorothioate link-ages in which a nonbridging phosphate oxygen atom was substituted with asulfur atom to protect oligonucleotides from serum- and nuclease-mediateddegradation, as previously described in detail (52). Uptake of ODNs by thecells occurred passively without any further pretreatment of the cells (52).Statistical analysis. All values are given as mean, with the bars showingstandard deviations. The means of groups were compared by analysis ofvariance using the Student’s t test to correct for multiple comparisons. P �0.05 was considered to be statistically significant.

RESULTS

Sustained NF-�B activation in vivo. Hyperglycemiapromotes leukocyte adhesion to the endothelium at leastin part through NF-�B–dependent upregulation of adhe-sion molecules (1,6,89,90). This implicates a hyperglyce-mia-associated continuous activation of NF-�B in bothleukocytes and endothelial cells. Consistently, the pres-

ence of activated NF-�Bp65 in PBMCs isolated fromdiabetic patients has been reported to depend on the stateof glycemic control (80). When PBMCs were isolated fromnewly manifested diabetic patients with poor glycemiccontrol (HbA1c �13), almost all PBMCs were recognizedby a monoclonal antibody that exclusively binds to acti-vated NF-�Bp65 (Fig. 1A, left) (13). Activated NF-�Bp65was localized within the cells and more pronounced in thenuclear region (Fig. 1A, left, insert), thereby indicating thatmost of the activated NF-�Bp65 was present in the nu-cleus. In contrast, minor staining in much fewer cells,indicating only a little activated NF-�Bp65, was observedin healthy control subjects (Fig. 1A, right). In these cells,staining for activated NF-�Bp65 was located outside thenucleus in the cytoplasm, close to the outer cell membrane(Fig. 1A, right, insert). Repetitive sampling of PBMC fromthree patients with poor metabolic control also showedactivated NF-�Bp65 in almost all cells over 3 days (Fig.1C). In contrast, no difference was found when PBMCs ofthe above diabetic patients and healthy control subjectswere analyzed for full-length I�B� (Fig. 1B and C). Anaverage of 86% PBMCs derived from diabetic patients werepositive for I�B�, similar to the 87% found in healthycontrol subjects (Fig. 1C). Consequently, the sustainedNF-�Bp65 activation observed in PBMCs of diabetic pa-tients (Fig. 1A and C) cannot be explained by a loss ofI�B�. Therefore, pathways of sustained NF-�B activationmust be present in vivo, overriding the I�B�-dependentautoregulatory loop, and allow activation of NF-�Bp65 inthe presence of substantial amounts of I�B�. This hypoth-esis is consistent with an upregulation of NF-�Bp65 mRNAin PBMCs of diabetic patients: When total RNA wasisolated from blood samples at the same time PBMCs wereprepared and subjected to RT-PCR for NF-�Bp65, NF-�Bp65 mRNA was markedly increased in PBMC of patientswith diabetes compared with PBMCs of healthy controlsubjects (Fig. 1D). RT-PCR for HPRT served as an internalcontrol and confirmed comparable RNA input in eachreaction (Fig. 1D).

Endothelial cells cannot be obtained from patients withdiabetes. To define whether a comparable NF-�B activa-tion associated with severe diabetes can also be observedin endothelial cells, we studied NF-�B activation in kid-neys of diabetic BB/OK rats (79), compared to kidneys ofage-matched euglycemic Sprague Dawley-rats. The BB/OKrats manifested diabetes at the age of 104 16 days, andthe duration of diabetes was 57 9 days, sufficient toallow significant AGE formation (91,92). A total of ninediabetic kidneys and five control kidneys were subjectedto immunocytochemistry, with identical results (Fig. 2).Using a polyclonal rabbit-derived antibody recognizingNF-�Bp65, antigen-pronounced cytoplasmic and nuclearimmunoreactive NF-�Bp65 was observed in the renalvascular endothelium of diabetic BB/OK rats (Fig. 2A, left,arrowhead), whereas vessels in nondiabetic control kid-neys demonstrated only minor staining for NF-�Bp65 (Fig.2A, right, arrowhead). When immunocytochemistry wasrepeated on consecutive sections of diabetic kidneys usingthe monoclonal antibody for activated NF-�Bp65, pro-nounced staining was found in large blood vessel endothe-lia and close to small blood vessels (Fig. 2B). In almost allof the endothelial cells, NF-�Bp65 staining was visible in



FIG. 1. Immunocytochemistry of NF-�Bp65 and I�B� in circulating cells of patients with diabetes and in healthy control subjects. A: Immunocytochemistry for

activated NF-�Bp65 in PBMCs of a patient with newly manifested type 1 diabetes (left) and in a healthy control subject (right) (magnification: �40). Positivity

for NF-�Bp65 is indicated by the dark brown color. Higher magnification (�200) of single cells demonstrated nuclear localization of NF-�Bp65 in PBMCs derived

from diabetic patients (left insert, arrow), whereas staining for NF-�Bp65 in control PBMCs is restricted to cytoplasmic areas close to the cell wall (right insert,

arrow). B: Immunocytochemistry for full-length I�B� in PBMCs of a patient with newly manifested type 1 diabetes (left) and a healthy control subject (right).

Positivity for I�B� is indicated by the dark brown color. C: Compared with healthy control subjects (n � 6), staining for activated NF-�Bp65 was significantly

increased in PBMCs of patients with diabetes (n � 6; P � 0.0006 at day 1, diabetes versus control). In contrast, there was no significant difference in I�B�expression (P � 0.846 at day 1, diabetes versus control). NF-�Bp65 and I�B� levels did not change within 3 days. D: PBMCs were isolated from four healthy

volunteers and four patients with diabetes; total RNA was prepared, and 1 �g of each preparation was analyzed by RT-PCR with primers specific for NF-�Bp65

(top) and HPRT (bottom), respectively. Representative RT-PCR results from two diabetic patients (lanes 1 and 2) and two control subjects (lanes 3 and

4) are shown at the left. The signal intensity of the NF-�Bp65 representing band and the HPRT band was quantified by densitometry, and the ratio of

NF-�Bp65 to HPRT was calculated for each patient (n � 4) and each control subject (n � 4). The results are summarized at the right side.



the cytoplasm and within the nuclear region (Fig. 2B,arrowheads). If a self-regulated, transient, protein synthe-sis–independent NF-�B activation were the only pathwaypresent in vivo, however, one would not expect all endo-thelial cells (Fig. 2B) or all PBMCs (Fig. 1A and D) to reactpositively. This implicates that protein synthesis–depen-dent pathways of sustained NF-�B activation may exist.RAGE-mediated sustained activation of NF-�B. Incu-bation of BAECs with tumor necrosis factor-� (TNF-�; 1nmol/l) induces only a transient activation of NF-�B,lasting �1 h (83 and data not shown), whereas exposure ofBAECs to AGE-albumin (Fig. 3A) (or in vitro synthesizedcarboxymethyllysine-modified albumin [see below]) or A�peptide (Fig. 3B) induced an NF-�B activation that lastedfor �6 days. Similar prolonged activation of NF-�B wasobserved with several different AGE-albumin prepara-tions, in which 36% of lysine residues were modified asdescribed in Table 1. Incubation with similar molar con-centrations of control albumin (Fig. 3C), heat-inactivated

AGE-albumin (data not shown), or a scrambled version ofA� (Fig. 3D), which does not bind to RAGE (29), did notresult in activation of NF-�B, excluding a general proteinoverload (93) or a nonspecific membrane effect as thecause for the observed persistent NF-�B activation. Fur-thermore, incubation of BAECs with AGE-albumin (or A�)for 1 day followed by washing and removal of the ligandsresulted in NF-�B activation still present 3 days later (datanot shown). Therefore, a transient ligand-receptor interac-tion rather than restimulation or carryover of an excess ofligands triggers sustained NF-�B activation. ProlongedNF-�B activation was not restricted to endothelial cellsbut was also observed in neuronal cells (Fig. 3E), SMCs(Fig. 3F), and THP-1 monocytic cells (Fig. 3G). In allexperiments, NF-�B binding activity was specifically com-peted by a 160-fold molar excess of the unlabeled NF-�Boligonucleotides (see Fig. 3, last lane in each gel) but notby nonspecific or mutated NF-�B oligonucleotides (datanot shown).

FIG. 2. Immunocytochemistry of NF-�Bp65 in resting cells. A: Nine diabetic kidneys and five control kidneys were examined by immunocyto-chemistry with a rabbit-derived polyclonal antibody for NF-�Bp65, which recognizes both the activated and the inactive protein. Selectedrepresentative stainings are shown. Blood vessel endothelia of diabetic kidneys demonstrated pronounced immunoreactivity for NF-�Bp65 (left),whereas blood vessel endothelia of control kidneys showed no NF-�Bp65 staining (right). Positivity for NF-�Bp65 is indicated by the brown color.B: Immunocytochemistry of the above diabetic kidneys using the monoclonal antibody for activated NF-�Bp65, demonstrating nucleartranslocation of activated NF-�Bp65 in the majority of endothelial cells (arrowheads). Positivity for NF-�Bp65 is indicated by the dark browncolor.



