G-protein-coupled receptor kinase-2 (GRK2) belongs to the GRK family of serine/threonine protein kinases critical in the regulation of G-protein-coupled receptors. Apart from this canonical role, GRK2 is also involved in several signaling pathways via distinct intracellular interactomes. In the present study, we examined the role of GRK2 in TNFα signaling in colon epithelial cell–biological processes including wound healing, proliferation, apoptosis, and gene expression. Knockdown of GRK2 in the SW480 human colonic cells significantly enhanced TNFα-induced epithelial cell wound healing without any effect on apoptosis/proliferation. Consistent with wound-healing effects, GRK2 knockdown augmented TNFα-induced matrix metalloproteinases (MMPs) 7 and 9, as well as urokinase plasminogen activator (uPA; factors involved in cell migration and wound healing). To assess the mechanism by which GRK2 affects these physiological processes, we examined the role of GRK2 in TNFα-induced MAPK and NF-κB pathways. Our results demonstrate that while GRK2 knockdown inhibited TNFα-induced IκBα phosphorylation, activation of ERK was significantly enhanced in GRK2 knockdown cells. Our results further demonstrate that GRK2 inhibits TNFα-induced ERK activation by inhibiting generation of reactive oxygen species (ROS). Together, these data suggest that GRK2 plays a critical role in TNFα-induced wound healing by modulating MMP7 and 9 and uPA levels via the ROS–ERK pathway. Consistent with in vitro findings, GRK2 heterozygous mice exhibited enhanced intestinal wound healing. Together, our results identify a novel role for GRK2 in TNFα signaling in intestinal epithelial cells.

Introduction

G-protein-coupled receptor kinases (GRKs) are serine/threonine kinases that include seven distinct proteins separated into three subfamilies based on their sequence homology and functional similarities. These include the retinal kinase (GRK1 and GRK7), the GRK2 (GRK2 and GRK3), and the GRK4 (GRK4, 5, and 6) subfamilies [1]. GRKs were initially described for their canonical regulation of G-protein-coupled receptor (GPCR) phosphorylation and desensitization. However, it is now clear that their role in cell signaling is not limited to this canonical function. GRK2 can interact with and regulate (or be regulated by) a wide variety of non-GPCRs and non-receptor substrates, including IGF-1R, insulin-R, EGFR, ERK, MEK, IκBα, and p38, which demonstrates its capacity to affect a diverse set of cellular functions through both phosphorylation-dependent and -independent mechanisms [2]. These non-canonical roles of GRK2 allow for this kinase to influence basic cellular processes such as inflammatory gene expression, cellular migration, and mitochondrial metabolism [3]. The role for GRK2 in these cellular functions has been extensively examined in cardiac, immune, and other cell types [2,4,5], but the role of GRK2 in intestinal epithelial cells is not well known.

Mucosal surfaces, including the intestinal tract, are lined by epithelial cells and are critical in defining and maintaining a barrier between the host and external environment. Dysregulation of this barrier integrity leads to inflammatory processes including inflammatory bowel disease (IBD) [69]. Intestinal epithelial cell healing (proliferation, differentiation, and migration) and other epithelial dynamics such as permeability become impaired in IBD [10]. In addition, patients with chronic IBD are also at increased risk for colitis-associated colon cancer. Thus, understanding the signaling mechanisms that regulate colonic epithelial cell biology is important for long-term drug development strategies for IBD and colitis-associated colon cancer.

IBD is associated with an increase in inflammatory cytokines, such as TNFα, and these cytokines have been shown to alter the ability of epithelial cells to repair damage to the monolayer. These changes are mediated by intracellular signaling through the regulation of NF-κB and MAPK pathways which affect epithelial cell apoptosis, proliferation, and migration from the crypts [11]. In previous studies, we and others showed that GRK2 is an important regulator of TNFα signaling in myeloid cells both in vitro and in vivo [1217]. However, the function of GRK2 in intestinal epithelial cells in the context of TNFα signaling processes is not known. Therefore, in the present study, we examined the role of GRK2 in colonic epithelial cells in terms of various cell biological responses including cell signaling pathways that are modulated by GRK2. Our studies underscore a critical role of GRK2 in intestinal epithelial cell wound healing in vitro. Part of this regulation occurs via the role of GRK2 in the TNFα-induced ROS–ERK pathway and subsequent MMP expression. Consistent with in vitro findings, we also demonstrate that intestinal wound healing is enhanced in GRK2 heterozygous knockout mice in vivo.

Materials and methods

Materials

Cell culture and knockdown of GRK2

Human colon epithelial cell line SW480 was obtained from ATCC and was maintained in Dulbecco's modified Eagle's medium with high glucose and supplemented with 10% (v/v) fetal bovine serum (Life Technologies) along with 5% penicillin/streptomycin (Life Technologies). These cells were grown at 37°C in a humidified 5% CO2 incubator. One hour prior to stimulation, cells were replenished with culture media without FBS and then stimulated with 20 ng/ml TNFα (PeproTech) for various time points as indicated. For knockdown of GRK2, 1 million cells were treated with 25 pmol of ‘SmartPool siRNA’ containing a mixture of four designed siRNAs targeting GRK2 or scrambled siRNA (Dharmacon) and electroporated using an Amaxa V electroporation instrument using the standard protocol provided by the supplier and solution V (Lonza) and program L-024. The cells were then incubated for 48 h to facilitate knockdown before being used for experiments.

Scratch wound-healing assay

SW480 cells were cultured in a 6-well plate and the confluent cells were used for an in vitro scratch wound assay [18]. Cells were treated with TNFα (20 ng/ml) or vehicle at time 0 h. To create the wound, a uniform scratch was created vertically in the culture dish using a 200 µl plastic pipette tip, and the plate was then incubated for 48 additional hours in the presence of mitomycin C (1 µg/ml) to inhibit proliferation. The scratch was assessed and photos of the wound were taken at 0, 16, 24, 32, and 48 h. Images were taken with the cellSens Software (Olympus), and the distance of the wound was calculated by measuring the distance between the leading edges of the wound using the ImageJ software. The effect of TNFα on wound closing was calculated by determining the amount of wound closed relative to the initial time point in the respective groups.

Proliferation assay

After siRNA transfection, SW480 cells were cultured in a 96-well plate at a concentration of 50 000/well for 24 or 48 h, and stimulated with TNFα (20 ng/ml) or vehicle for 24 or 48 h. The amount of proliferation was determined using the CyQuant Cell Proliferation Assay (Thermo Scientific). In short, after TNFα treatment, the media were removed and the plate was incubated at −80°C overnight. The following day, the plate was thawed and a DNA-binding dye in lysis buffer was added to each well and the fluorescence was read on a Tecan M2000 spectrophotometer at 485 nm excitation and 530 nm emission and compared with a standard curve.

Apoptosis assay

After siRNA transfection, SW480 cells were cultured in 6-well plates at a concentration of 1 million cells/well for 48 h and were treated with TNFα (20 ng/ml) for 24 or 48 h. Apoptosis was quantified using the Annexin V Apoptosis Detection Kit (eBioscience) according to the manufacturer's protocol. Cells were quantified using an LSR II flow cytometer and data were analyzed using the FlowJo software. Data were divided into four categories: Live (Annexin V ‘−’, Propidium Iodide (PI) ‘−’), Early Apoptotic (Annexin V ‘+’, PI ‘−’), Apoptotic (Annexin V ‘+’, PI ‘+’), and Necrotic/Dead (Annexin V ‘−’, PI ‘+’).

