Knockout of microRNA-21 attenuates alcoholic hepatitis through the VHL/NF-κB signaling pathway in hepatic stellate cells

Abstract

microRNA-21 (miRNA) is one of the most abundant miRNAs in chronic liver injuries including alcoholic liver injury. Previous studies have demonstrated that miR-21 plays a role in inflammation in the liver and functions in hepatic stellate cells (HSCs), which reside in the perisinusoidal space between sinusoidal endothelial cells and hepatocytes and regulate sinusoidal circulation. HSCs integrate cytokine-mediated inflammatory responses in the sinusoids and relay them to the liver parenchyma. Here, we showed that the activation of Von Hippel-Lindau (VHL) expression, by miR-21 knockout in vivo and anti-miR-21 or VHL overexpression in vitro, suppressed the production of proinflammatory cytokines, such as interleukin (IL)-6, monocyte chemoattractant protein-1, and IL-1β, in human HSCs during alcoholic liver injury. Sequence and functional analyses confirmed that miR-21 directly targeted the 3′-untranslated region of VHL. Immunofluorescence and real-time PCR analysis revealed that miR-21 depletion blocked NF-κB activation in human HSCs both in cultured HSCs as well as HSCs isolated from alcohol-related liver disease mice liver by laser capture microdissection. We also showed that conditioned medium from anti-miR-21-transfected HSCs suppressed human monocyte-derived THP-1 cell migration. Taken together, our study indicates that depletion of miR-21 may downregulate cytokine production in HSCs and macrophage chemotaxis during alcoholic liver injury and that the targeting of miR-21 may have therapeutic potential for preventing the progression of alcoholic liver diseases.

NEW & NOTEWORTHY This study demonstrates that silencing microRNA-21 can inhibit cytokine production and inflammatory responses in human hepatic stellate cells during alcoholic liver injury and that the targeting of microR-21 in hepatic stellate cells may have therapeutic potential for prevention and treatment of alcoholic liver diseases.

INTRODUCTION

Alcohol-related liver disease (ALD) is a major health concern, which encompasses a broad spectrum of liver injury ranging from simple steatosis through alcohol-related hepatitis to cirrhosis and hepatocellular carcinoma (13). The pathophysiology of ALD is complex, but inflammatory response is considered a key player, as gut microflora-derived lipopolysaccharides (LPS) and the proinflammatory cytokine tumor necrosis factor-α (TNF-α) play an important role in ALD development in patients and alcohol-fed animal models (39, 47), suggesting that alcohol-related inflammation contributes to ALD development and progression. Additionally, increased plasma LPS (endotoxemia) is a frequent event in patients with alcoholic steatohepatitis and cirrhosis, implicating a key role for LPS-induced inflammation in the pathogenesis of human ALD.

Hepatic sinusoids, connected directly to the portal circulation, serve as the first barrier against these inflammatory and noxious stimuli (12). Besides sinusoidal endothelial cells and Kupffer cells (liver macrophages), hepatic stellate cells (HSCs) are one of the key components. HSCs are located in the space of Disse, the abluminal side of the sinusoids between hepatic sinusoidal endothelium and hepatocytes. HSCs represent 5–8% of all liver cells and 13% of hepatic sinusoidal cells. The functional roles of HSCs include storage of vitamin A, synthesis of extracellular matrices and matrix-degrading metalloproteinases, and regulation of sinusoidal blood flow (37). The involvement of HSCs in liver fibrosis is well recognized and attracts much attention; nevertheless, their role in liver inflammation has been barely defined. In consideration of their anatomical position, HSCs may be critical for inflammatory signaling from the sinusoids. Human and mouse HSCs produce cytokines and chemokines upon aberrant stimuli such as LPS and other toxic substances, suggesting that HSCs can potentially regulate hepatic immune and inflammatory responses through their own signaling network.

microRNAs (miRNAs) are noncoding RNAs (ncRNAs) recently found to downregulate a large subset of human genes, and evidence suggests that some miRNAs have powerful cellular functions (15). Although a role for miRNAs in human liver disorders has been proposed, the molecular mechanisms by which miRNAs can modulate hepatic inflammation during acute and chronic alcoholic liver injuries are undefined. We have shown that several miRNAs are aberrantly expressed in human hepatocytes, HSCs, as well as cholangiocytes and in mouse ALD liver compared with normal control tissue (22, 23, 38). miR-21, the most upregulated miRNA in ethanol-feeding mice liver from our study, has been reported to emerge as a specific miRNA modulated in liver disease. miR-21 is involved in interleukin-6 (IL-6)/Stat3 signaling, which may subsequently modulate liver-associated inflammatory activities that were initially viewed as a host-protective mechanism to remove damaged hepatocytes (11, 29). Thus we aimed to define the role of knockout of miR-21 in HSCs during ALD by posing the following questions. First, is miR-21 expression altered in HSCs isolated from ethanol-exposed mice and ALD human liver tissues? Second, does modulation of miR-21 alter HSC response to inflammatory stimuli in vitro and in animals with ALD? Third, what is the upstream inflammatory stimulator for miR-21 in HSCs during ALD? Fourth, what downstream targets of miR-21 are involved HSC-associated inflammatory response in ALD?

MATERIALS AND METHODS

Materials, reagents, and plasmids.

Reagents were purchased from Sigma-Aldrich Chemical (St. Louis, MO) unless otherwise indicated. The rabbit polyclonal antibodies against VHL and Toll-like receptor-4 (TLR4) were purchased from Abcam (Cambridge, MA). The antibodies against IkBα, α-smooth muscle actin, vimentin, fibronectin, tissue inhibitor of metalloproteinase-3, matrix metalloproteinase (MMP)-2, MMP-9, and desmin were purchased from Santa Cruz Biotechnology (Dallas, TX). The precursor/inhibitor for miR-21 and the control miRNA precursor/inhibitor were purchased from Thermo Fisher Scientific (Waltham, MA). The plasmids encoding Lac Z and VHL (wt-VHL) were obtained from Addgene (Cambridge, MA). The RNeasy Mini kit for RNA purification and all selected PCR primers were purchased from Qiagen (Valencia, CA). The iScript cDNA Synthesis kit (170-8891) and iTaq Universal SYBR Green Supermix (172-5124) were purchased from Bio-Rad (Hercules, CA).

Cell lines.

Human HSCs were obtained from ScienCell (San Diego, CA) and cultured as recommended by the supplier. The human acute monocytic leukemia cell line THP-1 was purchased from ATCC (Manassas, VA) and maintained in Roswell Park Memorial Institute (RPMI) 1640 medium (Sigma) with 10% FCS without antibiotics.

