vascular endothelial cell specific nf- b suppression attenuates hypertension-induced renal damage

Vascular Endothelial Cell–Specific NF-�B SuppressionAttenuates Hypertension-Induced Renal Damage

Norbert Henke, Ruth Schmidt-Ullrich, Ralf Dechend, Joon-Keun Park, Fatimunnisa Qadri,Maren Wellner, Michael Obst, Volkmar Gross, Rainer Dietz, Friedrich C. Luft,

Claus Scheidereit, Dominik N. Muller

Abstract—Nuclear factor kappa B (NF-�B) participates in hypertension-induced vascular and target-organ damage. Wetested whether or not endothelial cell–specific NF-�B suppression would be ameliorative. We generated Cre/loxtransgenic mice with endothelial cell–restricted NF-�B super-repressor I�B��N (Tie-1-�N mice) overexpression. Weconfirmed cell-specific I�B��N expression and reduced NF-�B activity after TNF-� stimulation in primary endothelialcell culture. To induce hypertension with target-organ damage, we fed mice a high-salt diet and N(omega)-nitro-L-arginine-methyl-ester (L-NAME) and infused angiotensin (Ang) II. This treatment caused a 40-mm Hg blood pressureincrease in both Tie-1-�N and control mice. In contrast to control mice, Tie-1-�N mice developed a milder renal injury,reduced inflammation, and less albuminuria. RT-PCR showed significantly reduced expression of the NF-�B targetsVCAM-1 and ICAM-1, compared with control mice. Thus, the data demonstrate a causal link between endothelialNF-�B activation and hypertension-induced renal damage. We conclude that in vivo NF-�B suppression in endothelialcells stops a signaling cascade leading to reduced hypertension-induced renal damage despite high blood pressure. (CircRes. 2007;101:268-276.)

Key Words: hypertension � endothelium � NF-�B � target-organ damage

Hypertension is a risk factor for target-organ damage suchas cardiac and renal disease.1 Endothelial cell injury

initiates adhesion molecule and chemokine expression pro-moting inflammation that contributes to the pathogenesis ofhypertension-induced target organ damage.2–5 Patients withsevere hypertension and target-organ damage often showelevated angiotensin (Ang) II levels and reduced nitric oxide(NO) production. Ang II signaling blockade reduces bloodpressure and blunts the development and progression ofvascular disease in small and large vessels in experimentalanimal models and in humans. Ang II also elicits an inflam-matory response in both endothelial cells6 and vascularsmooth muscle cells.7,8 Numerous in vitro and in vivo studiesdemonstrated that Ang II activates the nuclear factor kappa B(NF-�B), a major transcription factor in mediating inflamma-tion and innate immunity.4,6–8 In resting cells, NF-�B residesinactive in the cytoplasm by forming complexes with theinhibitor of �B (I��) proteins. Exposure to extracellularstimuli like TNF-�, IL-1, reactive oxygen species, Ang II,and numerous other activators leads to rapid phosphorylation,site-specific ubiquitination, and subsequent degradation ofthe I�B proteins by the 26S proteasome.9 The resulting freeNF-�B molecules translocate to the nucleus and regulate

target gene expression. The targets include genes involved inthe control of cell proliferation, apoptosis, innate, and adap-tive immune response.10,11 NF-�B and inflammation play animportant role in the development of the target organ dam-age.4,12–17 However, inhibition of NF-�B signaling could alsobe detrimental.18,19 Furthermore, the ideal site for NF-�Binhibition in the vessel wall is unknown. To address the roleof NF-�B in the endothelium in hypertension-induced renaldamage, we used the Cre/lox technique to generate a novelmouse model with endothelial cell–restricted expression ofthe human NF-�B super-repressor I�B��N.14,20

Materials and Methods

MiceFor endothelial cell–restricted super-repressor I�B��N overexpres-sion, mice carrying loxP-flanked alleles encoding for the humanI�B��N (loxP-�N) were crossed with Cre knock-in mice expressingthe Cre recombinase under the control of endothelial-specific Tie-1promoter. Tie-1-Cre mice were a gift from Dr Erika Gustafsson.21

The loxP-�N mice were described previously.14,20 The Tie-1-Creknock-in mediates excision of loxP-flanked sequences in renal, cardiac,pulmonary, and aortic endothelial cells.21 The double knock-in micewere designated Tie-1-�N. All mice were bred on a C57Bl/6

Original received February 12, 2007; revision received June 3, 2007; accepted June 7, 2007.From the Medical Faculty of the Charite (N.H., R.D., M.W., R.D., F.C.L., D.N.M.), Franz Volhard Clinic, HELIOS Klinikum-Berlin; the

Max-Delbruck-Center for Molecular Medicine (R.S.-U., F.Q., M.O., V.G., F.C.L., C.S., D.N.M.), Berlin-Buch; and the Medical School of Hannover(J.-K.P.), Hannover, Germany.

Correspondence to Ruth Schmidt-Ullrich, PhD, Max-Delbruck-Center, Robert-Rossle Street 50, 13125 Berlin-Buch, Germany. [email protected]

© 2007 American Heart Association, Inc.

Circulation Research is available at http://circres.ahajournals.org DOI: 10.1161/CIRCRESAHA.107.150474

268 by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from

http://circres.ahajournals.org/












background for more than 7 generations. The Berlin Animal ReviewBoard approved all protocols (Reg. 0135/01 and G 0261/02).

Ang II/L-NAME Mouse ModelEight-week-old control mice (single transgenic Tie-1-Cre and loxP-�N, wild-type C57BL/6 mice) and Tie-1-�N double transgenic mice(n�21 for each group) received the NO synthase inhibitorN(omega)-nitro-L-arginine-methyl-ester (L-NAME). L-NAME wasgiven at 0.1 mg�mL�1 in the drinking water. Ang II (1.0mg�kg�1�day�1) was infused via ALZET mini pumps. The animalswere fed a high-sodium diet (1% sodium in the chow) for 14 days.22

Because C57BL/6, Tie-1-Cre, and loxP-�N mice showed no differ-ences in renal histology, albuminuria, cell infiltration, and geneexpression, these groups were combined and termed controls. Wemeasured 24-hour mouse urinary albumin by ELISA (CellTrend,Germany). Animals were killed on day 14 under ether anesthesia.Hearts and kidneys were washed, weighed, harvested, and storeduntil analysis.

