Intracellular RIG-I Signaling Regulates TLR4-Independent Endothelial Inflammatory Responses to Endotoxin

Jill Moser, Peter Heeringa, Rianne M. Jongman, Peter J. Zwiers, Anita E. Niemarkt, Rui Yan, Inge A. de Graaf, Ranran Li, Erzsébet Ravasz Regan, Philipp Kümpers, William C. Aird, Geerten P. van Nieuw Amerongen, Jan G. Zijlstra, Grietje Molema and Matijs van Meurs
J Immunol June 1, 2016, 196 (11) 4681-4691; DOI: https://doi.org/10.4049/jimmunol.1501819

Abstract

Sepsis is a systemic inflammatory response to infections associated with organ failure that is the most frequent cause of death in hospitalized patients. Exaggerated endothelial activation, altered blood flow, vascular leakage, and other disturbances synergistically contribute to sepsis-induced organ failure. The underlying signaling events associated with endothelial proinflammatory activation are not well understood, yet they likely consist of molecular pathways that act in an endothelium-specific manner. We found that LPS, a critical factor in the pathogenesis of sepsis, is internalized by endothelial cells, leading to intracellular signaling without the need for priming as found recently in immune cells. By identifying a novel role for retinoic acid–inducible gene-I (RIG-I) as a central regulator of endothelial activation functioning independent of TLR4, we provide evidence that the current paradigm of TLR4 solely being responsible for LPS-mediated endothelial responses is incomplete. RIG-I, as well as the adaptor protein mitochondrial antiviral signaling protein, regulates NF-κB–mediated induction of adhesion molecules and proinflammatory cytokine expression in response to LPS. Our findings provide essential new insights into the proinflammatory signaling pathways in endothelial cells and suggest that combined endothelial-specific inhibition of RIG-I and TLR4 will provide protection from aberrant endothelial responses associated with sepsis.

Introduction

Sepsis is the leading cause of death worldwide, and despite advances in early detection, mortality remains high mainly because the pathophysiology is still unresolved. Based on in vitro and experimental animal studies, TLR4 emerged as a promising target for therapeutic strategies for the treatment of sepsis, despite the fact that TLR4 was not associated with the development or the severity of sepsis in patients (1). Moreover, sepsis-induced organ failure was shown to be mediated by TLR4-independent signaling pathways depending on the organ investigated, and the role of TLR4 was shown to vary depending on the experimental animal model used (2–6). Not unexpectedly, clinical studies adopting TLR4 antagonists such as eritoran failed to show significant clinical benefits in patients with sepsis and were stopped prematurely (7, 8). The failure of these antagonists may have been due to the existence of an as yet unidentified TLR4-independent and/or signaling redundancy mechanisms.

Research investigating sepsis-mediated organ dysfunction has, until now, mainly focused on the deranged immune system as the principal cause for mortality. However, the endothelium is also an active contributor because loss of microvascular endothelial integrity, an early event in sepsis, is associated with exaggerated endothelial activation, disturbances in blood flow, vascular leakage, and other derangements synergistically contributing to multiple organ failure (9–11). Therapies aimed to inhibit endothelial inflammatory responses are expected to attenuate sepsis-associated multiple organ failure (12).

LPS, also known as endotoxin, is the key structural component of the cell wall of Gram-negative bacteria and a critical factor in the pathogenesis of sepsis. Endothelial cells are activated by LPS, resulting in the production of inflammatory cytokines, upregulation of adhesion molecules, loss of endothelial barrier integrity, leukocyte recruitment, and ultimately cellular injury resulting in organ failure (10, 13, 14). Concomitantly, LPS activates monocytes and macrophages to stimulate the production of proinflammatory cytokines, which in turn modulate endothelial responses. The current paradigm suggests that TLR4 is solely responsible for cellular responses induced by LPS (15, 16). However, recent findings in immune cells describe the existence of TLR4-independent LPS signaling pathways (17–19)

Retinoic acid–inducible gene I (RIG-I) is an intracellular receptor belonging to the DExD/H box family of RNA helicases that are best known for recognizing RNA species in the host cytoplasm (20–22). However, RIG-I has other intracellular functions unrelated to its role as a virus sensor (23–25). A functional role for RIG-I in endotoxin-induced endothelial responses has never been reported. The failure of TLR4-blocking drugs to improve survival among severe sepsis patients in clinical trials (7, 8) motivated us to investigate the intracellular receptor RIG-I for its possible role in mediating endothelial activation and leukocyte adhesion in response to LPS.

In this study, we reveal a new and unanticipated role for the cytosolic receptor RIG-I and its adaptor protein mitochondrial antiviral signaling protein (MAVS) as regulators of endothelial inflammatory responses to LPS. Importantly, we uncover a pathway by which RIG-I signaling regulates downstream NF-κB–mediated endothelial activation and inflammatory responses promoting leukocyte adhesion to the endothelium. Notably, our findings delineate a functional role for RIG-I–MAVS signaling in LPS-mediated endothelial activation that is distinct from TLR4-mediated endothelial responses to LPS. As such, it may have important implications for the simultaneous use of RIG-I and TLR4 as targets for combination strategies to dampen excessive endothelial inflammation associated with sepsis organ failure.

Materials and Methods

Mice

C57BL/6 (Harlan Laboratories) and 129 SV (Charles River Laboratories) mice were housed in a specific pathogen–free facility, maintained on chow and water ad libitum, and housed in temperature-controlled chambers (24°C) with a 12 h light/dark cycle. Male C57BL/6 mice were challenged via i.p. injection of LPS 0.5 mg/kg Escherichia coli, serotype O26:B6 (15,000 endotoxin units/g), or 1 mg/kg serotype O111:B4 as indicated (Sigma-Aldrich, St. Louis, MO). Mice were sacrificed 4, 8, and 24 h after LPS. Vehicle-injected mice were administered the same volume of saline i.p. Prior to sacrifice, mice were anesthetized with O2/isoflurane, blood was withdrawn via aortic puncture, and organs were harvested and snap frozen in liquid nitrogen and stored at −80°C until analysis. Additionally, C57BL/6 mice were i.p injected with 0.5 mg rat anti-mouse neutrophil (NIMPR14, Hycult Biotech) to selectively deplete neutrophils before LPS administration. Normal rat IgG (0.5 mg, Sigma-Aldrich) was injected as control. Twenty-four hours later mice were i.p. injected with LPS (0.5 mg/kg) and sacrificed 8 h later (26). Polymicrobial sepsis was induced by cecal ligation and puncture (CLP) as previously described (27). All animals were sacrificed under anesthesia (isoflurane/O2), after which heparinized blood was collected. All organs were harvested and immediately snap frozen on liquid nitrogen and stored at −80°C until analysis. All experimental procedures were performed according to Dutch laws and international guidelines on animal experimentation and were approved by the Animal Ethics Committee of the University of Groningen.