LPS stimulation/activation of HSCs.

At 24 h after incubation in DMEM with 1% FCS, human HSCs were stimulated with 100 ng/ml LPS (Imgenex, San Diego, CA) for 24 h.

Transfections.

Transfections were performed by nuclear electroporation using the Nucleofector system (Amaxa Biosystems, Koln, Germany). Fifty microliters of 100 nM microRNA precursor, antisense inhibitor, or controls (Ambion, Austin, TX) were added to 1 × 10⁶ cells suspended in 50 μl of Nucleofector solution at room temperature. The sequences of the miRNA precursors and inhibitors used can be obtained from Ambion. After electroporation, transfected cells were resuspended in culture medium containing 10% FBS for 48–72 h before study. All studies were performed in quadruplicate unless otherwise specified.

Animal models.

All animal experiments were performed in accordance with protocols approved by the Baylor Scott & White Health Institutional Animal Care and Use Committee. Six-week-old male miR-21 knockout mice (129S6-Mir21a^tm1Yoli/J; stock no: 016856, miR-21 null) and B6 controls (101043 B6129SF1/J) were purchased from Jackson Laboratories (Bar Harbor, ME) and housed in a temperature-controlled environment (22°C) with 12-h:12-h light-dark cycles. Male C57Bl/6J mice and TLR4 knockout mice were also purchased from The Jackson Laboratory and used for ethanol (EtOH) treatments and studies with miR-21 knockout/control groups. The mice (10 wk old, n = 20) were randomly divided into four groups: wild-type (WT) mice, ethanol-fed mice (Leiber-DeCarli liquid diet EtOH, ethanol feeding for 5 wk), miR-21 knockout mice, and ethanol-fed miR-21 knockout mice (for 5 wk). For chronic intragastric ethanol administration, mice were aseptically implanted with gastrostomy catheters as described previously (35, 43). A dose of liquid ethanol (5%) or control solution was infused for 5 wk (35). After 5 wk of treatment, mice were weighed and anesthetized. Livers were then excised and weighed, and portions were fixed in formalin, frozen in optimal cutting temperature medium (Sakura Finetek, Torrance, CA), snap-frozen in liquid nitrogen, and stored in −80°C for further use.

Human healthy control and steatohepatitis samples.

Healthy human liver (n = 4) and liver samples of steatohepatitis patients with heavy alcohol consumption (n = 4) were purchase from Xenotech (Kansas, KS). The samples were used for RNA extraction, frozen section slides, and protein extraction. The characteristics of patients are listed in Table 1.

Table 1.

Characteristics of liver donors from control and steatohepatitis patients with heavy alcohol consumption

Sample ID	Product No.	Diagnosis	Macro-Fat, %	Age, yr	Sex	Ethnicity	BMI	Alcohol Use	Alcohol Frequency	Diabetes
H1255	HHPL.NT	Normal	1–2	56	F	African American	25	No	N/A	No
H1293	HHPL.NT	Normal	0	52	F	Caucasian	29.1	No	N/A	No
H1296	HHPL.NT	Normal	0	46	M	Caucasian	21.1	No	N/A	No
H1299	HHPL.NT	Normal	0	17	F	Caucasian	20.6	Yes	Occasional	No
H1271	HHPL.HST	Steatohepatitis	75	41	M	Caucasian	23.5	Yes	Heavy	No
H0959	HHPL.HST	Steatohepatitis	40	48	M	Caucasian	32.2	Yes	Heavy	No
H1063	HHPL.HST	Steatohepatitis	80	43	M	Hispanic	19.5	Yes	Heavy	No
H1259	HHPL.HST	Steatohepatitis	20	64	M	Caucasian	33.2	Yes	Heavy	Yes

BMI, body mass index; M, male; F, female.

Isolation of mouse HSCs by laser capture microdissection and treatment of cultured human HSC lines.

Mouse HSCs were isolated by laser capture microdissection (LCM) as described (42) (using desmin as a marker of HSCs). The RNA from LCM-isolated HSCs were extracted with the Arcturus PicoPure RNA isolation kit (Thermo Fisher Scientific) according to the instructions provided by the vendor. The expression of IL-6, monocyte chemoattractant protein-1 (MCP-1), and IL-1β was measured in these cells by quantitative polymerase chain reaction (qPCR). The in vitro studies were performed in HSCs (Sciencell) that were seeded into six-well plates the day before transfection. The cells were transfected with precursors/inhibitors of miR-21 or control pre-miRNAs/anti-miRNAs using Lipofectamine RNAiMAX Transfection Reagent (Invitrogen, Life Technologies, Carlsbad, CA). Following culture with the transfection mix for 24 h, the cells were cultured in normal medium or medium containing 100 ng/ml LPS for 24 h. Cells were then harvested, and the expression of fibrosis markers was evaluated by qPCR or immunofluorescence. All studies were performed in quadruplicate unless otherwise specified.

Liver histological analysis and measurement of serum chemistry.

Liver histology was evaluated in mouse and human liver sections (4- to 5-μm thick) by hematoxylin and eosin staining. Hepatic fibrosis was assessed by Sirius Red staining in liver sections (4- to 5-μm thick) (10 different fields were analyzed from 3 different samples obtained from 3 different animals). Images were obtained by Leica scanner (Buffalo Grove, IL). Collagen deposition in liver sections with Sirius Red staining were quantified by using Olympus Image Pro-Analyzer software (Olympus, Tokyo, Japan). The pathology score was determined by the criteria proposed originally by Brunt et al. (7) for use in nonalcoholic steatohepatitis. In this criterion, the severity of parenchymal necroinflammatory activity is reflected by the grade of the lesion. Grade has been shown to correlate with the AST and ALD levels. Mild activity (grade 1) represents steatosis involving up to 66% of lobules and mild steatohepatitis, while severe grade 3 represents panlobular steatosis and florid steatohepatitis (7, 8). The serum levels of alanine aminotransferase (ALT) were measured by a Dimension RxL Max Integrated Chemistry System (Dade Behring, Deerfield, IL) located at Baylor Scott & White Health. The expression of VHL was detected in human liver sections by immunohistochemistry.

SuperArray quantitative PCR assay and qPCR analysis.