Telemetric Mean Arterial Pressure MeasurementsTelemetry system and Dataquest ART 2.0 software from DataSciences International were used to measure intracarotid bloodpressure in 4 to 5 mice per group. Briefly, a telemetry probe catheter(TA11-PA40; Data Sciences International) was inserted into thesurgically exposed carotid artery.23 After surgery, all mice wereallowed at least 7 days to recover. Before Ang II/L-NAME treat-ment, baseline values were continuously recorded for 3 days andaveraged. Days 10 to 12 after Ang II/L-NAME treatment, bloodpressure and heart rate were recorded and statistically analyzed.Systolic blood pressure (SBP), diastolic blood pressure (DBP), andheart rate were sampled at 100 Hz. Mean arterial pressure (MAP)was calculated as follows: MAP�DBP�(SBP-DBP)/3.

Immunohistochemistry and In Situ HybridizationIce-cold acetone-fixed cryosections (6 �m) were stained by indirectimmunofluorescence techniques as described earlier.17 The sectionswere incubated with the following antibodies: anti-F4/80, anti-CD4(both serotec, UK), anti-CD3, anti-CD8 (both BD Bioscience, Ger-many), anti-collagen III (Fitzgerald, Mass), anti-VCAM-1 (Santa-Cruz, Germany), anti-ICAM-1 (Dianova, Germany), anti-p65 (Ab-cam, UK), and anti-TNF-� (SantaCruz, Germany). For histology andin situ hybridization, paraffin sections were used. Masson trichromestain was performed by clinical routine protocol. Semiquantificationof tubular damage was performed as described in online methods.P65 semiquantification was evaluated with a scoring system1–5 basedon the number of p65� nuclei. Paraffin sections were also stainedwith sirius red to evaluate the degree of fibrosis. In situ hybridizationwas performed as described earlier.24 The murine I�B� (nt 1 to 1091,U36277/NM010907) cDNA was used. Sections were counterstainedwith 0.01% PyroninG and mounted with Entellan (Merck). Allpictures were taken with a Zeiss Axioplan 2 Imaging microscope/Axiophot camera.

Endothelial Cell CultureMurine lung and aortic endothelial cells were isolated with Dyna-beads coated with anti–platelet-endothelial cell adhesion molecule-1or anti–ICAM-2 monoclonal antibody as described previously.25,26

Endothelial cells were cultured with DMEM supplemented with 20%FBS, heparin, endothelial cell growth factor, nonessential aminoacids, and antibiotics.

Immunoblotting and Electrophoretic MobilityShift AssayEndothelial cells were stimulated with TNF-� (20 ng/mL; Biomol,Germany) as indicated. Whole cell extracts were prepared asdescribed earlier.27 electrophoretic mobility shift assay (EMSA) wasperformed with 5 �g whole cell protein as described previously.27

For immunoblotting, 20 �g whole cell lysate protein was used andrevealed with an I�B� antibody (Santa Cruz, Germany).

Quantitative TaqMan RT-PCRTotal kidney RNA was isolated using the TRIZOL protocol; RNAfrom cells with the RNeasy kit (Qiagen, Germany). TaqMan RT-PCR by SYBR Green analysis was performed as described earlier.28

ICAM-1 and VCAM-1 mRNA quantities were analyzed in duplicate.For quantification, the target sequences were normalized in relationto the mouse GAPDH product (Clonetech, Germany). Biotez (Berlin,Germany) synthesized the primers. The sequences are listed in theonline method section.

Statistical AnalysisDifferences between experimental groups were analyzed by use ofadjusted student t test or ANOVA for repeated measures, asappropriate. Data are reported as mean�SEM. Values of P�0.05were considered significant.

Results

Generation of Mice with Endothelial-SpecificNF-�B InhibitionThe breeding scheme of floxed I�B��N (loxP-�N) mice andheterozygous Tie-1–Cre knock-in mice to generate mice withendothelial cell–restricted I�B��N expression, namely Tie-1-�N mice, is given (Figure 1). I�B��N lacks the destructionbox containing the phosphorylation and ubiquitination sitesand acts as a super-repressor of NF-�B activity (supplementalFigure I, available online at http://circres.ahajournals.org). Toconfirm the correct functioning of the Tie-1–Cre line, Tie-1–Cre mice were mated with ROSA26Rflox/flox reporter mice andendothelial cell–specific LacZ staining pattern in the aortawas observed (data not shown). These results agree with theprevious description of the Tie-1-directed Cre expression.21,29

Recombination in Tie-1-�N mice was verified by immuno-blot analysis of whole cell extracts from isolated primaryendothelial cells (Figure 2A). I�B��N lacks the N-terminal70 amino acids and thus runs at a smaller molecular weight

Figure 1. Generation of double transgenic Tie-1-�N mice isshown. The cDNA of the human I�B� super-repressor I�B��Nis preceded by a stop codon flanked by 2 loxP sites. The floxedI�B��N construct was integrated in frame into the �-cateninlocus replacing exons 3 to 6.

