TLR9 Signaling and Promotes Neutrophil Adherence Through

Bacterial DNA Activates Endothelial Cells and Promotes Neutrophil Adherence through TLR9 Signaling

This information is current as Driss El Kebir, Levente József, Wanling Pan, Lili Wang and of September 26, 2021. János G. Filep J Immunol 2009; 182:4386-4394; ; doi: 10.4049/jimmunol.0803044 http://www.jimmunol.org/content/182/7/4386 Downloaded from

References This article cites 52 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/182/7/4386.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 26, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Bacterial DNA Activates Endothelial Cells and Promotes Neutrophil Adherence through TLR9 Signaling1

Driss El Kebir,2 Levente Jo´zsef,2 Wanling Pan, Lili Wang, and Ja´nos G. Filep3

TLR9 detects bacterial DNA (CpG DNA) and elicits both innate and adoptive immunity. Recent evidence indicates that TLR9 is expressed in more diverse cell types than initially thought. In this study, we report that HUVECs constitutively express TLR9 and selectively recognize unmethylated CpG motifs in bacterial DNA and synthetic immune stimulatory CpG oligodeoxynucleotides. HUVECs respond to CpG DNA with rapid phosphorylation of I␬B-␣ and NF-␬B-mediated gene transcription and surface expression of the adhesion molecules ICAM-1 and E-selectin independent of MAPK signaling. The telomere-derived TLR9 inhibitory oligonucleotide 5؅-TTT AGG GTT AGG GTT AGG G-3؅, agents that block endosomal acidiﬁcation such as chloroquine and baﬁlomycin A, and NF-␬B inhibitors abrogated CpG DNA-induced signaling. HUVEC activation by CpG DNA led to markedly

enhanced neutrophil adhesion under nonstatic conditions that was further enhanced when neutrophils were stimulated with CpG Downloaded from DNA. The adhesive interactions were blocked by Abs against CD18 and, to a lesser degree, by anti-E-selectin and anti-L-selectin ␤ Abs. Our ﬁndings demonstrate that bacterial DNA promotes 2 integrin and E-selectin-mediated HUVEC-neutrophil adherence, and indicate the ability of CpG DNA to initiate and/or maintain the inﬂammatory response. The Journal of Immunology, 2009, 182: 4386–4394.

oll-like receptors detect microbial molecular patterns and Accumulation of neutrophils at sites of infection or tissue injury http://www.jimmunol.org/ initiate innate and adoptive immune responses (1). By is a critical component of the inflammatory response. Neutrophil T eliciting potent inflammatory responses (2, 3), TLRs are recruitment is tightly regulated by coordinated expression of ad- also involved in cardiovascular diseases (4). Sustained inflamma- hesion molecules on both neutrophils and endothelial cells and tion due to chronic or recurrent infections has been linked to in- production of cytokines/chemokines (19). Accumulating evidence creased risk of cardiovascular diseases (5). TLR9, localized to in- indicates a role for bacterial DNA in the regulation of these events. tracellular compartments, recognizes viral and bacterial DNA, In mice, bacterial DNA or synthetic oligodeoxynucleotides which contains unmethylated CpG dinucleotides in certain base (ODNs)4 containing unmethylated CpG motifs promotes neutro- contexts (6, 7). Bacterial DNA (CpG DNA) released from prolif- phil accumulation at the primary sites of infection (8, 20–23). erating bacteria or following killing of bacteria may persist in tis- Consistent with enhanced neutrophil trafficking in vivo, CpG DNA by guest on September 26, 2021 sues in the absence of bacteria and participate in ongoing inflam- enhances neutrophil chemotaxis (12), induces IL-8 release (11, mation. Thus, bacterial DNA has been detected in the lung of 24), triggers L-selectin shedding and up-regulation of CD11b/ cystic fibrosis patients (8), coronary artery specimens (9), and in CD18 (Mac-1) (11, 12), and delays neutrophil apoptosis (13). the blood of critically ill patients who have had negative blood However, it is not known whether bacterial DNA could promote culture (10). The function of TLR9 in APCs has extensively been neutrophil adhesion to endothelial cells, a critical step in neutrophil studied, consistent with the role of these cells in immune surveil- trafficking into tissues. lance. TLR9 expression is, however, not restricted to APCs. For In this study, we investigated the impact of CpG DNA on neu- instance, human neutrophils (11–13), murine endothelial cells (14, trophil adherence to HUVECs. Here, we report that HUVECs con- 15), human dermal microvascular (16) and lymphatic endothelial stitutively express TLR9 and selectively recognize unmethylated cells (17), and endothelial cells of human atherosclerotic plaques CpG motifs in bacterial DNA and synthetic immune stimulatory express TLR9 (18). CpG DNA was reported to stimulate ICAM-1 CpG ODNs. CpG DNA activates HUVEC as well as neutrophils ␤ mRNA and protein expression and IL-8 release from murine en- and promotes 2 integrin and E-selectin-mediated neutrophil ad- dothelial cells (14, 15), whereas it failed to affect human lymphatic hesion, indicating its ability to initiate and/or maintain the inflam- endothelial cells (17) and even suppressed IL-8 production in hu- matory response in the absence of other bacterial constituents. man dermal microvascular endothelial cells (16). Materials and Methods Research Center, Maisonneuve-Rosemont Hospital and Department of Pathology and Bacterial and mammalian DNA Cell Biology, University of Montreal, Montreal, Quebec, Canada Escherichia coli DNA (strain B) and calf thymus DNA (Sigma-Aldrich) Received for publication September 15, 2008. Accepted for publication January were purified by extraction with phenol:chloroform:isoamyl alcohol (25: 26, 2009. 24:1, v/v/v) and ethanol precipitation. Heat-denaturated (single-stranded) The costs of publication of this article were defrayed in part by the payment of page genomic DNA was used in all experiments. For some experiments, E. coli charges. This article must therefore be hereby marked advertisement in accordance DNA was treated for 16 h at 37°C with CpG methylase SssI(2U/␮g DNA) with 18 U.S.C. Section 1734 solely to indicate this fact. in NE buffer 2 supplemented with 160 ␮M S-adenosylmethionine (New 1 This work was supported by Grant MOP-67054 (to J.G.F.) and a Doctoral Research Award (to J.L.) from the Canadian Institutes of Health Research. 2 D.E.K. and L.J. contributed equally to this work. 4 Abbreviations used in this paper: ODN, oligodeoxynucleotide; iODN, TLR9 inhib- 3 Address correspondence and reprint requests to Dr. Jańos G. Filep, Research Center, itory oligodeoxynucleotide; ctrlODN, control ODN. Maisonneuve-Rosemont Hospital, 5415 boulevard de l’Assomption, Montreal, Que- bec, Canada H1T 2M4. E-mail address: janos.g.fi[email protected] Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 www.jimmunol.org/cgi/doi/10.4049/jimmunol.0803044 The Journal of Immunology 4387