AGE-albumin–induced NF-�B binding activity (Fig. 4A,lane 2) was suppressed by the addition of an excess ofsRAGE (Fig. 4A, lane 3), a truncated form of the receptor(27,29,31,32,45,53,70), and thereby implicated that engage-ment of RAGE is central for activation of NF-�B. However,sRAGE might also act by inhibition of RAGE dimerizationor by competing for ligand binding to not-yet-identifiednon-RAGE receptors. To prove directly the involvement ofRAGE in AGE-albumin–induced sustained NF-�B activa-tion, BAECs were preincubated with antisense oligonucle-otides directed against RAGE, previously shown tosuppress RAGE expression (52). Sustained NF-�B activa-tion was reduced (but not entirely blocked) in the pres-ence of antisense RAGE oligonucleotides (Fig. 4A, lane 4),whereas oligonucleotides in the sense orientation had noeffect (data not shown) and thus confirmed the involve-ment of RAGE. Because not all cells uniformly internalizedthe oligonucleotides, the effect of antisense RAGE wasonly partial, whereas sRAGE and neutralizing anti-RAGEantibodies (data not shown) blocked all cell surfaceinteractions with RAGE.

To prove that the effects observed with AGEs synthe-sized in vitro can be compared with natural RAGE ligandsthat occur in vivo, we isolated CML-modified proteins(36–40,44,70) from erythrocytes from the above patientswith diabetes, as previously described, and assayed forNF-�B activating activity (70). Consistent with CML beingcurrently recognized as a major AGE rapidly formedwithin a short time in the presence of oxidative andcarbonyl stress (36–40), CML-modified proteins were al-ready present at diabetes onset and could be isolated. Acomparable induction of NF-�B binding activity was de-tected when BAECs were stimulated for 5 days either withdifferent ligands, such as A� (1 �mol/l; Fig. 3B) and S100B(400 nmol/l; Fig. 4B), or with 400 nmol/l CML-modifiedproteins (Fig. 4C; lanes 2, 4, and 6) as in vitro synthesized,highly modified CML-albumin (36) (Fig. 4C, lane 2), 400nmol/l patient-derived CML-modified proteins (Fig. 4C,lane 4), or 400 nmol/l in vitro synthesized minimallymodified CML-albumin (36) (Fig. 4C, lane 6). In contrast,CML-modified hippuric acid (Fig. 4C, lane 8) did not resultin significant induction of NF-�B binding activity. Consis-

FIG. 3. Ligands of RAGE induce sustained NF-�B DNA binding activity. A–D: BAECs were left untreated (0 h) or were stimulated withAGE-albumin (500 nmol/l; A) or A� peptide (1 �mol/l; 1–40; B), unmodified control-albumin (500 nmol/l; C), or scrambled A� peptide (1 �mol/l;40–1; D) for 12 h to 6 days. Nuclear extracts were prepared as described (see RESEARCH DESIGN AND METHODS) and assayed for NF-�B binding activity,monitored in EMSA. Radioactive labeled oligonucleotides, spanning the consensus NF-�B recognition motif, were incubated with equal proteinamounts of nuclear extracts, and complexes were separated onto nondenaturing 5% PAA gels. For confirming NF-�B binding, binding observedafter 72 h was competed with a 160-fold molar excess of cold consensus NF-�B oligonucleotides (cons.). The position of NF-�B is indicated by anarrow. E–G: Nuclear extract from hNT-neurons (E), SMC (F), or THP-1 cells (G), harvested before (0 h) or 6 days after AGE-albumin stimulation(500 nmol/l) (6 days) were subjected to NF-�B–specific EMSA. NF-�B binding was confirmed by competing the shift observed after 6 days witha 160-fold molar excess of cold consensus NF-�B oligonucleotides (cons.). The position of NF-�B is indicated by an arrow.



tent with a recent report (44), CML-mediated NF-�Bactivation was dependent on RAGE, because blockade ofRAGE with a threefold molar excess of sRAGE resulted inreduced NF-�B binding activity (Fig. 4C, lanes 3, 5, 7, and9).

The observations in cell culture led us to predict thatAGE/RAGE-dependent NF-�B activation would also occurin vivo. Mice transgenic for an NF-�B–controlled �-globintransgene (78) were used. Under physiological conditions,constitutive expression of the transgene is restricted tolymphoid tissues, in which its expression is driven byconstitutively present NF-�B(p50/c-Rel) heterodimers(78). Activation of the p50/p65 heterodimer, however, canconfer inducible transgene activation in all cell types (78).These mice received a single administration of 1,000 �g ofAGE-albumin (500 �g i.v., 500 �g i.p.) or 1,000 �g ofcontrol albumin (500 �g i.v., 500 �g i.p.). Tissue washarvested 6 days later. The AGE concentration used was inthe range of AGE serum levels determined in diabeticpatients with end-stage renal disease (94–98). As a resultof degradation and urinary AGE clearance (25,96–98),AGE concentrations present in these mice after 6 daysprobably gives rise to mouse serum levels even below AGEserum levels found in diabetic patients with good glycemiccontrol (40,94–96). Total RNA was isolated from kidneysand subjected to RT-PCR for the �-globin reporter gene(Fig. 5A). RNA isolated from the spleen served as controland demonstrated constitutive �-globin transcription in allanimals (data not shown). Despite the presence of endog-enous AGE-degrading systems (25,97,98), an AGE-depen-dent induction of the NF-�B–driven �-globin transcriptionwas still observed after 6 days (Fig. 5A, lane 2) andparalleled by an increase in NF-�Bp65 mRNA (data notshown). No signal could be seen in mice that were infusedwith control albumin (Fig. 5A, lane 1) or in untreatedcontrol animals (data not shown). The strong �-globintransgene signal in mice that were treated with AGE-albumin (Fig. 5A, lane 2) was reduced when mice weretreated with anti-RAGE IgGs (40 �g/mouse i.v; Fig. 5A,lane 3) or an excess of sRAGE (25 �g/mouse i.v.; Fig. 5A,lane 4) at days 0 and 3 after stimulation. RT-PCR for actinserved as an internal control and confirmed comparableRNA input in each reaction (Fig. 5A, bottom). In-parallel–performed EMSA with whole-cell extracts from thesekidneys demonstrated prominent NF-�B binding activity

only in mice that had received AGE-albumin (Fig. 5B, lane

2). NF-�B binding activity was significantly reduced in thekidney nuclear extracts of mice that had received anti-RAGE IgGs (40 �g/mouse i.v.; Fig. 5B, lane 3) or sRAGE(25 �g/mouse i.v.) (Fig. 5B, lane 4). Inhibition of �-globinexpression (Fig. 5A) and NF-�B binding activity (Fig. 5B)by RAGE antibodies, however, was lower than the inhibi-tion observed in the presence of sRAGE. Because sRAGEcompetes for AGE binding with all cellular AGE receptors,whereas blocking RAGE antibodies specifically inhibitligation to RAGE, this implicates that other AGE receptors(25,43,99) or not-yet-identified RAGE-like receptors mightalso contribute to AGE-mediated sustained NF-�B, thoughto a much lesser extent than RAGE. RAGE antibodiesinhibit �80% of the AGE-dependent NF-�B activation,thereby indicating that RAGE is the predominant receptorin mediating sustained NF-�B activation.AGE- and A�-induced sustained NF-�B activation

mediating sustained gene expression. The functional

FIG. 4. AGE-mediated induction of NF-�B DNA binding activity in vitrois dependent on RAGE. A: BAECs were left untreated (lane 1) or wereincubated for 6 days with AGE-albumin (500 nmol/l) (lane 2) or withAGE-albumin (500 nmol/l) in the presence of either a 10-fold molarexcess of sRAGE (lane 3) or 0.1 �mol/l antisense RAGE oligonucleo-tides (lane 4) before NF-�B binding activity was monitored in EMSA.For confirming NF-�B binding, binding observed at day 6 was competedwith a 160-fold molar excess of unlabeled NF-�B consensus oligonucle-otides (lane 5). The result shown is representative of three indepen-dent experiments. B: BAECs were incubated for 5 days with S100B(400 nmol/l) in the absence (lane 1) or presence (lane 2) of a threefoldmolar excess of sRAGE, and NF-�B binding activity was monitored inEMSA as above. C: BAECs were left untreated (lane 1) or incubatedfor 5 days with 400 nmol/l highly modified CML-albumin (lane 2),CML-modified proteins isolated from erythrocytes of a patient withdiabetes and poor glycemic control (HbA1c 14.3%; lane 4), minimallymodified CML-albumin (lane 6), or CML-modified hippuric acid (lane

8) in the absence (lanes 1, 2, 4, 6, 8) or presence (lanes 3, 5, 7, 9) of athreefold molar excess of sRAGE. NF-�B binding activity was moni-tored in EMSA. The NF-�B signal intensity was quantified by densi-tometry and is given below in the autoradiogram.

TABLE 1Characterization of AGE-albumin used throughout the experi-ments

Amino acidQuantity

(pmol/mg)

Unmodified lysine 688.360Fructoselysine 181.470Carboxymethyllysine 204.320Pyrraline 190Pentosidine 6Total 1,074.346

Lysine derivatization:

[sum of lysine derivates]� 100% � 35.9%

[sum of lysine derivates] [unmocified lysine]



significance of RAGE-dependent continuous NF-�B activa-tion was demonstrated in transient transfection studies inwhich BAECs were transfected with a luciferase expres-sion plasmid controlled by four tandem copies of theNF-�B consensus sequence. Luciferase expression inBAECs stimulated with unmodified control albuminserved as control and was regarded as baseline expression(Fig. 6A, column 1). Whereas longer incubation withnonglycated control albumin (500 nmol/l) did not changegene expression (Fig. 6A, column 2), stimulation withAGE-albumin (500 nmol/l) induced NF-�B–dependentgene expression over time, increasing to the highest levels48 h after stimulation (Fig. 6A, columns 3 and 4). A similar

increase in luciferase expression was observed in cellsstimulated with A� (1 �mol/l) but not in BAECs incubatedwith scrambled A� (1 �mol/l) (Fig. 6A, columns 5–8).Because of the experimental limitations of transient trans-fection studies, expression of the luciferase reporter genecould not be followed for �48 h. However, immunoblotanalysis for the NF-�B–controlled HO-1 (100) demon-strated that HO-1 expression remained elevated over 5days (Fig. 6B). A similar increase was observed with otherNF-�B–regulated genes, such as the procoagulant tissuefactor (101 and data not shown) or the NF-�B–autoregu-lated inhibitor I�B� (see below).