Henke et al Endothelial NF-�B 269

by guest on June 29, 2015http://circres.ahajournals.org/Downloaded from


range than endogenous I�B� (Figure 2A). The efficiency ofNF-�B inhibition in Tie-1-�N was confirmed by EMSA(Figure 2B and 2C). After TNF-� stimulation, cultured endo-thelial cells from Tie-1-�N showed a reduced NF-�B DNA-binding activity, compared with wild-type control cells (Figure2C). Residual induced NF-�B activity was most likely derivedfrom nonrecombinant endothelial cells, because the Cre recom-binase does not function 100%. Excision of loxP sites by Tie-1promoter-induced Cre is estimated to take place in 70 and 95%of endothelial cells depending on the organ.21 Cultured wild-typeendothelial cells may grow at a faster rate and eventuallyoutnumber Tie-1-�N endothelial cells. Interestingly, in situhybridization (Figure 2D) using a murine anti-sense RNA probeof the bona fide NF-�B target gene I�B� clearly demonstratedthat hypertensive control mice with severe renal damage showedstrong I�B� signals in various substructures of the kidney,

namely blood vessels, tubules, and glomeruli. The increasedI�B� expression at these sites can be used as a direct read-out forNF-�B activity. As expected, in all these structures we observedsignificantly less upregulation of I�B� expression in Tie-1-�Nmice and in sham-controls (Figure 2D). This result alreadysuggests that NF-�B activity in endothelial cells has a directeffect on NF-�B activation in the surrounding structures such astubules and glomeruli. We analyzed nuclear p65 immunoreac-tivity as a marker for NF-�B activation. Vascular endothelialcells from Ang II/L-NAME-treated control mice often showedp65� nuclei, whereas only a few endothelial cells were positivein from Ang II/L-NAME–treated Tie-1-�N mice (supplementalFigure IIA). Tubular p65 nuclei were significantly less frequentin mice with suppressed NF-�B signaling compared with AngII/L-NAME–treated control mice, but still (supplemental FigureIIB and IIC).

Figure 2. Reduced NF-�B activity in primary aorticendothelial cells of Tie-1-�N mice. A, Whole cellextracts of unstimulated or TNF�-stimulated primaryendothelial cells isolated from aortas from Tie-1-�N orC57BL/6 control mice were evaluated in Western blot-ting. The human I�B��N protein is readily detectablein the Tie-1-�N endothelial cells. B, EMSA analysis ofthe same endothelial cell extracts as in A. After TNF�stimulation endothelial cells of Tie-1-�N mice show areduced NF-�B DNA binding activity compared withwild-type controls. Extracts treated with antibodiesagainst p50 or p65, as indicated, reveal a completeupward shift, indicating that the NF-�B DNA bindingcomplex is composed of p50/p65 heterodimers. C,Quantification of EMSA results. Results aremean�SEM; n�4; *P�0.05 vs TNF-� treated controlcells. D, In situ hybridization using a DIG-labeledmurine I�B� antisense RNA probe. Strongly increasedendogenous I�B� mRNA expression in kidney sec-tions of Ang II/L-NAME–treated control mice (bluecolor) is indicative for high NF-�B activity in theseareas. Ang II/L-NAME–treated Tie-1-�N revealed con-siderably lower I�B� expression in vessels, tubules,and glomeruli.

270 Circulation Research August 3, 2007



Figure 3. Reduced endothelial NF-�B activity has no effect on blood pressure but reduces albuminuria. A, Telemetric blood pressurerecordings of control and Tie-1-�N mice. Tie-1-�N and control mice are normotensive before Ang II/L-NAME treatment (baseline) butseverely hypertensive 14 days after chronic Ang II/L-NAME administration. B, Baseline urinary albumin levels before (baseline) andafter 14 days of Ang II/L-NAME treatment. Albuminuria is significantly higher in controls compared with Tie-1-�N mice. Results aremean�SEM. *P�0.05 vs baseline; #P�0.05 vs Ang II/L-NAME–treated controls.

Figure 4. Hypertension-induced kidney damage is diminished in Tie-1-�N mice. A, Masson trichrome stain of kidney paraffin sectionsfrom Ang II/L-NAME–treated control and Tie-1-�N mice and from untreated controls (sham). Treated control mice have severe tubulardamage with tubular necrosis (black arrows), and partial occlusion of the tubular lumen by cellular debris (black frame), but also vascu-lar damage with increased perivascular fibrosis. In treated Tie-1-�N mice these histological changes are less severe. (T indicatestubules; V, vein; A, artery; G, glomerulus; yellow bar represents 200 �m.) B, Semiquantification of tubular damage is shown. Results aremean�SEM; n�6 to 8; *P�0.05 vs Ang II/L-NAME–treated controls, #P�0.05 vs Ang II/L-NAME–treated Tie-1-�N mice. C, Sirius redstains for evaluation of renal fibrosis.




NF-�B Activation Promotes Hypertension-InducedRenal DamageBlood pressure recordings revealed that control mice andTie-1-�N mice were normotensive before Ang II/L-NAMEtreatment (106�1 mm Hg versus 103�1 mm Hg, respec-tively; Figure 3A). Chronic Ang II/L-NAME administrationraised the mean arterial blood pressure (MAP) by 40 mm Hg.There was no difference in MAP between both groups(143�5 mm Hg for control and 143�4 mm Hg for Tie-1-�N;Figure 3A). Baseline urinary albumin excretion was alsosimilar between both groups (Figure 3B). However, after AngII/L-NAME treatment, albuminuria was significantly higherin control mice compared with Tie-1-�N mice (6876�1058versus 3022�737 �g/d, respectively, P�0.05; Figure 3B).

Masson trichrome stain of kidney tissue showed that AngII/L-NAME–treated control mice had severe tubular andvascular damage with increased matrix production, comparedwith control mice (Figure 4A). Semiquantification of tubulardamage supported this result (Figure 4B). Ang II/L-NAME–treated Tie-1-�N mice also showed histological changes.Nevertheless, matrix formation and tubular damage weresignificantly less pronounced in mice with suppressed NF-�Bsignaling compared with Ang II/L-NAME–treated controlmice (Figure 4A and 4B). Sirius red stains confirmed asignificant reduction of perivascular and interstitial fibrosis inAng II/L-NAME–treated Tie-1-�N mice (Figure 4C). Vas-cular damage was less pronounced in Ang II/L-NAME–treated Tie-1-�N compared with control mice (supplementalFigure IIIA). Suppressed NF-�B activity only protectedglomeruli to a minor degree. Glomerular collagen III depo-sition was partially reduced in Ang II/L-NAME–treatedTie-1-�N mice (supplemental Figure IIIB).

