Decrease of Intracellular Chloride Concentration Promotes Endothelial Cell Inflammation by Activating Nuclear Factor-κB PathwayNovelty and Significance
Recent evidence suggested that ClC-3 channel/antiporter is involved in regulation of nuclear factor (NF)-κB activation. However, the mechanism explaining how ClC-3 modulates NF-κB signaling is not well understood. We hypothesized that ClC-3-dependent alteration of intracellular chloride concentration ([Cl−]i) underlies the effect of ClC-3 on NF-κB activity in endothelial cells. Here, we found that reduction of [Cl−]i increased tumor necrosis factor-α (TNFα)-induced expression of intercellular adhesion molecule 1 and vascular cell adhesion molecule 1 and adhesion of monocytes to endothelial cells (P<0.05; n=6). In Cl− reduced solutions, TNFα-evoked IκB kinase complex β and inhibitors of κBα phosphorylation, inhibitors of κBα degradation, and NF-κB nuclear translocation were enhanced. In addition, TNFα and interleukin 1β could activate an outward rectifying Cl− current in human umbilical vein endothelial cells and mouse aortic endothelial cells. Knockdown or genetic deletion of ClC-3 inhibited or abolished this Cl− conductance. Moreover, Cl− channel blockers, ClC-3 knockdown or knockout remarkably reduced TNFα-induced intercellular adhesion molecule 1 and vascular cell adhesion molecule 1expression, monocytes to endothelial cell adhesion, and NF-κB activation (P<0.01; n=6). Furthermore, TNFα-induced vascular inflammation and neutrophil infiltration into the lung and liver were obviously attenuated in ClC-3 knockout mice (P<0.01; n=7). Our results demonstrated that decrease of [Cl−]i induced by ClC-3-dependent Cl− efflux promotes NF-κB activation and thus potentiates TNFα-induced vascular inflammation, suggesting that inhibition of ClC-3-dependent Cl− current or modification of intracellular Cl− content may be a novel therapeutic approach for inflammatory diseases.
ClC-3 has been found to be ubiquitously expressed in almost all eukaryotic cells, which functions as anion channel at cell plasma membrane or as Cl−/H+ antiporter in intracellular vesicles.1–5 Previous studies have demonstrated that ClC-3 plays an important role in the regulation of a variety of physiological activities, including cell volume, proliferation, differentiation, migration, apoptosis, insulin secretion, neuron excitability, and synaptic activation.4,6–13 Alterations of ClC-3 have been proposed to be associated with the development of macrophagic foam cell formation,14 carotid artery neointima formation,15 hypertension-induced cerebrovascular remodeling,16 and ischemia/reperfusion-induced myocardial damage.17
Interestingly, recent study in vascular smooth muscle cell has documented that several cytokines, including tumor necrosis factor-α (TNFα) and interleukin 1β, could activate chloride conductance, and this chloride current is dependent on ClC-3 expression.18 Moreover, ClC-3 has been found to be an essential regulator of nuclear factor (NF)-κB signaling, because ClC-3 deficiency obviously diminished NF-κB activity in mouse aortic smooth muscle cells.19 These results suggested that ClC-3 may play a critical role in inflammatory response. However, the direct link between ClC-3 and inflammation and the mechanism of how ClC-3 regulates NF-κB signaling remain elusive.
Chloride is the primary anion in the extracellular fluid. Chloride movement across the cell plasma membrane has been suggested to be involved in regulating cell volume, transepithelial fluid transport, smooth muscle cell contraction, and synaptic transmission.4,8,9,20,21 Notably, recent accumulating evidence has demonstrated that intracellular chloride concentration ([Cl−]i) was dynamically regulated. Obvious changes in [Cl−]i have been observed in hippocampal neurons during development, in T cells undergoing apoptosis, and in macrophages during foam cell formation.12,14,22 Moreover, previous work in Jurkat T cells showed that depletion of intracellular chloride blocked UV-C-induced cytochrome c release from mitochondria and thus inhibited cell apoptosis.22 These results suggested that the alterations of [Cl−]i play a critical role in a variety of physiological and pathological processes.
ClC-3 has been found to be an essential regulator of [Cl−]i5,16; however, whether the alteration of [Cl−]i contributes to the functional roles of ClC-3 is not clear. Our present study, therefore, aimed to investigate the effects of [Cl−]i, as well as ClC-3 genetic deficiency on TNFα-induced inflammatory response and NF-κB activation in endothelial cells. Our results demonstrated that decrease of [Cl−]i underlies, at least in part, the proinflammatory effects of ClC-3-dependent Cl− efflux.