RNA was isolated from liver tissues or cell lysates using TRIzol (Invitrogen) according to the manufacturer's protocol. The RNA was subsequently cleaned using the RNeasy kit (Qiagen, Valencia, CA) according to the manufacturer's protocol. The optional on-column DNase treatment was performed. Reverse transcription was done using 1 μg of RNA with RT² first strand kit according to the manufacturer's protocol (SABiosciences, Frederick, MD). Mouse liver tissues or normal human HSC cDNAs were analyzed using SuperArray plates (Human Chemokines & Receptors PCR Array, PAHS-022A). To validate the translational significance of the findings related to gene expression, mouse liver or human HSC samples were analyzed using real-time PCR. RT² qPCR primer assays (SABiosciences) or TaqMan miRNA PCR assays were used. Real-time PCR was performed using RT² SYBR Green/ROX qPCR master mix for a Stratagene Mx3005P real-time PCR system according to the manufacturer's protocol (SABiosciences). ROX was used as an endogenous reference, and data were analyzed using the PCR array data analysis template (SABiosciences). The comparative C_T method (ΔΔC_T) was used for quantification of gene expression. All samples were tested in triplicate, and average values were used for quantification. The genes related to miR-21/NF-κB associated inflammatory signaling were analyzed using Ingenuity Pathway Analysis (IPA) software (Ingenuity System; Qiagen, Redwood City, CA) for the functionally relevant pathway.

Immunofluorescence and Western blot assay.

The expressions of α-smooth muscle actin/vimentin/fibronectin/tissue inhibitor of metalloproteinase-3/MMP-2/MMP-9 in liver sections (6 to 8 μm thick) and the expressions of VHL and IkBα in cultured HSCs were evaluated by immunofluorescence (14). For the detection of VHL and IkBα expression, cells were seeded on glass coverslips. Before staining, the cells were washed with 1× PBS. Then, immunofluorescent staining for VHL and IkBα similar to staining in liver sections was performed. Following staining, images were obtained by using Leica AF 6000 Modular Systems. The protein of cultured HSCs was extracted with lysis buffer and quantified by the bicinchoninic acid method (Pierce Biotechnology, Rockford, IL). Then, the protein expression of VHL, IkBα, and β-actin was evaluated by immunoblots as described previously (42). Protein expression was visualized and quantified using the LI-COR Odyssey Infrared Imaging System (LI-COR Bioscience, Lincoln, NE).

Luciferase reporter assay.

Intact putative miR-21 recognition sequences from the 3′-untranslated region (UTR) of VHL (pMIR-VHL-WT-3′-UTR) or with random mutations (pMIR-VHL-mut-3′-UTR) were cloned downstream of the firefly luciferase reporter gene. HSCs were cotransfected with 1 μg of pMIR-VHL-WT or pMIR-VHL-MUT construct and 1 μg of pRL-TK Renilla luciferase expression construct with or without anti-miR-21/anti-miR-Con inhibitors using TransIT-siQUEST transfection reagent (Mirus, Madison, WI). Luciferase assays were performed 72 h after transfection using the Dual-Luciferase Reporter Assay system (Promega, Madison, WI).

ELISA assay.

Conditioned media were collected and stored at −80°C until use. Human IL-1β, IL-6, and MCP- 1 were measured using a Thermo Fisher Scientific/Invitrogen cytokine-ELISA kit according to the manufacturer’s protocols. Briefly, conditioned media were added to plates and incubated at room temperature for 4 h, followed by incubation with biotinylated monoclonal antibodies. Avidin-conjugated peroxidase was added to the plates, and enzyme activity was detected with SpectraMax iD5 Multi-Mode Microplate Reader from Molecular Devices (Sunnyvale, CA). The sensitivities of human IL-1β, IL-6, and MCP-1 for these ELISA kits were 5–300, 40–2,000, and 20–1,000 pg/ml, respectively.

Migration assay.

A migration assay was performed with polycarbonate membrane inserts with 5-μm pore size using a CytoSelect 24-well cell migration assay kit (Cell Biolabs, San Diego, CA) according to the manufacturer’s instructions. In brief, 5 × 10⁵ THP-1 cells were purchased from ATCC (TIB-202) and placed in the upper chamber, and conditioned medium from LPS-stimulated HSCs was added to the lower chamber. After 3 h of incubation in a cell culture incubator, migratory cells that had detached from the bottom side of the inserted membrane and cells that had migrated into the lower chamber were combined. The migratory cells were lysed and quantified using CyQuant GR Fluorescent Dye (Invitrogen). Fluorescence was measured using a SpectraMax iD5 Multi-Mode Microplate Reader from Molecular Devices with a 480/535 nm filter set. All migration assays were performed in triplicate.

Statistical analysis.

All data are expressed as means ± SE. The differences between groups were analyzed by Student’s unpaired t-test when two groups were compared or one-way ANOVA when more than two groups were compared by using SPSS 22.0 software. P < 0.05 was used to indicate statistically significant differences.

RESULTS

Expression of miR-21 is upregulated along with enhanced inflammatory responses in the livers from patients with steatohepatitis induced by heavy alcohol consumption.

miR-21 plays a critical role in both liver physiology and the pathogenesis of ALDs (21). qPCR revealed that the expression of miR-21 increased in the livers of patients with steatohepatitis induced by heavy alcohol consumption compared with healthy controls (Fig. 1A). Typical pathological changes were observed in liver sections from patients with steatohepatitis compared with healthy controls (Fig. 1B). IPA was performed based on the data from Inflammatory Response & Autoimmunity PCR Array to determine the cellular context of the differentially expressed signaling mechanisms related to miR-21-mediated liver injury. IPA analysis indicated that the inflammatory response was the most altered signaling through NF-κB-related pathological mechanisms (Fig. 1C). Real-time PCR analysis demonstrated the enhanced mRNA level of myeloperoxidase (MPO), a marker of neutrophil infiltration (48), as well as decreased levels of IκBα gene expression, a rapid, sensitive, and powerful method to quantify the transcriptional power of NF-κB (6), in livers from patients with steatohepatitis compared with healthy controls (Fig. 1, D and E). As well, the mRNA expression of inflammation markers (IL-6, MCP-1, and IL-1β) was increased in patients with steatohepatitis compared with healthy controls (Fig. 2A). Increased mRNA expression of TNF-α and LPS receptor TLR4 was also observed in livers from patients with steatohepatitis when compared with healthy controls (Fig. 2, B and C).

Fig. 1.