NF-�B Drives Renal TNF-�, VCAM-1, andICAM-1 Expression in Hypertensive MiceTNF-� expression was increased in damaged glomeruli,tubules, and the vessel wall from Ang II/L-NAME–treated

controls, significantly reduced in Ang II/L-NAME–treated Tie-1-�N mice, and hardly or not expressed in sham controls(Figure 5). RT-PCR analysis revealed that after Ang II/L-NAME treatment renal VCAM-1 (Figure 6A) and ICAM-1(Figure 6B) expression was significantly increased in controlmice, compared with Tie-1-�N mice. Nevertheless, bothgroups showed significantly increased adhesion moleculeexpression, compared with untreated normotensive sham-control mice (Figure 6A and 6B). Immunohistological anal-ysis of renal VCAM-1 (Figure 6C) and ICAM-1 (Figure 6D)confirmed our RT-PCR results. Interestingly, we observeddifferent expression patterns of VCAM-1 and ICAM-1 im-munoreactivity in large or medium arteries, small arterioles,and capillaries. Large and medium arteries, as well ascapillaries of treated Tie-1-�N mice showed a significantlyreduced adhesion molecule expression compared with con-trols. However, arterioles showed the same staining intensityas treated controls (data not shown). This result suggestsinsufficient Cre recombination in arterioles. ICAM-1 wasalso expressed constitutively in the interstitium. However,Ang II/L-NAME–treated Tie-1-�N mice showed a signifi-cantly reduced renal ICAM-1 immunoreactivity in the inter-stitium, glomerular bed (Figure 6D upper panel), and thevessel wall (Figure 6D lower panel) compared with AngII/L-NAME–treated controls.

Renal Inflammatory Cell Infiltration Is Controlledby NF-�BBecause of the increased VCAM-1 and ICAM-1 expressionin Ang II/L-NAME-treated animals, we analyzed immunecell infiltration in our treated mouse models. Using ananti-CD3 antibody, elevated T-cell infiltration is seen inkidneys of treated control animals and to a lesser degree inTie-1-�N animals (Figure 7A). The 2 T cell subpopulationsthat included CD4� (helper) and CD8� (cytotoxic) T-cells, as

Figure 5. Renal TNF-� production is greatly sup-pressed in hypertensive kidneys of Tie-1-�N mice.Cryosections of kidney tissue from Ang II/L-NAME–treated control and Tie-1-�N mice andfrom sham control mice were stained with an anti–TNF-� antibody (in red). High TNF� expression isseen in damaged tubules (T), glomeruli (G), andvessels (A) of kidneys of treated control mice.Tie-1-�N mice show reduced amounts of TNF-�protein, whereas in sham-controls antibody stain-ing is absent. (Upper panel: tubules and glomeru-lus; lower panel: vessel. Yellow bar represents50 �m).




well as macrophages (F4/80� cells), were primarily locatedaround blood vessels and to a lesser extent within theinterstitium. We quantified infiltrating monocytes/macro-phages, T cells (CD3�, CD4�), and CD8� T-cell subpopula-tions in stained renal sections (Figure 7B through 7E). Weobserved partially diminished renal immune cell infiltrationin Ang II/L-NAME–treated-Tie-1-�N compared with AngII/L-NAME–treated control mice (Figure 7B through 7E).Nevertheless, cell infiltration was still significantly increasedin treated Tie-1-�N mice, compared with untreated sham-controls (Figure 7B through 7E).

DiscussionThe surprising finding is that endothelial-specific NF-�Bsuppression suffices to ameliorate Ang II/L-NAMEhypertension-induced renal damage. The results demon-strate that inhibition of NF-�B signaling in a single celltype is able to reduce renal tubular, vascular, and to aminor extent also glomerular damage. Endothelial NF-�Binhibition did not affect the development of hypertension,indicating that the renal protection was blood pressure–independent. Our study provides the first in vivo evidencethat specific endothelial NF-�B inhibition results in a

Figure 6. Suppressed NF-�B activity leads to reducedrenal VCAM-1 and ICAM-1 expression after Ang-II–L-NAME treatment. Kidneys from treated and untreatedanimals were analyzed by real time RT-PCR forVCAM-1 (A) and ICAM-1 (B) expression. Results aremean�SEM; n�6 to 8; *P�0.05 vs Ang II/L-NAME–treated controls, #P�0.05 vs Ang II/L-NAME–treatedTie-1-�N mice. C, VCAM-1 immunoreactivity (red) ishigher in the vascular endothelium of treated controlimmunohistochemistry on kidney cryosections fromAng II/L-NAME–treated control and Tie-1-�N mice andsham-controls. VCAM-1 kidneys, compared with Tie-1-�N and sham kidneys. D, ICAM-1 immunofluores-cence staining (red) is detected in glomeruli (G) andinterstitium in kidneys of control-treated mice (upperpanel). A reduced signal is seen in Tie-1-�N and shamkidneys (upper panel). Vascular ICAM-1 (A, artery) isshown in the lower panel. (Yellow bar represents50 �m.)




diminished upregulation of several proinflammatoryNF-�B target genes including VCAM-1 and ICAM-1 andreduces renal inflammation.

Hypertension is a major risk factor for cardiovascularevents. Nevertheless, high blood pressure is not a uniquefactor in the pathogenesis of target organ damage. Wereported earlier that dexamethasone inhibited NF-�B ac-tivity and prevented renal damage in an Ang II– dependentrat model despite blood pressures exceeding 200 mm Hg.30

In the present study, Ang II/L-NAME–treated controls andTie-1-�N mice were severely hypertensive and both de-veloped renal damage. Nevertheless, endothelial-specificNF-�B suppression protected Tie-1-�N mice to a signifi-cant extent.