ROS assay

After siRNA transfection, SW480 cells were cultured in 6-well plates at a concentration of 1 million cells/well for 48 h and stimulated for 10 or 20 min with TNFα (20 ng/ml) or vehicle. ROS was measured using CellRox Green Reagent (Thermo Scientific). Briefly, the cells were pretreated with CellRox Green reagent (5 µM) for 30 min. The cells were then treated with TNFα and after the treatment washed three times with PBS. The cells were fixed with 3.7% formaldehyde for 15 min, washed three times with FACS buffer (0.5% BSA, 0.05% sodium azide, in 1× PBS), quantified using the LSR II flow cytometer, and analyzed using the FlowJo software. Data are expressed as mean fluorescent intensity.

Invasion assay

After siRNA transfection, SW480 cells were cultured in a 6-well plate to facilitate the siRNA-mediated knockdown of GRK2. After 48 h, the cells were removed from the plate and 20 000 cells were seeded in serum-free medium in the upper chamber of Transwell inserts coated with Matrigel (Cell BioLabs, Inc.), and cellular migration was determined per the manufacturer's protocol. In brief, serum-containing medium was inserted in the lower chamber and the cells were incubated for 48 h to allow for migration through the membrane in the presence or absence of TNFα (20 ng/ml). After 48 h, the migrated cells were detached using the manufacturer's detachment solution and lysed in lysis buffer. A fluorescent dye was added to the lysate and fluorescence was read on a Tecan M2000 spectrophotometer at 485 nm excitation and 530 nm emission.

RT-PCR

RNA was extracted from SW480 cells using the RNEasy Kit (Qiagen) and treated with DNase (LifeTech) as per the manufacturer's protocol. A total of 1 µg RNA was used for the template for single-stranded cDNA synthesis. Quantitative real-time PCR was performed for the various genes and the cDNA was amplified using SYBR Green Pro Master Mix, and the genes of interest were normalized to the corresponding hypoxanthine–guanine phosphoribosyltransferse (HPRT) controls as described before [19].

Western blot analysis

Sample preparation and western blot analysis were done as described before [20]. Briefly, cells were lysed in lysis buffer [20 mM Tris–HCl (pH 7.4), 1 mM EDTA, and 150 mM NaCl] containing 1% Triton X-100 and protease inhibitors (protease inhibitor cocktail, Roche Diagnostics) and phosphatase inhibitors. Insoluble material was removed by centrifugation, and protein concentration was determined by the Bradford assay. Proteins in the cellular lysates were separated by 10% SDS–PAGE and the gel was electroblotted against the nitrocellulose membrane (Thermo Fisher Scientific). Following this transfer, the blot was probed with specific primary antibodies for pERK1/2, total ERK-2, p38, pp38, JNK1/2, pJNK1/2, and tubulin (Santa Cruz) as well as IκBα and pIκBα (Cell Signaling) diluted in LiCor blocking buffer (LiCor) and followed by antirabbit or antimouse IR dye-conjugated secondary antibodies (LiCor). The blot was scanned and analyzed with an Odyssey LiCor instrument along with its software Odyssey.

Mice

GRK2 heterozygous mice were obtained from Jackson Laboratory (kindly donated by Dr Robert Lefkowitz, Duke University). Animals were bred at Michigan State University by breeding GRK2 heterozygous mice with wild-type (WT) mice. Note that homozygous knockout of GRK2 are embryonically lethal [21]. All animals were housed in a specific pathogen-free facility maintained at 22–24°C with a 12 h light–dark cycle and were given mouse chow and water ad libitum. All animals were performed with age- and sex-matched mice (males) between 8 and 10 weeks of age. All animal procedures were approved by the Michigan State University IACUC and conformed to the NIH guidelines.

Statistical analysis

All values are represented as mean ± SEM. All data were tested for statistical analysis for statistical significance using unpaired Student's t-test (two-sample comparisons) and analysis of variance with the Tukey post hoc test (more than two-sample comparisons). The analysis was done using the Prism GraphPad software. A P-value of <0.05 was considered significant.

Results

GRK2 suppresses TNFα-induced epithelial wound closure

The present studies were designed to investigate the role of GRK2 in TNFα signaling in intestinal epithelial cells. In other studies, TNFα has been shown to significantly modulate the intestinal wound-healing response and protect from intestinal epithelial apoptosis [11]. To this end, we performed a scratch wound-healing assay with SW480 cells (human colonic epithelial cell line) to identify the role of GRK2 in TNFα-induced wound healing. For this, cells were transfected with either control siRNA or GRK2 siRNA and plated as described in the Materials and Methods section. GRK2 knockdown was ∼80–90% compared with controls (Figure 1A,B). The wound was induced by a scratch assay, and wound closure was monitored up to 48 h. Proliferation was inhibited using mitomycin C; therefore, any cells in the wound space are the result of cell migration and not due to cell proliferation. The scratch-induced wound size was similar in each experimental group at time 0. The effect of TNFα on wound closing was calculated by determining the amount of wound closed through quantification of the gap distance using the ImageJ software and comparing the amount of wound closed with the initial time point in the respective groups. We observed that the amount of wound closure was similar between untreated and TNFα-treated groups in the control siRNA-transfected cells, at all the time points tested. However, in the GRK2 knockdown cells, wound size was markedly reduced in response to TNFα (Figure 1C,D). This suggests that GRK2 normally inhibits TNFα-induced wound healing and, therefore, its knockdown unmasks the healing potential of TNFα. Importantly, the concentration of TNFα used in these experiments had no effect on SW480 proliferation or apoptosis between the control and the GRK2 knockdown cells (Figure 2A,B). To further understand whether this phenomenon was kinase activity-dependent, we performed the wound-healing experiments in the presence of a GRK2 kinase inhibitor, paroxetine. Paroxetine binds to the active site of GRK2, inhibiting its kinase function [22]. Pretreatment of SW480 cells with paroxetine (10 µM) did not affect wound healing in the absence or presence of TNFα, suggesting that our observations with GRK2 knockdown are probably kinase-independent (data not shown). Together, our results indicate that knockdown of GRK2 in intestinal epithelial cells accelerates the TNFα-induced wound-healing process.

GRK2 inhibits wound healing in SW480 cells.

Figure 1.
GRK2 inhibits wound healing in SW480 cells.

(A) SW480 cells were transfected with either scrambled siRNA (siCtrl) or GRK2 siRNA (siGRK2) and incubated for 48 h to facilitate knockdown of GRK2. Representative western blot and their quantification showing levels of GRK2 after siCtrl or siGRK2 treatment. Tubulin is shown as a loading control. (B) Gene expression levels of GRK2 in cells treated with either siCtrl or siGRK2 (n = 5). (C) Confluent SW480 cells were wounded through the generation of a linear scratch and the cells were photographed at 0, 16, 24, 40 and 48 h. At the time of wounding, mitomycin C was added to prevent proliferation. The cells were treated with TNFα (20 ng/ml) at the onset of the linear scratch. Differences between the amount of wound closed at each time point between untreated and treated cells for both siCtrl and siGRK2 cells were calculated through the ImageJ Software and by identifying the leading edge of the wound. Data expressed as a fold increase from 16 h for the respective groups (n = 9). (D) Representative image from each group shows the area of wound closure at 0 and 48 h at 4× magnification. Mean ± SEM, ***P < 0.001, ****P < 0.0001.