Cell culture

HUVEC (Lonza, Breda, the Netherlands) and human aortic endothelial cells (HAEC; Life Technologies, Bleiswijk, the Netherlands) were cultured as described previously (28, 29). Conditionally immortalized human glomerular endothelial cells, a gift from S. Satchell (University of Bristol, Bristol, U.K.), were incubated for 24 h at 33°C, followed by 5 d at 37°C at 5% CO2/95% air before starting an experiment. Primary kidney cortex peritubular endothelial cells were isolated and cultured after collagenase treatment using anti-CD31 Ab-conjugated beads as described previously (30). Human immortalized glomerular endothelial cells, a gift from Dr. Simon Satchell, were cultured as previously described (31). Immortalized HL60 leukocytes, provided by G. Fey (University of Erlangen, Erlangen, Germany), were maintained in RPMI 1640 medium supplemented with 10% FBS. All cells were cultured at the University Medical Center Groningen Endothelial Cell Facility.

Endothelial cell treatment with LPS, TNF-α, triphosphate dsRN, and blocking agents

Cells were stimulated for 4 h with LPS (1 μg/ml) or TNF-α (Boehringer Ingelheim, Ingelheim, Germany) (10 ng/ml) unless indicated otherwise. LPS used in this study included E. coli serotype O26:B6, E. coli serotype O111:B4, FITC-conjugated LPS (E. coli serotype O111:B4) (Sigma-Aldrich), and ultrapure LPS E. coli serotype O111:B4 (InvivoGen, Toulouse, France). LPS serotype O26:B6 was used in experiments unless otherwise stated. Alternatively, cells were incubated with 5′ triphosphate dsRNA/LyoVec for either 4 or 24 h (InvivoGen). For TLR4 blocking experiments, HUVEC were pretreated with 20–25 μg/ml LEAF purified anti-human CD284 (TLR4) Ab (clone HTA 125) (BioLegend, San Diego, CA) 30 min before addition of LPS. For LPS inhibition experiments, HUVEC were pretreated with polymyxin B (InvivoGen) at different concentrations (50–400 μg/ml) 15 min before addition of LPS.

Small interfering RNA–mediated gene silencing

TLR4, RIG-I, MAVS, and RelA (NF-κB p65) were knocked down using FlexiTube small interfering RNA (siRNA) sequences for Tlr4, Ddx58, and Visa, respectively (Qiagen, Venlo, the Netherlands). Allstars negative control siRNA (Qiagen) was used as a negative control for all RNA interference experiments. Transfection was performed using Lipofectamine 2000 (Life Technologies) according to the manufacturer’s instructions. A second siRNA targeted to our genes of interest was used to confirm the results of siRNA experiments (Supplemental Fig. 1). Knockdown of our genes of interest did not diminish endothelial cell viability as assessed microscopically and by determining the total number of viable cells using flow cytometry (data not shown).

Gene expression analysis by quantitative RT-PCR

Cryosections of mouse kidney were prepared and subjected to laser microdissection as described before (26) Total RNA was isolated using a RNeasy Micro Plus kit (Qiagen) according to the manufacturer’s protocols. Total RNA was isolated from in vitro experiments and cryosections, and integrity was checked and cDNA synthesized as previously described (31). Quantitative RT-PCR (RT-qPCR) was performed using the ViiA 7 system (Applied Biosystems/Life Technologies). Duplicate real-time PCR analyses were executed for each sample, and the obtained threshold cycle (CT) values were averaged. Gene expression was normalized to the expression of housekeeping gene GAPDH, yielding the ∆CT value. The relative mRNA level was calculated by 2−∆CT. Assay-on-demand primers (Applied Biosystems) were used in this study and are described in Supplemental Table I. Analysis of primary kidney cortex endothelial cells was performed using Illumina expression arrays as described previously (31).

Western blot analysis

Endothelial cell lysates were prepared using RIPA buffer containing protease inhibitor mixture and phosphatase inhibitors (Roche, Almere, the Netherlands) and subjected to SDS-PAGE and immunoblot analysis according to standard procedures and using the following Abs: mouse anti-human RIG-I (Enzo Life Sciences, Raamsdonksveer, the Netherlands), rabbit anti-human MAVS (Bethyl Laboratories, Montogomery, AL), rabbit anti-human VCAM-1 (Santa Cruz Biotechnology, Heidelberg, Germany), rabbit anti-IκBα (06-494), and rabbit anti-actin (MAB1501) (Merck Millipore, Darmstadt, Germany). Protein bands were visualized by Luminata Forte Western HRP substrate (Merck Millipore) using HRP-conjugated secondary Abs and exposure to x-ray film Super KX (Fujifilm, Rotterdam, the Netherlands). Protein amounts were quantified by calculating the intensity ratio of the protein of interest/loading control (actin) corrected to background using Quantity One (Bio-Rad Laboratories software).

ELISA

Human IL-6 and IL-8 were determined in the supernatants of HUVEC where indicated using either IL-6 and IL-8 DuoSet (R&D Systems, Oxon, U.K.) or IL-6 and IL-8 ELISA MAX ELISA kits (BioLegend), following the manufacturers’ protocols.

Endothelial leukocyte adhesion assay

HUVEC monolayers were transfected with TLR4, RIG-I, MAVS, or scrambled siRNA. Forty-eight hours after transfection, cells were challenged with LPS for 4 h. Leukocytes were labeled with Hoechst 33342 (Life Technologies) for 10 min, washed, and resuspended in RPMI 1640 containing 1% FBS. The labeled cells were allowed to attach to HUVEC monolayers for 1 h. After extensive washing, the cells were trypsinized and resuspended in RPMI 1640 containing 1% FBS, and the number of adhered HL60 cells was determined by flow cytometry using a MACSQuant analyzer (Miltenyi Biotec, Leiden, the Netherlands).

Precision-cut human kidney tissue slice preparation and incubation

Human kidney tissue was obtained from tumor-free surgical waste from three male patients (age between 60 and 66 y) subjected to renal carcinoma surgery, with normal renal function. Precision-cut tissue slices were cut and prepared as previously described (26).

Immunofluorescence

Immunofluorescence microscopy was performed essentially as described previously (32). Briefly, cells were washed in cold PBS and fixed by 2% paraformaldehyde (20 min on ice) followed by 0.2% Triton X-100 and incubation for 5 min at room temperature (RT). Cells were washed twice with PBS and incubated with 3% BSA in PBS for 30 min at RT. Subsequently, the cells were incubated with primary Abs diluted in wash buffer (PBS containing 0.5% BSA and 0.05% Tween 20) for 1 h at RT. Primary Abs were mouse anti-TLR4 (clone HTA 125) (BioLegend) or rabbit anti-p65(NF-κB) (Cell Signaling Technology, Danvers, MA). The cells were washed three times with was buffer and incubated with secondary Abs, goat anti-mouse biotin or goat anti-rabbit biotin (SouthernBiotech), for 1 h at RT. The cells were again washed three times with wash buffer before incubating with Alexa Fluor 555–streptavidin (Life Technologies) for 1 h at RT. Cells were mounted in Aqua-PolyMount mounting medium containing DAPI (1.5 μg/ml) (Polysciences, Warrington, PA). Fluorescence images were taken with a Leica DM/RXA fluorescence microscope equipped with a Leica DFC450C digital camera (Leica, Wetzlar, Germany) and Leica LAS V4.2 image overlay software. All images were taken with equal exposure times.

Immunohistochemistry

Immunohistochemical staining was performed on formalin-fixed, paraffin-embedded kidney tissue slides that were deparaffinized and rehydrated. Endogenous peroxidase activity was blocked, and Ags were retrieved by boiling the sections for 15 min in 10 mM Tris/1 mM EDTA (pH 9.0). The tissue sections were subsequently incubated with mouse mAb against RIG-I for 1 h at RT, followed by incubation with the secondary Ab rabbit anti-mouse HRP for 45 min, after which peroxidase activity was detected with 3-amino-9-ethylcarbazole (Dako). Sections were counterstained with Mayer’s hematoxylin (Merck, Darmstadt, Germany).