England Biolabs). Methylated DNA was purified as above. All DNA prep- 15-␮g nuclear extracts. Binding is expressed as OD following correction arations contained Ͻ5 ng of LPS/mg of DNA by Limulus assay. with binding to the mutant sequence GGCATAGGTCC. Phosphorylation of I␬B-␣ at Ser32 and Ser36 relative to total I␬B-␣ protein was monitored Culture of HUVECs using 20 ␮g of cytosolic protein with the CASE kit for I␬B-␣ S32/S36 (SuperArray). The amounts of phosphorylated I␬B-␣ are expressed as OD. HUVECs (cryopreserved primary cells; Cascade Biologics) were cultured in Medium 200 containing Low Serum Growth Supplement (Cascade RNase protection assay Biologics). For multiple RNase protection assays, HUVECs were lysed with 50 ␮lof TLR9 expression lysis/denaturation solution (Ambion). 32P-Labeled antisense RNA probes were generated using templates for IL-8, MCP-1, ICAM-1, E-selectin, L32, Nonpassaged primary culture of HUVECs and untreated confluent mono- and GAPDH (RiboQuant; BD Pharmingen), and the assays were performed layers of HUVECs (passages two and three) were lysed in situ. Protein with the direct protect kit (Ambion) as described previously (28). extracts were subjected to Western blotting using mouse anti-human TLR9 mAb 26C593 (Calbiochem). The results were confirmed using rat anti- Isolation and culture of neutrophils human TLR9 Ab eB72-1665 (eBioscience). In additional experiments, untreated HUVECs were detached, permeabilized with Permeabilization Venous blood (anticoagulated with sodium heparin, 50 U/ml) was obtained Buffer (eBioscience), and stained with R-PE-conjugated anti-TLR9 Ab from healthy volunteers who had denied taking any medication for at least eB72-1665 or a class-matched irrelevant Ab (eBioscience). Fluorescence 2 wk. The Clinical Research Committee of the Maisonneuve-Rosemont was assessed with a FACScan flow cytometer and CellQuestPro software Hospital approved the experimental protocols. Isolated neutrophils (5 ϫ (BD Biosciences) (13). To visualize cytosolic location of TLR9, HUVECs 106 cells/ml, purity Ͼ96%, viability Ͼ98%) (24) were resuspended in were grown on coverslips, fixed in 4% paraformaldehyde for 5 min, and HBSS supplemented with 10% autologous plasma, placed on a rotator, permeabilized with Permeabilization Buffer (eBioscience). After washing preincubated for 30 min with the endosomal acidification inhibitor bafilo- with PBS, cells were blocked with 1% BSA in PBS for 1 h, incubated with mycin A (30 nM) or chloroquine (5 ␮g/ml), the MAPK inhibitor PD98059 Downloaded from R-PE-conjugated anti-TLR9 Ab eB72-1665, or an irrelevant Ab (50 ␮M), the PI3K inhibitor wortmannin (100 nM), or the p38 MAPK (eBioscience) for 90 min and counterstained with the nuclear stain 4Ј,6- inhibitor SB203580 (2 ␮M) and then challenged with CpG DNA at 37°C diamidino-2-phenylindole. Images were captured under a Leica DMRI flu- in a humidified atmosphere with 5% CO2. orescence microscope equipped with a digital camera (Retiga EX; QIM- AGING) and OpenLab software. For TLR9 gene expression, total RNA Neutrophil adhesion molecules ϫ 5 was isolated from 5 10 HUVECs using TRIzol reagent (Invitrogen). Resting and treated polymorphonuclear neutrophils were stained with cDNA was prepared with superscript reverse transcriptase (Invitrogen). FITC-conjugated anti-human L-selectin mAb DREG-56 (BD Pharmingen) http://www.jimmunol.org/ The following primers were used for subsequent PCR analysis: TLR9, and PE-labeled anti-CD11b mAb Leu-15 (BD Biosciences). Immunofluo- sense 5Ј-TGGTGTTGAAGGACAGTTCTCTC-3Ј and antisense 3Ј- Ј Ј rescence (10,000 cells for each sample) was analyzed with a FACScan flow CACTCGGAGGTTTCCCAGC-5 and GAPDH, sense 5 -GAAGGT cytometer (29). Adhesion molecule expression is expressed as the percent- GAAGGTCGGAGTC-3Ј and antisense 3Ј-GAAGATGGTGATGG Ј age of fluorescence of resting cells following correction with the fluores- GATTTC-5 . The corresponding 70- and 225-bp products were amplified cence intensity of an appropriately labeled, class-matched irrelevant Ab. enzymatically by 40 repeated cycles, then subjected to electrophoresis on 15% acrylamide and stained with ethidium bromide. Western blotting HUVEC stimulation Protein extracts prepared from 107 neutrophils were resolved by SDS- PAGE and transferred to polyvinylidene difluoride membranes. Blots were Confluent monolayers of HUVECs (passages three through five) in 24- or blocked with 5% skimmed milk powder in TBS plus Tween 20 and probed by guest on September 26, 2021 96-well microplates were incubated with CpG DNA, calf thymus DNA, with Abs to phosphorylated and total ERK1/2 and p38 MAPK (Cell Sig- methylated CpG DNA, or one of the following synthetic immune stimu- naling Technologies) (13). latory CpG ODNs (all from InvivoGen, bases in uppercase letters are phosphodiester, those in lowercase are phosphorothioate linkage): type A ODN Neutrophil-endothelial cell adhesion assay 2336, 5Ј-gggGACGAC:GTCGTGgggggg-3Ј; type B ODN 2006, 5Ј- tcgtcgttttgtcgttttgtcgtt-3Ј; and type C ODN 2395, 5Ј-tcgtcgttttcggcgc: The adhesion assays were performed under nonstatic conditions (29). In gcgccg-3Ј (7). In some experiments, HUVECs were pretreated for 30 min brief, confluent HUVEC monolayers in 96-well microplates were cultured with bafilomycin A (30 nM), chloroquine (5 ␮g/ml), PD98059 (50 ␮M), with CpG DNA (0.25–16 ␮g/ml) or LPS (1 ␮g/ml, a positive control) for ϫ 5 wortmannin (100 nM), SB203580 (2 ␮M), the TLR9 inhibitory ODN 5Ј- 4 h at 37°C in a 5% CO2 atmosphere, washed extensively, and 2 10 51 tttagggttagggttaggg-3Ј (iODN, 6.6 or 20 ␮M; InvivoGen) (25), a negative human Cr-labeled-neutrophils in 100 ␮l were then added. In some ex- control ODN 5Ј-tgctgctgcttgcaagcagcttgat-3Ј (ctrlODN; InvivoGen), or the periments, CpG DNA (0.025–1.6 ␮g/ml) was added back along with neu- selective NF-␬B inhibitors BAY 11-7082 (10 ␮M) (26) or Ro 106-9920 (5 trophils to CpG DNA-treated HUVECs. Some experiments were repeated ␮M) (27), then cultured with CpG DNA. In additional experiments, re- using function-blocking anti-E-selectin mAb ENA-2 (10 ␮g/ml, purified Ј ␮ sponsiveness of nonpassaged primary culture of HUVECs to CpG ODN F(ab )2; Monosan), anti-L-selectin DREG-56 mAb (20 g/ml), anti-CD18 was also assessed. At the indicated times, culture supernatants were col- mAb L130 (10 ␮g/ml), or the irrelevant Ab MOPC-21 (20 ␮g/ml, all from lected and the cells were processed as described below. BD Biosciences). HUVECs were incubated with neutrophils for 30 min at 37°C on an orbital shaker at 90 rpm. Loosely adherent or unattached cells Endothelial cell adhesion molecule expression were removed by washing, and the endothelial monolayer and the adherent neutrophils were lysed. The number of adhered neutrophils in each exper- After incubation for 4 h, HUVECs were detached with EDTA (0.1% in iment was calculated from the radioactivity of a control sample. PBS) and stained for ICAM-1 or E-selectin using fluorescent dye-conjugated anti-ICAM-1 (BD Pharmingen) and anti-E-selectin (Serotec) Abs Statistical analysis (28). Nonspecific binding was evaluated by the use of appropriately labeled mouse IgG1. Immunofluorescence (10,000 cells for each sample) was an- Results are presented as mean Ϯ SEM. Statistical comparisons were made alyzed with a FACScan flow cytometer using CellQuestPro software. by ANOVA using ranks (Kruskal-Wallis test) followed by Dunn’s multiple contrast hypothesis test to identify differences between various treatments. MCP-1 and IL-8 production Values of p Ͻ 0.05 were considered significant. The concentrations of MCP-1 and IL-8 in culture supernatants were determined by selective ELISAs (OptEIA; BD PharMingen). Intraassay and Results interassay coefficients of variation were typically Ͻ4 and Ͻ6%, TLR9 expression in HUVECs respectively. By Western blot analysis, we found that both primary nonpassaged NF-␬B and I␬B-␣ as well as passaged HUVECs constitutively express TLR9 (Fig. 1A). This was confirmed with intracellular flow cytometry using Nuclear and cytosolic fractions were prepared with a NE-PER Nuclear and Cytoplasmic Extraction Kit (Pierce). Binding of NF-␬B/p65 to the immo- permeabilized HUVECs (Fig. 1B). No specific staining was de- bilized ␬B consensus sequence GGGGTATTTCC was assayed with tected when nonpermeabilized HUVECs were stained with the anti- ELISA (Mercury TransFactor NF-␬B/p65 kit; BD Biosciences) using TLR9 Ab (Fig. 1B). Intracellular localization of TLR9 was also 4388 BACTERIAL DNA AND NEUTROPHIL-ENDOTHELIAL ADHESION Downloaded from http://www.jimmunol.org/