Figure 1.
GRK2 inhibits wound healing in SW480 cells.

(A) SW480 cells were transfected with either scrambled siRNA (siCtrl) or GRK2 siRNA (siGRK2) and incubated for 48 h to facilitate knockdown of GRK2. Representative western blot and their quantification showing levels of GRK2 after siCtrl or siGRK2 treatment. Tubulin is shown as a loading control. (B) Gene expression levels of GRK2 in cells treated with either siCtrl or siGRK2 (n = 5). (C) Confluent SW480 cells were wounded through the generation of a linear scratch and the cells were photographed at 0, 16, 24, 40 and 48 h. At the time of wounding, mitomycin C was added to prevent proliferation. The cells were treated with TNFα (20 ng/ml) at the onset of the linear scratch. Differences between the amount of wound closed at each time point between untreated and treated cells for both siCtrl and siGRK2 cells were calculated through the ImageJ Software and by identifying the leading edge of the wound. Data expressed as a fold increase from 16 h for the respective groups (n = 9). (D) Representative image from each group shows the area of wound closure at 0 and 48 h at 4× magnification. Mean ± SEM, ***P < 0.001, ****P < 0.0001.

GRK2 does not affect TNFα-induced apoptosis or proliferation in SW480 cells.

Figure 2.
GRK2 does not affect TNFα-induced apoptosis or proliferation in SW480 cells.

(A) After siRNA transfection, SW480 cells were cultured in 6-well plates at a concentration of 1 million cells/well and were treated +/− TNFα (20 ng/ml) for 24 and 48 h in order to induce apoptosis. Data are expressed as % of total cells quantified by flow cytometry (n = 3). (B) Proliferation was assessed by measuring the total DNA from either siCtrl or siGRK2-treated cells +/− TNFα (20 ng/ml) for 24 and 48 h. Equal numbers of cells were plated and the DNA concentration was measured after TNFα treatment (n = 6). Mean ± SEM.

Figure 2.
GRK2 does not affect TNFα-induced apoptosis or proliferation in SW480 cells.

(A) After siRNA transfection, SW480 cells were cultured in 6-well plates at a concentration of 1 million cells/well and were treated +/− TNFα (20 ng/ml) for 24 and 48 h in order to induce apoptosis. Data are expressed as % of total cells quantified by flow cytometry (n = 3). (B) Proliferation was assessed by measuring the total DNA from either siCtrl or siGRK2-treated cells +/− TNFα (20 ng/ml) for 24 and 48 h. Equal numbers of cells were plated and the DNA concentration was measured after TNFα treatment (n = 6). Mean ± SEM.

GRK2 differentially modulates TNFα signaling in SW480 cells

Previous studies have shown that GRK2 can influence TNFα signaling in macrophages [12]. Our results here suggest that GRK2 can significantly affect TNFα-induced wound healing in epithelial cells. To examine the signaling mechanisms by which this occurs, we treated control and GRK2 siRNA-transfected SW480 cells with TNFα for various time points (0–90 min) and examined the cell signaling pathways. TNFα stimulation led to increases in phosphorylation of IκBα and ERK1/2 in the control siRNA-treated cells (Figure 3A,B). Surprisingly, these effects were differentially affected in the GRK2 knockdown cells. TNFα-induced IκBα phosphorylation was significantly inhibited in the GRK2 knockdown cells, whereas ERK1/2 phosphorylation was markedly enhanced (Figure 3A,B). These results demonstrate opposing roles of GRK2 in MAPK and NFκB pathways, with GRK2 augmenting the NFκB pathway and simultaneously inhibiting the ERK pathway. Other signaling pathways including JNK and p38 were not consistently affected by GRK2 knockdown (data not shown). Overall, these results suggest that GRK2 is an important regulator of ERK1/2 and IκBα signaling pathways in intestinal epithelial cells.

GRK2 differentially affects the TNFα signaling pathway in SW480 colon epithelial cells.

Figure 3.
GRK2 differentially affects the TNFα signaling pathway in SW480 colon epithelial cells.

(A) Representative blots show protein levels analyzed by western blot and loading controls. GRK2 knockdown up-regulates pERK and down-regulates pIκBα. ERK and Tubulin are loading controls. (B) Quantification of western blots for pERK/ERK and pIκBa/IκBa. Means ± SEM, *P < 0.05, **P < 0.01, compared with the corresponding siCtrl group (n = 5).

Figure 3.
GRK2 differentially affects the TNFα signaling pathway in SW480 colon epithelial cells.

(A) Representative blots show protein levels analyzed by western blot and loading controls. GRK2 knockdown up-regulates pERK and down-regulates pIκBα. ERK and Tubulin are loading controls. (B) Quantification of western blots for pERK/ERK and pIκBa/IκBa. Means ± SEM, *P < 0.05, **P < 0.01, compared with the corresponding siCtrl group (n = 5).

GRK2 alters TNFα-induced gene expression

Both ERK1/2 and NFκB pathways are potent regulators of gene transcription; therefore, we sought to identify which inflammatory and matrix-modifying genes are altered by GRK2 knockdown at 6 and 24 h post-TNFα treatment. TNFα-induced expressions of genes that encode extracellular matrix degradation, including matrix metalloproteinase 7 (MMP7), matrix metalloproteinase 9 (MMP9), as well as urokinase plasminogen activator (uPA), were all markedly enhanced in the GRK2 knockdown cells when compared with control cells at 24 h (Figure 4A); however, after 6 h of stimulation, we observed a mixed response in inflammatory gene expression where TNFα-induced expression of CXCL8 and TNFα was markedly enhanced in the GRK2 knockdown cells (compared with control siRNA cells) and IL-6 expression was significantly decreased in GRK2 knockdown compared with control cells (Figure 4A). After 24 h of TNFα treatment, there essentially was no difference in the expression of these three genes between control and GRK2 knockdown groups.

GRK2 differentially regulates TNFα-induced gene expression in SW480 cells.

Figure 4.
GRK2 differentially regulates TNFα-induced gene expression in SW480 cells.

(A) GRK2 knockdown alters TNFα-induced gene expression in SW480 colon epithelial cells transfected with siCtrl or siGRK2. Gene expression levels were assessed by analyzing mRNA levels using real-time PCR after TNFα stimulation (20 ng/ml) for 0 (untreated), 6 and 24 h, normalized to HPRT and then expressed as % maximal expression (n = 5). (B) Cells were either treated with siCtrl or siGRK2 and stimulated with TNFα (24 h) in the presence or absence of MEK inhibitor U0126. Gene expression levels of the indicated genes were assessed and normalized to HPRT and then expressed as % maximal expression (n = 3). (C) Following siRNA transfection, cells were seeded in the upper chamber of a Transwell insert coated with Matrigel. Following 48 h incubation with TNFα, migrated cells were detached and lysed and quantified using a fluorescent dye and expressed as fold control (n = 3). Means ± SEM, P < 0.05, ***P < 0.001.