Statistical analysis

Statistical analysis of results was performed by a two-tailed unpaired Student t test, assuming equal variances to compare two replicate means, or one-way ANOVA followed by Bonferroni post hoc analysis to compare multiple replicate means. All statistical analyses were performed using GraphPad Prism software v.5. Differences were considered significant when p < 0.05.

Results

Endothelial activation is partly dependent on TLR4 signaling

First, we investigated whether TLR4 is an essential molecular determinant for LPS-mediated activation in endothelial cells because recent reports found TLR4-independent LPS signaling in macrophages (17–19). Upon blocking TLR4 using a neutralizing Ab at concentrations that were enough to completely inhibit TLR4 responses in macrophages (13), we found only partial inhibition of LPS-mediated endothelial activation as demonstrated by persistent high upregulation of E-selectin and VCAM-1 mRNA (Fig. 1A). An explanation for the lack of inhibition may be due to the intracellular localization of TLR4 in endothelial cells (13), which we confirmed by microscopic analysis (Fig. 1B). To examine the role of intracellular TLR4, we next investigated the effect of TLR4 knockdown on endothelial responses to LPS. LPS-induced upregulation of endothelial adhesion molecules, as well as inflammatory cytokines and chemokines, at both mRNA and protein levels were still only partially inhibited in TLR4 knockdown cells (Fig. 1C–E). Importantly, this effect was independent of LPS serotype and purity. A second siRNA targeting TLR4 was used to confirm these results (Supplemental Fig. 1). In line with the effects on inflammatory phenotype, leukocyte adherence to LPS-treated TLR4 knockdown endothelial cells was only partly inhibited using LPS serotypes 026:B6 and 0111:B4, but completely inhibited when stimulating with ultrapure LPS (Fig. 1F). Confirming intracellular localization, FITC-labeled LPS was confined to the cytoplasmic compartment in quiescent state (Fig. 1G). From these data we conclude that TLR4 signaling plays only a partial role in the regulation of LPS-mediated endothelial activation and that alternative signaling mechanisms must be required to propagate LPS-mediated inflammatory responses.

FIGURE 1.
FIGURE 1.

Endothelial responses to LPS are partially TLR4-independent. (A) mRNA expression levels of E-selectin and VCAM-1 in LPS-treated HUVEC pretreated with anti-TLR4 blocking Ab. (B) Fluorescent immunostaining for TLR4 (yellow) merged with DAPI nuclear staining (blue) of HUVEC treated with LPS for 4 h, original magnification ×400. (C) Percentage residual mRNA levels of genes in control and 4 h LPS-treated HUVEC transfected with either siScr or siTLR4. Different serotypes of LPS were used. (D) siScr- and siTLR4-transfected HUVEC were vehicle-treated (−) or LPS-treated (+) for 4 h and subjected to immunoblot analysis for VCAM-1 protein. Actin was used as a reference for protein loading. The immunoblots are representative of three independent experiments. Densitometric analysis of protein bands represent the mean ± SD of three independent experiments. (E) IL-6 and IL-8 cytokine concentrations in the medium from siScr- and siTLR4-transfected HUVEC treated with vehicle (−) or LPS (+) for 4 h, as determined by ELISA. (F) Percentage of HL60 leukocytes adherent to confluent monolayers of HUVEC that were transfected with siScr or siRIG-I and treated with LPS for 4 h. Different serotypes of LPS were used. (G) Confocal microscopy of HUVEC 4 h after addition of FITC-labeled LPS (O111:B4) and stained with DAPI, original magnification ×400. Bars represent the mean ± SD. Data are representative of three independent experiments. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

RIG-I mediates LPS-induced endothelial activation and leukocyte adhesion

Because LPS localized in the cytoplasm of endothelial cells, we hypothesized that a cytosolic receptor may regulate TLR4-independent endothelial activation induced by LPS. One such receptor is DExD/H box RNA helicase RIG-I. RIG-I is expressed in various cell types and tissues and induced by stimuli such as IFN-γ, LPS, or IL-1β (25). This suggests that it may be involved in other cellular processes in addition to its role in antiviral responses. RIG-I expression was upregulated in response to LPS in HUVEC, confirming previously reported findings (33), as well as in HAEC, and in human immortalized glomerular endothelial cells (Fig. 2A, 2B). The temporal upregulation of RIG-I gene expression upon LPS stimulation was paralleled by that of the adhesion molecule VCAM-1 (Fig. 2A). In primary renal microvascular endothelial cells, RIG-I mRNA expression was retained under culture conditions (Fig. 2C), and when exposed to laminar shear stress, endothelial RIG-I expression remained unchanged (Fig. 2D). Hence, altered flow states do not affect RIG-I mRNA levels.

FIGURE 2.
FIGURE 2.

RIG-I mediates endothelial inflammation and leukocyte adhesion. (A) Kinetics of RIG-I and VCAM-1 mRNA expression in HAEC and HUVEC upon treatment with LPS as assessed by RT-qPCR. (B) RIG-I mRNA expression in human immortalized glomerular endothelial cells upon LPS stimulation for 2 and 4 h. (C) Microvascular endothelial cells were isolated from mouse kidney cortex and placed in culture for 3 or 5 d. t = 0 represents the initial endothelial cell fraction immediately after isolation without being placed in culture. RIG-I mRNA expression relative to vascular endothelial cadherin (VEcad) was determined from Illumina array data. Bars represent the mean ± SD of three independent isolations. (D) HUVEC were kept under laminar shear stress (LSS, 20 dyne/cm2) for 48 h and next subjected to 8 h flow cessation before RIG-I mRNA expression analysis by RT-qPCR. (E) Percentage residual mRNA levels of indicated genes in 4 h LPS-treated HUVEC transfected with either siScr or siRIG-I. Different serotypes of LPS were used to stimulate the cells as specified. (F) Whole-cell lysates from siScr- and siRIG-I–transfected HUVEC were vehicle-treated (−) or LPS-treated (+) for 4 h and subjected to immunoblot analysis for RIG-I and VCAM-1. Actin blotting was used as a reference for protein loading. The immunoblot is representative of three independent experiments. Densitometric analyses of protein bands represent the mean ± SD of three independent experiments. (G) IL-6 and IL-8 cytokine concentrations in the cell culture medium of siScr- and siRIG-I–transfected HUVEC treated with vehicle (−) or LPS (+) for 4 h were determined by ELISA. (H) mRNA expression levels of genes in HUVEC treated with LPS (1 μg/ml) or transfected with RIG-I–specific agonist, 5′ 3p-dsRNA/LyoVec (2 μg/ml) for 4 and 24 h. (I) Percentage of HL60 leukocytes that adhered to confluent monolayers of HUVEC that were transfected with siScr, siRIG-I, or siNF-κB p65 and 48 h later stimulated with vehicle (control) or LPS for 4 h. Bars represent mean ± SD. All data are representative of three independent experiments. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

Next, we evaluated the consequences of exogenously abrogated RIG-I levels on the induction of endothelial adhesion molecules and inflammatory cytokines by LPS. The data clearly show that LPS-induced upregulation of endothelial adhesion molecules E-selectin, VCAM-1, and ICAM-I was markedly inhibited in the absence of RIG-I (Fig. 2E). Additionally, upregulation of inflammatory cytokines IL-6 and IL-8, as well as cyclooxygenase-2 and CXCL10, was inhibited when RIG-I was knocked down in HUVEC treated with LPS (Fig. 2E), which was confirmed at the protein level (Fig. 2F, 2G). A second siRNA targeting RIG-I was also found to significantly inhibit LPS-mediated endothelial activation, confirming these results (Supplemental Fig. 1). Importantly, the protective effect of RIG-I knockdown was not dependent on LPS serotype or purity, thereby excluding endothelial activation as a result of LPS contaminants (Fig. 2E).