FIGURE 1. TLR9 expression in HUVECs. A, Primary nonpassaged HUVECs and HUVEC (passage three) lysates were subjected to immunoblotting with anti-TLR9 mAb 26C593. Human PBMC and RBC lysates served as positive and negative controls, respectively. B, Intracellular localization of TLR9. Following detachment, nonpermeabilized and permeabilized HUVECs (passage three) were stained with R-PE-conjugated anti-human TLR9 Ab eB72-1665 or its isotype control (IgG). Immunostaining was analyzed by ﬂow cytometry. C, HUVECs (passage three) were seeded onto coverslips, ﬁxed, and permeabilized and then incubated with R-PE-conjugated anti-TLR9 Ab eB72-1665 followed by staining with 4Ј,6-diamidino-2-phenylindole (DAPI).

Images were captured using a Leica ﬂuorescence microscope. D, TLR9 gene expression. RT-PCR was performed on primary nonpassaged HUVECs, by guest on September 26, 2021 HUVECs in passage three, PBMC, and neutrophils. The PCR products were resolved on 15% acrylamide gels by electrophoresis and stained with ethidium bromide. The results are representative of four determinations. RFU, Relative ﬂuorescence intensity units.

visualized with immunofluorescence microscopy (Fig. 1C). Fur- immune stimulatory synthetic CpG ODNs. Type A ODNs, char- thermore, conventional RT-PCR showed expression of the TLR9 acterized by a phosphodiester central CpG-containing palindromic gene in both primary and passaged HUVECs (Fig. 1D). motif and a phosphorothioate 3Ј poly(G) ring, induce high IFN-␣ production in dendritic cells, but are weak stimulators of TLR9- CpG DNA activates HUVECs coupled NF-␬B signaling (7). Type B ODNs, containing a full We next investigated whether CpG DNA could activate HUVECs. phosphorothioate backbone, strongly activate B cells, but are weak Culture of HUVECs with CpG DNA for 4 h resulted in concen- inducers of IFN-␣ production (7). Type C ODNs combine features tration-dependent increases in the overall expression as well as in of both types A and B (7). The type A ODN 2336 was a consid- the percentage of HUVECs expressing E-selectin and ICAM-1 erably less potent inducer of HUVEC activation than the type B (Fig. 2A). CpG DNA also stimulated IL-8 and MCP-1 release (Fig. ODN 2006 or the type C ODN 2395 (Fig. 4). On a molar basis, 2B). We detected significant induction with 2 ␮g/ml, which peaked ODN 2006 and ODN 2395 appeared to be equally potent; how- at 16 ␮g/ml. Unlike CpG DNA, methylated CpG DNA and calf ever, at the concentrations tested, both ODNs evoked smaller thymus DNA did not activate HUVECs (Fig. 2C). Furthermore, changes in ICAM-1 expression and IL-8 release than CpG DNA. LPS at 25 pg/ml (the maximum level of contamination in our DNA Because transcription of ICAM-1, E-selectin, IL-8, and MCP-1 preparations) did not produce detectable changes (data not shown). genes require activation of NF-␬B (30), we prepared HUVEC cy- CpG DNA evoked similar changes in ICAM-1 expression and IL-8 tosolic and nuclear extracts and assessed cytosolic I␬B-␣ phos- production in primary nonpassaged HUVECs as in HUVECs at phorylation and DNA binding by NF-␬B/p65 with ELISAs. CpG passages three through five (Fig. 2D). DNA evoked rapid phosphorylation of I␬B-␣ and increases in To confirm the involvement of TLR9 in mediating CpG DNA DNA binding by NF-␬B/p65 (Fig. 5A). These actions were con- activation of HUVECs, we used a TLR9 inhibitory ODN (25). centration dependent (Fig. 5B). By contrast, calf thymus DNA and Pretreatment of HUVECs with iODN markedly inhibited CpG methylated CpG DNA failed to activate NF-␬B (Fig. 5B). Inhibi- DNA-induced adhesion molecule expression and cytokine release tion of endosomal acidification partially prevented NF-␬B activa- (Fig. 3A). The ctrlODN was without detectable effects (Fig. 3A). tion by CpG DNA (Fig. 5C). Preincubation of HUVECs with the Inhibition of endosomal acidification with bafilomycin A or chlo- NF-␬B inhibitors BAY 11-7082 or Ro 106-9920 inhibited 60– roquine attenuated by ϳ40% of HUVEC responses to CpG DNA 80% of CpG DNA-induced ICAM-1 and E-selectin expression (Fig. 3B). We also tested HUVEC responses to various types of (Fig. 5D), indicating that the increased nuclear accumulation of The Journal of Immunology 4389 Downloaded from