Figure 4.
GRK2 differentially regulates TNFα-induced gene expression in SW480 cells.

(A) GRK2 knockdown alters TNFα-induced gene expression in SW480 colon epithelial cells transfected with siCtrl or siGRK2. Gene expression levels were assessed by analyzing mRNA levels using real-time PCR after TNFα stimulation (20 ng/ml) for 0 (untreated), 6 and 24 h, normalized to HPRT and then expressed as % maximal expression (n = 5). (B) Cells were either treated with siCtrl or siGRK2 and stimulated with TNFα (24 h) in the presence or absence of MEK inhibitor U0126. Gene expression levels of the indicated genes were assessed and normalized to HPRT and then expressed as % maximal expression (n = 3). (C) Following siRNA transfection, cells were seeded in the upper chamber of a Transwell insert coated with Matrigel. Following 48 h incubation with TNFα, migrated cells were detached and lysed and quantified using a fluorescent dye and expressed as fold control (n = 3). Means ± SEM, P < 0.05, ***P < 0.001.

Because MMPs are important in cell migration and MMP9 has been shown to be regulated via the ERK pathway, we investigated whether the TNFα-induced MMP9 in SW480 cells is ERK-dependent. To this end, we treated control and GRK2 knockdown cells with the MEK1/2 inhibitor U0126 (10 µM, 40-min pretreatment) prior to TNFα treatment. Inhibition of the MEK–ERK pathway abolished TNFα-induced MMP9 expression, confirming that MMP9 expression is driven by the ERK pathway in both control and GRK2 knockdown groups (Figure 4B). We also examined the transcripts of MMP7 and uPA, but observed that their regulation is not ERK-dependent. We attempted to inhibit the IκBα pathway using the IKK inhibitor BMS-345541, but the combination of the inhibition of the pro-survival NFκB and the stimulation with TNFα was detrimental to cell survival in culture at both 6 and 24 h (data not shown). To further confirm that an increase in MMP9 activity is important in migration through an extracellular matrix, we subjected the SW480 cells to a Matrigel invasion assay. Control and GRK2 knockdown SW480 cells were seeded in the upper chamber of a Transwell insert coated with Matrigel and allowed to migrate toward serum for 48 h in the presence or absence of TNFα (20 ng/ml). As shown, GRK2 knockdown cells treated with TNFα showed enhanced degradation of the Matrigel layer and a subsequent increase in migration toward the serum stimulus (Figure 4C). Recently, groups have demonstrated that, in other epithelial cell models, MMP9 is critical for the onset and activation of epithelial migration and wound repair [23], suggesting that the effect of GRK2 on ERK1/2 activation and MMP9 expression may be responsible for the increase in TNFα-induced wound healing in these cells.

GRK2 localizes in the mitochondria and inhibits ROS production

In previous studies, we have shown that GRK2 can directly regulate TNFα-induced IκBα phosphorylation [12]. Moreover, because the cellular phenotype (MMP expression and wound healing, [2325]) appears to be regulated by the ERK pathway, we focused on determining the mechanisms by which GRK2 regulates the TNFα-induced ERK pathway. To define how GRK2 knockdown enhanced ERK1/2 phosphorylation, we initially focused on known GRK2-binding partners with respect to the ERK pathway (mainly MEK1/2). However, our immunoprecipitation approach did not identify MEK1/2 as a binding partner for GRK2 in these cells (data not shown). Therefore, we pursued an unbiased mass spectrometry approach to identify the various partners of GRK2 in the SW480 cells. Our results indicated that GRK2 may interact with some mitochondrial proteins in this cell line (data not shown).

To further determine if GRK2 can localize to the mitochondria, we isolated mitochondria from untreated and TNFα-treated SW480 cells and subjected them to SDS–PAGE followed by western blotting for GRK2. We found that GRK2 is detectable in the mitochondria even under basal conditions as well as at 10 min and 3 h after TNFα stimulation, indicating that GRK2 is not only present in the mitochondria but remains there even under inflammatory conditions (Figure 5A). To further assess the functional consequences of GRK2 being present in the mitochondria, we measured the generation of reactive oxygen species (ROS) in control and GRK2 knockdown cells. We focused on ROS because previous studies have shown that ROS generation can lead to ERK activation [26]. Therefore, we treated control and GRK2 siRNA-transfected SW480 cells with TNFα and examined ROS production using flow cytometry. There was a modest increase in TNFα-induced ROS production in control cells over the time points tested; however, in the GRK2 knockdown cells, ROS production was significantly enhanced and this was more evident at 10 min following TNFα treatment (Figure 5B,C).

GRK2 is present in the mitochondria in SW480 cells and regulates TNFα-induced ROS generation.

Figure 5.
GRK2 is present in the mitochondria in SW480 cells and regulates TNFα-induced ROS generation.

Isolated mitochondria show GRK2 localization in mitochondria in the absence and presence of TNFα (20 ng/ml). (A) Representative western blot identifies the location of GRK2 in different cellular fractions. To see the specificity of fractionation, tubulin is used for cytoplasmic fraction control and VDAC is used for mitochondrial fraction (n = 3). (B) GRK2 knockdown enhances ROS expression. SW480 cells were treated with either siCtrl or siGRK2 and after 0, 10 or 20 min of TNFα stimulation; ROS production was measured and analyzed using flow cytometry. Data are expressed as % maximal expression (n = 3). (C) Representative histogram of ROS expression after 10 min of TNFα expression. Lighter color is siCtrl and darker is siGRK2 (n = 3). Means ± SEM, *P < 0.05.

Figure 5.
GRK2 is present in the mitochondria in SW480 cells and regulates TNFα-induced ROS generation.

Isolated mitochondria show GRK2 localization in mitochondria in the absence and presence of TNFα (20 ng/ml). (A) Representative western blot identifies the location of GRK2 in different cellular fractions. To see the specificity of fractionation, tubulin is used for cytoplasmic fraction control and VDAC is used for mitochondrial fraction (n = 3). (B) GRK2 knockdown enhances ROS expression. SW480 cells were treated with either siCtrl or siGRK2 and after 0, 10 or 20 min of TNFα stimulation; ROS production was measured and analyzed using flow cytometry. Data are expressed as % maximal expression (n = 3). (C) Representative histogram of ROS expression after 10 min of TNFα expression. Lighter color is siCtrl and darker is siGRK2 (n = 3). Means ± SEM, *P < 0.05.