An expanded Materials and Methods section is available in the online-only Data Supplement.
All of the experimental procedures were approved by the Sun Yat-Sen University Committee for Animal Research and conformed to the Guide for the Care and Use of Laboratory Animals of the National Institute of Health in China. ClC-3 heterozygous mice (ClC-3+/−) were kindly provided by Dr Dean Burkin (Nevada Transgenic Center, University of Nevada School of Medicine). ClC-3 knockout (KO) (ClC-3−/−) mice were prepared as described previously.7,23,24 Genotypes of the mice were examined by polymerase chain reaction on tail DNA (Figure S1 in the online-only Data Supplement).
Human umbilical vein endothelial cells (HUVECs) were isolated and cultured as described previously.25 The study protocol was approved by the medical research ethics committee of Sun Yat-Sen University. Informed consent was obtained from all subjects, and the experiments were conducted according to the principles expressed in the Declaration of Helsinki.
Perforated whole-cell patch experiments were performed as described previously.18 Adhesion of monocytes to endothelial cells was examined by using Calcein-AM labeled human acute monocytic leukemia cells or mouse monocytes. Quantitative real–time-polymerase chain reaction was performed using SYBR green fluorescence. Immunohistochemistry was performed by using the streptavidin-biotin-peroxidase complex system, according to the manufacturer’s instructions (SABC peroxidase kit). Myeloperoxidase activity was measured as described previously.26 Small interfering RNA transfection, Western blot, and [Cl−]i measurement were preformed as described previously.5,16,27 Chloride-reduced medium was prepared by replacing chloride with gluconate.
All data were expressed as mean±SEM. Statistical analysis was determined by an unpaired 2-tailed Student t test or 1-way ANOVA followed by Bonferroni multiple comparison post hoc test with a 95% CI. Values of P<0.05 were considered significant.
Reduction of [Cl−]i Potentiated TNFα-Induced Inflammatory Response and NF-κB Activation in HUVECs
We found that TNFα (10 ng/mL) treatment decreased [Cl−]i from 33.3±2.6 (n=52) to 23.1±1.9 mmol/L (n=30) in HUVECs (P<0.05). To examine whether the drop of [Cl−]i is involved in TNFα-induced inflammation, we prepared the cell culture medium with reduced chloride concentration to decrease the intracellular chloride content and investigated the inflammatory response in these solutions. In HUVECs, medium Cl− and low Cl− solution decreased the [Cl−]i to 26.8±2.1 (n=36) and 16.5±1.6 mmol/L (n=40), respectively (Figure S2). In the medium Cl− or low Cl− solution, TNFα-induced expression of intercellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1) was enhanced obviously compared with that in normal Cl− solution (Figure 1A and 1B and Figure S3; n=4–6). Meanwhile, TNFα-induced adhesion of human acute monocytic leukemia cell monocytes to HUVECs was also increased in Cl− reduced solution (Figure 1C; n=6).
We next investigated whether reduction of [Cl−]i affected NF-κB activation. Nuclear translocation of NF-κB subunit p65 was triggered within 30 minutes after TNFα stimulation in normal Cl− solution (Figure 1D; n=5). In low Cl− medium, p65 translocation started from 15 minutes and reached a maximum at 30 minutes after TNFα treatment (Figure 1E; n=6).
These findings suggested that reduction of [Cl−]i enhanced NF-κB activation and thus increased TNFα-induced inflammation.
Lowering [Cl−]i Promoted NF-κB Activation through Modulating IKKβ-IκBα Signaling Pathway
To explore the mechanism how decreases of [Cl−]i enhanced NF-κB activation, we investigated the effects of Cl− reduced medium on TNFα-induced degradation of inhibitors of κBα (IκBα) and phosphorylation of IκBα and IκB kinase complex β (IKKβ). Our results showed that TNFα induced IκBα degradation from 15 minutes and reached a maximum at 30 minutes in normal Cl− solution. In low Cl− solution, TNFα induced more significant degradation of IκBα, which started from 5 minutes and reached a maximum at 15 minutes (Figure 2A, n=6). TNFα induced phosphorylation of IκBα in low Cl− solution from 5 minutes after treatment, which preceded the increase of IκBα phosphorylation in normal Cl− solution (Figure 2B, n=6). In addition, TNFα-induced IKKβ phosphorylation in low Cl− solution also preceded the onset of IKKβ phosphorylation in normal Cl− solution. Moreover, IKKβ phosphorylation in low Cl− solution maintained longer than that in normal Cl− solution (Figure 2C, n=5). These results indicated that the increases of IKKβ and IκBα phosphorylation and IκBα degradation contributed to the enhanced inflammatory response in low Cl− solution.