FIGURE 2. CpG DNA activates HUVECs. HUVEC monolayers were cultured for 4 h with CpG DNA. A, ICAM-1 and E-selectin expression was assessed by flow cytometry and is expressed as relative fluorescence intensity (RFU) after subtracting nonspecific immunostaining. Positive cells represent the percentage of HUVECs that stained positive for the indicated adhesion molecule. B, IL-8 and MCP-1 were measured in culture supernatant. C, http://www.jimmunol.org/ Methylated CpG DNA (Met-CpG-DNA) and thymus DNA (16 ␮g/ml) do not activate HUVECs. Results are mean Ϯ SEM for six to seven experiments. p Ͻ 0.001 vs untreated (vehicle). D, CpG DNA evoked similar responses in nonpassaged primary HUVECs (two batches ,ءءء p Ͻ 0.01; and ,ءء ;p Ͻ 0.05 ,ء of cells, measurements in duplicate) and HUVECs (passages three through five) (n ϭ 6).

NF-␬B correlates with HUVEC activation. Pharmacological (Fig. 7A). Statistically signiﬁcant changes were detectable with 0.8 blockade of the ERK, PI3K and p38 MAPK pathways did not ␮g/ml CpG DNA and reached an apparent maximum at 3 ␮g/ml. affect HUVEC responses to CpG DNA (Fig. 5E). Calf thymus DNA (Fig. 7A) or methylated CpG DNA (data not

We performed RNase protection assays on RNA extracted from shown) were without detectable effects. by guest on September 26, 2021 HUVECs after4hofincubation with CpG DNA. Consistent with Pretreatment of neutrophils with baﬁlomycin A or chloro- the observations at the protein levels, CpG DNA, but not calf thy- quine almost completely prevented CpG DNA-induced changes mus DNA, stimulated transcription of ICAM-1, E-selectin, IL-8, in CD11b and L-selectin expression (Fig. 7B), indicating that and MCP-1 genes (Fig. 6). endosomal acidiﬁcation of CpG DNA is a necessary step for initiating intracellular signaling for modulating adhesion mol- CpG DNA modulation of CD11b and L-selectin on human ecule expression on neutrophils. CpG DNA induced rapid phos- neutrophils phorylation of both ERK1/2 and p38 MAPK (Fig. 7C). Consis- Consistent with previous studies (11, 12), we found that CpG DNA tently, PD98059 and, to a lesser degree SB203580 inhibited induced concentration-dependent down-regulation of L-selectin CpG DNA-evoked changes in neutrophil adhesion molecule ex- and up-regulation of CD11b expression on the neutrophil surface pression (Fig. 7D).

FIGURE 3. TLR9 inhibition (A) and blockers of endosomal acidification (B) attenuate HUVEC activation by CpG-DNA. HUVECs were pretreated with iODN, ctrlODN, bafilomycin A (30 nM), or chloroquine (5 ␮g/ml) for 30 min and then cultured for 4 h with CpG DNA (8 ␮g/ml). ICAM-1 and E-selectin expression was assessed by flow cytometry following detachment of the cells. IL-8 and MCP-1 levels in the culture medium were determined with ELISAs. Results are mean Ϯ SEM for four to seven experiments. †, p Ͻ 0.05 and ††, p Ͻ 0.01 vs CpG DNA. 4390 BACTERIAL DNA AND NEUTROPHIL-ENDOTHELIAL ADHESION

FIGURE 4. HUVEC responses to immune stimulatory CpG ODNs. HUVECs were cultured with ODN 2336, ODN 2006, ODN 2395, or ctrlODN for 4 h. ICAM-1 expression was assessed by ﬂow cytometry following detachment of the cells. IL-8 levels in the culture medium were p Ͻ 0.05 vs FIGURE 6. Effects of CpG DNA on ICAM-1, E-selectin, MCP-1, IL-8 ,ء .(determined with ELISA. Results are mean Ϯ SEM (n ϭ 4 ctrl ODN. mRNA, L32, and GAPDH mRNA expression. HUVECs were cultured for

4 h with CpG DNA (4–16 ␮g/ml) or thymus DNA (16 ␮g/ml). Left lane Downloaded from indicates control (medium only). Shown is a representative multiprobe CpG DNA promotes neutrophil adhesion to HUVECs RNase protection assay of four independent experiments. To address the biological signiﬁcance of CpG DNA-induced adhesion molecule expression, we performed neutrophil-HUVEC ad- ICAM-1 expression. CpG DNA signiﬁcantly enhanced neutrophil hesion assays. We found that culture of HUVECs with CpG DNA attachment at 2 ␮g/ml and reached an apparent maximum at 16 resulted in marked increases in the number of adherent neutrophils ␮g/ml. The number of adherent neutrophils was further enhanced http://www.jimmunol.org/ (Fig. 8A), consistent with the changes observed in E-selectin and when the adhesion assay was performed in the presence of CpG by guest on September 26, 2021