Increase in ROS is responsible for increase in ERK1/2 phosphorylation

ROS is produced by a variety of cell types, but is normally viewed as harmful to both cells and tissues. Over the last few decades, the functional capabilities of ROS have been expanding and ROS has been shown to play alternative roles other than causing harmful effects. ROS generation in non-phagocytotic cells (at low concentrations) is recognized as ubiquitous intracellular messengers that can activate mitogen-associated signal transduction pathways, including ERK1/2 [27,28]. To determine if the observed increase in ROS generation was causative for the increases in phosphorylated ERK1/2 in SW480 cells, we treated these cells with a superoxide inhibitor, N-acetylcysteine (NAC), to inhibit the production of ROS and demonstrate the link between ROS and ERK1/2 activation. Control and GRK2 knockdown SW480 cells were pretreated with NAC (100 µM) for 30 min and then stimulated with TNFα for 30 min. In the groups untreated with NAC, we observed that knocking down GRK2 caused a significant increase in the pERK1/2 (as expected), but in the presence of NAC, ERK activation was suppressed especially in the GRK2 knockdown cells (Figure 6A). These data demonstrate that the enhanced ERK activation observed in the GRK2 knockdown cells following TNFα stimulation is due to increase in ROS generation. To further link ROS generation to the wound-healing response, we repeated the wound-healing assay in control and GRK2 knockdown cells pretreated with NAC and untreated or treated with TNFα for 48 h. NAC significantly inhibited the effect of GRK2 knockdown on TNFα-induced wound closure as shown in Figure 6B. Together, these data support the fact that GRK2 inhibits TNFα-induced wound healing in these cells via the ROS pathway.

ROS scavenging prevents the effect of GRK2 on ERK activation and wound closure.

Figure 6.
ROS scavenging prevents the effect of GRK2 on ERK activation and wound closure.

(A) Treatment with NAC prevents TNFα-induced increase in pERK in siGRK2 cells. Cells were untreated or treated with TNFα and NAC as shown. TNFα treatment was performed for 30 min. pERK/ERK levels were assayed as described in the Materials and Methods section. Representative blot is shown in the top and quantitation in the bottom; expressed as fold siCtrl for either no treatment or NAC treatment (n = 3). Means ± SEM, *P < 0.05, **P < 0.01. (B) Treatment with NAC prevents TNFα-induced wound closure in siGRK2 cells. Wound-healing experiments were done as described in the Materials and Methods section. siControl and siGRK2 cells were untreated or pretreated with NAC followed by treatment (or not with TNFα) for 48 h as shown (N = 3). ***P < 0.01.

Figure 6.
ROS scavenging prevents the effect of GRK2 on ERK activation and wound closure.

(A) Treatment with NAC prevents TNFα-induced increase in pERK in siGRK2 cells. Cells were untreated or treated with TNFα and NAC as shown. TNFα treatment was performed for 30 min. pERK/ERK levels were assayed as described in the Materials and Methods section. Representative blot is shown in the top and quantitation in the bottom; expressed as fold siCtrl for either no treatment or NAC treatment (n = 3). Means ± SEM, *P < 0.05, **P < 0.01. (B) Treatment with NAC prevents TNFα-induced wound closure in siGRK2 cells. Wound-healing experiments were done as described in the Materials and Methods section. siControl and siGRK2 cells were untreated or pretreated with NAC followed by treatment (or not with TNFα) for 48 h as shown (N = 3). ***P < 0.01.

GRK2 heterozygous mice are protected in intestinal wound-healing model

Our in vitro wound-healing assay indicated that knocking down GRK2 in intestinal epithelial SW480 cells improved their ability to close the wound in the presence of TNFα. A common characteristic of dextran sodium sulfate (DSS)-induced colitis in mice is gross epithelial damage and high levels of TNFα in the colon [29,30]. To determine if our in vitro results would correlate to an in vivo model, we used mice that were heterozygous for GRK2 and subjected them to a 2.5% DSS colitis. Both WT and heterozygous GRK2 mice (GRK2+/−) were administered 2.5% DSS in drinking water. Mice were scored daily for weight loss and external signs of disease measured as the Disease Activity (DAI, indicated by weight loss, loose and bloody stools, hunched posture, crusty eyes, and ruffled hair coat) [31]. After 7 days of DSS, the mice were given clean water and allowed to heal from intestinal inflammation for an additional 5 days. After DSS treatment, DAI increased after day 7 but recovered after day 12 and were back to normal by day 13 in the WT mice (Figure 7A). Interestingly, the DAI was significantly decreased in the GRK2+/− group. This was also evident in the histopathology where GRK2 heterozygous knockout mice showed marked protection (Figure 7B). GRK2, therefore, is protective in in vivo wound healing, perhaps through alterations of the epithelial response to TNFα and its influence on wound healing.

Intestinal wound healing is enhanced in GRK2 heterozygous mice.

Figure 7.
Intestinal wound healing is enhanced in GRK2 heterozygous mice.

WT and GRK2 heterozygous mice were subjected to 2.5% DSS for the indicated times. (A) Disease activity index was determined as described in the text (n = 6–8). (B) Histology of the colon from mice subjected to intestinal inflammation at day 13 (n = 6–8). Means ± SEM, *P < 0.05.

Figure 7.
Intestinal wound healing is enhanced in GRK2 heterozygous mice.

WT and GRK2 heterozygous mice were subjected to 2.5% DSS for the indicated times. (A) Disease activity index was determined as described in the text (n = 6–8). (B) Histology of the colon from mice subjected to intestinal inflammation at day 13 (n = 6–8). Means ± SEM, *P < 0.05.

Discussion

Since the discovery of GRK2, it has been shown as a key regulator of GPCR phosphorylation and desensitization. As it has continued to be studied, its role has expanded and it has been shown to have many different functions both dependent and independent of its catalytic activity [3]. Furthermore, GRK2 has been shown to regulate non-GPCRs as well as a variety of intracellular targets, leading to many cellular roles for this kinase. In the present study, we investigated these non-receptor regulatory capabilities of GRK2 in intestinal epithelial cells in response to TNFα stimulation.

In previous studies, GRK2 has been demonstrated to play a negative role in wound healing that is also associated with increased levels of TNFα. Measuring levels of inflammatory cytokines in the hearts of mice lacking β1-adrenergic receptor (B1KO) and expressing βARKct (a GRK2 inhibitor), there was an increase in TNFα, IL-6, and IL-1β 24 h after undergoing myocardial infarction, which is indicative of increased healing-associated inflammation [32]. Alternatively, GRK2 was shown to positively mediate sphingosine-1 phosphate-induced ERK1/2 activation and wound closing in HEK293 cells as well as mouse embryonic fibroblasts [33]. In our studies with colonic epithelial cells, GRK2 inhibited TNFα-induced ERK1/2 phosphorylation and wound closure, and thus, the role of GRK2 in wound healing per se is different between the two studies. In the above study, GRK2 regulates ERK signaling through modifications of the scaffolding ability of GIT1 through direct interactions. This modification alters the Rac/PAK/MEK/ERK pathway leading to eventual changes in migration. While these differences in how GRK2 regulates the ERK pathway could be attributed to different cell types and ligands, the role of GRK2 in the SW480 cells offers an alternative pathway of ERK regulation through modification of TNFα−-induced ROS production that acts as a mitogenic signaling factor to stimulate ERK. Thus, the critical factor between both of these studies is the activation of the ERK pathway. The role of the ERK pathway in wound healing is becoming more and more apparent through studies in other epithelial cell types, and several studies have now shown ERK to be a critical regulator of wound healing and cellular migration [23,24]. Importantly, activation of the ERK pathway has been linked to improved wound-healing responses in a variety of different cell types, including colonic epithelial cells [34]. Thus, understanding the mechanisms by which the ERK pathway is regulated in colonic epithelial cells could lead to the identification of better therapeutic targets for intestinal inflammation. In this context, our results using GRK2 suggest that this may be one additional mechanism by which the ERK pathway and its consequent effect on wound healing are regulated. Thus, we show for the first time that GRK2 inhibits the TNFα-induced wound-healing response in colonic epithelial cells.