FIG. 5. AGE-mediated induction of NF-�B DNA binding activity in vivois dependent on RAGE. A: Mice transgenic for an NF-�B–controlled�-globin reporter gene (78) were infused once with 1,000 �g ofAGE-albumin (500 �g i.v.; 500 �g i.p.) or control albumin (500 �g i.v.;500 �g i.p.). At time points 0 and 3 days, mice were treated with RAGEIgGs (40 �g/mouse i.v.) or sRAGE (25 �g/mouse i.v.). After 6 days, micewere killed and total RNA from the kidney was prepared and 2 �g wereanalyzed by RT-PCR for �-globin transgene and �-actin expression(�-globin transgene [top, 440 bp] and �-actin [bottom, 720 bp]). B:Nuclear extracts were prepared from the other kidney of the above�-globin transgenic mice. Ten micrograms of each nuclear extract wereanalyzed for NF-�B (p50/p65) binding activity in EMSA using a 32P-radiolabeled oligonucleotide with the high-affinity NF-�B(p50/p65)consensus sequence. The inducible NF-�B complex is indicated by anarrow. For confirming NF-�B–binding, kidney nuclear extracts fromAGE-treated transgenic mice was competed with a 160-fold molarexcess of unlabeled NF-�B consensus oligonucleotides (lane 5).

FIG. 6. Sustained activation of NF-�B by AGE-albumin and A� inducesNF-�B–dependent gene expression. A: BAECs were transiently trans-fected with the plasmid NF-�B-Luc, which contains four tandem copiesof the NF-�B consensus sequence fused to a TATA-like promoterregion from the herpes simplex virus thymidine kinase promoter usingthe Effectene transfection reagent. One hour after application of DNA,cells were stimulated with control albumin (500 nmol/l), AGE-albumin(500 nmol/l), scrambled A� (1 �mol/l), or A� (1 �mol/l) for either 6 or48 h. After harvest, luciferase activity was determined in the celllysates and normalized for transfection efficiency by the amount of�-gal activity expressed by the co-transfected control plasmid pSV-�-Gal (83). Corrected values were expressed as relative luciferase (LUC)units and are given as percentage of luciferase expression determinedin BAECs, stimulated with control albumin for 6 h. Three independentexperiments were performed six times with identical results. Theresults are given � SD. Statistical analysis was performed using atwo-tailed Student’s t test. B: Immunoblot analysis of HO-1 in cyto-plasmic extracts from BAECs that were untreated (0 h) or treated withAGE-albumin (500 nmol/l) for the times indicated demonstrated time-dependent induction of HO-1. HO-1 (molecular weight 39 kDa) isindicated by an arrow. The result shown is representative of threeindependent experiments. The HO-1 signal intensity was quantified bydensitometry and is given below in the autoradiogram. Similar datawere obtained after stimulation with 1 �mol/l A� (data not shown).



AGE and A� induce increased nuclear translocation

of NF-�Bp65. To define the NF-�B proteins induced byAGE-albumin and A� in cultured endothelial cells, weperformed supershift experiments using antibodies spe-cific for NF-�Bp65 and NF-�Bp50, and the supershiftedbands were quantified by densitometry. In unstimulatedcells, 90% of the densitometrically measured binding ac-tivity was due to NF-�Bp50, whereas NF-�Bp65 contrib-uted only 10% (Fig. 7A). These amounts changed after 24 has a result of a relative increase of NF-�Bp65 compared toNF-�Bp50 (Fig. 7A and B). The increased contribution ofNF-�Bp65 to the DNA binding complex continued overtime: 3 days after AGE-albumin (Fig. 7A and B) or A�induction (data not shown), NF-�B anti-p50 and anti-p65both yielded prominent supershifts. At later time points,however, the portion of NF-�Bp50 decreased, whereasonly the contribution of NF-�Bp65 to the observed shiftfurther increased, reaching a maximum 5 days after theinitial stimulus and being still evident at day 6 (Fig. 7B).These data implicate an increased availability or stabilityof nuclear NF-�Bp65 at later time points.

Immunoblot analysis confirmed an increase in nuclearNF-�Bp65 after prolonged induction by AGE-albumin (Fig.8A). Despite persistent high nuclear NF-�Bp65 levels, wedid not observe a loss of NF-�Bp65 immunoreactivity inthe cytoplasm (Fig. 8B). The increase of NF-�Bp65 antigenobserved after 3 days in the cytoplasm (Fig. 8B) correlatedwith an increase in I�B� (Fig. 8C) and I�B� (Fig. 8D) andalso with the observation in diabetic patients (Fig. 1). Adecrease of cytoplasmic I�B� and I�B� was evident onlywithin the first 48 h of AGE-albumin induction (Fig. 8C and

FIG. 7. Relative contribution of NF-�Bp50 and NF-�Bp65 to the bindingcomplexes formed during AGE-albumin–mediated sustained NF-�Bbinding activity monitored by supershifting experiments. Character-ization of the NF-�B subunits, contributing to the observed shiftformed at the NF-�B consensus sequence, was performed by adding 2.5�g of an irrelevant IgG or of anti–NF-�Bp50 and anti–NF-�Bp65antibodies into the binding reactions with nuclear extracts of unstimu-lated BAECs or of BAECs stimulated with AGE-albumin for 24 h to 6days. Signals obtained were quantified by densitometry. A: Signalintensity determined for the NF-�B binding complex in the presence ofa nonspecific IgG was taken as 100% at each time point investigatedand compared with signal intensity for NF-�B binding determined inthe presence of anti-p50 or anti-p65 antibodies. Because of super-shifted complexes and blockade of NF-�B binding, the NF-�B bindingcomplex was significantly reduced in the presence of anti-p65 andanti-p50 antibodies. The extent of reduction served as a measure tocalculate the relative contribution of NF-�Bp65 and NF-�Bp50 to eachNF-�B binding complex. Two different experiments were evaluatedwith identical results, and the data of one representative experimentare shown. Similar data were obtained after stimulation with 1 �mol/lA� (data not shown). B: Ratio formed by the amount of NF-�Bp65divided by the amount of NF-�Bp50 contributing to the NF-�B bindingactivity observed over time.

FIG. 8. AGE-albumin induces sustained translocation of NF-�Bp65 attime points at which I�B proteins already reappeared in the cytoplasm.A and B: Immunoblot analysis of nuclear (A) and cytoplasmic (B)extracts of BAECs, stimulated with AGE-albumin for the times indi-cated, demonstrated AGE-albumin–dependent prolonged transloca-tion of NF-�Bp65 antigen into the nucleus (A). The NF-�Bp65–specificcomplex is indicated by an arrow. The NF-�B signal intensity wasquantified by densitometry and is given on the right side of eachautoradiogram. The result shown is representative of six independentexperiments. Similar data were obtained after stimulation with 1�mol/l A� (data not shown). C: Immunoblot analysis of I�B� incytoplasmic extracts from BAECs that were untreated (0 h) or treatedwith AGE-albumin (500 nmol/l) for the times indicated demonstrated adecrease in cytoplasmic I�B� antigen between 12 and 48 h. TheI�B�-specific complex is indicated by an arrow. The I�B� signalintensity was quantified by densitometry and is given on the right sideof the autoradiogram. The result shown is representative of fourindependent experiments. Similar data were obtained after stimula-tion with 1 �mol/l A� (data not shown). D: Immunoblot analysis ofI�B� in cytoplasmic extracts from BAECs that were untreated (0 h) ortreated with AGE-albumin (500 nmol/l) for the times indicated dem-onstrated a decrease of cytoplasmic I�B� antigen between 12 and 48 h.The I�B�-specific complex is indicated by an arrow. The I�B� signalintensity was quantified by densitometry and is given on the right sideof the autoradiogram. The result shown is representative of fourindependent experiments. Similar data were obtained after stimula-tion with 1 �mol/l A� (data not shown).



D). The reappearance of I�B� and I�B� and the simulta-neous increase in nuclear and cytoplasmic NF-�Bp65suggest that mechanisms such as de novo protein synthe-sis of NF-�B might be operative in the maintenance of

sustained NF-�B activity. Similar data were obtained afterstimulation with 1 �mol/l A� (data not shown).

Because degradation of I�B� also depends on its phos-phorylation state, recombinant I�B�-GST fusion proteinwas coincubated with cell extracts from control or AGE-albumin–treated BAECs in the presence of [�-32P]ATP.Increased phosphorylation of I�B� was evident between24 and 48 h (Fig. 9), demonstrating that AGE-albuminstimulates a kinase activity able to phosphorylate I�B�within the first 48 h of induction, which corresponds to theloss of I�B� during this time (Fig. 8C). Although thisimplicates that phosphorylation might tag I�B� for subse-quent degradation, it cannot be excluded that phosphory-lation might also have other effects.AGE- and A�-induced sustained NF-�B activation is

dependent on newly synthesized NF-�Bp65. Twelvehours after AGE-albumin stimulation of BAECs, RT-PCRanalysis demonstrated an increase in NF-�Bp65 mRNA(Fig. 10A, left), occurring parallel to the increase inNF-�Bp65 protein observed in immunoblot analysis (Fig.8A and B). This time span corresponds to the increasedNF-�Bp65 mRNA observed in diabetic patients (Fig. 1D).mRNA levels reached a maximum between 72 h and 5 daysand slightly decreased after 6 days (Fig. 10A, left). Controlalbumin did not induce NF-�Bp65 mRNA over time (Fig.10A, right). Intrinsic labeling of BAECs with L-[35S]methi-onine was performed to prove whether de novo proteinsynthesis of NF-�Bp65 also occurred in response to RAGE-

FIG. 9. AGE-albumin induces an I�B� phosphorylating kinase activity.BAECs were left untreated (0 h) or were stimulated with AGE-albumin(500 nmol/l) for 12 h to 6 days. Total cell extracts were prepared andincubated with recombinant I�B� for 30 min at 37°C in the presence of10 �Ci [�-32P]ATP. Subsequently, the reaction was immunoprecipi-tated with I�B�-specific antibodies and analyzed by SDS-PAGE andautoradiography.