FIGURE 5. Role of NF-␬B and MAPKs in CpG DNA-induced HUVEC activation. HUVEC monolayers were cultured for the indicated times (A)orfor4h(B–E) with CpG DNA, methylated CpG DNA, or calf thymus DNA (8 ␮g/ml) with or without baﬁlomycin A (30 nM), chloroquine (5 ␮g/ ml), BAY 11-7082 (10 ␮M), Ro 106-9920 (5 ␮M), PD98059 (50 ␮M), wortmannin (200 nM), or SB203580 (2 ␮M). A and B, Time- and concentration-dependent activation of NF-␬B. Nuclear extracts were prepared and DNA binding of NF-␬B was detected by ELISA using an immobilized ␬B consensus sequence and is expressed as the OD. Binding to a scrambled sequence served as a control. Phosphorylation of I␬B-␣ relative to total I␬B-␣ protein was monitored with ELISA. Endosomal acidiﬁcation inhibitors (C) and NF-␬B inhibitors (D) attenuate HUVEC activation by CpG DNA. E, Lack of effect of ERK, PI3K, and p38 MAPK blockade. The results are mean Ϯ SEM of ;p Ͻ 0.05 ,ء .three to six experiments -p Ͻ 0.001 vs unstimu ,ءءء ;p Ͻ 0.01 ,ءءء lated; and ††, p Ͻ 0.01 vs CpG DNA. The Journal of Immunology 4391 Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 7. Mechanism of CpG DNA-induced changes in L-selectin and CD11b expression on human neutrophils. Neutrophils were preincubated with bafilomycin A (30 nM), chloroquine (5 ␮g/ml), PD98059 (50 ␮M), wortmannin (200 nM), or SB203580 (2 ␮M) for 30 min at 37°C and FIGURE 8. then challenged with CpG DNA (3.2 ␮g/ml) for 30 min. A, Concentration- CpG DNA promotes neutrophil adhesion to HUVECs. Confluent HUVEC monolayers were cultured in medium only (control) dependent actions of CpG DNA. Inhibition of neutrophil responses to CpG 51 DNA by endosomal acidification inhibitors (B) and MAPK inhibitors (C). or challenged with CpG DNA or LPS for 4 h, then Cr-labeled poly- A B Adhesion molecule expression is presented as percentage of control (i.e., morphonuclear neutrophils (PMN) without ( ) or along with ( ) CpG the mean fluorescence intensity of neutrophils incubated in medium only). DNA were added and incubated with HUVECs for 30 min at 37°C at 90 C p Ͻ 0.05; rpm. , Inhibition of CpG DNA-stimulated neutrophil adhesion by ,ء .Results are mean Ϯ SEM of six independent experiments p Ͻ 0.001 vs control; and †, p Ͻ 0.05 vs CpG DNA. function-blocking anti-E-selectin, anti-CD18, and anti-L-selectin ,ءءء ;p Ͻ 0.01 ,ءء ␮ D, Phosphorylation of ERK1/2 and p38 MAPK. The immunoblots are rep- mAbs. HUVECs were cultured with CpG DNA (8 g/ml, 4 h) and resentative of four independent experiments. neutrophil adherence was assayed in the presence of mAbs. Irrelevant Ab MOPC-21 served as a negative control. Results are mean Ϯ SEM of p Ͻ 0.001 ,ءءء ;p Ͻ 0.01 ,ءء ;p Ͻ 0.05 ,ء .four to seven experiments vs control (medium only); †, p Ͻ 0.05; and ††, p Ͻ 0.01 vs CpG DNA DNA (Fig. 8B), indicating that CpG DNA activated both neutro- without mAbs (solid columns). phils and HUVECs. Because multiple receptors are involved in mediating neutrophil adherence to HUVECs under nonstatic conditions (31) and CpG of CpG DNA (Fig. 8C), consistent with CpG DNA-induced down- DNA affected adhesion molecule expression on both neutrophils regulation of L-selectin and up-regulation of CD11b/CD18 expres- and HUVECs, we assessed the contribution of L-selectin, E-selec- sion on neutrophils. ␤ tin, and 2 integrins to the binding interaction by using function- blocking mAbs. Neutrophil adhesion to CpG DNA-activated Discussion HUVECs were blocked by Abs against CD18 (71 Ϯ 8%, n ϭ 4), The present data demonstrate that bacterial DNA is a potent acti- E-selectin (33 Ϯ 7%), and L-selectin (26 Ϯ 7%) (Fig. 8C). The vator of key responses of human endothelial cells in amplifying the combination of these Abs inhibited neutrophil adhesion by 87 Ϯ inflammatory response. CpG DNA and unmethylated CpG motifs 3% (Fig. 8C). The anti-CD18 mAb was more effective, whereas in particular evoked endothelial cell activation that is associated the anti-L-selectin mAb was less effective in inhibiting neutrophil with increased nuclear accumulation of NF-␬B and subsequent adherence when the adhesion assay was performed in the presence up-regulation of adhesion molecules ICAM-1 and E-selectin and 4392 BACTERIAL DNA AND NEUTROPHIL-ENDOTHELIAL ADHESION enhanced chemokine production, thereby promoting neutrophil ad- TLR9-mediated NF-␬B activation in immune cells (7). Intrigu- herence to endothelial cells. Because injection of bacterial DNA or ingly, iODN produced higher degrees of inhibition than endo- synthetic CpG ODNs enhances accumulation of neutrophils at the somal acidification inhibitors in HUVECs. Consistent with our sites of injection (8, 20–23), our observations bear directly on the in vitro observations, CpG DNA fails to induce local neutrophil molecular mechanisms mediating bacterial DNA-evoked recruit- accumulation in TLR9-deficient or MyD88-deficient mice (22) ment of inflammatory cells. and silencing TLR9 with small interfering RNA markedly re- Expanding previous observations with murine vascular and hu- duces the number of infiltrating neutrophils (39). Of note, re- man lymphatic endothelial cells (14, 15, 17), we found that pri- cent observations indicate that cytosolic DNA sensor(s), such as mary nonpassaged as well as passaged HUVECs also express DAI (DNA-dependent activator of IFN regulatory factors also functional TLR9. We used permeabilized cells, flow cytometry, called DLM-1 or Z-DNA-binding protein 1), can initiate an and immunofluorescence microscopy to confirm intracellular lo- innate immune response independently of TLR9 (40–42). Fur- cation of TLR9. Thus, in addition to APCs (7, 32), epithelial cells thermore, TLR9 may recognize more than CpG (32). It remains (33), and neutrophils (11, 13), various types of endothelial cells are to be investigated whether HUVECs or other types of endothe- capable of recognizing bacterial DNA. lial cells express DAI or other cytosolic DNA sensors. Our results indicate that differences in DNA methylation pat- Our results revealed differences in downstream signaling terns would enable HUVECs to selectively recognize and respond pathways triggered by CpG DNA in HUVECs and neutrophils. to bacterial DNA. Methylation of cytosines in CpG dinucleotides Thus, CpG DNA activation of HUVECs is mediated through in bacterial DNA resulted in complete loss of its HUVEC activat- NF-␬B independent of activation of p38 MAPK. CpG DNA ing activity, and calf thymus DNA did not reproduce the actions of evoked phosphorylation of I␬B-␣ and concomitant nuclear ac- Downloaded from CpG DNA. Neutrophil recruitment into the lungs can be signifi- cumulation and DNA binding of NF-␬B/p65. Pharmacological cantly reduced by methylation of bacterial DNA, further pointing blockade of NF-␬B activation markedly, although never com- toward unmethylated CpG motifs as inducers of the inflammatory pletely, inhibited CpG DNA-stimulated expression of ICAM-1 response (8). CpG dinucleotides are commonly methylated in and E-selectin, indicating involvement of transcription factors mammalian DNA and inhibitory sequences flanking CpG motifs in other than NF-␬B. Thrombin and modified C-reactive protein mammalian DNA can also mask its stimulatory effects (7, 34). The induce endothelial NF-␬B activation through a p38 MAPK-de- http://www.jimmunol.org/ CpG DNA actions were not due to endotoxin contamination, since pendent mechanism (28, 43, 44). However, this pathway is not culture of HUVECs with 25 pg/ml LPS (the highest level of LPS involved in mediating the CpG DNA actions, because the p38 detected in our DNA preparations) did not produce detectable MAPK inhibitor SB203580 at a concentration used in previous effects. studies failed to affect HUVEC responses to CpG DNA. By Endosomal acidification of bacterial DNA is a critical step to contrast, SB203580 markedly attenuated CpG DNA-induced initiate intracellular signaling through binding to TLR9 in immune ICAM-1 and IL-8 expression in murine endothelial cells (14), cells (35, 36). We previously found that pretreatment of neutro- ϩ suggesting species differences in CpG DNA-triggered endothe- phils with bafilomycin A, which inhibits vacuolar H -ATPase (37)