TNFα Activated a ClC-3-Dependent Chloride Current in Endothelial Cells
In HUVECs, perfusion of TNFα (10 ng/mL) induced an outwardly rectifying current under isotonic solutions. This current showed slow inactivation at positive potentials (≥ +60 mV). The reversal potential of this current was −0.4±1.8 mV, which was near to the equilibrium potential for Cl− (0 mV) in our experimental conditions (Figure S4). Reduction of extracellular Cl− concentration from 133 to 44 mmol/L shifted the reversal potential to 25.3±2.1 mV, indicating that this current is mainly carried by Cl− (n=6). TNFα-activated chloride current was remarkably inhibited by the chloride channel blockers 4,4’-diisothiocyanatostilbene- 2,2’-disulphonic acid (DIDS; 100 μmol/L), 5-Nitro-2-(3-henylpropylamino) benzoic Acid (NPPB; 100 μmol/L), and tamoxifen (10 µmol/L) (Figure S4; n=6).
One recent study in vascular smooth muscle cells suggested that cytokine-activated chloride current is dependent on ClC-3 expression.18 Here, we also found that knockdown of ClC-3 with ClC-3 small interfering RNA transfection remarkably decreased TNFα-activated chloride current in HUVECs (Figure 3A and Figures S5 and S6A). Moreover, our results showed that TNFα can activate an outwardly rectifying chloride current in mouse aortic endothelial cells (MAECs) isolated from ClC-3+/+ mice; however, this chloride current was not observed in ClC-3−/− MAECs (Figure 3B and Figure S6B). These results demonstrated that ClC-3 is necessary for TNFα-activated chloride currents in endothelial cells. This ClC-3-dependent chloride current was also observed after interleukin-1β (Figure S7) and angiotensin II (data not shown) treatment.
Chloride Channel Blockers Inhibited TNFα-Induced Inflammatory Response in HUVECs
To understand the functions of TNFα-activated chloride current in endothelial cells, we examined the effects of the chloride channel blockers on TNFα-evoked inflammation. In HUVECs, DIDS (100 µmol/L), NPPB (100 µmol/L), or tamoxifen (10 µmol/L) pretreatment significantly inhibited TNFα-induced expression of ICAM-1 and VCAM-1 both at mRNA (Figure S8) and protein levels (Figure S9A and S9B; n=6 in each group). In addition, DIDS, NPPB, or tamoxifen remarkably reduced TNFα-induced adhesion of monocyte to HUVECs (Figure S9C and S9D; n=5). These observations suggested that TNFα-activated chloride current is involved in endothelial cell inflammation.
ClC-3 Deficiency Attenuated TNFα-Induced Inflammatory Response in Endothelial Cells
ClC-3 is required for TNFα-activated chloride current, so we next examined whether TNFα-induced inflammation was altered in ClC-3-deficient endothelial cells. In HUVECs, knockdown of ClC-3 had no significant effects on basal ICAM-1 and VCAM-1 expression; however, ClC-3 knockdown remarkably attenuated TNFα-induced expression of ICAM-1 and VCAM-1 (Figure 4A and 4B, n=6; Figure S10A and S10B; n=4) and adhesion of monocytes to HUVECs (Figure 4C; n=6).
To further determine the functional role of ClC-3 in vascular inflammation, we compared TNFα-induced inflammatory response in MAECs isolated from ClC-3+/+ mice with those from ClC-3−/− mice. The expression of ICAM-1 and VCAM-1 and the adhesion of monocyte to MAECs in basal condition were not significantly different between the 2 groups. However, ClC-3 KO dramatically attenuated TNFα-evoked increases of ICAM-1 and VCAM-1 expression and monocyte to MAEC adhesion (Figure 4 and Figure S10C and S10D; n=4–6). Moreover, we found that heterogenous expression of ClC-3 in ClC-3−/− MAECs restored TNFα-induced expression of ICAM-1 and VCAM-1 and adhesion of monocyte to MAECs (Figure S11; n=5).
These findings highlighted that ClC-3 plays an important role in modulating TNFα-induced endothelial cell inflammation.