NF-�B plays an important role in the pathogenesis ofcardiovascular disease. Cardiac NF-�B inhibition by a decoytechnique reduced the extent of myocardial infarction afterreperfusion.31 In renal allografts, pretreatment of donor kid-neys with NF-�B decoys led to reduced NF-�B activity andexpression of VCAM-1 leading to reduced cell infiltration.32

Boyle et al showed that treatment with pyrrolidine dithiocar-bamate (PDTC), a compound which acts as an antioxidant

and prevents ubiquitination and degradation of the NF-�Binhibitor I�B�, blocked endothelial cell E-selectin expressionin vitro and reduced cell infiltration in LPS-infused rabbits.33

We reported earlier that NF-�B inhibition by PDTC amelio-rated Ang II–induced cardiac hypertrophy and renal andvascular damage.4 In fact, PDTC probably not only blocksNF-�B activation, because it also inhibits the ubiquitin ligase�-TrCP, which is involved in other signaling pathways.34

Therefore, antioxidative, NF-�B, or other signaling pathwaysmight have accounted for the effect in vivo. New strategieswith cell-type specific NF-�B inhibition as described in thisand other recent articles can circumvent this problem. Freundet al14 were the first to show that in mice with cardiomyo-cyte-specific NF-�B suppression, cardiac hypertrophy wassignificantly reduced after Ang II treatment. Herrmann et al15

used a mouse stroke model and showed that suppressed I�Bkinase (IKK)-2 activity in neurons markedly reduced cerebralinfarct size. The animal models described above considerNF-�B as a detrimental factor in the pathogenesis of cardio-vascular disease.

NF-�B also regulates cell survival and plays a fundamentalrole in normal cell physiology. Thus, NF-�B inhibition may

Figure 7. Decreased cell infiltration in hyperten-sive kidney tissue of Tie-1-�N mice. A, A repre-sentative photomicrograph of T-cell infiltrationillustrated by anti-CD3 antibody staining (in red) ofkidney sections from Ang II/L-NAME–treated con-trols, Tie-1-�N mice, and sham-controls. B,Monocyte/macrophage (F4/80�) infiltration in AngII--L-NAME–treated control and Tie-1-�N miceand in sham controls. C through E, T-cell infiltra-tion: T cells (CD3), CD4� helper T, and CD8� cyto-toxic T. Cells were counted per view field. Resultsare mean�SEM; n�6 to 8; *P�0.05 vs Ang II/L-NAME–treated controls, #P�0.05 vs Ang II/L-NAME–treated Tie-1-�N mice.




also cause harmful effects. For instance, NF-�B inhibition inmacrophages increased atherosclerosis in LDL receptor–deficient mice.18 Furthermore, deletion of the IKK-NF-�Bpathway components results in massive hepatic apoptosis,defects in hematopoiesis, and death in utero.35–37 A transgenicmouse model similar to the Tie-1-�N mice was recentlyestablished by Kisseleva et al and differed from ours in theoverexpression of a mutated I�B� under the control of theTie-2 promoter.19 In these mice, endothelial cell NF-�Binhibition increased the sensitivity to LPS-induced toxemiaand altered vascular integrity. Whether or not the 2 modelshave an identical transmission of the modified I�B� in thesame cell type is difficult to determine without a directcomparison; such a comparison would be interesting andperhaps revealing. One challenging goal for the future will bethe elucidation of protective versus harmful effects of NF-�Bin cardiovascular disease. These effects seem to be cell- andtime-dependent. More detailed studies are needed to investi-gate this important controversy.

In contrast to atherosclerosis, hypertension-induced vascu-lar injury is not generally considered as an inflammatory andimmune disorder.38 We observed increased VCAM-1 andICAM-1 expression on the endothelium with cell infiltrationin hearts and kidneys of hypertensive double-transgenic ratswith high Ang II. Direct renin inhibition, Ang II receptor andendothelin receptor blockade, aspirin, or dexamethasonetreatment all reduced NF-�B activation and amelioratedinflammatory target organ damage in that rat model.17,30,39,40

Our earlier studies and the data presented here support theefficacy of antiinflammatory approaches to cardiovasculardiseases. In our Tie-1-�N mice, NF-�B inhibition in a singlecell type, namely endothelial cells, promoted renal protection.As a consequence of suppressed endothelial NF-�B activity,reduced VCAM-1 and ICAM-1 expression led to stronglydecreased immune cell infiltration in the surrounding tissues.These infiltrating inflammatory cells synthesize and secretevarious cytokines that could contribute locally to organdamage.

We focused our attention on the endothelium because thiscell layer has been particularly implicated in target-organdamage through endothelial dysfunction mediated by nitricoxide synthase dysfunction, Ang II, and aldosterone.41 Wepurposely selected a model that we were sure would causesufficient target-organ damage in mice.22 Our model withhigh Ang II and low nitric oxide availability mimics a patternseen in patients with heart failure, diabetes mellitus, andarteriosclerosis, as reviewed elsewhere.42 Nevertheless, ourstudy should not be generalized to situations in which nitricoxide synthase is still active. A better understanding of thecomplexity of NF-�B in the regulation of beneficial anddetrimental effects is needed. Therefore, uniquely targetedmodels such as the one reported here can be extended to othercell types. The approach could give new insights into bothpathogenesis and novel therapeutic options.

AcknowledgmentsWe thank Petra Berkefeld, Jana Czychi, Karin Ganzel, JulietteBergemann, May-Britt Kohler, Gabriele N’diaye, Mathilde Schmidt,and Sarah Ugowski for their excellent technical assistance.

Sources of FundingThe studies were supported by a grant-in-aid from the EuropeanUnion (EuReGene). The Deutsche Forschungsgemeinschaft (DFG)supported D.N.M. and F.C.L, and D.N.M. is a Helmholtz fellow.

DisclosuresNone.

References1. Nadar SK, Tayebjee MH, Messerli F, Lip GY. Target organ damage in

hypertension: pathophysiology and implications for drug therapy. CurrPharm Des. 2006;12:1581–1592.