To further exclude the possibility that RNA contaminants of the LPS used were the underlying cause of the observed effects, we transfected HUVEC with the RIG-I agonist triphosphate dsRNA (3p-dsRNA) (20). This did not induce endothelial activation nor did it induce proinflammatory endothelial responses at 4 h after transfection (Fig. 2H). In contrast, RIG-I mRNA expression was highly upregulated, with concomitant induction of CXCL10 mRNA levels, only at 24 h after transfection with 3p-dsRNA, which was in line with previous findings (34). However, despite RIG-I upregulation, the expression of adhesion molecules E-selectin, VCAM-1, and ICAM-1, as well as of IL-6 and IL-8, was not significantly induced 24 h after transfection with 3p-dsRNA. Likewise, HUVEC stimulated with extracellular poly(I:C) for 4 h resulted in a large increase in RIG-I mRNA expression but hardly induced endothelial activation (data not shown). Additionally, RIG-I upregulation and the induction of endothelial activation was abolished by the presence of polymyxin B, an inhibitor of LPS, demonstrating that LPS, not impurities in the LPS, is responsible for the RIG-I upregulation (Supplemental Fig. 2).

To further elucidate the role of endothelial RIG-I, we investigated the functional consequences of RIG-I knockdown on LPS-induced leukocyte adhesion to the endothelium. In line with the gene expression and protein data, significantly fewer HL60 cells adhered to LPS-treated RIG-I–deficient HUVEC compared with LPS-treated control cells, an effect similar to that induced by defective NF-κB signaling via RelA knockdown (Fig. 2I). Taken together, these data strongly imply a specific role for RIG-I as a regulator of leukocyte recruitment by controlling crucial endothelial inflammatory responses to LPS independent of its RNA sensing functions.

RIG-I and TLR4 are two separate pathways that regulate endothelial activation in response to LPS

We next explored whether TLR4 and RIG-I are part of the same pathway or function independently to regulate endothelial responses to LPS. To examine this, we employed siRNA to inhibit TLR4 and/or RIG-I expression in HUVEC. LPS-induced upregulation of ICAM-1 may be regulated by both TLR4 and RIG-I to a similar extent, and double knockdown did not result in further inhibition of ICAM-1 expression, suggesting a common downstream mechanism. In contrast, single knockdown of RIG-I or TLR4 resulted in significant inhibition of E-selectin, VCAM-1, IL-6, and IL-8 mRNA expression in LPS-treated cells, and double knockdown almost completely abolished LPS endothelial proinflammatory responses (Fig. 3A). In line with the gene expression data, significantly fewer HL60 cells adhered to LPS-treated RIG-I or TLR4-deficient HUVEC compared with LPS-treated control cells. Knockdown of both RIG-I and TLR4 resulted in abolishment of leukocyte adhesion to the LPS-treated endothelium (Fig. 3B). These findings suggest that RIG-I and TLR4 function independently in the regulation of LPS-mediated inflammatory responses in endothelial cells.

FIGURE 3.
FIGURE 3.

RIG-I regulates TLR4-independent responses to LPS. (A) Gene expression levels of E-selectin, VCAM-1, ICAM-1, IL-6, IL-8, TLR4, and RIG-I as measured by RT-qPCR in vehicle (control) and 4 h LPS-treated HUVEC transfected with either siScr, siTLR4, siRIG-I, or double knockdown using siTLR4 and siRIG-I simultaneously. (B) Percentage of HL60 leukocytes that adhered to confluent monolayers of HUVEC that were, prior to LPS stimulation for 4 h, transfected with siScr, siRIG-I, siTLR4, or both siRIG-I and siTLR4 (n = 3). Data are representative of two independent experiments. Bars represent mean ± SD. ***p ≤ 0.001.

RIG-I upregulation in the microvascular compartments of the kidney

To determine the in vivo relevance of endothelial RIG-I signaling in vivo, we stained RIG-I protein in kidney sections from mice that were treated with LPS for 4 h. As shown in Fig. 4A, RIG-I localized in the microvascular endothelial compartments. In endotoxemia, renal RIG-I mRNA was upregulated after LPS administration irrespective of the LPS serotype used (Fig. 4B, 4C). Similar to our observations in in vitro cell culture systems, this was paralleled by changes in the expression of the adhesion molecule VCAM-1 and the renal damage marker neutrophil gelatinase–associated lipocalin (NGAL; Fig. 4B). Moreover, upregulation of RIG-I expression occurred in all microvascular compartments of the mouse kidney (Fig. 4D). To exclude the possibility that the increase in renal RIG-I mRNA levels by LPS was due to infiltrating neutrophils expressing RIG-I (35), we depleted neutrophils prior to LPS administration (26). This depletion did not block the upregulation of renal RIG-I mRNA by LPS (Fig. 4E), thus verifying that renal RIG-I upregulation after systemic LPS administration is a response of cells intrinsic to the kidney. Besides this model of acute kidney injury, renal RIG-I, VCAM-1, and NGAL gene expression were also increased at 16 h after CLP (Fig. 4F). That microvascular RIG-I is upregulated in mice is suggestive of its involvement in aberrant endothelial responses associated with sepsis-mediated acute kidney injury. To translate our findings to the human situation, we made use of precision-cut human kidney slices. Stimulation of the tissue slices with LPS for 4 h resulted in upregulation of renal RIG-I and concurrently endothelial activation molecules VCAM-1 and E-selectin (Fig. 4G). Thus, in vivo murine models of endotoxemia and polymicrobial sepsis as well as ex vivo human renal tissue slices exposed to LPS suggest a possible role for RIG-I in the microvascular pathogenesis associated with sepsis.

FIGURE 4.
FIGURE 4.