ClC-3 Deficiency Decreased NF-κB Activation in Endothelial Cells
In HUVECs, TNFα treatment for 30 minutes induced remarkable translocation of p65 subunit to the nuclei (Figure 5). Knockdown of ClC-3 decreased p65 in the nuclei after TNFα incubation (Figure 5A and Figure S12A). To further confirm the effects of ClC-3 expression on NF-κB activation, we compared the TNFα-induced p65 nuclear translocation in ClC-3−/− MAECs with that in ClC-3+/+ MAECs. Consistent with the results in ClC-3 knockdown HUVECs, ClC-3 KO drastically reduced TNFα-induced p65 translocation from the cytoplasma to the nuclei in MAECs (Figure 5B and Figure S12B). These data suggested that modulation of the NF-κB activation underlies, at least in part, the proinflammatory effects of ClC-3 in endothelial cells.
ClC-3 Deletion Reduced Vascular Inflammation In Vivo
Similar to the results in MAECs in vitro, the basal expression levels of ICAM-1 and VCAM-1 in the aorta were no differences between ClC-3−/− mice and their ClC-3+/+ littermates. After intraperitoneal injection of TNFα (30 µg/kg) for 72 hours, the expression of ICAM-1 and VCAM-1 in aorta was remarkably increased in ClC-3+/+ mice. However, TNFα-induced increases of these adhesion molecules in ClC-3−/− mice were reduced obviously (Figure 6 and Figures S13 and S14; n=5–6). Moreover, histological examination revealed that the inflammation in lung and liver was dramatically attenuated in ClC-3−/− mice (Figure S15). Myeloperoxidase activity, an indicator of neutrophil infiltration, in the lung tissue was also significantly decreased in ClC-3−/− mice (Figure S15; n=6). The data further supported the critical role of ClC-3 in the regulation of vascular inflammatory response.
Chloride was once supposed to be an inert ion; however, recent growing evidence suggested that [Cl−]i is dynamically modulated, and alteration of intracellular Cl− is involved in several physiological processes, such as T-cell apoptosis and neuron excitability.12,22 In HUVECs, we observed that TNFα treatment decreased [Cl−]i. Reduction of [Cl−]i dramatically potentiated TNFα-induced expression of ICAM-1 and VCAM-1 and adhesion of monocytes to HUVECs. The data suggested that reduction of [Cl−]i underlies, at least in part, TNFα-induced inflammatory response. Our findings were consistent with a previous study in renal cortical thick ascending limb of Henle cells, which reported that the expression of cyclooxygenase 2, a critical enzyme for inflammatory response, was upregulated in low Cl− solution.28 These results together supported that the Cl− plays an essential role in inflammation.
NF-κB signaling pathway is rapidly activated after exposure to proinflammatory inducer, and a variety of proinflammatory cytokines and adhesion molecules have been proposed to be the direct targets of NF-κB.29,30 So NF-κB was thought to be a critical mediator of inflammatory disorders. In resting state, NF-κB is held in the cytoplasma with low basal transcriptional activity through interaction with IκB proteins. On stimulation, the IKKs rapidly induce IκB phosphorylation at N-terminal sites. Phosphorylated IκB then undergoes ubiquitin-dependent degradation, which exposes the nuclear localization sequence of NF-κB, causing its nuclear translocation.29,30 In this study, we found that reduction of [Cl−]i increased TNFα-induced IKKβ and IκBα phosphorylation, promoted TNFα-induced IκBα degradation, and enhanced NF-κB nuclear translocation, indicating that [Cl−]i is involved in regulation of the activity of the NF-κB signaling pathway. Our present findings are the first to our knowledge to demonstrate that NF-κB signaling pathway can be activated by decrease of [Cl−]i. Our results here, together with recent evidence demonstrating that the activities of several intracellular kinases, ion channels, and transporters are dependent on or regulated by Cl−, suggested that Cl− may function as a key intracellular second messenger.8,28,31,32
Chloride channel is one of the major routes for chloride transport across the cell membrane. Notably, previous work in mouse aortic smooth muscle cells has shown that TNFα and interleukin-1β could activate a chloride conductance, and this current is dependent on ClC-3 expression.18 Consistent with this study, we found that TNFα, interleukin-1β, and angiotensin II could activate a chloride current in endothelial cells. In HUVECs transfected with ClC-3 small interfering RNA, this chloride conductance was decreased obviously. Moreover, this current disappeared in ClC-3−/− MAECs. These results indicated that ClC-3 is required for the activation of this chloride current in endothelial cells.