The ERK1/2 pathway has a diverse set of roles within the cell including regulation of cell growth, survival, and the regulation of cell motility [35]. A necessary mechanism for cellular migration and wound healing is the capacity for matrix degradation through the localized expression of matrix MMPs. One suggested mechanism for the ability of ERK1/2 to alter migration and wound healing is the up-regulation of MMPs for extracellular matrix remodeling [36]. We provide evidence that MMP7, MMP9, and uPA (another ECM remodeler) are all elevated in GRK2 knockdown cells and that the increase in MMP9 expression especially is ablated by inhibition of the ERK pathway. Other studies demonstrate that MMP9 is critical to the onset and activation of epithelial migration and wound repair in both SW480 cells and alternative cell types such as HEK293 cells [2325]. Consistent with those studies, our data have shown that GRK2 knockdown cells have increased invasive capabilities through a Matrigel layer. Taken together, our studies suggest that GRK2 regulation of the ERK pathway in SW480 cells modulates MMP9 and consequent epithelial cell migration.

It is possible that MMP7 and uPA (as well as the inflammatory cytokines, CXCL8, IL-6, and TNFα) are regulated through the changes observed in NFκB signaling or other unidentified pathways. Our laboratory has previously confirmed that GRK2 regulates the expression of CXCL8 and IL-6 through alterations of NFκB in murine macrophages [17]. In the present study, we focused on mechanisms and consequences by which GRK2 regulates the ERK pathway. Thus, regulation of MMP9 through pERK1/2 and the importance of this MMP in published studies indicate that it may be sufficient to drive the phenotype we observe in our system.

Previously, we showed that GRK2 could directly influence TNFα signaling through binding to IκBα [12]. Additionally, GRK2 has been shown to interact with other signaling proteins such as PI3K [37], AKT [38], GIT [39], and MEK1 [40]. Based on these previous studies, we hypothesized that GRK2 would interact with an unknown component of the TNFα signaling pathway to influence regulation of ERK1/2 in SW480 cells. Interestingly, our preliminary data suggest that GRK2 may be interacting with different mitochondrial proteins. Indeed, when we isolated the mitochondrial fraction from these cells, we observed that GRK2 is present in the mitochondria both basally and after TNFα stimulation. This ability for GRK2 to be present in the mitochondria has been shown by other groups in both fibroblasts [41] as well as macrophages [42] and cardiac myocytes [43], but has never been shown in colonic epithelial cells. Previous work investigating the role of GRK2 in the mitochondria showed that knocking down GRK2 increases the mitochondrial generation of ROS in response to lipopolysaccharide in macrophages [42]. Our results on ROS generation in SW480 cells, following TNFα stimulation, are consistent with these findings. Our results on ROS generation and ERK activation are consistent with previous studies demonstrating a strong link between ROS and the ERK pathway [26,44]. In line with this, treatment with superoxide inhibitor completely abolished the enhanced ERK activation observed in GRK2 knockdown cells. Together, these results indicate that GRK2 is present both in the cytoplasm and in the mitochondria under normal and inflammatory conditions in colonic epithelial cells and this presence regulates ROS generation and subsequent modulation of the ERK-mediated wound-healing response.

IBD is a chronic inflammatory state characterized by persistent damage to the epithelial barrier in the colon. A common characteristic of DSS-induced colitis model in mice is gross epithelial damage and high levels of TNFα in the colon. To determine if our in vitro results with SW480 cells can be correlated to the in vivo pathophysiology, we used mice that were heterozygous for GRK2 and subjected them to a wound-healing DSS model. Our data indicate that knocking down GRK2 does, in fact, protect against intestinal inflammation, consistent with the in vitro data. It should, however, be noted that in vivo intestinal inflammation involves several different cell types. Even though the injury is initiated by DSS at the level of the intestinal epithelial cells, pathogenesis of inflammation involves participation of many different cell types including immune cells. Since GRK2 is expressed in many different cell types, it is possible that the effect observed in the heterozygous mice is not restricted to its role in intestinal epithelial cells. These certainly are important questions for future studies.

Although very little is known about the role of GRK2 in the colon or in colitis, GRK2 has been implicated in several diseases including multiple sclerosis [14], Alzheimer's disease [45], and rheumatoid arthritis as well as sepsis and endotoxemia [15]. In addition, others have examined the role of other GRKs, including GRK6, in a DSS-induced colitis model and shown that mice deficient in GRK6 have enhanced immune cell infiltration and enhanced severity of colitis [46]. These results suggest that the role of the various GRKs in colitis models could be different and warrant further detailed investigation. Altogether, these studies show a regulatory capacity for GRKs in the colon and a vital role of GRK2 in inflammation, which merits future work on the mechanism of protection and the role of GRK2 in colitis.

In conclusion, our studies unravel a critical role for GRK2 in TNFα signaling in colonic epithelial cells and a potential role for GRK2 in intestinal inflammation. Through regulation of ROS-generated ERK1/2 phosphorylation, GRK2 knockdown enhances the ability of SW480 cells to close a wound possibly through increased MMP9 expression. These studies suggest GRK2 to be a possible therapeutic target in intestinal inflammation.

Abbreviations

     
  • DAI

    Disease Activity Score

  •  
  • DSS

    dextran sodium sulfate

  •  
  • GPCRs

    G-protein-coupled receptors

  •  
  • GRK2

    G-protein-coupled receptor kinase-2

  •  
  • HPRT

    hypoxanthine–guanine phosphoribosyltransferse

  •  
  • MMPs

    metalloproteinases

  •  
  • PI

    propidium iodide

  •  
  • ROS

    reactive oxygen species

  •  
  • uPa

    urokinase plasminogen activator

  •  
  • WT

    wild type

Acknowledgments

We thank the University Laboratory Animal Resources for taking excellent care of our animals, and the Histopathology Laboratory for their excellent service.

Funding

We gratefully acknowledge the support from NIH (grants HL095637, AI099404, and AR056680 to N.P.).