RIG-I upregulation in the microvascular compartments of the kidney. (A) RIG-I staining of control mice and mice treated with LPS (1 mg/kg) for 4 h. A, arteriole; G, glomerulus; V, venule. Scale bars, 50 μm. (B) C57BL/6 mice were i.p. injected with vehicle (control) or LPS (0.5 mg/kg) and the kidneys were harvested at the time points indicated. Renal mRNA expression of RIG-I, VCAM-1, and NGAL was determined by RT-qPCR. Bars represent mean ± SD of five mice per group. (C) C57BL/6 mice were given an i.p. injection of LPS (serotype O111:B4) (1 mg/kg) or vehicle (control) and their kidneys were harvested at 4 h after LPS challenge. Renal mRNA expression of RIG-I was determined by RT-qPCR. (D) mRNA expression levels of RIG-I in the different mouse kidney microvascular compartments under basal conditions (right panel) and at 4 h after LPS challenge (left panel) (n = 3). (E) C57BL/6 mice were pretreated with anti-NIMP Ab or control anti-IgG Ab 24 h prior to LPS (O26:B6) challenge (0.5 mg/kg). Kidneys were harvested 8 h after LPS challenge, and the mRNA levels of RIG-I were determined by RT-qPCR. Control IgG Ab did not affect neutrophil count as determined by flow cytometry (data not shown). Bars represent mean ± SD of five mice per group. (F) Renal mRNA expression of RIG-I, VCAM-1, and NGAL was determined by RT-qPCR in the kidneys of sham (control) or CLP-operated mice. Mice were sacrificed 16 h after the CLP procedure. Bars represent mean ± SD of five mice per group. (G) RIG-I, E-selectin, and VCAM-1 mRNA expression in human kidney tissue in a controlled, ex vivo precision-cut tissue slice culture system. After incubation for 4 h in medium with vehicle or LPS (50 μg/ml), slices were harvested and processed for mRNA expression analysis (n = 3). Bars represent mean ± SD and depict combined data of three independent experiments. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

MAVS also mediates endothelial inflammatory signaling

Upon viral activation of macrophages, intracellular RIG-I binds to its adaptor protein MAVS, which promotes the activation of NF-κB, AP-1, and various IFN regulatory factors (36, 37). We therefore examined whether MAVS is also required to propagate RIG-I–mediated signaling in endothelial cells. In HUVEC the upregulation of E-selectin, VCAM-1, IL-6, IL-8, but not ICAM-1, by LPS stimulation was significantly inhibited when MAVS was silenced (Fig. 5A–C), a finding that was not dependent on LPS serotype or purity (Supplemental Fig. 3). A second siRNA targeted to MAVS was used to confirm these results (Supplemental Fig. 1). Functionally, siRNA knockdown of MAVS resulted in significantly less HL60 leukocytes adhering to LPS-treated endothelial monolayers (Fig. 5E). Thus, along with RIG-I, MAVS is required to regulate endothelial activation.

FIGURE 5.
FIGURE 5.

MAVS is required for RIG-I–mediated inflammatory signaling in endothelial cells. (A) Gene expression levels of MAVS, endothelial adhesion molecules, and cytokines as measured by RT-qPCR in control and 4 h LPS-treated HUVEC transfected with either siScr or siMAVS. (B) Whole-cell lysates from siScr- and siMAVS-transfected HUVEC were vehicle (−) or LPS (+) treated for 4 h and subjected to immunoblot analysis for MAVS and VCAM-1. Actin was used as a reference for protein loading. Representative immunoblots from three independent experiments are shown. Densitometric analysis of protein bands represent the mean ± SEM of three independent experiments. (C) IL-6 and IL-8 cytokine concentrations in cell culture medium from siScr- and siMAVS-transfected HUVEC treated with vehicle (−) or LPS (+) for 4 h were determined by ELISA. (D) Percentage of HL60 leukocytes that adhered to confluent monolayers of HUVEC that were transfected with siScr or siMAVS and treated with vehicle (control) or LPS for 4 h. Bars represent mean ± SD. All data are representative of three independent experiments. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

LPS-induced endothelial activation is not caused by autocrine IFN-β production in vitro

To investigate the downstream RIG-I signaling pathways mediating endothelial activation, we examined the expression levels of IFN-β in endothelial cells and investigated whether IFN-β expression was induced upon LPS and/or TNF-α stimulation. IFN-β is known to be produced and secreted as a result of RIG-I signaling, and it may bind to the IFN receptor in an autocrine manner to direct JAK-STAT signaling and the expression of IFN-stimulated genes (38). We found very low expression levels of IFN-β and no significant increase in IFN-β gene expression in HUVEC stimulated with either LPS or TNF-α (data not shown). Thus, RIG-I–mediated endothelial cell activation by LPS is not a result of autocrine IFN-β production.

RIG-I mediates TNF-α–induced endothelial activation whereas MAVS only mediates E-selectin and IL-8 expression in response to TNF-α

Based on our results, we wondered whether RIG-I was required for the regulation of endothelial activation by a different inflammatory stimulus. Endothelial RIG-I expression was upregulated by TNF-α stimulation, which was paralleled by VCAM-1 upregulation (Fig. 6A). Similar to LPS, RIG-I silencing inhibited TNF-α–induced endothelial activation and proinflammatory cytokine expression, thus identifying a more central role for RIG-I as a regulator of endothelial inflammation (Fig. 6B). In contrast, MAVS knockdown only resulted in inhibition of TNF-α–induced E-selectin and IL-8 expression (Fig. 6B). Taken together, these data suggest that RIG-I also mediates TNF-α–induced expression of inflammatory genes. However, MAVS only mediates TNF-α–induced induction of E-selectin and IL-8.

FIGURE 6.
FIGURE 6.

RIG-I mediates TNF-α–induced endothelial inflammation whereas MAVS only mediates E-selectin and IL-8 expression in response to TNF-α. (A) Kinetics of RIG-I and VCAM-1 mRNA expression in HUVEC and HAEC upon treatment with vehicle (control) and TNF-α (10 ng/ml) as assessed by RT-qPCR. (B) Gene expression levels of E-selectin, VCAM-1, ICAM-1, IL-6, IL-8, RIG-I, and MAVS as measured by RT-qPCR in 4 h vehicle (control) and TNF-α–treated HUVEC transfected with either siScr, siRIG-I, or siMAVS (n = 4). Bars represent mean ± SD of two independent experiments. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

RIG-I–MAVS signaling mediates NF-κB proinflammatory signaling in response to LPS

Because NF-κB plays an important role in regulating endothelial responses to LPS (12, 39), we examined whether RIG-I exerts its effects via NF-κB activation. IκBα levels rapidly decreased in siScr cells after 30 min LPS exposure, whereas siRIG-I– or siMAVS–treated HUVEC were devoid of IκBα degradation (Fig. 7A). Translocation of the NF-κB p65 subunit from the cytoplasm to the nucleus was also significantly reduced in endothelial cells that lack RIG-I or MAVS compared with control cells (Fig. 7B). Thus, RIG-I–MAVS signaling regulates LPS-mediated endothelial activation via the NF-κB pathway.

FIGURE 7.
FIGURE 7.

RIG-I–MAVS signaling mediates NF-κB proinflammatory signaling in response to LPS. (A) The amounts of RIG-I, MAVS, and IκBα protein levels in siScr-, siRIG-I–, and siMAVS-transfected HUVEC after vehicle (−) or 30 min LPS (+) stimulation were determined by immunoblotting whole-cell extracts. Actin was used as a reference for protein loading. Representative immunoblots from three independent experiments are shown. Densitometric analyses of protein bands represent the mean ± SEM of three independent experiments. ***p ≤ 0.001. (B) HUVEC transfected with either siScr, siRIG-I, or siMAVS and treated with vehicle (control) or LPS for 4 h were subjected to immunostaining for p65 (RelA) protein (left panel). The percentage of cells with p65 NF-κB subunit localized in the nucleus was quantified (right panel). At least 250 cells from each sample were analyzed. Bars represent mean + SD, representative of three independent experiments. ***p ≤ 0.001.