In the vessels, ClC-3-dependent chloride current has been documented to be a critical regulator of cell volume, proliferation, migration, and apoptosis in vascular smooth muscle cells.8,9,13,15 The expression of ClC-3 has been detected in endothelial cells.33 However, its functions remain elusive. ClC-3 is required for cytokines-induced chloride current in endothelial cells, so we examined the effects of this chloride conductance on vascular inflammation. Our results showed that inhibition of the chloride current with chloride channel blockers or knockdown of ClC-3 remarkably reduced TNFα-induced ICAM-1 and VCAM-1 expression and monocytes to HUVECs adhesion. Moreover, we further observed that TNFα-induced endothelial inflammation was attenuated obviously in MAECs isolated from ClC-3−/− mice. Heterogenous expression of ClC-3 in ClC-3−/− MAECs restored TNFα-induced inflammatory responses. Furthermore, TNFα-induced ICAM-1 and VCAM-1 expression in aorta and neutrophil infiltration into the lung and liver were drastically reduced in ClC-3−/− mice compared with their wild-type littermates. Importantly, we found that ClC-3 deficiency profoundly suppressed TNFα-induced nuclear translocation of p65 subunit of NF-κB. Our results were in agreement with a recent study that showed that ClC-3 KO inhibited NF-κB activity in mouse aortic smooth muscle cells.19 All these findings indicated that ClC-3-dependent chloride conductance is involved in regulating NF-κB activation and thus the inflammatory responses.
Our present study provided evidence that ClC-3-dependent Cl− efflux contributes to TNFα-induced endothelial cell inflammation. ClC-3-dependent Cl− efflux evokes decrease of [Cl−]i, which underlies the proinflammatory effects of ClC-3-dependent Cl− conductance by activating the NF-κB pathway. The data suggested that modulation of intracellular Cl− content may be a novel strategy to prevent inflammatory diseases.
Sources of Funding
This work was supported by National Basic Research Program of China (2009CB521903), National Natural Science Foundation of China (30973536 and 30873060), Natural Science Foundation of Guangdong Province (No. 8151008901000009), and the Fundamental Research Funds for the Central Universities (09ykpy81).
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.112.198648/-/DC1.
- Received May 10, 2012.
- Revision received June 2, 2012.
- Accepted August 22, 2012.
- © 2012 American Heart Association, Inc.
- Zhou JG,
- Ren JL,
- Qiu QY,
- He H,
- Guan YY
- Chu X,
- Filali M,
- Stanic B,
- Takapoo M,
- Sheehan A,
- Bhalla R,
- Lamb FS,
- Miller FJ Jr.
- Liu YJ,
- Wang XG,
- Tang YB,
- Chen JH,
- Lv XF,
- Zhou JG,
- Guan YY
- Matsuda JJ,
- Filali MS,
- Moreland JG,
- Miller FJ,
- Lamb FS
- Miller FJ Jr.,
- Filali M,
- Huss GJ,
- Stanic B,
- Chamseddine A,
- Barna TJ,
- Lamb FS
- Heimlich G,
- Cidlowski JA
- Yamamoto-Mizuma S,
- Wang GX,
- Liu LL,
- Schegg K,
- Hatton WJ,
- Duan D,
- Horowitz TL,
- Lamb FS,
- Hume JR
- Shi XL,
- Wang GL,
- Zhang Z,
- Liu YJ,
- Chen JH,
- Zhou JG,
- Qiu QY,
- Guan YY
- Haas M,
- McBrayer D,
- Lytle C
- Moriguchi T,
- Urushiyama S,
- Hisamoto N,
- Iemura S,
- Uchida S,
- Natsume T,
- Matsumoto K,
- Shibuya H
- Barakat AI,
- Leaver EV,
- Pappone PA,
- Davies PF
Novelty and Significance
What Is New?
Reduction of [Cl−]i activates IKKβ-IκBα-NF-κB signaling pathway and promotes endothelial cell inflammatory response.
ClC-3 deficiency inhibits NF-κB activation and attenuates vascular inflammation.
What Is Relevant?
Vascular inflammation contributes to the development of hypertension and hypertension-associated cardiovascular diseases.
Inhibition of ClC-3-dependent chloride efflux reduces vascular inflammation.
Our findings demonstrate that ClC-3-dependent Cl− efflux induces decrease of [Cl−]i, which underlies the proinflammatory effects of ClC-3-dependent Cl− conductance by activation of NF-κB pathway. The data suggested that inhibition of the ClC-3-dependent Cl− current or modulation of intracellular Cl− content may be a novel strategy to prevent inflammatory diseases.