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Gurevich
,
E.V.
,
Tesmer
,
J.J.G.
,
Mushegian
,
A.
and
Gurevich
,
V.V
. (
2012
)
G protein-coupled receptor kinases: more than just kinases and not only for GPCRs
.
Pharmacol. Ther.
133
,
40
69
doi:
2
Packiriswamy
,
N.
and
Parameswaran
,
N
. (
2015
)
G-protein-coupled receptor kinases in inflammation and disease
.
Genes Immun.
16
,
367
377
doi:
3
Evron
,
T.
,
Daigle
,
T.L.
and
Caron
,
M.G
. (
2012
)
GRK2: multiple roles beyond G protein-coupled receptor desensitization
.
Trends Pharmacol. Sci.
33
,
154
164
doi:
4
Penela
,
P.
,
Murga
,
C.
,
Ribas
,
C.
,
Lafarga
,
V.
and
Mayor
, Jr,
F
. (
2010
)
The complex G protein-coupled receptor kinase 2 (GRK2) interactome unveils new physiopathological targets
.
Br. J. Pharmacol.
160
,
821
832
doi:
5
Woodall
,
M.C.
,
Ciccarelli
,
M.
,
Woodall
,
B.P.
and
Koch
,
W.J.
(
2014
)
G protein-coupled receptor kinase 2: a link between myocardial contractile function and cardiac metabolism
.
Circ. Res
.
114
,
1661
1670
doi:
6
Sansonetti
,
P.J
. (
2004
)
War and peace at mucosal surfaces
.
Nat. Rev. Immunol.
4
,
953
964
doi:
7
Kagnoff
,
M.F.
and
Eckmann
,
L
. (
1997
)
Epithelial cells as sensors for microbial infection
.
J. Clin. Invest.
100
,
6
10
doi:
8
Xavier
,
R.J.
and
Podolsky
,
D.K
. (
2007
)
Unravelling the pathogenesis of inflammatory bowel disease
.
Nature
448
,
427
434
doi:
9
Gribar
,
S.C.
,
Richardson
,
W.M.
,
Sodhi
,
C.P.
and
Hackam
,
D.J
. (
2008
)
No longer an innocent bystander: epithelial Toll-like receptor signaling in the development of mucosal inflammation
.
Mol. Med.
14
,
645
659
doi:
10
Okamoto
,
R.
and
Watanabe
,
M
. (
2005
)
Cellular and molecular mechanisms of the epithelial repair in IBD
.
Dig. Dis. Sci.
50
,
S34
S38
doi:
11
Leppkes
,
M.
,
Roulis
,
M.
,
Neurath
,
M.F.
,
Kollias
,
G.
and
Becker
,
C
. (
2014
)
Pleiotropic functions of TNF-α in the regulation of the intestinal epithelial response to inflammation
.
Int. Immunol.
26
,
509
515
doi:
12
Patial
,
S.
,
Luo
,
J.
,
Porter
,
K.J.
,
Benovic
,
J.L.
and
Parameswaran
,
N
. (
2011
)
G-protein-coupled-receptor kinases mediate TNFα-induced NF-κB signalling via direct interaction with and phosphorylation of IκBα
.
Biochem. J.
425
,
169
180
doi:
13
Patial
,
S.
and
Parameswaran
,
N
. (
2010
)
Tumor necrosis factor-α signaling in macrophages
.
Crit. Rev. Eukaryot. Gene Expr.
20
,
87
103
doi:
14
Vroon
,
A.
,
Kavelaars
,
A.
,
Limmroth
,
V.
,
Lombardi
,
M.S.
,
Goebel
,
M.U.
,
Van Dam
,
A.-M.
et al. 
(
2005
)
G protein-coupled receptor kinase 2 in multiple sclerosis and experimental autoimmune encephalomyelitis
.
J. Immunol.
174
,
4400
4406
doi:
15
Parvataneni
,
S.
,
Gonipeta
,
B.
,
Packiriswamy
,
N.
,
Lee
,
T.
,
Durairaj
,
H.
and
Parameswaran
,
N
. (
2011
)
Role of myeloid-specific G-protein coupled receptor kinase-2 in sepsis
.
Int. J. Clin. Exp. Med.
4
,
320
330
PMID:
[PubMed]
16
Durairaj
,
H.
,
Steury
,
M.D.
and
Parameswaran
,
N
. (
2015
)
Paroxetine differentially modulates LPS-induced TNFα and IL-6 production in mouse macrophages
.
Int. Immunopharmacol.
25
,
485
492
doi:
17
Patial
,
S.
,
Saini
,
Y.
,
Parvataneni
,
S.
,
Appledorn
,
D.M.
,
Dorn
,
G.W.
,
LaPres
,
J.J.
et al. 
(
2011
)
Myeloid-specific GPCR kinase-2 negatively regulates NF-κB1p105-ERK pathway and limits endotoxemic shock in mice
.
J. Cell Physiol.
226
,
627
637
doi:
18
Gao
,
C.
,
Negash
,
S.
,
Guo
,
H.T.
,
Ledee
,
D.
,
Wang
,
H.
and
Zelenka
,
P
. (
2002
)
CDK5 regulates cell adhesion and migration in corneal epithelial cells
.
Mol. Cancer Res.
1
,
12
24
PMID:
[PubMed]
19
Loniewski
,
K.
,
Shi
,
Y.
,
Pestka
,
J.
and
Parameswaran
,
N
. (
2008
)
Toll-like receptors differentially regulate GPCR kinases and arrestins in primary macrophages
.
Mol. Immunol.
45
,
2312
2322
doi:
20
Parameswaran
,
N.
,
Pao
,
C.S.
,
Leonhard
,
K.S.
,
Kang
,
D.S.
,
Kratz
,
M.
,
Ley
,
S.C
et al.  et al.  (
2006
)
Arrestin-2 and G protein-coupled receptor kinase 5 interact with NFκB1 p105 and negatively regulate lipopolysaccharide-stimulated ERK1/2 activation in macrophages
.
J. Biol. Chem.
281
,
34159
34170
doi:
21
Jaber
,
M.
,
Koch
,
W.J.
,
Rockman
,
H.
,
Smith
,
B.
,
Bond
,
R.A.
,
Sulik
,
K.K.
et al. 
(
1996
)
Essential role of β-adrenergic receptor kinase 1 in cardiac development and function
.
Proc. Natl Acad. Sci. U.S.A.
93
,
12974
12979
doi:
22
Thal
,
D.M.
,
Homan
,
K.T.
,
Chen
,
J.
,
Wu
,
E.K.
,
Hinkle
,
P.M.
,
Huang
,
Z.M.
et al. 
(
2012
)
Paroxetine is a direct inhibitor of G protein-coupled receptor kinase 2 and increases myocardial contractility
.
ACS Chem. Biol.
7
,
1830
1839
doi:
23
Bove
,
P.F.
,
Hristova
,
M.
,
Wesley
,
U.V.
,
Olson
,
N.
,
Lounsbury
,
K.M.
and
van der Vliet
,
A
. (
2008
)
Inflammatory levels of nitric oxide inhibit airway epithelial cell migration by inhibition of the kinase ERK1/2 and activation of hypoxia-inducible factor-1α
.
J. Biol. Chem.
283
,
17919
17928
doi:
24
Kim
,
H.C.
,
Kim
,
Y.S.
,
Oh
,
H.-W.
,
Kim
,
K.
,
Oh
,
S.-S.
,
Kim
,
J.-T.
et al. 
(
2014
)
Collagen Triple Helix Repeat Containing 1 (CTHRC1) acts via ERK-dependent induction of MMP9 to promote invasion of colorectal cancer cells
.
Oncotarget
5
,
519
529
doi:
25
Luo
,
Y.
,
Liang
,
F.
and
Zhang
,
Z.-Y.
(
2009
)