2. Endemann DH, Schiffrin EL. Endothelial dysfunction. J Am Soc Nephrol.2004;15:1983–1992.

3. Li JJ, Fang CH, Hui RT. Is hypertension an inflammatory disease? MedHypotheses. 2005;64:236–240.

4. Muller DN, Dechend R, Mervaala EM, Park JK, Schmidt F, FiebelerA, Theuer J, Breu V, Ganten D, Haller H, Luft FC. NF-kappaBinhibition ameliorates angiotensin II-induced inflammatory damage inrats. Hypertension. 2000;35:193–201.

5. Ruiz-Ortega M, Esteban V, Ruperez M, Sanchez-Lopez E,Rodriguez-Vita J, Carvajal G, Egido J. Renal and vascular hypertension-induced inflammation: role of angiotensin II. Curr Opin NephrolHypertens. 2006;15:159–166.

6. Pueyo ME, Gonzalez W, Nicoletti A, Savoie F, Arnal JF, Michel JB.Angiotensin II stimulates endothelial vascular cell adhesion molecule-1via nuclear factor-kappaB activation induced by intracellular oxidativestress. Arterioscler Thromb Vasc Biol. 2000;20:645–651.

7. Ruiz-Ortega M, Lorenzo O, Ruperez M, Konig S, Wittig B, Egido J.Angiotensin II activates nuclear transcription factor kappaB throughAT(1) and AT(2) in vascular smooth muscle cells: molecular mech-anisms. Circ Res. 2000;86:1266–1272.

8. Zhang L, Cheng J, Ma Y, Thomas W, Zhang J, Du J. Dual pathways fornuclear factor kappaB activation by angiotensin II in vascular smoothmuscle: phosphorylation of p65 by IkappaB kinase and ribosomal kinase.Circ Res. 2005;97:975–982.

9. Collins T, Cybulsky MI. NF-kappaB: pivotal mediator or innocentbystander in atherogenesis? J Clin Invest. 2001;107:255–264.

10. Hayden MS, Ghosh S. Signaling to NF-kappaB. Genes Dev. 2004;18:2195–2224.

11. Karin M, Greten FR. NF-kappaB: linking inflammation and immunity tocancer development and progression. Nat Rev Immunol. 2005;5:749–759.

12. Brand K, Page S, Walli AK, Neumeier D, Baeuerle PA. Role of nuclearfactor-kappa B in atherogenesis. Exp Physiol. 1997;82:297–304.

13. de Winther MP, Kanters E, Kraal G, Hofker MH. Nuclear factor kappaBsignaling in atherogenesis. Arterioscler Thromb Vasc Biol. 2005;25:904–914.

14. Freund C, Schmidt-Ullrich R, Baurand A, Dunger S, Schneider W, LoserP, El-Jamali A, Dietz R, Scheidereit C, Bergmann MW. Requirement ofnuclear factor-kappaB in angiotensin II- and isoproterenol-inducedcardiac hypertrophy in vivo. Circulation. 2005;111:2319–2325.

15. Herrmann O, Baumann B, de Lorenzi R, Muhammad S, Zhang W,Kleesiek J, Malfertheiner M, Kohrmann M, Potrovita I, Maegele I, BeyerC, Burke JR, Hasan MT, Bujard H, Wirth T, Pasparakis M, SchwaningerM. IKK mediates ischemia-induced neuronal death. Nat Med. 2005;11:1322–1329.

16. Luedde T, Assmus U, Wustefeld T, Meyer zu Vilsendorf A, Roskams T,Schmidt-Supprian M, Rajewsky K, Brenner DA, Manns MP, PasparakisM, Trautwein C. Deletion of IKK2 in hepatocytes does not sensitize thesecells to TNF-induced apoptosis but protects from ischemia/reperfusioninjury. J Clin Invest. 2005;115:849–859.

17. Muller DN, Heissmeyer V, Dechend R, Hampich F, Park JK, Fiebeler A,Shagdarsuren E, Theuer J, Elger M, Pilz B, Breu V, Schroer K, Ganten D,Dietz R, Haller H, Scheidereit C, Luft FC. Aspirin inhibits NF-kappaBand protects from angiotensin II-induced organ damage. FASEB J. 2001;15:1822–1824.

18. Kanters E, Pasparakis M, Gijbels MJ, Vergouwe MN, Partouns-HendriksI, Fijneman RJ, Clausen BE, Forster I, Kockx MM, Rajewsky K, Kraal G,Hofker MH, de Winther MP. Inhibition of NF-kappaB activation inmacrophages increases atherosclerosis in LDL receptor-deficient mice.J Clin Invest. 2003;112:1176–1185.




19. Kisseleva T, Song L, Vorontchikhina M, Feirt N, Kitajewski J, SchindlerC. NF-kappaB regulation of endothelial cell function during LPS-inducedtoxemia and cancer. J Clin Invest. 2006;116:2955–2963.

20. Schmidt-Ullrich R, Aebischer T, Hulsken J, Birchmeier W, Klemm U,Scheidereit C. Requirement of NF-kappaB/Rel for the development ofhair follicles and other epidermal appendices. Development. 2001;128:3843–3853.

21. Gustafsson E, Brakebusch C, Hietanen K, Fassler R. Tie-1-directedexpression of Cre recombinase in endothelial cells of embryoid bodiesand transgenic mice. J Cell Sci. 2001;114:671–676.

22. Oestreicher EM, Martinez-Vasquez D, Stone JR, Jonasson L, Roubsan-thisuk W, Mukasa K, Adler GK. Aldosterone and not plasminogen acti-vator inhibitor-1 is a critical mediator of early angiotensin II/NG-nitro-L-arginine methyl ester-induced myocardial injury. Circulation. 2003;108:2517–2523.

23. Gross V, Obst M, Kiss E, Janke J, Mazak I, Shagdarsuren E, Muller DN,Langenickel TH, Grone HJ, Luft FC. Cardiac hypertrophy and fibrosis inchronic L-NAME-treated AT2 receptor-deficient mice. J Hypertens.2004;22:997–1005.