FIG. 10. AGE-albumin induces accumulation of NF-�Bp65 mRNA and de novo synthesis of NF-�Bp65 antigen in endothelial cells. A: BAECs wereleft untreated (0 h) or were stimulated with AGE-albumin (500 nmol/l, left) or control albumin (500 nmol/l, right) for 12 h to 6 days. Total RNAwas prepared, and 2 �g of each preparation was analyzed by RT-PCR with primers specific for NF-�Bp65 (top, 680 bp) and �-actin (bottom, 720bp), respectively (for details, see RESEARCH DESIGN AND METHODS). The left lane (SM) shows the 564- to 947-bp fragment of a -DNA EcoRI-HindIIIdigest. The signals obtained in AGE-albumin–stimulated BAECs were analyzed by densitometry and are expressed as the ratio of NF-�Bp65 to�-actin. B: BAECs were left untreated (0 h) or were stimulated with AGE-albumin (500 nmol/l) for 24 h, 72 h, or 6 days in the absence (left) orpresence (right) of 0.1 �mol/l antisense p65 oligonucleotides. Four hours before harvest, the medium was changed to methionine-free mediumfor 1 h. Thereafter, medium was changed again to DMEM supplemented with 7 �Ci/ml L-[35S]methionine. At the end of incubation, intrinsic labeledcells were harvested, lysed, and subjected to NF-�Bp65–specific immunoprecipitation (for details, see RESEARCH DESIGN AND METHODS). Incorpora-tion of L-[35S]methionine into NF-�Bp65–precipitated material was determined in a �-counter. The experiment was repeated twice with similarresults.



mediated cell stimulation (Fig. 10B). Subsequent immuno-precipitation of the labeled proteins with antibodies toNF-�Bp65 demonstrated an AGE-albumin–induced in-crease in anti–NF-�Bp65–precipitated material, reaching amaximum 72 h after stimulation (Fig. 10B). Consistentwith the results in EMSA (Fig. 3), immunoblot (Fig. 8A andB), and RT-PCR (Fig. 10A), the amount of NF-�Bp65–precipitated material decreased at day 6 but was stillthreefold higher than in unstimulated control cells (Fig.10B, left). No significant induction of NF-�Bp65–precipi-tated material was observed (Fig. 10b, right) when BAECswere preincubated in the presence of antisense NF-�Bp65oligonucleotides before induction with AGE-albumin.Therefore, de novo NF-�Bp65 transcription and expres-sion seem to be an important step in maintaining sustainedNF-�B activation.

AGE-albumin–dependent NF-�B activation was reducedat time points later than 72 h but not at the earlier timepoints (EMSA) (Fig. 11) as a result of blocking mRNAsynthesis by pulsing BAECs for 3 h with actinomycin D (15�g/ml) before harvest. Pulsing for 3 h with cycloheximide(1 �g/�l) reduced NF-�B activation only after �3 days(data not shown), further confirming that sustained NF-�Bactivation is dependent, at least in part, on de novo mRNAsynthesis. However, these reagents uniformly suppressNF-�Bp65 and I�B� synthesis, whereas the latter has beendemonstrated to contribute to sustained NF-�B activation(64–67).

Therefore, we applied oligonucleotides directed againstNF-�Bp65 to prove directly the contribution of newlysynthesized NF-�Bp65 to the observed NF-�B bindingactivity. Applied 24 h before harvest, antisense NF-�Bp65oligonucleotides reduced AGE-albumin–dependent NF-�Bactivation at time points later than 24 h of AGE-albuminstimulation (EMSA) (Fig. 12A, top, lanes 4–8) but notearlier (Fig. 12A, top, lanes 2 and 3). Sense (EMSA) (Fig.12A, bottom) or scrambled oligonucleotides (data notshown) had no effect. Specificity of oligonucleotides wasconfirmed in immunoblot experiments, which demon-strated reduction of AGE-albumin–induced NF-�Bp65 ex-pression, but not of NF-�Bc-Rel expression in the

presence of antisense p65 oligonucleotides (data notshown).

Consistently, AGE-albumin–induced NF-�B binding ac-tivity and NF-�Bp65 mRNA were not affected by sense p65oligonucleotides (Fig. 12B, left, lane 2), whereas upregu-lation of NF-�B binding activity and NF-�Bp65 mRNA wassignificantly reduced in the presence of antisense p65oligonucleotides (Fig. 12B, left, lane 3). In contrast, NF-�Bbinding activity induced by 3 h of stimulation with 1 nmol/lTNF� was not reduced in the presence of antisense p65oligonucleotides (Fig. 12B, right, lane 6), and RT-PCRfrom BAECs stimulated with TNF� (1 nmol/l, 3 h) in thepresence of sense (Fig. 12B, right, lane 5) or antisense p65oligonucleotides (Fig. 12B, right, lane 6) failed to detectNF-�Bp65 mRNA (Fig. 12B, right, lane 5).

DISCUSSION

The NF-�B system triggers a self-terminating acute-phaseresponse in situations in which a rapid protein synthesis–independent activation of defense mechanisms is criticalfor survival (1–11). Recent studies, however, describe thepresence of activated NF-�B in injured arteries (102),atherosclerotic plaques (6,13), Crohn’s disease bowel tis-sue (61–63,103), and circulating mononuclear cells ofpatients with septicemia (82) and of diabetic patients witha new onset of diabetes or a long history of diabetes (Fig.1) (80,81). Under these conditions, NF-�B seems to beactivated in almost all cells, indicating the possibility thatNF-�B activation also occurs over longer time periods invivo. Here we show that sustained NF-�B activation in theabsence of significantly decreased I�B� expression isassociated with an increase in NF-�Bp65 mRNA in PBMCsof patients with type 1 diabetes (Fig. 1) and in the renalendothelium of diabetic rats (Fig. 2). In vitro studies showthat ligand-RAGE interaction results in sustained NF-�Bactivation dependent on de novo synthesis of NF-�Bp65 inall cell types tested.

To our knowledge, RAGE-mediated sustained NF-�Bactivation provides, for the first time, evidence that pro-longed NF-�B activation, which is dependent on newlysynthesized NF-�Bp65 mRNA and results in perpetuatedNF-�B–dependent gene expression, occurs as a cellularresponse to a nonviral stimulus. Recently, upregulation ofNF-�Bp65 mRNA was demonstrated in hypoxia-exposedoligodendrocytes (104); however, no detectable expres-sion of NF-�B target genes was observed in this model(104). Consistent with the in vitro data, the in vivoexperiments presented show that NF-�B–dependent �-glo-bin reporter gene expression is elevated for 6 days in micetreated with a single dose of AGE-albumin (Fig. 5). Fur-thermore, the animal model supports the hypothesis thatRAGE is central for sustained NF-�B activation in vivo(Fig. 5). This animal model does not, however, allow us todefine whether the observed increase in NF-�Bp65– andNF-�B–driven gene expression results from cells in whichNF-�B activation is sustained or from cells that have beenrecently activated by primary or secondary activationthrough AGE-albumin–induced cytokines or immune mod-ulators. Reactive oxygen species (ROS) produced by themitochondrial respiratory chain have been described asone major mediator of hyperglycemia-dependent NF-�Bactivation (16). Although monocytes do not contain mito-

FIG. 11. AGE-albumin induction of NF-�B activation at time pointslater than 2 days is dependent on mRNA. BAECs were stimulated withAGE-albumin (500 nmol/l) for the times indicated. Three hours beforeharvest, cells received a pulse of actinomycin D (15 �g/ml). Nuclearextracts were prepared as described and assayed for NF-�B bindingactivity in EMSA using radioactive labeled oligonucleotides, spanningthe consensus NF-�B recognition motif. For confirming NF-�B binding,the observed shift was competed with a 160-fold molar excess of coldconsensus NF-�B oligonucleotides (last lane). To exclude the possibil-ity that EtOH influenced NF-�B binding activity, BAECs were leftuntreated (0 h) or were stimulated with AGE-albumin (500 nmol/l) forthe times indicated in the presence of 0.4% EtOH (final concentrationsused in actinomycin D experiments; right). The position of NF-�B isindicated by an arrow.