lial signaling pathways. Bacterial DNA induces IL-8 release by guest on September 26, 2021 or chloroquine, which accumulates in and increases the pH in en- from epithelial cells by a MAPK-dependent NF-␬B-indepen- dosomes (35), rendered the apoptotic machinery and IL-8 gene dent pathway (33). Thus, the mechanism of CpG DNA-induced transcription in neutrophils unresponsive to CpG DNA (13, 24). In IL-8 expression may also be cell specific. In neutrophils, CpG this study, we showed that bafilomycin A and chloroquine prevented CpG DNA-induced L-selectin shedding and up-regulation DNA-evoked changes in adhesion molecule expression were of CD11b/CD18 expression. A CpG motif-independent pathway mediated through activation of the ERK and, to a lesser degree, has also been suggested to mediate neutrophil responses to bacte- of the p38 MAPK pathways. The mechanisms responsible for rial DNA (12). However, in this study, CpG DNA was used at a the cell and species-specific differences in CpG DNA signaling Ͼ30-fold higher concentration (100 ␮g/ml) than in the present are not clear at present and require additional investigations. study. We also found that bafilomycin A and chloroquine attenu- The biological significance of adhesion molecule expression ated HUVEC responses to CpG DNA, although the degree of in- was confirmed by the marked increase of adhesion of neutrophils hibition was considerably lower than in neutrophils. These obser- to CpG DNA-activated HUVECs under nonstatic conditions. CpG vations may reflect cell-specific maturation of CpG DNA. DNA-evoked neutrophil-HUVEC adherence was predominantly Alternatively, TLR9 may receive a stimulatory signal from CpG mediated through up-regulation of ICAM-1 and E-selectin expres- DNA before reaching the endosomes. Indeed, TLR9 has been sug- sion on endothelial cells. Significantly higher numbers of neutro- gested to sense CpG in the endoplasmic reticulum or endoplasmic phils adhered to CpG DNA-activated HUVECs when the adhesion reticulum-derived structures (38). assay was performed in the presence rather than in the absence of The present results with iODN lend further support to the CpG DNA, indicating that CpG DNA can promote adhesion by notion that CpG DNA predominantly signals through TLR9 in activating both HUVECs and neutrophils. No adhesion experi- HUVECs. This suppressive ODN contains the potent TLR9 in- ments were performed with neutrophils preincubated with CpG hibitory sequence TTAGGG multimers found in mammalian DNA since by up-regulating CD11b/CD18 expression, bacterial telomeres (25). iODN diffuses into cells and disrupts the colo- DNA may induce homotypic neutrophil adhesion, thereby making calization of CpG DNA with TLR9 in endosomal vesicles with- interpretation of the results difficult. out affecting cellular uptake (25). Our data also show that In the present study, we did not address the functional sig- HUVECs respond to immune stimulatory CpG ODNs. The type nificance of CpG DNA-induced IL-8 and MCP-1 expression. B ODN 2006 and type C ODN 2395 appeared to be equally IL-8 is a key regulator of neutrophil trafficking and activation potent, whereas the type A ODN 2336 was a considerably (45), whereas MCP-1 along with IL-8 directs recruitment of weaker inducer of HUVEC activation. Although the molecular monocytes (46). Our observations would imply enhanced mechanisms responsible for this discrepancy remain elusive, chemokine production as an additional mechanism by which our findings are consistent with the known differences in the CpG DNA may contribute to inflammatory cell trafficking capability of type A ODNs vs type B and C ODNs to induce in vivo. The Journal of Immunology 4393