The expression of microRNA-21 (miR-21), neutrophil infiltration, and inflammation are increased in livers from heavy alcohol consumers with steatohepatitis. A: expression of miR-21 was upregulated in the livers from steatohepatitis patients with heavy alcohol consumption compared with healthy controls by Taqman real-time PCR analysis (n = 4). B: histological changes was detected by hematoxylin and eosin (H&E) staining (original magnification: ×10; scale bar = 100 μm). C: Ingenuity Pathway Analysis software analysis based on showed that several genes implicated in inflammatory response are regulated by miR-21, including IL-6, monocyte chemoattractant protein-1 (MCP-1), IL-1β, as well as NF-κB signaling mediators such as NF-κB, MYD88, von Hippel-Lindau (VHL), Toll-like receptor-4 (TLR4), and TNF-α. D and E: enhanced neutrophil infiltration and NF-κB activation were verified in heavy alcohol drinkers with steatohepatitis compared with healthy controls by real-time PCR analysis of myeloperoxidase (MPO), the marker of neutrophil infiltration, and IkBα, the negative regulator of NF-κB signaling pathway. *P < 0.05 vs. healthy controls.

Fig. 2.

Aberrant expressed inflammatory cytokines in the livers from heavy alcohol drinkers with steatohepatitis. A and B: total RNA was isolated from steatohepatitis patients with heavy alcohol consumption compared with healthy controls, and real-time PCR analysis was performed as described in materials and methods. The mRNA expression of inflammation markers [IL-6, monocyte chemoattractant protein-1 (MCP-1), and IL-1β; A] and cytokine (TNFα; B) was increased in steatohepatitis patients compared with healthy controls. C: increased expression of LPS receptor Toll-like receptor-4 (TLR4) in heavy alcohol consumers with steatohepatitis was verified by immunohistochemistry analysis compared with healthy controls (original magnification: ×20; scale bar = 100 μm). D: total RNA was collected from hepatocytes (CK8), cholangiocytes (CK19), hepatic stellate cells (HSCs; desmin), and macrophages (F4/80) isolated from control and ethanol-treated mice liver by laser capture microdissection using specific markers, and Taqman real-time PCR assay was carried out to detect miR-21 expression. Ethanol feeding significantly increased the hepatic expression of miR-21 in hepatocytes, cholangiocytes, HSCs, and macrophages when compared with control mice. *P < 0.05 vs. healthy controls.

Ethanol feeding increases miR-21 expression and miR-21 depletion decreases inflammatory response in isolated HSCs in ethanol-fed mice.

Ethanol feeding significantly increased the hepatic expression of miR-21 in hepatocytes (CK8), cholangiocytes (CK19), HSCs (desmin), and macrophages (F4/80) by LCM using specific markers when compared with control mice (Fig. 2D). Furthermore, miR-21 expression was significantly decreased in miR-21 knockout (miR-21^−/−) mice fed EtOH compared with WT EtOH-fed mice (Fig. 3A). In EtOH-fed miR-21^−/− mice there was decreased serum ALT levels compared with WT EtOH-fed mice (Fig. 3B). Furthermore, in EtOH-fed mice, liver cell swelling, irregular nucleus size, shrinkage in varying degrees, and fatty degeneration of a large number of liver cells were all observed (Fig. 3C). These pathological changes and scores were reduced in EtOH-fed miR-21^−/− mice compared with WT EtOH-fed mice (Fig. 3D), along with the significant reductions of hepatic stellate activations by immunofluorescence staining of HSC markers (Fig. 4).

Fig. 3.

microRNA-21 (miR-21) depletion attenuates alcoholic liver injury in ethanol-fed mice. A: miR-21 expression was assessed by Taqman real-time PCR assay in wild-type (WT) and miR-21 KO mice with ethanol treatment relative to controls (n = 5). Ethanol feeding significantly increased the hepatic expression of miR-21 compared with WT control mice. Furthermore, miR-21 expression was significantly decreased in miR-21 KO (miR-21^−/−) mice fed EtOH compared with WT EtOH-fed mice. B:iIn EtOH-fed miR-21^−/− mice there was decreased serum ALT levels compared with WT EtOH-fed mice. C: in EtOH-fed mice, liver cell swelling, irregular nucleus size, shrinkage in varying degrees, and fatty degeneration of a large number of liver cells were all observed. D: pathological changes and scores were reduced in EtOH-fed miR-21^−/− mice compared with WT EtOH-fed mice. The results shown represent the means ± SE from 4 independent experiments. *P < 0.05, relative to controls. #P < 0.05, relative to EtOH-fed mice.

Fig. 4.

microRNA-21 (miR-21) depletion inhibits hepatic stellate cell activation during alcoholic liver injury. A and B: alcoholic liver injury was induced by 8 wk of ethanol feeding in wild-type (WT) and miR-21 knockout (KO) mouse and reduced expressions of hepatic stellate cell activation markers α-smooth muscle acting (α-SMA), tissue inhibitor of metalloproteinase-3 (TIMP3), matrix metalloproteinase (MMP-2; A), as well as the mesenchymal cell proliferation markers vimentin and fibronectin were observed by immunofluorescence staining in EtOH-treated miR-21^−/− mouse livers when compared with EtOH control. Original magnifications: ×200.

Human HSCs produce proinflammatory cytokines after LPS stimulation.

LPS was the first identified and best understood inflammatory inducer in the pathogenesis of ALD (39, 47). Next, we examined whether human HSCs responds to the TLR4 ligand LPS (24). The expression levels of miR-21, IL-6, MCP-1, and IL-1β were examined and found to be upregulated in HSCs treated with LPS for 24 h (Fig. 5A). We also confirmed the downregulation of protein levels of IκBα and VHL (Fig. 5B). TNF-α is an important cytokine in the liver as well, and its expression is triggered by LPS in HSCs during alcoholic liver injury (45). TNF-α mRNA expression and secretion were also upregulated at 12 h after the addition of LPS, but they were decreased to basal level at 24 h poststimulation (Fig. 5, C and D). Together, these results suggest that LPS induced inflammatory cytokines in HSCs, supporting the concept that HSCs possess a functional VHL/NF-κB signaling pathway.

Fig. 5.

Regulation of inflammatory response by LPS and in human hepatic stellate cells (HSCs). A: total RNA was isolated from human HSCs treated with LPS (100 ng/ml, 24 h) relative to controls. Real-time PCR analysis confirmed enhanced expressions of microRNA-21 (miR-21) as well as the inflammatory cytokines IL-6, monocyte chemoattractant protein-1 (MCP-1), and IL-1β in LPS-treated HSCs when compared with controls. B: immunofluorescence studies were carried out in control and LPS-treated human HSCs. The expressions of negative regulators of NF-κB signaling pathway, IkBα, and von Hippel-Lindau (VHL), as well as the inflammatory cytokine MCP-1 were detected by immunofluorescence staining. C and D: regulation of TNF-α by LPS in human HSCs. Total mRNA (C) and protein levels (D) of TNF-α were detected in LPS treated HSCs relative to controls at 12 and 24 h poststimulation by real-time PCR and ELISA assay. TNF-α mRNA expression and secretion were upregulated at 12 h after the addition of LPS, but they were decreased to basal level at 24 h poststimulation. *P < 0.05, relative to the controls. #P < 0.05 vs. LPS at 12 h.