Discussion

Endothelial cells are crucial sentinels in the onset of inflammatory responses (10, 40). Previous studies found that blockade of endothelial-specific NF-κB signaling was sufficient to abolish endothelial inflammatory activation and protect mice from endotoxemia and sepsis (12). Hence, therapies targeted to inhibit endothelial inflammatory responses are expected to attenuate sepsis-associated multiple organ failure. The existing paradigm proposes that TLR4 is solely responsible for endothelial responses induced by LPS. In the present study, we provide evidence that for optimal endothelial activation induction by LPS, two cytosolic receptors are required. We have identified RIG-I as a key molecule in a TLR4-independent pathway that functions concomitantly with TLR4 to regulate endothelial inflammatory responses to LPS. In vivo, RIG-I is located in the microvasculature of the kidney in mice, and in mouse models and in a human ex vivo tissue model of endotoxemia and sepsis, RIG-I is temporarily upregulated. In addition to RIG-I, MAVS acts to promote endothelial activation induced by LPS. Overall, our findings suggest a much more complex organization of events promoting endothelial inflammation than has to date been considered. Moreover, our findings may explain, in part, the failure of TLR4-blocking drugs in recent clinical trials aimed at improving survival among sepsis patients and may have important implications for future treatment strategy development.

In this study, we found that LPS was internalized into endothelial cells, which distinctly differed from the internalization of LPS in macrophages in that no priming was required. The precise mechanism of LPS internalization into endothelial cells is not known but is thought to be mediated by a TLR4-independent scavenger receptor pathway (13, 41). Intracellular recognition of LPS may provide a defense mechanism to ensure that low levels of LPS, which are normally present in our blood and thus constantly exposing the endothelium, do not initiate endothelial inflammatory responses. In contrast, high concentrations of LPS in the setting of severe sepsis may lead to internalization, thereby eliciting endothelial activation. Hence, the level of exposure to LPS and whether it is internalized may represent thresholds for endothelial inflammatory activation. Based on this finding, we hypothesized that endothelial cytoplasmic pattern recognition receptors must mediate proinflammatory transcriptional responses to LPS. Intracellular TLR4 localization confirmed previous studies (13). However, in addition to its unique feature of being situated intracellularly in endothelial cells, TLR4 knockdown only partly inhibited LPS-induced endothelial activation and inflammatory cytokine production, and it had a limited inhibitory effect on leukocyte adhesion to the endothelium. These results led to the hypothesis that endothelial responses to LPS must be regulated by an additional intracellular mechanism.

Our studies revealed that LPS-induced endothelial activation and leukocyte adhesion to the endothelium are largely controlled by RIG-I and MAVS, an effect that was independent of E. coli LPS serotype. Under various conditions, RIG-I mRNA levels were upregulated. Additionally, exposure to ultrapure LPS hardly affected endothelial RIG-I mRNA levels, whereas RIG-I as well as MAVS knockdown still significantly inhibited ultrapure LPS-induced endothelial activation. This suggests that the presence of RIG-I and MAVS in the cytoplasm is sufficient to regulate proinflammatory responses to LPS and that transcriptional upregulation of RIG-I is not per se a prerequisite for endothelial activation to occur.

The present study has identified RIG-I as a rapid responder in endotoxemia and sepsis models in vivo and demonstrated that RIG-I plays a crucial role in mediating endothelial inflammation in vitro. Moreover, our results revealed that TLR4 and RIG-I represent two independent pathways that function concurrently to regulate NF-κB–mediated endothelial activation. In addition to LPS-mediated endothelial responses, a more fundamental role for RIG-I in the regulation of proinflammatory endothelial responses is plausible because we found that RIG-I also mediated TNF-α–induced endothelial activation. MAVS, alternatively, was found to only regulate E-selectin and IL-8 expression in TNF-α–stimulated HUVEC. Hence, RIG-I mediates TNF-α–induced expression of adhesion molecules VCAM-1 and ICAM-1, and IL-6 by a mechanism independent of MAVS, whereas TNF-α induction of E-selectin and IL-8 requires both RIG-I and MAVS. RIG-I may thus provide a valuable therapeutic target because it regulates both LPS and TNF-α–mediated endothelial inflammatory activation.

These discoveries fill an important gap in our knowledge of how endothelial cells respond to proinflammatory stimulation. Further studies are required to determine the specific roles of endothelial TLR4 and RIG-I in regulating endothelial responses to LPS in vivo that ideally make use of endothelial cell–specific conditional knockout mouse models. Based on the data generated in this study and in light of the urgent need for new therapies for the treatment of sepsis beyond organ function support, we propose that combined endothelial-specific inhibition of RIG-I and TLR4 will provide protection from aberrant endothelial responses associated with sepsis-mediated organ dysfunction.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Henk E. Moorlag of the University Medical Center Groningen Endothelial Cell Facility for excellent technical support. W. Schaafsma (Department of Neuroscience, Section of Medical Physiology, University Medical Center Groningen) is acknowledged for providing kidney tissue from LPS (E. coli O111:B4)–treated mice. We thank Kate McIntyre (University Medical Center Groningen) for editing the manuscript.

Footnotes

  • G.M. and M.v.M. initiated the study; J.M. conceived and directed the project; J.M., P.H., R.M.J., P.J.Z., A.E.N., R.L., R.Y., E.R.R., G.M., and M.v.M. designed and/or performed the experiments; J.M., R.M.J., P.J.Z., A.E.N., R.L., and E.R.R. analyzed the data; M.v.M. and P.K. carried out the in vivo studies; I.A.d.G. performed the ex vivo experiments; J.M., P.H., G.M., M.v.M., W.C.A., G.P.v.N.A., and J.G.Z. discussed the data; and J.M. and G.M. wrote the manuscript, which was read and commented on by all authors prior to finalization by J.M. and G.M.

  • This work was supported by the Research Foundation of the Department of Critical Care, University Medical Center Groningen, as well as by the Foundation for Improvement of Critical Care Water en Lucht.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    CLP
    cecal ligation and puncture
    HAEC
    human aortic endothelial cell
    MAVS
    mitochondrial antiviral signaling protein
    NGAL
    neutrophil gelatinase–associated lipocalin
    3p-dsRNA
    triphosphate dsRNA
    RIG-I
    retinoic acid–inducible gene-I
    RT
    room temperature
    RT-qPCR
    quantitative RT-PCR
    siRNA
    small interfering RNA.

  • Received August 12, 2015.
  • Accepted April 1, 2016.