Our results may have important implications in innate immu- 17. Pegu, A., S. Qin, B. A. Fallert Junecko, R. E. Nisato, M. S. Pepper, and nity. Our data demonstrate that CpG DNA activates NF-␬Bin T. A. Reinhart. 2008. Human lymphatic endothelial cells express multiple functional TLRs. J. Immunol. 180: 3399–3405. human endothelial cells, the driving force in the initiation and 18. Edfeldt, K., J. Swedenborg, G. K. Hansson, and Z. Q. Yan. 2002. Expression of progression of inflammation (30, 47, 48). Thus, CpG DNA Toll-like receptors in human atherosclerotic lesions: a possible pathway for through TLR9 promotes neutrophil adherence to endothelial plaque activation. Circulation 105: 1158–1161. 19. Ley, K., C. Laudanna, M. I. Cybulsky, and S. Noursargh. 2007. Getting to the site cells, induces cytokine release (11, 13), and suppresses neutro- of inflammation: the leukocyte adhesion cascade updated. Nat. Rev. Immunol. 7: phil apoptosis (13), indicating that bacterial DNA is capable of 678–689. triggering and/or maintaining inflammation even in the absence 20. Deng, J. C., T. A. Moore, M. W. Newstead, X. Zeng, A. M. Krieg, and T. J. Standiford. 2004. CpG oligodeoxynucleotides stimulate protective innate of other bacterial constituents. Treatment of mice with inhibi- immunity against pulmonary Klebsiella infection. J. Immunol. 173: tory TLR9 ODN protects against sepsis-induced mortality (49) 5148–5155. and acute kidney injury (50). Intriguingly, TLR9-deficient mice 21. Weighardt, H., C. Feterowski, M. Veit, M. Rump, H. Wagner, and B. Holzmann. are highly susceptible to Gram-negative and Gram-positive bac- 2000. Increased resistance against acute polymicrobial sepsis in mice challenged with immunostimulatory CpG oligodeoxynucleotides is related to an enhanced terial pneumonia (51, 52), whereas markedly resistant to septic innate effector cell response. J. Immunol. 165: 4537–4543. peritonitis (49). One potential explanation for this apparent dis- 22. Johnson, A. C., F. P. Heinzel, E. Diaconu, Y. Sun, A. G. Hise, D. Golenbock, crepancy is that a restricted degree of TLR9 stimulation may J. H. Lass, and E. Perlman. 2005. Activation of Toll-like receptor (TLR)2, TLR4, and TLR9 in the mammalian cornea induces MyD88-dependent corneal inflam- initiate protective immunity, whereas widespread TLR9 activa- mation. Invest. Ophthalmol. Vis. Sci. 46: 589–595. tion, including endothelial TLR9, may lead to an exaggerated 23. Knuefermann, P., G. Baumgarten, A. Koch, M. Schwederski, M. Velten, inflammatory response. H. Ehrentraut, J. Mersmann, R. Meyer, A. Hoeft, K. Zacharowski, and C. Grohe´. 2007. CpG oligonucleotide activates Toll-like receptor 9 and causes lung inflam- In summary, our results indicate that human endothelial cells mation in vivo. Respir. Res. 8: 72. Downloaded from recognize and respond to unmethylated CpG motifs in bacterial 24. Jo´zsef, L., T. Khreiss, D. El Kebir, and J. G. Filep. 2006. Activation of TLR-9 DNA. Bacterial DNA through activation of both endothelial cells induces IL-8 secretion through peroxynitrite signaling in human neutrophils. and neutrophils promotes ␤ integrin and E-selectin-mediated neu- J. Immunol. 176: 1195–1202. 2 25. Gursel, I. M. Gursel, H. Yamada, K. J. Ishii, F. Takeshita, and D. M. Klinman. trophil adhesion and thereby may contribute to triggering and/or 2003. Repetitive elements in mammalian telomeres suppress bacterial DNA-in- aggravation of the inflammatory response. duced immune activation. J. Immunol. 171: 1393–1400. 26. Izban, K. F., M. Ergin, J.-Z. Qin, R. L. Martinez, R. J. Pooley, Jr., S. Saeed, and http://www.jimmunol.org/ S. Alkan. 2000. Constitutive expression of NF-␬B is a characteristic feature of Disclosures mycosis fungoides: implications for apoptosis resistance and pathogenesis. Hum. Pathol. 31: 1482–1490. The authors have no financial conflict of interest. 27. Swinney, D. C., Y.-Z. Xu, L. E. Scarafia, I. Lee, A. Y. Mak, G.-F. Gan, C. S. Ramesha, M. A. Mulkins, J. Dunn, O.-Y. So, et al. 2002. A small molecule ubiquitination inhibitor blocks NF-␬B-dependent cytokine expression in cells and References rats. J. Biol. Chem. 277: 23573–23581. 1. Akira, S., S. Uematsu, and O. Takeuchi. 2006. Pathogen recognition and innate 28. Khreiss, T., L. Jo´zsef, L. A. Potempa, and J. G. Filep. 2004. Conformational immunity. Cell 124: 783–801. rearrangement in C-reactive protein is required for proinflammatory actions on 2. Akira, S., and K. Takeda. 2004. Toll-like receptor signalling. Nat. Rev. Immunol. human endothelial cells. Circulation 109: 2016–2022. 4: 499–511. 29. Zouki, C., M. Beauchamp, C. Baron, and J. G. Filep. 1997. Prevention of in vitro 3. Janeway, C. A., and R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. neutrophil adhesion to endothelial cells through shedding of L-selectin by C-re- by guest on September 26, 2021 Immunol. 20: 197–216. active protein and peptides derived from C-reactive protein. J. Clin. Invest. 100: 4. Hansson, G. K., and P. Libby. 2006. The immune response in atherosclerosis: a 522–529. double-edged sword. Nat. Rev. Immunol. 6: 508–519. 30. Hayden, M. S., and S. Gosh. 2004. Signaling to NF-␬B. Genes Dev. 18: 5. Zhu, J., A. A. Quyyumi, J. E. Norman, G. Csako, M. A. Waclawiw, 2195–2224. G. M. Shearer, and S. E. Epstein. 2000. Effects of total pathogen burden on 31. Spertini, O., F. W. Luscinskas, G. S. Kansas, J. M. Munro, J. D. Griffin, coronary artery disease risk and C-reactive protein levels. Am. J. Cardiol. 85: M. A. Gimbrone, Jr., and T. F. Tedder. 1991. Leukocyte adhesion molecule-1 140–146. (LAM-1, L-selectin) interacts with an inducible endothelial cell ligand to support 6. Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, leukocyte adhesion. J. Immunol. 147: 2565–2573. K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll-like receptor 32. Ishii, K. J., and S. Akira. 2006. Innate immune recognition of, and regulation by, recognizes bacterial DNA. Nature 408: 740–745. DNA. Trends Immunol. 27: 525–532. 7. Krieg, A. M. 2002. CpG motifs in bacterial DNA and their immune effects. Annu. 33. Akhtar, M., J. L. Watson, A. Nazli, and D. M. McKay. 2003. Bacterial DNA Rev. Immunol. 20: 709–760. evokes epithelial IL-8 production by a MAPK-dependent, NF-␬B-independent 8. Schwartz, D. A., T. J. Quinn, P. S. Thorne, S. Sayeed, A.-K. Yi, and A. M. Krieg. pathway. FASEB J. 17: 1319–1321. 1997. CpG motifs in bacterial DNA cause inflammation in the lower respiratory 34. Stacey, K., G. Young, F. Clark, D. Sester, T. Roberts, S. Naik, M. J. Sweet, and tract. J. Clin. Invest. 100: 68–73. D. Hume. 2003. The molecular basis for the lack of immunostimulatory activity 9. Lehtiniemi, J., P. J. Karhunen, S. Goebeler, S. Nikkari, and S. T. Nikkari. 2005. of vertebrate DNA. J. Immunol. 170: 3614–3620. Identification of different bacterial DNAs in human coronary arteries. Eur. 35. Yi, A. K., R. Tuetken, T. Redford, M. Waldschmidt, J. Kirsch, and A. M. Krieg. J. Clin. Invest. 35: 13–16. 1998. CpG motifs in bacterial DNA activate leukocytes through the pH-depen- 10. Ratanarat, R., S. Cazzavillan, Z. Ricci, M. Rassu, C. Segala, M. de Cal, D. Cruz, dent generation of reactive oxygen species. J. Immunol. 160: 4755–4761. V. Corradi, S. Manfro, E. Roessler, N. Levin, and C. Ronco. 2007. Usefulness of a molecular strategy for the detection of bacterial DNA in patients with severe 36. Ha¨cker, G., V. Redecke, and H. Ha¨cker. 2002. Activation of the immune system sepsis undergoing continuous renal replacement therapy. Blood Purif. 25: by bacterial CpG-DNA. Immunology 105: 245–251. 106–111. 37. Yoshimori, T., A. Yamamoto, Y. Morijama, M. Futai, and Y. Tashiro. 1991. ϩ 11. Hayashi, F., T. K. Means, and A. D. Luster. 2003. Toll-like receptors stimulate Bafilomycin A1, a specific inhibitor of vacuolar-type H -ATPase, inhibits acid- human neutrophil function. Blood 102: 2660–2669. ification and protein degradation in lysosomes of cultured cells. J. Biol. Chem. 12. Trevani, A. S., A. Chorny, G. Salamone, M. Vermeulen, R. Gamberale, 266: 17707–17712. J. Schettini, S. Raiden, and J. Geffner. 2003. Bacterial DNA activates human 38. Brinkmann, M. M., E. Spooner, K Hoebe, B. Beutler, H. L. Ploegh, and neutrophils by a CpG-independent pathway. Eur. J. Immunol. 33: 3164–3174. Y. M. Kim. 2007. The interaction between ER membrane protein UNC93B and 13. Jo´zsef, L., T. Khreiss, and J. G. Filep. 2004. CpG motifs in bacterial DNA delay TLR3, 7, and 9 is crucial for TLR signaling. J. Cell Biol. 177: 265–275. apoptosis of neutrophil granulocytes. FASEB J. 18: 1776–1778. 39. Huang, X., R. P. Barrett, S. A. McClellan, and L. D. Hazlett. 2005. Silencing 14. Li, J., Z. Ma, Z.-L. Tang, T. Stevens, B. Pitt, and S. Li. 2004. CpG DNA- Toll-like receptor-9 in Pseudomonas aeruginosa keratitis. Invest. Ophthalmol. mediated immune response in pulmonary endothelial cells. Am. J. Physiol. 287: Vis. Sci. 46: 4209–4216. L552–L558. 40. Hochrein, H., B. Schlatter, M. O’Keeffe, C. Wagner, F. Schmitz, M. Schiemann, 15. Martin-Armas, M., J. Simon-Santamaria, I. Pettersen, U. Moens, B. Smedsrod, S. Bauer, M. Suter, and H. Wagner. 2004. Herpes simplex virus type-1 induces and B. Sveinbjornsson. 2008. Toll-like receptor 9 (TLR9) is present in murine IFN-␣ production via Toll-like receptor 9-dependent and –independent pathways. liver sinusoidal endothelial cells (LSEC) and mediates the effect of CpG-oligo- Proc. Natl. Acad. Sci. USA 101: 11416–11421. nucleotides. J. Hepatol. 44: 939–946. 41. Ishii, K. J., C. Coban, H. Kato, K. Takahashi, Y. Torii, F. Takeshita, H. Ludwig, 16. Fitzner, N., S. Claubergt, F. Essmann, J. Liebmann, and V. Kolb-Bachofen. 2008. G. Sutter, K. Suzuki, H. Hemmi, et al. 2006. Toll-like receptor-independent an- Human skin endothelial cells can express all 10 TLR genes and respond to re- tiviral response induced by double-stranded B-form DNA. Nat. Immunol. 7: spective ligands. Clin. Vaccine Immunol. 15: 138–146. 40–48. 4394 BACTERIAL DNA AND NEUTROPHIL-ENDOTHELIAL ADHESION