24. Schmidt-Ullrich R, Tobin DJ, Lenhard D, Schneider P, Paus R, Schei-dereit C. NF-kappaB transmits Eda A1/EdaR signalling to activate Shhand cyclin D1 expression, and controls post-initiation hair placode downgrowth. Development. 2006;133:1045–1057.

25. Lim YC, Garcia-Cardena G, Allport JR, Zervoglos M, Connolly AJ,Gimbrone MA Jr, Luscinskas FW. Heterogeneity of endothelial cellsfrom different organ sites in T-cell subset recruitment. Am J Pathol.2003;162:1591–1601.

26. Rikitake Y, Liao JK. Rho-kinase mediates hyperglycemia-induced plas-minogen activator inhibitor-1 expression in vascular endothelial cells.Circulation. 2005;111:3261–3268.

27. Krappmann D, Wulczyn FG, Scheidereit C. Different mechanisms controlsignal-induced degradation and basal turnover of the NF-kappaB inhibitorIkappaB alpha in vivo. Embo J. 1996;15:6716–6726.

28. Fiebeler A, Nussberger J, Shagdarsuren E, Rong S, Hilfenhaus G,Al-Saadi N, Dechend R, Wellner M, Meiners S, Maser-Gluth C, Jeng AY,Webb RL, Luft FC, Muller DN. Aldosterone synthase inhibitor ame-liorates angiotensin II-induced organ damage. Circulation. 2005;111:3087–3094.

29. Bjarnegard M, Enge M, Norlin J, Gustafsdottir S, Fredriksson S,Abramsson A, Takemoto M, Gustafsson E, Fassler R, Betsholtz C. En-dothelium-specific ablation of PDGFB leads to pericyte loss and glomer-ular, cardiac and placental abnormalities. Development. 2004;131:1847–1857.

30. Muller DN, Shagdarsuren E, Park JK, Dechend R, Mervaala E, HampichF, Fiebeler A, Ju X, Finckenberg P, Theuer J, Viedt C, Kreuzer J,

Heidecke H, Haller H, Zenke M, Luft FC. Immunosuppressive treatmentprotects against angiotensin II-induced renal damage. Am J Pathol. 2002;161:1679–1693.

31. Morishita R, Sugimoto T, Aoki M, Kida I, Tomita N, Moriguchi A,Maeda K, Sawa Y, Kaneda Y, Higaki J, Ogihara T. In vivo transfectionof cis element “decoy” against nuclear factor- kappaB binding siteprevents myocardial infarction. Nat Med. 1997;3:894–899.

32. Vos IH, Govers R, Grone HJ, Kleij L, Schurink M, De Weger RA,Goldschmeding R, Rabelink TJ. NFkappaB decoy oligodeoxynucleotidesreduce monocyte infiltration in renal allografts. Faseb J. 2000;14:815–822.

33. Boyle EM Jr, Kovacich JC, Canty TG Jr, Morgan EN, Chi E, Verrier ED,Pohlman TH. Inhibition of nuclear factor-kappa B nuclear localizationreduces human E-selectin expression and the systemic inflammatoryresponse. Circulation. 1998;98(Suppl):II282–II288.

34. Hayakawa M, Miyashita H, Sakamoto I, Kitagawa M, Tanaka H, YasudaH, Karin M, Kikugawa K. Evidence that reactive oxygen species do notmediate NF-kappaB activation. Embo J. 2003;22:3356–3366.

35. Gerondakis S, Grossmann M, Nakamura Y, Pohl T, Grumont R. Geneticapproaches in mice to understand Rel/NF-kappaB and IkappaB function:transgenics and knockouts. Oncogene. 1999;18:6888–6895.

36. Karin M, Ben-Neriah Y. Phosphorylation meets ubiquitination: thecontrol of NF-�B activity. Annu Rev Immunol. 2000;18:621–663.

37. Makris C, Godfrey VL, Krahn-Senftleben G, Takahashi T, Roberts JL,Schwarz T, Feng L, Johnson RS, Karin M. Female mice heterozygous forIKK gamma/NEMO deficiencies develop a dermatopathy similar to thehuman X-linked disorder incontinentia pigmenti. Mol Cell. 2000;5:969–979.

38. Ross R. Atherosclerosis–an inflammatory disease. N Engl J Med. 1999;340:115–126.

39. Mervaala EM, Muller DN, Park JK, Schmidt F, Lohn M, Breu V,Dragun D, Ganten D, Haller H, Luft FC. Monocyte infiltration andadhesion molecules in a rat model of high human renin hypertension.Hypertension. 1999;33:389 –395.

40. Muller DN, Mervaala EM, Schmidt F, Park JK, Dechend R, Genersch E,Breu V, Loffler BM, Ganten D, Schneider W, Haller H, Luft FC. Effectof bosentan on NF-kappaB, inflammation, and tissue factor in angiotensinII-induced end-organ damage. Hypertension. 2000;36:282–290.

41. Struthers AD, MacDonald TM. Review of aldosterone- and angiotensinII-induced target organ damage and prevention. Cardiovasc Res. 2004;61:663–670.

42. Schulman IH, Zhou MS, Raij L. Interaction between nitric oxide andangiotensin II in the endothelium: role in atherosclerosis and hyper-tension. J Hypertens. 2006;Suppl 24:S45–S50.




Online Supplement Henke et al. Vascular endothelial cell-specific NF-κB suppression attenuates…

Expanded Methods

Semi-quantification of tubular damge

Semiquantitative scoring of tubular damage was performed by quantification of

Masson Tri-chrome stains (n=5 mice per group) with the following categories:

5+: severe tubular damage with tubular necrosis in >90% of the tubules including the

presence of partial occlusion of the tubular lumen by cellular debris

4+: severe tubular damage with tubular necrosis in 75-90% of the tubules including the


3+: severe tubular damage with tubular necrosis in 50-75% of the tubules without the


2+: tubular damage with tubular necrosis in 25-50% of the tubules

1+: tubular damage with tubular necrosis in <25% of the tubules

0: no tubular changes

Results are mean±SEM. * p<0.05 vs. Ang II/L-NAME-treated controls, # p<0.05 vs. Ang

II/L-NAME-treated Tie-1-∆N mice.