PRL1 promotes cell migration and invasion by increasing MMP2 and MMP9 expression through Src and ERK1/2 pathways
.
Biochemistry
48
,
1838
1846
doi:
26
Iovine
,
B.
,
Iannella
,
M.L.
,
Nocella
,
F.
,
Pricolo
,
M.R.
and
Bevilacqua
,
M.A
. (
2012
)
Carnosine inhibits KRAS-mediated HCT116 proliferation by affecting ATP and ROS production
.
Cancer Lett.
315
,
122
128
doi:
27
Zhang
,
J.
,
Jin
,
N.
,
Liu
,
Y.
and
Rhoades
,
R.A
. (
1998
)
Hydrogen peroxide stimulates extracellular signal-regulated protein kinases in pulmonary arterial smooth muscle cells
.
Am. J. Respir. Cell Mol. Biol.
19
,
324
332
doi:
28
Chen
,
K.C.-W.
,
Zhou
,
Y.
,
Xing
,
K.
,
Krysan
,
K.
and
Lou
,
M.F
. (
2004
)
Platelet derived growth factor (PDGF)-induced reactive oxygen species in the lens epithelial cells: the redox signaling
.
Exp. Eye Res.
78
,
1057
1067
doi:
29
Neurath
,
M.F
. (
2014
)
New targets for mucosal healing and therapy in inflammatory bowel diseases
.
Mucosal Immunol.
7
,
6
19
doi:
30
Iizuka
,
M.
and
Konno
,
S.
(
2011
)
Wound healing of intestinal epithelial cells
.
World J. Gastroenterol.
17
,
2161
2171
doi:
31
Sharma
,
D.
,
Malik
,
A.
,
Steury
,
M.D.
,
Lucas
,
P.C.
and
Parameswaran
,
N
. (
2015
)
Protective role of β-arrestin2 in colitis through modulation of T-cell activation
.
Inflamm. Bowel Dis.
21
,
2766
2777
doi:
32
Salazar
,
N.C.
,
Vallejos
,
X.
,
Siryk
,
A.
,
Rengo
,
G.
,
Cannavo
,
A.
,
Liccardo
,
D.
et al. 
(
2013
)
GRK2 blockade with βARKct is essential for cardiac β2-adrenergic receptor signaling towards increased contractility
.
Cell Commun. Signal.
11
,
64
doi:
33
Penela
,
P.
,
Ribas
,
C.
,
Aymerich
,
I.
,
Eijkelkamp
,
N.
,
Barreiro
,
O.
,
Heijnen
,
C.J.
et al. 
(
2008
)
G protein-coupled receptor kinase 2 positively regulates epithelial cell migration
.
EMBO J.
27
,
1206
1218
doi:
34
Murray
,
D.
,
Morrin
,
M.
and
McDonnell
,
S.
(
2004
)
Increased invasion and expression of MMP-9 in human colorectal cell lines by a CD44-dependent mechanism
.
Anticancer Res.
24
,
489
494
PMID:
[PubMed]
35
Vial
,
E.
and
Pouysségur
,
J
. (
2004
)
Regulation of tumor cell motility by ERK mitogen-activated protein kinases
.
Ann. N. Y. Acad. Sci.
1030
,
208
218
doi:
36
Mccawley
,
L.J.
,
Li
,
S.
,
Wattenberg
,
E.V.
and
Hudson
,
L.G
. (
1999
)
Sustained activation of the mitogen-activated protein kinase pathway. A mechanism underlying receptor tyrosine kinase specificity for matrix metalloproteinase-9 induction and cell migration
.
J. Biol. Chem.
274
,
4347
4353
doi:
37
Naga Prasad
,
S.V.
,
Barak
,
L.S.
,
Rapacciuolo
,
A.
,
Caron
,
M.G.
and
Rockman
,
H.A.
(
2001
)
Agonist-dependent recruitment of phosphoinositide 3-kinase to the membrane by beta-adrenergic receptor kinase 1. A role in receptor sequestration
.
J. Biol. Chem.
276
,
18953
18959
doi:
38
Liu
,
S.
,
Premont
,
R.T.
,
Kontos
,
C.D.
,
Zhu
,
S.
and
Rockey
,
D.C
. (
2005
)
A crucial role for GRK2 in regulation of endothelial cell nitric oxide synthase function in portal hypertension
.
Nat. Med.
11
,
952
958
doi:
39
Premont
,
R.T.
,
Claing
,
A.
,
Vitale
,
N.
,
Freeman
,
J.L.R.
,
Pitcher
,
J.A.
,
Patton
,
W.
et al. 
(
1998
)
β2-Adrenergic receptor regulation by GIT1, a G protein-coupled receptor kinase-associated ADP ribosylation factor GTPase-activating protein
.
Proc. Natl Acad. Sci.
95
,
14082
14087
doi:
40
Jiménez-Sainz
,
M.C.
,
Murga
,
C.
,
Kavelaars
,
A.
,
Jurado-Pueyo
,
M.
,
Krakstad
,
B.F.
,
Heijnen
,
C.J.
et al. 
(
2006
)
G protein-coupled receptor kinase 2 negatively regulates chemokine signaling at a level downstream from G protein subunits
.
Mol. Biol. Cell
17
,
25
31
doi:
41
Fusco
,
A.
,
Santulli
,
G.
,
Sorriento
,
D.
,
Cipolletta
,
E.
,
Garbi
,
C.
,
Dorn
,
G.W.
et al. 
(
2012
)
Mitochondrial localization unveils a novel role for GRK2 in organelle biogenesis
.
Cell. Signal.
24
,
468
475
doi:
42
Sorriento
,
D.
,
Fusco
,
A.
,
Ciccarelli
,
M.
,
Rungi
,
A.
,
Anastasio
,
A.
,
Carillo
,
A.
et al. 
(
2013
)
Mitochondrial G protein coupled receptor kinase 2 regulates proinflammatory responses in macrophages
.
FEBS Lett.
587
,
3487
3494
doi:
43
Chen
,
M.
,
Sato
,
P.Y.
,
Chuprun
,
J.K.
,
Peroutka
,
R.J.
,
Otis
,
N.J.
,
Ibetti
,
J.
et al. 
(
2013
)
Prodeath signaling of G protein-coupled receptor kinase 2 in cardiac myocytes after ischemic stress occurs via extracellular signal-regulated kinase-dependent heat shock protein 90-mediated mitochondrial targeting
.
Circ. Res.
112
,
1121
1134
doi:
44
Samavati
,
L.
,
Monick
,
M.M.
,
Sanlioglu
,
S.
,
Buettner
,
G.R.
,
Oberley
,
L.W.
and
Hunninghake
,
G.W
. (
2002
)
Mitochondrial KATP channel openers activate the ERK kinase by an oxidant-dependent mechanism
.
Am. J. Physiol. Cell Physiol
283
,
C273
C281
doi:
45
Obrenovich
,
M.E.
,
Palacios
,
H.H.
,
Gasimov
,
E.
,
Leszek
,
J.
and
Aliev
,
G.
(
2009
)
The GRK2 overexpression is a primary hallmark of mitochondrial lesions during early alzheimer disease
.
Cardiovasc. Psychiatry Neurol.
2009
,
1
14
doi:
46
Eijkelkamp
,
N.
,
Heijnen
,
C.J.
,
Lucas
,
A.
,
Premont
,
R.T.
,
Elsenbruch
,
S.
,
Schedlowski
,
M
et al.  et al.  (
2007
)
G protein-coupled receptor kinase 6 controls chronicity and severity of dextran sodium sulphate-induced colitis in mice
.
Gut
56
,
847
854
doi:

Supplementary data