Anti-miR-21 treatment inhibits cytokine production in HSCs in vitro.

We next examined whether anti-miR-21 inhibits the production of inflammatory cytokines induced by LPS in HSCs. IL-6, MCP-1 and IL-1β mRNA expression levels were significantly reduced in HSCs transfected with miR-21 inhibitor compared with HSCs transfected with anti-miR-Con (Fig. 6A). The modulation of IL-6 and MCP-1 by miR-21 in conditioned medium was also verified by ELISA assay. Our results demonstrated that IL-6 and MCP-1 were reduced in conditioned medium from the anti-miR-21-transfected HSCs compared with control cells (Fig. 6, B and C). IL-1β in conditioned medium from each group of HSCs was as follows: the untransfected: 14.53 ± 2.98 pg/ml; anti-miR-Con-transfected: 12.34 ± 3.85 pg/ml; and anti-miR-21-transfected: 6.44 ± 1.27 pg/ml.

Fig. 6.

Anti-microRNA-21 (anti-miR-21) treatment inhibits cytokine production in LPS-activated human hepatic stellate cells (HSCs) and suppresses the migration of THP-1 cells. A: effects of anti-miR-21 on the expression of cytokine mRNA. HSCs were transfected with 50 nM control anti-miRNA (Anti-miR-Con) or 50 nM Anti-miR-21. After incubation in DMEM with 1% FCS, the cells were treated with 100 ng/ml LPS for 24 h. Cellular RNAs were isolated, and the expressions of IL-6, monocyte chemoattractant protein-1 (MCP-1), and IL-1β mRNA levels were examined by real-time RT-PCR. GAPDH mRNA was used for normalization. B and C: effects of anti-miR-21 on the expression of cytokines at protein level. Conditioned medium was collected from LPS-stimulated HSCs, and ELISA was performed to assess IL-6 and MCP-1 expression. D: migration of the human monocyte cell line THP-1 was suppressed by conditioned media from anti-miR-21-transfected HSCs. A total of 5 × 10⁵ THP-1 cells were placed in the upper chamber, and conditioned medium from LPS-stimulated HSCs transfected with Anti-miR-Con or Anti-miR-21 was added to the lower chamber. After 3 h of incubation, THP-1 cells that migrated toward the lower chamber were detected by fluorescent dye. Relative fluorescence units (AFU) are indicated (vs. 1%FCS). All results are shown as means ± SE. A minimum of 3 replicates were performed for each set of experiments to compile the data as presented. *P < 0.05, relative to Anti-miR-Con group.

Conditioned media from anti-miR-21-transfected HSCs suppress human monocyte-derived THP-1 cell migration.

We next examined the functional role of the anti-miR-21-mediated inhibition of inflammatory cytokine expression. Because MCP-1 functions as a chemoattractant cytokine for monocytes, macrophages, and Kupffer cells (2, 9, 49), which seem to contribute to ethanol associated inflammation in the liver, we focused on cell migration of the human monocyte-derived cell line THP-1. We performed in vitro migration assays to determine whether conditioned media from anti-miR-21-transfected HSCs suppress the migration of THP-1 cells. For this experiment, conditioned media from HSCs transfected with anti-miR-21 or anti-miR-Con were evaluated. The migration of THP-1 cells was markedly decreased (~30%) when conditioned media from HSCs transfected with anti-miR-21 were used (Fig. 6D). The number of migrated THP-1 cells was 3,650, 3,490, or 2,680/well when conditioned medium from HSCs transfected with mock control, anti-miR-Con, or anti-miR-21, respectively, was used. These data suggest that the inhibition of inflammatory cytokine production together with anti-miR-21 in HSCs inhibits the migration of monocytes and monocyte-derived cells, supporting the concept that miR-21 contributes to hepatic inflammation by promoting cytokine production and the recruitment of immune cells to the liver.

miR-21 modulates NF-κB signaling pathway in HSCs.

To further advance mechanistic insights into the role of miR-21 in innate immunity, including its effects on cytokines and the NF-κB signaling pathway, we assessed the gene expression profile of HSCs. We performed a Human Chemokines & Receptors PCR Array (selected based on IPA with the focus of NF-κB signaling-associated gene list) to identify miR-21 target genes in LPS-treated anti-miR-21-transfected HSCs compared with LPS treated anti-miR-Con-transfected HSCs (Fig. 7, A and B). Out of 84 NF-κB-signaling pathway-associated genes, 11 genes (13.1%) were upregulated by 2-fold or greater in anti-miR-21-transfected HSCs compared with anti-miR-Con-transfected HSCs in the presence of LPS stimulation (Fig. 7A; n = 4), and only 1 gene (1.2%) was upregulated by 2-fold or greater in anti-miR-21-transfected HSCs compared with anti-miR-Con-transfected HSCs in the presence of LPS stimulation (Fig. 7B; n = 4, P < 0.05). The inhibition of several key mediators of NF-κB pathway, such as NF-κB1, MYD88, and TLR4, was observed from the IPA analysis based on the PCR array data (Fig. 7C).

Fig. 7.

microRNA-21 (miR-21) modulates NF-κB signaling pathway in LPS-activated hepatic stellate cells (HSCs) in vitro. A and B: expression levels of key mediators of NF-κB signaling pathway are altered in anti-miR-21-treated HSCs after LPS stimulation. Relative gene expression profile between anti-miR-21-treated HSCs after LPS stimulation vs. anti-miR controls is shown. The expression of a panel of diverse inflammation-associated genes was evaluated by real-time PCR assay using Human Chemokines & Receptors PCR Array (PAHS-022 from SABiosciences), which was selected based on Ingenuity Pathway Analysis with the focus of NF-κB signaling associated gene list). A: gene expression relative to GAPDH was plotted as the Volcano Plots, depicting the relative expression levels (Log10) for selected genes in anti-miR-Con vs. anti-miR-21. B: the relative expression levels and P values for each gene in the related samples were also plotted against each other in the scatterplot. The key mediators of NF-κB signaling pathway, NF-κB1, Toll-like receptor-4 (TLR4), MYD88, and von Hippel-Lindau (VHL) are the most altered genes in anti-miR-21-treated HSCs after LPS stimulation. Data represent mean from 3 separate experiments. C: Ingenuity Pathway Analysis based on PCR array discoveries showed that miR-21 may target VHL and subsequently alter the NF-κB signaling pathway.

miR-21 directly regulates VHL expression.