FIG. 12. AGE-albumin induction of NF-�B activation at time points later than 2 days is dependent on NF-�Bp65 mRNA de novo synthesis. A:BAECs were stimulated with AGE-albumin (500 nmol/l) for the times indicated. Twenty-four hours before harvest, a pulse of 0.1 �mol/l antisense(top) or sense p65 oligonucleotides (bottom) was added to the medium. Uptake was in a passive manner, as previously described in detail (52).NF-�B binding activity was assayed in EMSA as above; the position of NF-�B is indicated by an arrow. The experiment was repeated twice, withsimilar results, and the densitometric analysis of both experiments is summarized below in the autoradiogram. The position of NF-�B is indicatedby an arrow. B: BAECs were stimulated with AGE-albumin (500 nmol/l) for 6 days (left) or TNF-� (1 nmol/l) for 3 h (right). Twenty-four hoursbefore harvest, a pulse of 0.1 �mol/l antisense or sense p65 oligonucleotides was added to the medium. From one-half of the cells, nuclear extractswere prepared and NF-�B binding activity was assayed in EMSA as above (top). Intensity of NF-�B binding activity was quantified bydensitometry (middle). The second half of cells were used for RNA isolation, and 2 �g of each preparation was subsequently subjected to cDNAsynthesis and RT-PCR for NF-�Bp65 and �-actin (bottom).



chondria, ligation to RAGE results in antioxidant-inhibit-able NF-�B activation (data not shown). Therefore,additional routes of ROS production also seem to beinvolved. A recent study demonstrated that upon AGEengagement of RAGE, NADPH-oxidase is one majorsource of ROS in endothelial cells (105) and thus impli-cates that multiple sources of oxidative stress contributeto RAGE-mediated NF-�B activation (43,105). Thiore-doxin, which is rapidly activated by oxidative stress, hasbeen demonstrated to activate NF-�B by c-Jun NH2-termi-nal kinase–mediated redox-dependent degradation ofI�B� in human pulmonary artery endothelial cells (106).Because oxidative stress also induces nuclear transloca-tion of thioredoxin, thioredoxin might further cooperatewith redox-factor 1 to facilitate the DNA binding of NF-�B,which requires reduced cysteines in the NF-�Bp65 subunit(106,107). In addition, ligands of RAGE might induceNF-�B coactivators, such as CBP/p300 or HMGY (108),that support and enhance sustained NF-�B activation.

One pathway of RAGE-dependent cellular perturbationincludes oxidative stress–mediated activation of p21ras

and subsequent activation of mitogen-activated proteinkinases, encoded by the extracellular signal-regulated ki-nases (ERK) 1 and 2 (43,109). A recent study in SMCsdemonstrated that ERK-1 and ERK-2 promote NF-�B ac-tivity independent of I�B� degradation, most probably bymodulating NF-�B binding to the DNA (110). Additionalstudies are required to elucidate whether ligands of RAGEalso modulate NF-�B DNA recognition. Besides ERK sig-nal transduction pathways, phosphorylation of I�B� attyrosine residue 42 has been demonstrated to mediateNF-�B translocation in the absence of I�B� degradation(111). Although preliminary data in our laboratory indicatethat AGE-albumin stimulation increases phosphotyrosinemodifications on I�B� and I�B� in vitro (A.B., P.P.N.,unpublished observations), we do not know whether thesemodifications also contribute to RAGE-mediated sustainedNF-�B activation. The time course of the AGE-inducedphosphorylating activity (Fig. 9) implicates that phosphor-ylation might tag I�B� for subsequent degradation. How-ever, we did not characterize the locus of AGE-dependentI�B� phosphorylation and thus do not know whetherphosphorylation occurs at serine residues, thereby initiat-ing subsequent I�B� degradation (4,7,9–11), or at tyrosineresidues, thereby promoting NF-�B translocation in theabsence of I�B� degradation (111). Future studies willaddress this issue. New synthesis of I�B� is evident at timepoints later than 3 days (Fig. 8D). Newly synthesized I�B�has been shown to complex with NF-�B without inhibitingits nuclear translocation or transactivation capacity (64–67). Therefore, it is likely that the newly generated I�B�prevents NF-�B from being trapped by newly synthesizedI�B� and thus might also contribute to RAGE-dependentsustained NF-�B activation. However, additional studiesare required to evaluate the role of I�B� in mediatingsustained NF-�B activation induced by ligands of RAGEand to define the extent of sustained NF-�B activation inthe absence of newly synthesized I�B�.

From the data presented here, we conclude that RAGE-dependent sustained NF-�B activation occurs in twophases. In the first phase, lasting up to 48 h, free NF-�Bp65protein translocates to the nucleus as a result of stimulus-

induced degradation of I�B� and I�B�. This phase isindependent of mRNA and protein synthesis, becauseneither actinomycin D nor cycloheximide or antisenseNF-�Bp65 oligonucleotides blocked NF-�B activation(Figs. 11 and 12). Because I�B� expression itself is con-trolled by NF-�B (9,11), the increase in nuclear NF-�Bp65results in elevated I�B� gene transcription and expres-sion, which replenishes the pool of cytoplasmic I�B� attime points later than 48 h (Fig. 8C). The second phase ofNF-�B activation occurs at time points later than 48 h andis associated with elevated NF-�Bp65 mRNA levels, asobserved in AGE-albumin–stimulated cells (Figs. 10 and12), AGE-albumin–treated mice (data not shown), anddiabetic patients (Fig. 1D). Inhibition of mRNA and/orprotein synthesis by actinomycin D, cycloheximide, orantisense NF-�Bp65 oligonucleotides results in loss ofsustained NF-�B activation (Figs. 11 and 12), therebyconfirming that new synthesis rather than stabilization ofNF-�Bp65 mRNA is critical for perpetuating and maintain-ing NF-�B activation. New synthesis of NF-�Bp65 resultsin a constantly growing pool of free NF-�Bp65 in vitro andin vivo, which provides newly generated transcriptionallyactive NF-�Bp65 that can override the NF-�B–dependentI�B�-autoregulatory loop. Thus, delineation of mecha-nisms underlying increased NF-�Bp65 transcription andexpression upon engagement of RAGE might be crucial tounderstanding sustained NF-�B activation. The hypothesisthat newly synthesized NF-�Bp65 plays an essential role insustained NF-�B activation is further supported by studiesof human cytomegalovirus infections (112), in which per-sistent NF-�B activation, critical to the viral life cycle, isalso mediated through induction of NF-�Bp65 mRNA (112).

The RAGE promoter itself contains two functionalNF-�B sites (113), and RAGE expression has been demon-strated to be induced in the presence of AGE-albumin(52,113). Therefore, it seems likely that the sustainedNF-�B activation described here results in elevated RAGEexpression and, in turn, upregulation of the receptorensures that sustained NF-�B activation is not only main-tained but also amplified. Therefore, the data suggest alink between acute activation of NF-�B observed in re-sponse to many stimuli, such as LPS and cytokines andsustained activation of NF-�B observed in vivo in chronicdisorders. The nature of such mechanisms may be criticalin settings where endogenous negative feedback path-ways, those responsible for returning cellular behavior tohomeostasis, are replaced by an upwardly spiraling cycleof cellular perturbation due, in part, to sustained NF-�Bp65 transcription and inappropriate NF-�B activation.However, complete and persistent NF-�B inhibition hasbeen linked directly to apoptosis (20–23), and sustainedNF-�B activation might represent a mechanism necessaryto ensure survival in the presence of an ongoing proinflam-matory challenge accompanying chronic disorders (16,24,114,115). Additional studies are required to define whetherthe sustained NF-�B activation described here has to beunderstood as supportive for complications or as a pro-tective mechanism in the course of diabetes.

ACKNOWLEDGMENTS

This work was supported in part by grants from theDeutsche Forschungsgemeinschaft (P.P.N., M.S., M.H.);



the Bundesministerium fur Forschung und Technik (B.W.,M.K.); the Fortune program (A.B., P.P.N.) and the IZKFprogram of the University of Tubingen (P.P.N.); StiftungVerum (P.P.N.); the University of Heidelberg (A.B.); andASTA-Medica (A.B., P.P.N.). P.P.N. performed this workduring the tenure of a Schilling professorship. A.M.S. andD.S. were supported by grants from the U.S. Public HealthService, the American Heart Association (New York affil-iate), and the Juvenile Diabetes Foundation.

Part of this work was presented at the 17th meeting ofthe International Society for Thrombosis and Hemostasis(ISTH), Florence, Italy, 10–16 June 1997; at the 42ndmeeting of the German Society for Endocrinology,Freiburg, Germany, 4–7 March 1998; and at the 33rdmeeting of the German Diabetes Society, Leipzig, Ger-many, 21–23 May 1998.

We thank Dr. T. Wirth (Wurzburg, Germany) for provid-ing the �-globin transgenic mice and Dr. I. Besenthal(Tubingen, Germany) for help in identifying patients withnewly manifested diabetes. TNF-� was a gift from Knoll(Ludwigshafen, Germany). Anti-RAGE antibodies werekindly provided by Dr. M.A. Shearman (Merck, Sharpe &Dome, Essex, U.K.). The authors also thank M. Kanitz(Heidelberg, Germany), S. Langer (Dresden, Germany),and A. Oehmichen (Dresden, Germany) for excellent tech-nical assistance and B. Andrassy for editorial help.