Primer Sequences

Mouse GAPDH:

mGAPDH 154f 5´GGCAAATTCAACGGCACAGT 3´

mGAPDH 223r 5´ AGATGGTGATGGGCTTCCC 3´

mGAPDH 177T 6-Fam AAGGCCGAGAATGGGAAGCTTGTCATC Tamra

Acidic ribosomal phosphoprotein P0 (36B4):

36B4 113f 5´AGCAGATGCAGCAGATCCG 3´

36B4 172r 5´TCTTGCCCATCAGCACCAC 3´


VCAM-1:

VCAM-1 90f 5ĆCC TGA ATA CAA AAC GAT CGC 3´

VCAM-1 155r 5ĆAG CCC GTA GTG CTG CAA G 3´

ICAM-1:

ICAM-1 754f 5ÁTA TAC CTG GAG CTG GGA GGC 3´

ICAM-1 823r 5´TGG CTG ACA CAG AGT CAC TGC 3´

Expanded Results

Histological and Immunohistochemical Analyses

We analyzed nuclear p65 immunoreactivity as a marker for NF-κB activation. Vascular

endothelial cells from Ang II/L-NAME treated control mice often showed p65+ nuclei, while

only a few endothelial cells were positive in from Ang II/L-NAME treated Tie-1-∆N mice

(Online Fig IIA). A representative photomicrograph focusing on tubular p65 is shown (Online

Fig. IIIA). Ang II/L-NAME-treated Tie-1-∆N mice also showed tubular p65

immunoreactivity in addition to their tubular damage. Nevertheless, tubular p65 nuclei were

significantly less frequent in mice with suppressed NF-κB signaling compared to Ang II/L-

NAME-treated control mice (Online Fig. IIIA-B). The renal damage was not limited to the

tubular system, since vascular changes occurred in a similar manner. A representative

photograph focusing on vascular alterations is shown (Online Fig. IIIA). Ang II/L-NAME-

treated control mice often showed severely damaged endothelium and media hypertrophy

including media fibrosis. In contrast, vessels from Ang II/L-NAME treated Tie-1-∆N mice

were less affected. The degree of damage was lower and also the endothelial and media

damage was less pronounced compared to Ang II/L-NAME-treated control mice. However,


Ang II/L-NAME treated Tie-1-∆N mice showed significantly more damage compared to

sham-controls, which was not surprising, since those mice had about 40 mm Hg higher MAP.

Both Ang II/L-NAME treated groups showed glomerular collagen III expression

compared to sham-controls. However, Ang II/L-NAME treated Tie-1-∆N mice had a slightly

reduced expression (Online Fig. IIIB).


Legends for online figures

Online Fig. I. The above scheme demonstrates the 2 systems, endogenous IκBα (left panel)

and IκBα∆N (right panel, knock-in), present side by side in endothelial cells of Tie-1-∆N

mice. IκBα∆N has very high protein stability and a long half-life because it lacks the N-

terminal phosphorylation and ubiquitination sites. This situation prevents degradation of

IκBα∆N upon stimulation with TNFα, or others.27 However, endogenous IκBα∆N still binds

NF-κB very efficiently. In our system, IκBα∆N was constitutively produced under the control

of the endognous β-catenin promoter.20

Online Fig. IIA. A representative photograph of p65 stained kidney sections showing nuclear

p65 immunoreactivity (red). More p65+ nuclei were observed in the vascular endothelium of

treated control kidneys, compared to Tie-1-∆N and sham kidneys.

Online Fig. IIB. A representative photograph of tubular p65. Tubular p65 nuclei were

significantly less frequent in mice with suppressed NF-κB signaling compared to Ang II/L-

NAME-treated control mice. Quantification is given in Online Fig. IIC. Results are

mean±SEM. * p<0.05 vs. Ang II/L-NAME-treated controls, # p<0.05 vs. Ang II/L-NAME-

treated Tie-1-∆N mice.

Online Fig. IIIA. A representative photograph of Masson trichrome stained kidney sections

showing vascular damage. The renal damage was not limited to the tubular system, since

vascular changes occurred in a similar manner. Ang II/L-NAME-treated control mice show

severely damaged endothelium and media hypertrophy including media fibrosis. In contrast,

vessels from Ang II/L-NAME treated Tie-1-∆N mice are less affected. The degree of damage

is lower and also the endothelial and media damage is less pronounced compared to Ang II/L-

NAME-treated control mice. However, as expected, an increase of the mean arterial pressure

(MAP) of about 40mm Hg in Ang II/L-NAME treated Tie-1-∆N mice still produced

significantly more vascular damage compared to sham-controls.


Online Fig. IIIB. Collagen III stains (in red) for the evaluation of glomerular matrix

formation. Both Ang II/L-NAME treated groups show glomerular collagen III expression

compared to sham-controls. However, Ang II/L-NAME treated Tie-1-∆N mice have a

reduced expression.

Scheidereit and Dominik N. MullerMaren Wellner, Michael Obst, Volkmar Gross, Rainer Dietz, Friedrich C. Luft, Claus

Norbert Henke, Ruth Schmidt-Ullrich, Ralf Dechend, Joon-Keun Park, Fatimunnisa Qadri,Renal Damage

B Suppression Attenuates Hypertension-InducedκSpecific NF-−Vascular Endothelial Cell

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 2007 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/CIRCRESAHA.107.1504742007;101:268-276; originally published online June 21, 2007;Circ Res.

http://circres.ahajournals.org/content/101/3/268World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org/content/suppl/2007/06/21/CIRCRESAHA.107.150474.DC1.htmlData Supplement (unedited) at:

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


http://circres.ahajournals.org/content/101/3/268

http://circres.ahajournals.org/content/suppl/2007/06/21/CIRCRESAHA.107.150474.DC1.html

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


vascular endothelial cell specific nf- b suppression attenuates hypertension-induced renal damage

Documents