42. Takaoka, A., Z. C. Wang, M. K. Coi, H. Yanai, H. Negishi, T. Ban, Y. Lu, 48. Ye, X., J. Ding, X. Zhou, G. Chen, and S. F. Liu. 2008. Divergent roles of M. Miyagishi, T. Kodama, K. Honda, et al. 2007. DAI (DLM-1/ZBP1) is a endothelial NF-␬B in multiple organ injury and bacterial clearance in mouse cytosolic DNA sensor and an activator of innate immune response. Nature 448: models of sepsis. J. Exp. Med. 205: 1303–1315. 501–505. 49. Plitas, G., B. M. Burt, H. M. Nguyen, Z. M. Bamboat, and R. P. Dematteo. 2008. 43. Marin, V., C. Famarier, S. Gres, S. Kaplanski, M. S. Su, C. A. Dinarello, and Toll-like receptor 9 inhibition reduces mortality in polymicrobial sepsis. J. Exp. G. Kaplanski. 2001. The p38 mitogen-activated protein kinase pathway plays a Med. 205: 1277–1283. critical role in thrombin-induced endothelial chemokine production and leukocyte recruitment. Blood 98: 667–673. 50. Yasuda, H., A. Leelahavanichkul, S. Sunoda, James W. Dear, Y. Takahashi, 44. Rahman, A., K. N. Anwar, S. Uddin, N. Xu, R. D. Ye, L. C. Platanias, and S. Ito, X. Hu, H. Zhou, K. Doi, R. Childs, et al. 2008. Chloroquine and inhibition A. B. Malik. 2001. Protein kinase C-␦ regulates thrombin-induced ICAM-1 gene of Toll-like receptor 9 protect from sepsis-induced acute kidney injury. expression in endothelial cells via activation of p38 mitogen-activated protein Am. J. Physiol. 294: F1050–F1058. kinase. Mol. Cell Biol. 21: 5554–5565. 51. Bhan, U., N. W. Lukacs, J. J. Osterholzer, M. W. Newstead, X. Zeng, 45. Luster, A. D. 1998. Chemokines: chemotactic cytokines that mediate inflamma- T. A. Moore, T. R. McMillan, A. M. Krieg, S. Akira, and T. J. Standiford. 2007. tion. N. Engl. J. Med. 338: 436–445. TLR9 is required for protective innate immunity in Gram-negative bacterial 46. Gerszten, R. E., E. A. Garcia-Zepeda, Y. C. Lim, M. Yoshida, H. A. Ding, pneumonia: a role for dendritic cells. J. Immunol. 179: 3937–3946. M. A. Gimbrone, Jr., A. D. Luster, F. W. Luscinskas, and A. Rosenzweig. 1999. MCP-1 and IL-8 trigger firm adhesion of monocytes to vascular endothelium 52. Albinger, B., S. Dahlberg, A. Sandgren, F. Wartha, K. Beiter, H. Katsuragi, under flow conditions. Nature 398: 718–723. S. Akira, S. Normark, and B. Henriques-Normark. 2007. Toll-like receptor 9 acts 47. Liu, S. F., and A. B. Malik. 2006. NF-␬B activation as a pathologic mechanism at an early stage in host defence against pneumococcal infection. Cell. Microbiol. of septic shock and inflammation. Am. J. Physiol. 290: L622–L645. 9: 633–644. Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021