REFERENCES

1. Barnes PJ, Karin M: Nuclear Factor-�B—a pivotal transcription factor inchronic inflammatory diseases. N Engl J Med 336:1066–1071, 1997

2. May MJ, Ghosh S: Rel/NF-kappa B, and I kappa B proteins: an overview.Semin Cancer Biol 8:63–73, 1997

3. Baeuerle PA, Baltimore D: NF-�B: ten years after. Cell 87:13–20, 19964. Verma I, Stevenson JK, Schwarz EM, van Antwerp D, Miyamoto S:

Rel/NF-�B/I�B family: intimate tales of association and dissociation.Genes Dev 9:2723–2735, 1995

5. Baeuerle PA, Henkel T: Function and activation of NF-�B in the immunesystem. Annu Rev Immunol 12:141–179, 1994

6. Collins T: Endothelial nuclear factor-kappa B and the initiation of theatherosclerotic lesion. Lab Invest 68:499–508, 1993

7. Read MA, Whitley MZ, Williams AJ, Collins T: NF-kappa B, and I kappa Balpha: an inducible regulatory system in endothelial activation. J Exp Med

179:503–512, 19948. Mercurio F, Manning AM: NF-�B as a primary regulator of the stress

response. Oncogene 18:6163–6171, 19999. Beg AA, Baldwin AS: The I�B proteins: multifunctional regulators of

Rel/NF-�B transcription factors. Genes Dev 7:2064–2070, 199310. Alkalay I, Yaron A, Hatzubai A, Orian A, Chiechanover A, Ben-Neriah Y:

Stimulation-dependent I�B�: phosphorylation marks the NF-�B inhibitorfor degradation via the ubiquitin-proteasome pathway. Proc Natl Acad Sci

U S A 92:10599–10603, 199511. de-Martin R, Vanhove B, Cheng Q, Hofer E, Csizmadia V, Winkler H, Bach

FH: Cytokine-inducible expression in endothelial cells of an I kappa Balpha-like gene is regulated by NF kappa B. EMBO J 12:2773–2779, 1993

12. Ruef J, Peter K, Nordt TK, Runge MS, Kubler W, Bode C: Oxidative stressand atherosclerosis: its relationship to growth factors, thrombus forma-tion and therapeutic approaches. Thromb Haemost 82 (Suppl. 1):32–37,1999

13. Brand K, Page S, Rogler G, Bartsch A, Brandl R, Knuechel R, Page M,Kaltschmidt C, Baeuerle P, Neumeier D: Activated transcription factor-kappa B is present in the atherosclerotic lesion. J Clin Invest 97:1715–1722, 1996

14. Baynes JW: Role of oxidative stress in development of complications indiabetes. Diabetes 40:405–412, 1991

15. Giugliano D, Ceriello A, Paolisso G: Oxidative stress and diabetic vascularcomplications. Diabetes Care 19:257–267, 1996

16. Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, Kaneda Y,Yorek MA, Beebe D, Oates PJ, Hammes HP, Giardino I, Brownlee M:Normalizing mitochondrial superoxide production blocks three pathwaysof hyperglycaemic damage. Nature 404:787–789, 2000

17. Munch G, Schinzel R, Loske C, Wong A, Durany N, Li JJ, Vlassara H, SmithMA, Perry G, Riederer P: Alzheimer’s disease—synergistic effects ofglucose deficit, oxidative stress and advanced glycation endproducts.J Neural Transm 105:439–461, 1998

18. Yan SD, Yan SF, Chen X, Fu J, Chen M, Kuppusamy P, Smith MA, PerryG, Godman GC, Nawroth P, Zweier J, Stern D: Non-enzymatically glycatedtau in Alzheimer’s disease induces neuronal oxidant stress resulting incytokine gene expression and release of amyloid �-peptide. Nat Med

1:693–699, 199519. Nunomura A, Perry G, Pappola MA, Wade R, Hirai K, Chiba S, Smith MA:

RNA oxidation is a prominent feature of vulnerable neurons in Alzhei-mer’s disease. J Neurosci 19:1959–1964, 1999

20. Beg AA, Baltimore D: An essential role of NF-�B in preventing TNF�-induced cell death. Science 274:782–784, 1996

21. Badrichani AZ, Stroka DM, Bilbao G, Curiel DT, Bach FH, Ferran C: Bcl-2and Bcl-Xl serve an anti-inflammatory function in endothelial cellsthrough inhibition of NF-�B. J Clin Invest 103:543–553, 1999

22. Van Antwerp D, Martin S, Jafri T, Green D, Verma I: Suppression ofTNF�-induced apoptosis by NF-�B. Science 274:787–789, 1996

23. Bach FH, Hancock WW, Ferran C: Protective genes expressed in endo-thelial cell: a regulatory response to injury. Immunol Today 18:483–486,1997

24. Schmidt AM, Yan SD, Stern DM: The dark side of glucose. Nat Med

1:1002–1004, 199525. Vlassara H, Bucala R, Striker L: Pathogenetic effects of advanced glyco-

sylation: biochemical, biologic and clinical implications for diabetes andaging. Lab Invest 70:138–151, 1994

26. Bierhaus A, Hofmann MA, Ziegler R, Nawroth PP: The AGE/RAGEpathway in vascular disease and diabetes mellitus. Part I. The AGE-concept. Cardiovasc Res 37:586–600, 1998

27. Hofmann MA, Drury S, Fu C, Wu Q, Taguchi A, Lu Y, Avila C, KambhamN, Slattery T, Beach D, McClary J, Nagashima M, Morser J, Bierhaus A,Neurath M, Nawroth P, Stern D, Schmidt AM: RAGE mediates a novelproinflammatory axis: the cell surface receptor for S100/calgranulinpolypeptides. Cell 97:889–901, 1999

28. Yan SD, Zhu H, Zhu A, Golabek A, Du H, Rohrer A, Yu J, Soto C, SchmidtAM, Stern D, Kindy M: Receptor-dependent cell stress and amyloidaccumulation in systemic amyloidosis. Nat Med 6:643–651, 2000

29. Yan SD, Chen X, Fu J, Chen M, Zhu H, Roher A, Slattery T, Zhao L,Nagahima M, Morser J, Migheli A, Nawroth P, Stern D, Schmidt AM:RAGE and amyloid-� peptide neurotoxicity in Alzheimer’s disease. Na-

ture 382:685–691, 199630. Smith MA, Taneda S, Richey PL, Miyata S, Yan SD, Stern D, Sayre LM,

Monnier VM, Perry G: Advanced Maillard reaction end products areassociated with Alzheimer disease pathology. Proc Natl Acad Sci U S A

91:5710–5714, 199431. Wautier JL, Zoukourian C, Chappey O, Wautier MP, Guillausseau PJ, Cao

R, Hori O, Stern DM, Schmidt AM: Receptor-mediated endothelial celldysfunction in diabetic vasculopathy. Soluble receptor for advancedglycation end products blocks hyperpermeability in diabetic rats. J Clin

Invest 97:238–243, 199632. Schmidt AM, Hori O, Chen JX, Li JF, Crandall J, Zhang J, Cao R, Yan SD,

Brett J, Stern DM: Advanced glycation endproducts interacting with theirendothelial receptor induce expression of vascular cell adhesion mole-cule-1 (VCAM-1) in cultured human endothelial cells and in mice. J Clin

Invest 96:1395–1403, 199533. Schmidt AM, Crandall J, Hori O, Cao R, Lakatta E: Elevated plasma levels

of vascular cell adhesion molecule-1 (VCAM-1) in diabetic patients withmicroalbuminuria: a marker of vascular dysfunction and progressivevascular disease. Br J Haematol 92:747–750, 1996

34. Palinski W, Koschinsky T, Butler SW, Miller E, Vlassara H, Cerami A:Immunological evidence for the presence of advanced glycation endproducts in atherosclerotic lesions of euglycemic rabbits. Arterioscler

Thromb Vasc Biol 15:571–582, 199535. Miyata T, Inagi R, Ilida Y, Sato N, Yamada N, Oda O, Maeda K, Seo H:

Involvement of beta-2-microglobulin modified with advanced glycationend products in the pathogenesis of hemodialysis associated amyloidosis.J Clin Invest 93:521–528, 1994

36. Schleicher ED, Wagner E, Nerlich AG: Increased accumulation of theglycoxidation product Ne-(carboxymethyl)lysine in human tissues indiabetes and aging. J Clin Invest 99:457–468, 1997

37. Anderson MM, Requena JR, Crowley JR, Thorpe SR, Heinecke JW: Themyeloperoxidase system of human phagocytes generates Nepsilon-(car-boxymethyl)lysine on proteins: a mechanism for producing advancedglycation end products at sites of inflammation. J Clin Invest 104:103–13,1999



38. Suzuki D, Miyata T: Carbonyl stress in the pathogenesis of diabeticnephropathy. Intern Med 38:309–314, 1999

39. Sebokova K, Blazicek P, Syrova D, Krvosikowa Z, Spustova V, Heidland A,Schinzel R: Circulating advanced glycation end products levels in ratsrapidly increase with acute renal failure. Kidney Int Suppl 78:S58–S62,2001

40. Berg TJ, Clausen JT, Torjesen AA, Dahl-Jorgensen K, Bangstad HJ,Hanssen KF: The advanced glycation endproduct Nepsilon-(carboxy-methyl)lysine is increased in serum from children and adolescents withtype 1 diabetes. Diabetes Care 21:1997–2002, 1998

41. Yan SD, Schmidt AM, Anderson GM, Zhang J, Brett J, Zou YS, Pinsky D,Stern D: Enhanced cellular oxidant stress by the interaction of advancedglycation end products with their receptors/binding proteins. J Biol Chem

269:9889–9897, 199442. Schmidt AM, Hori O, Brett J, Yan SD, Wautier JL, Stern DM: Cellular

receptors for advanced glycation end products: implications for inductionof oxidant stress and cellular dysfunction in the pathogenesis of vascularlesions. Arterioscler Thromb 14:1521–1528, 1994

43. Schmidt AM, Yan SD, Wautier JL, Stern D: Activation of receptor foradvanced glycation end products: a mechanism for chronic vasculardysfunction in diabetic vasculopathy and atherosclerosis. Circ Res 84:489–497, 1999

44. Kislinger T, Fu C, Huber B, Qu W, Taguchi A, Yan SD, Hofmann M, YanAF, Pischetsrieder M, Stern D, Schmidt AM: N�-(carboxymethyl)lysineadducts of proteins are ligands for receptor for advanced glycation endproducts that activate cell signaling pathways and modulate gene expres-sion. J Biol Chem 274:31740–31749, 1999