References

  1. ↵
    1. Feterowski C.,
    2. K. Emmanuilidis,
    3. T. Miethke,
    4. K. Gerauer,
    5. M. Rump,
    6. K. Ulm,
    7. B. Holzmann,
    8. H. Weighardt
    . 2003. Effects of functional Toll-like receptor-4 mutations on the immune response to human and experimental sepsis. Immunology 109: 426–431.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Deng M.,
    2. M. J. Scott,
    3. P. Loughran,
    4. G. Gibson,
    5. C. Sodhi,
    6. S. Watkins,
    7. D. Hackam,
    8. T. R. Billiar
    . 2013. Lipopolysaccharide clearance, bacterial clearance, and systemic inflammatory responses are regulated by cell type-specific functions of TLR4 during sepsis. J. Immunol. 190: 5152–5160.
    OpenUrlAbstract/FREE Full Text
    1. Dear J. W.,
    2. H. Yasuda,
    3. X. Hu,
    4. S. Hieny,
    5. P. S. Yuen,
    6. S. M. Hewitt,
    7. A. Sher,
    8. R. A. Star
    . 2006. Sepsis-induced organ failure is mediated by different pathways in the kidney and liver: acute renal failure is dependent on MyD88 but not renal cell apoptosis. Kidney Int. 69: 832–836.
    OpenUrlCrossRefPubMed
    1. Cunningham P. N.,
    2. Y. Wang,
    3. R. Guo,
    4. G. He,
    5. R. J. Quigg
    . 2004. Role of Toll-like receptor 4 in endotoxin-induced acute renal failure. J. Immunol. 172: 2629–2635.
    OpenUrlAbstract/FREE Full Text
    1. Sha T.,
    2. Y. Iizawa,
    3. M. Ii
    . 2011. Combination of imipenem and TAK-242, a Toll-like receptor 4 signal transduction inhibitor, improves survival in a murine model of polymicrobial sepsis. Shock 35: 205–209.
    OpenUrlCrossRefPubMed
  6. ↵
    1. van Lieshout M. H.,
    2. T. van der Poll,
    3. C. van’t Veer
    . 2014. TLR4 inhibition impairs bacterial clearance in a therapeutic setting in murine abdominal sepsis. Inflamm. Res. 63: 927–933.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Rice T. W.,
    2. A. P. Wheeler,
    3. G. R. Bernard,
    4. J. L. Vincent,
    5. D. C. Angus,
    6. N. Aikawa,
    7. I. Demeyer,
    8. S. Sainati,
    9. N. Amlot,
    10. C. Cao,
    11. et al
    . 2010. A randomized, double-blind, placebo-controlled trial of TAK-242 for the treatment of severe sepsis. Crit. Care Med. 38: 1685–1694.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Opal S. M.,
    2. P. F. Laterre,
    3. B. Francois,
    4. S. P. LaRosa,
    5. D. C. Angus,
    6. J. P. Mira,
    7. X. Wittebole,
    8. T. Dugernier,
    9. D. Perrotin,
    10. M. Tidswell,
    11. et al,
    12. ACCESS Study Group
    . 2013. Effect of eritoran, an antagonist of MD2-TLR4, on mortality in patients with severe sepsis: the ACCESS randomized trial. JAMA 309: 1154–1162.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Lee W. L.,
    2. A. S. Slutsky
    . 2010. Sepsis and endothelial permeability. N. Engl. J. Med. 363: 689–691.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Shapiro N. I.,
    2. W. C. Aird
    . 2011. Sepsis and the broken endothelium. Crit. Care 15: 135.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Gomez H.,
    2. C. Ince,
    3. D. De Backer,
    4. P. Pickkers,
    5. D. Payen,
    6. J. Hotchkiss,
    7. J. A. Kellum
    . 2014. A unified theory of sepsis-induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics, and the tubular cell adaptation to injury. Shock 41: 3–11.
    OpenUrlPubMed
  12. ↵
    1. Ye X.,
    2. J. Ding,
    3. X. Zhou,
    4. G. Chen,
    5. S. F. Liu
    . 2008. Divergent roles of endothelial NF-κB in multiple organ injury and bacterial clearance in mouse models of sepsis. J. Exp. Med. 205: 1303–1315.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Dunzendorfer S.,
    2. H. K. Lee,
    3. K. Soldau,
    4. P. S. Tobias
    . 2004. Toll-like receptor 4 functions intracellularly in human coronary artery endothelial cells: roles of LBP and sCD14 in mediating LPS responses. FASEB J. 18: 1117–1119.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Shapiro N. I.,
    2. P. Schuetz,
    3. K. Yano,
    4. M. Sorasaki,
    5. S. M. Parikh,
    6. A. E. Jones,
    7. S. Trzeciak,
    8. L. Ngo,
    9. W. C. Aird
    . 2010. The association of endothelial cell signaling, severity of illness, and organ dysfunction in sepsis. Crit. Care 14: R182.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Hoshino K.,
    2. O. Takeuchi,
    3. T. Kawai,
    4. H. Sanjo,
    5. T. Ogawa,
    6. Y. Takeda,
    7. K. Takeda,
    8. S. Akira
    . 1999. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J. Immunol. 162: 3749–3752.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Poltorak A.,
    2. X. He,
    3. I. Smirnova,
    4. M. Y. Liu,
    5. C. Van Huffel,
    6. X. Du,
    7. D. Birdwell,
    8. E. Alejos,
    9. M. Silva,
    10. C. Galanos,
    11. et al
    . 1998. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282: 2085–2088.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Hagar J. A.,
    2. D. A. Powell,
    3. Y. Aachoui,
    4. R. K. Ernst,
    5. E. A. Miao
    . 2013. Cytoplasmic LPS activates caspase-11: implications in TLR4-independent endotoxic shock. Science 341: 1250–1253.
    OpenUrlAbstract/FREE Full Text
    1. Shi J.,
    2. Y. Zhao,
    3. Y. Wang,
    4. W. Gao,
    5. J. Ding,
    6. P. Li,
    7. L. Hu,
    8. F. Shao
    . 2014. Inflammatory caspases are innate immune receptors for intracellular LPS. Nature 514: 187–192.
    OpenUrlPubMed
  19. ↵
    1. Kayagaki N.,
    2. M. T. Wong,
    3. I. B. Stowe,
    4. S. R. Ramani,
    5. L. C. Gonzalez,
    6. S. Akashi-Takamura,
    7. K. Miyake,
    8. J. Zhang,
    9. W. P. Lee,
    10. A. Muszyński,
    11. et al
    . 2013. Noncanonical inflammasome activation by intracellular LPS independent of TLR4. Science 341: 1246–1249.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Hornung V.,
    2. J. Ellegast,
    3. S. Kim,
    4. K. Brzózka,
    5. A. Jung,
    6. H. Kato,
    7. H. Poeck,
    8. S. Akira,
    9. K. K. Conzelmann,
    10. M. Schlee,
    11. et al
    . 2006. 5′-Triphosphate RNA is the ligand for RIG-I. Science 314: 994–997.
    OpenUrlAbstract/FREE Full Text
    1. Yoneyama M.,
    2. M. Kikuchi,
    3. K. Matsumoto,
    4. T. Imaizumi,
    5. M. Miyagishi,
    6. K. Taira,
    7. E. Foy,
    8. Y. M. Loo,
    9. M. Gale Jr..,
    10. S. Akira,
    11. et al
    . 2005. Shared and unique functions of the DExD/H-box helicases RIG-I, MDA5, and LGP2 in antiviral innate immunity. J. Immunol. 175: 2851–2858.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Yoneyama M.,
    2. M. Kikuchi,
    3. T. Natsukawa,
    4. N. Shinobu,
    5. T. Imaizumi,
    6. M. Miyagishi,
    7. K. Taira,
    8. S. Akira,
    9. T. Fujita