In general, miRNAs inhibit gene expression; therefore, we focused on genes upregulated by anti-miR-21. In particular, PCR array analysis showed that anti-miR-21 upregulated VHL mRNA expression (3.38-fold; Fig. 7, A and B), and TargetScan Human database as well as out IPA analysis (Fig. 7C) revealed that VHL mRNA is a putative target of miR-21.

To validate whether miR-21 regulates VHL mRNA expression, we performed real-time PCR analysis of HSCs transfected with anti-miR-21 or anti-miR-Con. We confirmed the significant upregulation of VHL mRNA expression in the HSCs transfected with anti-miR-21 (Fig. 8A).

Fig. 8.

microRNA-21 (miR-21) regulates expression of von Hippel-Lindau (VHL). A and B: VHL is increased in anti-miR-21-treated hepatic stellate cells (HSCs) with or without LPS stimulation. mRNA and protein levels between anti-miR-21/pre-miR-21-treated HSCs with or without LPS stimulation vs. anti-miR/pre-miR control-treated HSCs were compared by real-time PCR (A) and immunocytochemical analysis (B). Inhibition (overexpression) of miR-21 significantly upregulated (downregulated) the expression of VHL with or without LPS stimulation. C and D: schematic of predicted miR-21 binding site in the 3′-untranslated region (UTR) of human VHL. C: luciferase reporter constructs containing the miR-21 recognition sequence from the 3′-UTR of VHL inserted downstream of the luciferase gene were generated. VHL-WT contains the intact sequence, whereas VHL-MUT contained the sequence with random nucleotide changes. Reporter constructs were cotransfected with either the anti-miR-21 inhibitor or anti-miR-Con in normal HSCs. The expression of firefly luciferase activity was normalized to that of Renilla luciferase activity for each sample. D: the increases in relative firefly luciferase activity in the presence of the miR-21 inhibitor indicate the presence of a miR-21-modulated target sequence in the 3′-UTR of VHL. Data represent the means of 8 separate experiments. *P < 0.05, relative to anti-miR-Con group. #P < 0.05 relative to anti-miR-21-VHL-WT group.

To examine whether miR-21 inhibits VHL expression at the protein level, we also performed immunofluorescence analysis of HSCs transfected with Pre-miR-21 or Pre-miR-C. A significant downregulation of VHL expression in the HSCs expressing miR-21 was observed at the protein level (Fig. 8B).

To confirm that miR-21 directly regulates VHL expression, we examined whether its overexpression inhibited the activity of a luciferase reporter construct containing VHL 3′-UTR (wild) or VHL 3′-UTR (mutant) (Fig. 8C). Anti-miR-21 treatment significantly increased the luciferase activity of VHL 3′-UTR (wild) but not the VHL 3′-UTR (mutant) (Fig. 8D), suggesting that the mutation in the seed sequence prevented the binding of miR-21 to 3′-UTR (Fig. 8C). Moreover, we observed an increase in the phosphorylation status of NF-κB in miR-21-transfected HSCs (Fig. 8E). Taken together, these data demonstrate that VHL is a direct target of miR-21.

Overexpression of VHL inhibits cytokine production in HSCs.

After the specificities of wt-VHL and Lac Z were validated (Fig. 9, A and B), the expression levels of IL-6, MCP-1, and IL-1β mRNA in HSCs transfected with wt-VHL or Lac Z were analyzed using real-time RT-PCR. Overexpression of VHL significantly decreased the expression levels of IL-6, MCP-1, and IL-1β mRNAs (Fig. 9C). The effects of the overexpression of VHL on IL-6 and MCP-1 expression were also confirmed by ELISA. The protein levels of IL-6 and MCP-1 were significantly decreased in the conditioned medium from the wt-VHL-transfected HSCs (Fig. 9D). IL-1β in the conditioned medium from the untransfected, Lac Z-transfected, or wt-VHL-transfected HSCs was undetectable, 5.33 pg/ml, or undetectable, respectively.

Fig. 9.

Overexpression of von Hippel-Lindau (VHL) inhibits cytokine production in hepatic stellate cells (HSCs) and suppresses migration of THP-1 cells. A: overexpression of VHL mRNA expression in human HSCs. HSCs were transfected with 10 μg Lac Z or wt-VHL constructs obtained from Addgene. After incubation in DMEM with 1% FCS for 24 h, the cells were treated with 100 ng/ml LPS for 24 h. Cellular RNAs were isolated, and VHL mRNA was examined by real-time RT-PCR. GAPDH was used for normalization. All results are shown as means ± SE for 4 separate experiments. B: overexpression of VHL protein expression in HSCs. Lysates from Lac Z- and wt-VHL-transfected HSCs were labeled with antibody against VHL and β-actin. β-Actin was used as the internal control. The representative images and ratios from 4 separate experiments are displayed. C: effect of VHL overexpression on mRNA expression of cytokines. Cellular RNA was isolated from LPS stimulated HSCs with wt-VHL or Lac Z transfection, and the mRNA expression levels of IL-6, monocyte chemoattractant protein-1 (MCP-1), and IL-1β mRNA were examined by real-time RT-PCR. GAPDH was used for normalization. D: effect of VHL overexpression on protein expression of cytokines. Conditioned medium from LPS-stimulated HSCs was collected, and ELISA was performed to assess the expression of IL-6 and MCP-1. E: migration of THP-1 cells was suppressed by conditioned medium from wt-VHL-transfected HSCs. A total of 5 × 10⁵ THP-1 cells were placed into the upper chamber, and conditioned medium from LPS-stimulated HSCs transfected with Lac Z or wt-VHL constructs was added to the lower chamber. After 3 h of incubation, migration of THP-1 cells toward the lower chamber was detected with a fluorescent dye. Relative fluorescence units (AFU) are indicated (vs. Lac Z control). All results are shown as means ± SE. A minimum 3 of replicates were performed for each set of experiments to compile the data as presented. *P < 0.05, relative to Laz Z controls.

Conditioned media from wt-VHL-transfected HSCs suppress human monocyte-derived THP-1 cell migration.