45. Park L, Raman KG, Lee KJ, Lu Y, Ferran LJ Jr, Chow WS, Stern D, SchmidtAM: Suppression of accelerated diabetic atherosclerosis by the solublereceptor for advanced glycation endproducts. Nat Med 4:1025–1031, 1998

46. Ritthaler U, Deng Y, Zhang Y, Greten J, Abel M, Sido B, Allenberg J, OttoG, Roth H, Bierhaus A, Ziegler R, Schmidt AM, Wahl P, Stern DM,Nawroth PP: Expression of receptors for advanced glycation end prod-ucts in peripheral occlusive vascular disease. Am J Pathol 146:688–694,1995

47. Abel M, Ritthaler U, Zhang Y, Deng Y, Schmidt AM, Greten J, Sernau T,Wahl P, Andrassy K, Ritz E, Waldherr R, Stern DM, Nawroth PP:Expression of receptors for advanced glycosylated end products in renaldisease. Nephrol Dial Transplant 10:1662–1667, 1995

48. Greten J, Zhang Y, Wiesel R, Ritz E, Ziegler R, Wahl P, Nawroth PP:Expression of receptors for advanced glycation end products in uremia.Nephrol Dial Transplant 11:786–790, 1996

49. Taguchi A, Blood GC, 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, HoriO, Ogawa S, Stern DM, Schmidt AM: Blockade of RAGE-amphoterinsignaling suppresses tumour growth and metastasis. Nature 405:354–360,2000

50. Brett J, Schmidt AM, Yan SD, Zou YS, Weidman E, Pinsky D, NowygrodR, Neeper M, Przysiecki C, Shaw A, Migheli AD, Stern D: Survey of thedistribution of a newly characterized receptor for advanced glycation endproducts in tissues. Am J Pathol 143:1699–1712, 1993

51. Bierhaus A, Chevion S, Chevion M, Quehenberger P, Hofmann M, IllmerT, Luther T, Berentshtein E, Tritschler H, Muller M, Ziegler R, NawrothPP: Advanced glycation endproducts (AGEs) induced activation of NF-�Bis suppressed by �-lipoic acid in cultured endothelial cells. Diabetes

46:1481–1490, 199752. Bierhaus A, Illmer T, Kasper M, Luther T, Quehenberger P, Tritschler H,

Wahl P, Ziegler R, Muller M, Nawroth PP: Advanced glycation endprod-ucts (AGEs) mediated induction of tissue factor in cultured endothelialcells is dependent on RAGE. Circulation 96:2262–2271, 1997

53. Wautier JL, Wautier MP, Schmidt AM, Anderson GM, Hori O, ZoukourianC, Capron L, Chappey O, Yan SD, Brett J, Guillausseau P-J, Stern DM:Advanced glycation end products (AGEs) on the surface of diabeticerythrocytes bind to the vessel wall via a specific receptor inducingoxidant stress in the vasculature: a link between surface-associated AGEsand diabetic complications. Proc Natl Acad Sci U S A 91:7742–7746, 1994

54. Emmerich F, Meiser M, Hummel M, Demel G, Foss HD, Jundt F, MathasS, Krappmann D, Scheidereit C, Stein H, Dorken B: Overexpression of Ikappa B alpha without inhibition of NF-�B activity and mutations in theI kappa B alpha gene in Reed-Sternberg cells. Blood 94:3129–3134, 1999

55. Mori N, Fujii M, Ikeda S, Yamada Y, Tomonaga M, Ballard DW, YamamotoN: Constitutive activation of NF-kappaB in primary adult T-cell leukemiacells. Blood 93:2360–2368, 1999

56. Sovak MA, Bellas RE, Kim DW, Zanieski GJ, Rogers AE, Traish AM,Sonenshein GE: Aberrant nuclear factor-kappaB/Rel expression and thepathogenesis of breast cancer. J Clin Invest 100:2952–2960, 1997

57. Tabary O, Escotte S, Couetil JP, Hubert D, Dusser D, Puchelle E, JacquotJ: Genistein inhibits constitutive and inducible NF�B activation anddecreases IL-8 production by human cystic fibrosis bronchial gland cells.Am J Pathol 155:473–481, 1999

58. Krappmann D, Emmerich F, Kordes U, Scharschmidt E, Dorken B,Scheidereit C: Molecular mechanisms of constitutive NF-kappaB/Relactivation in Hodgkin/Reed-Sternberg cells. Oncogene 18:943–953, 1999

59. Bourcier T, Sukhova G, Libby, P: The nuclear factor kappa-B signalingpathway participates in dysregulation of vascular smooth muscle cells invitro and in human atherosclerosis. J Biol Chem 272:15817–215824, 1997

60. Neurath MF, Pettersson S, Meyer zum Buschenfelde KH, Strober W: Localadministration of antisense phosphorothioate oligonucleotides to the p65subunit of NF-�B abrogates established experimental colitis in mice. Nat

Med 2:998–1004, 199661. Thiele K, Bierhaus A, Autschbach F, Hofmann M, Stremmel W, Thiele H,

Ziegler R, Nawroth PP: Cell specific effects of glucocorticoid treatment onthe NF-�Bp65/I�B� system in patients with Crohn’s disease. Gut 45:693–704, 1999

62. Yang F, de Villiers WJ, Lee EY, McClain CJ, Varilek GW: Increased nuclearfactor-kappaB activation in colitis of interleukin-2-deficient mice. J Lab

Clin Med 134:378–385, 199963. Hauf N, Goebel W, Fiedler F, Sokolovic Z, Kuhn M: Listeria monocyto-

genes infection of P388D1 macrophages results in a biphasic NF-�B(RelA/p50) activation induced by lipoteichoic acid and bacterial phospho-lipases and mediated by I�B� and I�B� degradation. Proc Natl Acad Sci

U S A 94:9394–9399, 199764. Thompson JE, Phillips RJ, Erdjument-Bromage H, Tempst P, Ghosh S:

I�B-� regulates the persistent response in a biphasic activation of NF-�B.Cell 80:573–582, 1995

65. Johnson DR, Douglas I, Jahnke A, Ghosh S, Pober JS: A sustainedreduction in I�B-� may contribute to persistent NF-�B activation inhuman endothelial cells. J Biol Chem 271:16317–16322, 1996

66. Cheshire JL, Baldwin AS Jr: Synergistic activation of NF-�B by tumornecrosis factor alpha and gamma interferon via enhanced I�B� degrada-tion and de novo I�B� degradation. Mol Cell Biol 17:6746–6754, 1997

67. Suyang H, Phillips R, Douglas I, Ghosh S: Role of unphosphorylated,newly synthesized I�B� in persistent activation of NF-�B. Mol Cell Biol

16:5444–5449, 199668. Costello R, Lipcey C, Algarte M, Cerdan C, Baeuerle P, Olive D, Imbert J:

Activation of primary human T-lymphocytes through CD2 plus CD28adhesion molecules induces long-term nuclear expression of NF-kB. Cell

Growth Differ 4:329–339, 199369. Kahn-Perles B, Lipcey C, Lecine P, Olive D, Imbert J: Temporal and

subunit-specific modulations of the Rel/NF-kB transcription factorsthrough CD28 costimulation. J Biol Chem 272:21774–21783, 1997

70. Quehenberger P, Bierhaus A, Fasching P, Muellner C, Klevesath M, HongM, Stier G, Sattler M, Schleicher E, Speiser W, Nawroth PP: Endothelin-1transcription is under control of nuclear factor-�B in AGE-stimulatedcultured endothelial cells. Diabetes 49:1561–1570, 2000

71. Henle T, Walter H, Krause I, Klostermeyer H: Efficient determination ofindividual Maillard compounds in heat-treated milk products by aminoacid analysis. Int Dairy J 1:125–135, 1991

72. Henle T, Schwarzenbolz U, Klostermeyer H: Detection and quantificationof pentosidine in foods. Z Lebensm Unters Forsch 204:95–98, 1997

73. Finot PA, Deutsch R, Bujard E: The extent of the Maillard reaction duringthe processing of milk. Prog Food Nutr Sci 5:345–355, 1981

74. Schleicher E, Wieland OH: Specific quantitation by HPLC of protein(lysine) bound glucose in human serum albumin and other glycosylatedproteins. J Clin Chem Clin Biochem 19:81–87, 1981

75. Henle T, Walter AW, Klostermeyer H: Simultaneous determination ofprotein-bound Maillard products by ion-exchange chromatography andphotodiode array detection. In Maillard Reactions in Chemistry, Food,

and Health. Labuza TP, Reineccius GA, Monnier VM, O’Brien J, BaynesJW, Eds. London, The Royal Society of Chemistry, 1994, p. 195–200

76. Henle T, Schwarzenbolz U, Walter AW, Klostermeyer H: Protein-boundMaillard compounds in foods: analytical and technological aspects. In The

Maillard Reaction in Foods and Medicine. O’Brien J, Nursten HE,Crabbe MJC, Ames JM, Eds. London, The Royal Society of Chemistry,1998, p. 178–183

77. Henle T, Bachmann A: Synthesis of pyrraline. Z Lebensm Unters Forsch

202:72–75, 199678. Lernbecher T, Muller U, Wirth T: Distinct NF-kappa B/Rel transcription

factors are responsible for tissue-specific and inducible gene activation.Nature 365:767–770, 1993

79. Kloting I, Vogt L: BB/O(ttawa)K(arlsburg) rats: features of a subline ofdiabetes-prone BB rats. Diabetes Res 18:79–87, 1991