    . 2004. The RNA helicase RIG-I has an essential function in double-stranded RNA-induced innate antiviral responses. Nat. Immunol. 5: 730–737.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Hatakeyama M.,
    2. T. Imaizumi,
    3. F. Terasaki,
    4. F. Mori,
    5. K. Tanji,
    6. F. Sato,
    7. H. Kijima,
    8. H. Suma,
    9. K. Wakabayashi,
    10. H. Yoshida,
    11. et al
    . 2007. Interferon-γ upregulates retinoic acid-inducible gene-I in human pericardial mesothelial cells. Acta Cardiol. 62: 553–557.
    OpenUrlCrossRefPubMed
    1. Kubota K.,
    2. H. Sakaki,
    3. T. Imaizumi,
    4. H. Nakagawa,
    5. A. Kusumi,
    6. W. Kobayashi,
    7. K. Satoh,
    8. H. Kimura
    . 2006. Retinoic acid-inducible gene-I is induced in gingival fibroblasts by lipopolysaccharide or poly IC: possible roles in interleukin-1β, -6 and -8 expression. Oral Microbiol. Immunol. 21: 399–406.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Liu F.,
    2. J. Gu
    . 2011. Retinoic acid inducible gene-I, more than a virus sensor. Protein Cell 2: 351–357.
    OpenUrlCrossRefPubMed
  26. ↵
    1. van Meurs M.,
    2. N. F. Kurniati,
    3. F. M. Wulfert,
    4. S. A. Asgeirsdottir,
    5. I. A. de Graaf,
    6. S. C. Satchell,
    7. P. W. Mathieson,
    8. R. M. Jongman,
    9. P. Kümpers,
    10. J. G. Zijlstra,
    11. et al
    . 2009. Shock-induced stress induces loss of microvascular endothelial Tie2 in the kidney which is not associated with reduced glomerular barrier function. Am. J. Physiol. Renal Physiol. 297: F272–F281.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. David S.,
    2. J. K. Park,
    3. Mv. Meurs,
    4. J. G. Zijlstra,
    5. C. Koenecke,
    6. C. Schrimpf,
    7. N. Shushakova,
    8. F. Gueler,
    9. H. Haller,
    10. P. Kümpers
    . 2011. Acute administration of recombinant Angiopoietin-1 ameliorates multiple-organ dysfunction syndrome and improves survival in murine sepsis. Cytokine 55: 251–259.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Li R.,
    2. A. Aslan,
    3. R. Yan,
    4. R. M. Jongman,
    5. J. Moser,
    6. P. J. Zwiers,
    7. H. E. Moorlag,
    8. J. G. Zijlstra,
    9. G. Molema,
    10. M. van Meurs
    . 2015. Histone deacetylase inhibition and IκB kinase/nuclear factor-κB blockade ameliorate microvascular proinflammatory responses associated with hemorrhagic shock/resuscitation in mice. Crit. Care Med. 43: e567–e580.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Leus N. G.,
    2. E. G. Talman,
    3. P. Ramana,
    4. P. S. Kowalski,
    5. T. E. Woudenberg-Vrenken,
    6. M. H. Ruiters,
    7. G. Molema,
    8. J. A. Kamps
    . 2014. Effective siRNA delivery to inflamed primary vascular endothelial cells by anti-E-selectin and anti-VCAM-1 PEGylated SAINT-based lipoplexes. Int. J. Pharm. 459: 40–50.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Jin E.,
    2. J. Liu,
    3. J. Suehiro,
    4. L. Yuan,
    5. Y. Okada,
    6. V. Nikolova-Krstevski,
    7. K. Yano,
    8. L. Janes,
    9. D. Beeler,
    10. K. C. Spokes,
    11. et al
    . 2009. Differential roles for ETS, CREB, and EGR binding sites in mediating VEGF receptor 1 expression in vivo. Blood 114: 5557–5566.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Kurniati N. F.,
    2. R. M. Jongman,
    3. F. vom Hagen,
    4. K. C. Spokes,
    5. J. Moser,
    6. E. R. Regan,
    7. G. Krenning,
    8. J. R. Moonen,
    9. M. C. Harmsen,
    10. M. M. Struys,
    11. et al
    . 2013. The flow dependency of Tie2 expression in endotoxemia. Intensive Care Med. 39: 1262–1271.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Moser J.,
    2. H. Kool,
    3. I. Giakzidis,
    4. K. Caldecott,
    5. L. H. Mullenders,
    6. M. I. Fousteri
    . 2007. Sealing of chromosomal DNA nicks during nucleotide excision repair requires XRCC1 and DNA ligase IIIα in a cell-cycle-specific manner. Mol. Cell 27: 311–323.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Imaizumi T.,
    2. S. Aratani,
    3. T. Nakajima,
    4. M. Carlson,
    5. T. Matsumiya,
    6. K. Tanji,
    7. K. Ookawa,
    8. H. Yoshida,
    9. S. Tsuchida,
    10. T. M. McIntyre,
    11. et al
    . 2002. Retinoic acid-inducible gene-I is induced in endothelial cells by LPS and regulates expression of COX-2. Biochem. Biophys. Res. Commun. 292: 274–279.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Asdonk T.,
    2. I. Motz,
    3. N. Werner,
    4. C. Coch,
    5. W. Barchet,
    6. G. Hartmann,
    7. G. Nickenig,
    8. S. Zimmer
    . 2012. Endothelial RIG-I activation impairs endothelial function. Biochem. Biophys. Res. Commun. 420: 66–71.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Berger M.,
    2. C. Y. Hsieh,
    3. M. Bakele,
    4. V. Marcos,
    5. N. Rieber,
    6. M. Kormann,
    7. L. Mays,
    8. L. Hofer,
    9. O. Neth,
    10. L. Vitkov,
    11. et al
    . 2012. Neutrophils express distinct RNA receptors in a non-canonical way. J. Biol. Chem. 287: 19409–19417.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Xu L. G.,
    2. Y. Y. Wang,
    3. K. J. Han,
    4. L. Y. Li,
    5. Z. Zhai,
    6. H. B. Shu
    . 2005. VISA is an adapter protein required for virus-triggered IFN-β signaling. Mol. Cell 19: 727–740.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Seth R. B.,
    2. L. Sun,
    3. C. K. Ea,
    4. Z. J. Chen
    . 2005. Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-κB and IRF3. Cell 122: 669–682.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Dixit E.,
    2. J. C. Kagan
    . 2013. Intracellular pathogen detection by RIG-I-like receptors. Adv. Immunol. 117: 99–125.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Kowalski P. S.,
    2. P. J. Zwiers,
    3. H. W. Morselt,
    4. J. M. Kuldo,
    5. N. G. Leus,
    6. M. H. Ruiters,
    7. G. Molema,
    8. J. A. Kamps
    . 2014. Anti-VCAM-1 SAINT-O-Somes enable endothelial-specific delivery of siRNA and downregulation of inflammatory genes in activated endothelium in vivo. J. Control. Release 176: 64–75.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Yano K.,
    2. P. C. Liaw,
    3. J. M. Mullington,
    4. S. C. Shih,
    5. H. Okada,
    6. N. Bodyak,
    7. P. M. Kang,
    8. L. Toltl,
    9. B. Belikoff,
    10. J. Buras,
    11. et al
    . 2006. Vascular endothelial growth factor is an important determinant of sepsis morbidity and mortality. J. Exp. Med. 203: 1447–1458.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Dunzendorfer S.,
    2. H. K. Lee,
    3. K. Soldau,
    4. P. S. Tobias
    . 2004. TLR4 is the signaling but not the lipopolysaccharide uptake receptor. J. Immunol. 173: 1166–1170.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section