We confirmed that the overexpression of VHL decreased IL-6, MCP-1, and IL-1β expression in HSCs. Next, we examined the effects of VHL overexpression on monocyte migration by in vitro migration assays. For this experiment, conditioned medium from HSCs transfected with Lac Z or wt-VHL was evaluated. The migration of THP-1 cells was markedly decreased (~60%) when conditioned medium from HSCs transfected with wt-VHL was used (Fig. 9E). The number of migrated THP-1 cells was 35,100, 38,700, or 25,900/well when conditioned medium from HSCs transfected with mock, Lac Z or wt-VHL, respectively, was used. These data suggest that the inhibition of inflammatory cytokine production together with the overexpression of VHL in HSCs reduces the migration of monocytes and monocyte-derived cells, supporting the notion that miR-21 promotes cytokine production in HSCs and the recruitment of immune cells to the liver at least in part due to the suppression of VHL.

Regulation of VHL expression and NF-κB activation by miR-21 depletion in vivo during alcoholic liver injury.

Significant VHL loss has also been observed in the livers of steatohepatitis patients with heavy alcohol consumption compared with healthy controls (Fig. 10A). VHL loss was shown to activate NF-κB, a family of transcription factors and cytokine produced by activated leukocytes and many other cell types that is involved in systemic inflammation. Activation of NF-κB by various cellular stimuli involves the subsequent degradation of its inhibitor, IkBα. Therefore, we examined the expression of VHL and activation of NF-κB involved in inflammatory responses in control and EtOH-exposed mouse liver tissues with miR-21/TLR4 depletion. Compared with the HSCs from control liver tissue isolated by LCM using the HSC-specific marker Desmin, the expression of VHL and IkBα was significantly decreased in HSCs from TLR4 knockout mice liver with ethanol exposure (Fig. 10, B and C). Reduced expression of VHL and IkBα in HSCs was also associated with the morbidity of TLR4 knockout mice with ethanol exposure, which was 30% higher than the EtOH control group (n = 5). To confirm the functional effect and relevance of miR-21-dependent modulation of VHL and IkBα expression in HSCs in vivo, we assessed the effect of miR-21 depletion on VHL and IkB-α expression in EtOH-exposed mouse liver. Knockout of miR-21 in EtOH-exposed mice increased VHL and IkB-α expression in total liver tissues as well as the isolated HSCs by LCM (Fig. 10, D–E). Double staining immunofluorescence and immunohistochemistry analyses have confirmed the enhanced IkBα phosphorylation (Fig. 11A, top) and degradation (Fig. 11B, bottom) as well as NF-κB (p65 subunit) translocation (Fig. 11B) in HSCs in ethanoltreated mice liver and the recovery effects by miR-21 depletion to partially inhibit HSC NF-κB activation (Fig. 11, A and B). Furthermore, as we showed before, the serum alanine aminotransferase level was significantly decreased after miR-21 depletion, along with reduced inflammatory responses in ethanol-exposed mouse liver sections (Fig. 3). These findings link miR-21 associated NF-κB transcription factors with putative mediators of inflammation in EtOH-exposed mice liver and suggest that deregulated expression of miR-21/NF-κB contributes to liver inflammation during acute (alcoholic hepatitis) or chronic liver disease (steatosis and steatohepatitis) through VHL-associated mechanisms.

Fig. 10.

Regulation of von Hippel-Lindau (VHL) Expression and NF-κB activation by microRNA-21 (miR-21)/Toll-like receptor-4 (TLR4) depletion in vivo during alcoholic liver injury. A: expression of VHL was downregulated in the livers from steatohepatitis patients with heavy alcohol consumption compared with healthy controls by immunohistochemistry analysis. The representative images from 4 separate experiments are displayed. B and C: alcoholic liver injury was induced by 8 wk of ethanol feeding in wild-type (WT) and TLR4 knockout (KO) mouse and enhanced expressions of VHL (B) and IkBα (C) were observed by real-time PCR analysis in EtOH-treated TLR4^−/− mouse livers when compared with EtOH control. D–F: VHL and IkBα expressions were detected in wild-type and miR-21 knockout mouse with ethanol feeding relative to controls; immunofluorescence (D), real-time PCR [in isolated HSCs by laser capture microdissection (LCM); E], and immunohistochemistry (F) analyses were carried out. Knockout of miR-21 in EtOH-exposed mice increased VHL and IkBα expression in total liver tissues as well as the isolated hepatic stellate cells (HSCs) by LCM. *P < 0.05, relative to control group; #P < 0.05, relative to miR-21 KO group. Original magnifications: × 200.

Fig. 11.

Regulation of IkBα phosphorylation, degradation, and NF-κB nuclear translocation by microRNA-21 (miR-21) depletion in hepatic stellate cells in vivo during alcoholic liver injury. A: alcoholic liver injury was induced by 8 wk of ethanol feeding in wild-type (WT) and miR-21 knockout (KO) mouse and phosphorylation of IkBα (top) and degradation of IkBα (bottom) were detected by double staining immunofluorescence analysis using HSC specific marker Desmin plus phosphor- or total IkBα antibodies in EtOH-treated miR-21 knockout mouse livers when compared with EtOH control. B: NF-κB nuclear translocation was detected in EtOH-treated miR-21 knockout mouse livers when compared with EtOH control by double staining immunohistochemistry analysis. Multiple antigen labeling was performed in the same tissue section using the VECTASTAIN System (Vector Laboratories, Burlingame, CA). Specific enzyme substrates were incubated in sections to develop contrasting optimal color (NF-κB, gray/black; desmin, brown). The representative images from 4 separate experiments are displayed. Original magnifications, ×400.

DISCUSSION

Chronic liver inflammation (steatosis and steatohepatitis) has been observed in various types of liver injuries including ALD and subsequently results in hepatic fibrosis and cirrhosis (13, 20). Gut-derived LPS-activated TLR4 signaling also contributes to hepatic inflammation and fibrosis of the liver (3, 32, 46). In this study, we described the role of depletion of miR-21 by gene knockout in the inhibition of inflammatory responses that are associated with ALD progression. We have shown that miR-21 is increased in mouse livers with ALD after the activation of LPS/TLR4 signaling in vivo, and it is overexpressed in ethanol-treated HSCs compared with controls. We demonstrated that LPS/TLR4 contributes to alcoholic liver injury and tissue repairing through miR-21 by modulating inflammatory responses in HSCs. Some of these effects are mediated through VHL/NF-κB signaling pathways, the well-characterized regulator genes of steatosis and steatohepatitis that are also involved in ALD. Increased expression of miR-21 was shown by in situ hybridization during liver injury and ALD, and a similar role for miR-21 has been postulated in systemic inflammatory responses (5, 33, 44). The concomitant miR-21-dependent activation of inflammation genes such as NF-κB in hepatic cells can facilitate not only the diseases progression but also tissue recovery (36). These findings taken together support a functional role for miR-21 in promoting liver inflammation during the development of ALD.