80. Hofmann M, Schiekofer S, Kanitz M, Klevesath MS, Joswig M, Lee V,Morcos M, Tritschler H, Ziegler R, Wahl P, Bierhaus A, Nawroth PP:Insufficient glycemic control increases NF-�B binding activity in periph-eral blood mononuclear cells isolated from patients with type 1 diabetes.Diabetes Care 21:1310–1316, 1998

81. Hofmann M, Schiekofer S, Isermann B, Kanitz M, Henkels M, Joswig M,Treusch A, Morcos M, Weiss T, Borcea V, Abdel Khalek AK, Amiral J,Tritschler H, Ritz E, Wahl P, Ziegler R, Bierhaus A, Nawroth PP:Peripheral blood mononuclear cells isolated from patients with diabeticnephropathy demonstrate increased activation of the oxidative-stresssensitive transcription factor NF-�B. Diabetologia 42:222–232, 1999

82. Bohrer H, Qiu F, Zimmermann T, Zhang Y, Illmer T, Mannel D, BottigerBW, Stern DM, Waldherr R, Saeger H-D, Ziegler R, Bierhaus A, Martin E,Nawroth PP: Role of NF-�B in the mortality of sepsis. J Clin Invest

100:972–985, 199783. Bierhaus A, Zhang Y, Deng Y, Mackman N, Quehenberger P, Haase M,

Luther T, Muller M, Bohrer H, Greten J, Martin E, Baeuerle PA, WaldherrR, Kisiel W, Ziegler R, Stern DM, Nawroth PP: Mechanism of the TNF�mediated induction of endothelial tissue factor. J Biol Chem 270:26419–26432, 1995

84. Bierhaus A, Zhang Y, Quehenberger P, Luther T, Haase M, Muller M,Mackman N, Ziegler R, Nawroth PP: The dietary pigment curcuminreduces endothelial tissue factor gene expression by inhibiting binding ofAP-1 to the DNA and activation of NF-�B. Thromb Haemost 77:772–782,1997

85. Rodriguez MS, Michalopopulus I, Aenzana-Seisdedos F, Hay RT: Induc-ible degradation of I�B� in vitro and in vivo requires the acidic c-terminaldomain of the protein. Mol Cell Biol 15:2413–2419, 1995

86. Peng HP, Libby P, Liao JK: Induction and stabilisation of I�B� by nitricoxide mediates inhibition of NF-�B. J Biol Chem 270:14214–14219, 1995

87. Nawroth PP, Stern DM: Modulation of endothelial cell hemostatic prop-erties by tumor necrosis factor. J Exp Med 163:740–745, 1986

88. Lin J, Liliensiek B, Kanitz M, Schimanski U, Bohrer H, Waldherr R, MartinE, Kauffmann G, Ziegler R, Nawroth PP: Molecular cloning of genesdifferentially regulated by TNF-alpha in bovine aortic endothelial cells,fibroblasts and smooth muscle cells. Cardiovasc Res 38:802–813, 1998

89. Morigi M, Angioletti S, Imberti B, Donadelli R, Micheletti G, Figliuzzi M,Remuzzi A, Zoja C, Remuzzi G: Leukocyte-endothelial interaction isaugmented by high glucose concentrations and hyperglycemia in aNF-�B-dependent fashion. J Clin Invest 101:1905–1915, 1998

90. Dosquet C, Weill D, Wautier JL: Molecular mechanism of blood monocyteadhesion to vascular endothelial cells. Nouv Rev Fr Hematol 34 (Suppl.):S55–S59, 1992

91. Yang CW, Vlassara H, Peten EP, He CJ, Striker GE, Striker LJ: Advancedglycosylation endproducts upregulate gene expression found in diabeticglomerular disease. Proc Natl Acad Sci U S A 91:9436–9440, 1994

92. Vlassara H, Striker LJ, Teichberg H, Fuh YM, Li M, Steffes M: Advancedglycosylation endproducts induce glomerular sclerosis and albuminuriain normal rats. Proc Natl Acad Sci U S A 91:11704–11708, 1994

93. Pahl HL, Baeuerle PA: The ER-overload response: activation of NF-kappaB. Trends Biochem Sci 2:63–67, 1997

94. Berg TJ, Dahl-Jorgensen K, Torjesen PA, Hanssen KF: Increased serumlevels of advanced glycation end products (AGEs) in children andadolescents with IDDM. Diabetes Care 20:1006–1008, 1997

95. Wolffenbuttel BHR, Giordano D, Founds HW, Bucala R: Long-termassessment of glucose control by haemoglobin-AGE measurement. Lan-

cet 347:513–515, 199696. Makita Z, Radoff S, Rayfield E, Yang Z, Skolnik E, Friedmann EA, Cerami

A, Vlassara H: Advanced glycosylation endproducts in patients withdiabetic nephropathy. N Engl J Med 325:836–882, 1991

97. Koschinsky T, He CJ, Mitsuhashi T, Bucala R, Liu C, Buenting C,Heitmann K, Vlassara H: Orally absorbed reactive glycation products

(glycotoxins): an environmental risk factor in diabetic nephropathy. Proc

Natl Acad Sci U S A 94:6474–6479, 199798. Li Y-M, Steffes M, Donnelly T, Liu C, Fuh H, Basgen J, Bucala R, Vlassara

H: Prevention of cardiovascular and renal pathology of aging by theadvanced glycation inhibitor aminoguanidine. Proc Natl Acad Sci U S A

93:3902–3907, 199699. Stitt AW, Li YM, Gardiner TA, Bucala R, Archer DB, Vlassara H: Advanced

glycation end products (AGEs) colocalize with AGE receptors in theretinal vasculature of diabetic and of AGE-infused rats. Am J Pathol

150:523–539, 1997100. Willis D, Moore AR, Frederick R, Willoughby DA: Heme oxygenase: a

novel target for the modulation of the inflammatory response. Nat Med

2:87–90, 1996101. Esposito C, Gerlach H, Brett J, Stern D, Vlassara H: Endothelial receptor-

mediated binding of glucose-modified albumin is associated with in-creased monolayer permeability and modulation of cell surface coagulantproperties. J Exp Med 170:1387–1407, 1989

102. Lindner V, Collins T: Expression of NF-�B and I�B-� by aortic endothe-lium in an arterial injury model. Am J Pathol 148:427–438, 1996

103. Rogler G, Brand K, Vogl D, Page S, Hofmeister R, Andus T, Knuechel R,Baeuerle P, Scholmerich J, Gross V: Nuclear factor �B is activated inmacrophages and epithelial cells of inflamed mucosa. Gastroenterology

115:357–369, 1998104. Yin K, Juurlink BH: Regulation of expression of nuclear factor kappa B rel

A in oligodendrocytes: effects of hypoxia. Neuroreport 11:1877–1881, 2000105. Wautier MP, Chappey O, Corda S, Stern DM, Schmidt AM, Wautier JL:

Activation of NADPH oxidase by RAGE links oxidant stress to alteredgene expression via RAGE. Am J Physiol Endocrinol Metab 280:E685–E694, 2001

106. Das KC: c-Jun NH2-terminal kinase-mediated redox-dependent degrada-tion of I�B: role of thioredoxin in NF-�B activation. J Biol Chem

276:4662–4670, 2001107. Hirota K, Matsui M, Iwata S, Nishiyama A, Mori K, Yodoi J: AP-1

transcriptional activity is regulated by a direct association betweenthioredoxin and Ref-1. Proc Natl Acad Sci U S A 94:3633–3638, 1997

108. Sheppard KA, Rose DW, Haque ZK, Kurokawa R, McInerney E, Westin S,Thanos D, Rosenfeld MG, Glass CK, Collins T: Transcriptional activationby NF-kappaB requires multiple coactivators. Mol Cell Biol 19:6367–6378,1999

109. Lander H, Tauras J, Ogiste J, Moss R, Schmidt AM: Activation of RAGEtriggers a MAP kinase pathway regulated by oxidant stress. J Biol Chem

272:17810–17814, 1997110. Nishibe T, Parry G, Ishida A, Aziz M, Murray J, Patel Y, Rahman S, Strand

K, Saito K, Saito Y, Hammond WP, Savidge GF, Mackman N, Wijelath ES:Oncostatin M promotes biphasic tissue factor expression in smoothmuscle cells: evidence for Erk1/2 activation. Blood 97:692–699, 2001

111. Abu-Am Y, Ross FP, McHugh KP, Livolsi A, Peyron JF, Teitelbaum SL:Tumor necrosis factor-� activation of nuclear transcription factor �B inmarrow macrophages is mediated by c-SRC tyrosine phosphorylation ofI�B�. J Biol Chem 273:29417–29423, 1998

112. Yurochko AD, Mayo ME, Poma EE, Baldwin AS, Huang ES: Induction ofthe transcription factor Sp1 during human cytomegalovirus infectionmediates upregulation of the p65 and p105/p50 NF-kappaB promoters.J Virol 71:4638–4648, 1997

113. Li J, Schmidt AM: Characterization and functional analysis of the pro-moter of RAGE, the receptor for advanced glycation end products. J Biol

Chem 272:16498–16506, 1997114. Perry G, Nunomura A, Lucassen P, Lassmann H, Smith MA: Apoptosis and

Alzheimer’s disease (Letter). Science 282:1268–1269, 1999115. Smith MA, Joseph JA, Perry G. Arson: Tracking the culprit in Alzheimer’s

disease. Ann N Y Acad Sci 924:35–38, 2000



diabetes-associated sustained activation of the ...george perry,7 ann-marie schmidt,5 david m....

Documents