Liver inflammation is regulated by various cytokines and chemokines, which control the migration and activities of hepatocytes, cholangiocytes, Kupffer cells, HSCs, endothelial cells and other circulating immune cells (31, 32, 40). NF-κB is a transcriptional regulator of genes involved in immunity, inflammatory response, and cell fate and function (17). It seems to play a role in steatohepatitis in ALD and nonalcoholic steatohepatitis model mice and contributes liver injuries by recruiting monocytes to the liver (34). Kupffer cells, injured hepatocytes, cholangiocytes, and activated HSCs have an activated status of NF-κB (19). miR-21-deficient mice show decreased hepatic inflammation through the downregulation of NF-κB in HSCs, directly by the targeting of VHL and indirectly by responding to the inflammation underlying hepatocyte injury. In the present study, we showed another mechanism of NF-κB activation by miR-21 through inhibition of VHL, suggesting the critical role of HSCs in hepatic inflammation during ALD.

Recent studies have identified the biochemical link between VHL loss and heightened NF-κB activity (19, 30). Specifically, it has been shown that biallelic inactivating mutations of VHL induce NF-κB activity through the accumulation of hypoxia-inucible factor-α (HIF-α) expression (1). In turn, HIFα drives expression of TGF-α, which consequently activates an EGFR/PI3K/AKT/IKKα/NF-κB signaling cascade that is critical to the inflammation and fibrosis response. The biochemical connection between VHL/HIF-α and NF-κB may be potentially relevant not only to the HSCs, which manifest biallelic VHL inactivation, but also to any hepatic parenchymal and nonparenchymal cells that undergo hypoxia with consequent accumulation of HIF-α. Interestingly, hypoxia also activates NF-κB by enhancing DNA binding and increasing transcriptional activity of NF-κB in other cell types, including hepatocytes and cholangiocytes (25, 28). Hypoxia-induced NF-κB activation may also be operative in other cells inside the liver. For instance, hypoxia has been shown to activate NF-κB in vascular endothelial cells and macrophages (10). Interestingly, the infiltration by macrophages is associated with a poor prognosis in many human liver disorders, including human ALD, and it is conceivable that factors elaborated by macrophages in response to NF-κB, such as IL-6, IL-8, and VEGF, may drive liver injuries through induction of angiogenesis or perhaps direct effects on inflammatory responses and malignant transformation. Given the importance of macrophages and angiogenesis in physiological responses to inflammation, wound healing, ischemia, and other disease states characterized by hypoxia, the role of hypoxia-induced NF-κB may extend beyond the scope of various hepatic pathogenesis.

Our findings identify a previously unrecognized mechanism for direct regulation of VHL/NF-κB signaling in HSCs, involving ncRNA in ALD. The abusive consumption of alcohol can cause serious cellular injuries that often lead to liver inflammation, fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) (4). Aberrant miR-21/NF-κB signaling has been implicated in many human diseases, including ALDs (16, 18, 26, 41). Recent developments indicate that ethanol induces alterations of the miR-21/NF-κB signaling pathway, particularly on enhanced expression of IL-6-mediated Stat3 pathways (11, 27). This has opened up a new area of interest in ethanol research and is providing novel insight into the actions of ethanol at the nucleosomal level in relation to gene expression, pathophysiological consequences, and injury recovery/liver regeneration. Although miR-21/NF-κB signaling has been tightly linked to liver injuries and disease outcome in many hepatic disorders, including human ALDs, its application to ethanol-dependent ncRNA expression in HSCs and liver inflammation is novel. A better understanding of how ethanol interacts with specific cytokines to contribute to aberrant ncRNA expression will clearly advance the field and increase our understanding of the mechanisms of HSC-mediated liver inflammation involved in the development of ALDs. Genomic scanning approaches to identify transcriptional/translational factors and their modified targets of HSCs in ALD are lacking, but such strategies could identify other novel targets that could be genetically or epigenetically modified in ALD.

With chronic alcohol abuse, early and reversible liver inflammation occurs in the form of fatty liver long before the onset of clinically symptomatic and irreversible form of hepatitis and cirrhosis. Currently, the mechanisms and regulation of steatosis and steatohepatitis by HSCs during ALDs are poorly understood, and the HSC-associated putative biomarkers and regulatory molecules link to liver inflammation remains tenuous and unproven. In this study we characterized the role of depletion of miR-21 in ethanol-induced altered VHL/NF-κB inflammation signaling in HSCs and its progression to ALD. The data presented here may have direct application to the future translational research for an improved diagnosis and treatment of inflammatory responses in ALD patients.

GRANTS

This work was supported by Veterans Affairs (VA) Merit Award 1I01BX001724 and National Institutes of Health (NIH) Grant R21-AA-025997 (to F. Meng), Dr. Nicholas C. Hightower Centennial Chair of Gastroenterology from Baylor Scott & White, VA Research Career Scientist Award and VA Merit Award 5I01BX000574 (to G. Alpini), VA Merit Award 5I01BX002192 (to S. Glaser), and NIH Grants DK-058411, DK-076898, DK-115184, DK-110035, and AA-025157 (to G. Alpini, F. Meng, and S. Glaser). Portions of this work were supported by VA Merit Award 1I01BX003031 from the US. S. Department of Veteran’s affairs, Biomedical Laboratory Research and Development Service and NIH Grant R01-DK-108959 (to H. Francis).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors. This material is the result of work supported with resources and the use of facilities at the Central Texas Veterans Health Care System, Temple, TX. The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Veterans Affairs or the United States Government.

AUTHOR CONTRIBUTIONS

N.W., K.M., and F.M. conceived and designed research; N.W., K.M., T.Z., S.R.L. L.H., D.C., and T.A. performed experiments; N.W., K.M., T.Z., and F.M. analyzed data; C.W., H.F., S.G., G.A., and F.M. interpreted results of experiments; N.W., K.M., T.Z. prepared figures; N.W., K.M., and F.M. drafted manuscript; H.F., S.G., G.A., and F.M edited and revised manuscript; N.W., K.M., S.R.L., L.H., T.Z., D.C., S.G., H.F., T.A., G.A., and F.M. approved the final version of the manuscript.