Caveolin-1 is essential for protecting against binge drinking-induced liver damage through inhibiting reactive nitrogen species

Authors

  • Lei Gao,

    1. School of Chinese Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China
    2. School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China
    Search for more papers by this author
  • Yingchun Zhou,

    1. School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China
    Search for more papers by this author
  • Weichao Zhong,

    1. School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China
    Search for more papers by this author
  • Xiaohua Zhao,

    1. School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China
    Search for more papers by this author
  • Chun Chen,

    1. School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China
    Search for more papers by this author
  • Xingmiao Chen,

    1. School of Chinese Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China
    2. Research Center of Heart, Brain, Hormone & Healthy Aging, the University of Hong Kong, Hong Kong SAR, China
    Search for more papers by this author
  • Yong Gu,

    1. School of Chinese Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China
    Search for more papers by this author
  • Jianping Chen,

    1. School of Chinese Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China
    Search for more papers by this author
  • Zhiping Lv,

    Corresponding author
    1. School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China
    • Address reprint request to: Jiangang Shen, M.D., Ph.D., School of Chinese Medicine, Research Center of Heart, Brain, Hormone & Healthy Aging, The University of Hong Kong, 10 Sassoon Road, Pokfulam, Hong Kong SAR, China 999077. E-mail: shenjg@hku.hk; fax: +852 28725476; or Zhiping Lv, School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, Guangdong, China. E-mail: lzp48241@126.com; fax: +020 61648244.

    Search for more papers by this author
    • These authors contributed equally to this work.

  • Jiangang Shen

    Corresponding author
    1. School of Chinese Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China
    2. Research Center of Heart, Brain, Hormone & Healthy Aging, the University of Hong Kong, Hong Kong SAR, China
    • Address reprint request to: Jiangang Shen, M.D., Ph.D., School of Chinese Medicine, Research Center of Heart, Brain, Hormone & Healthy Aging, The University of Hong Kong, 10 Sassoon Road, Pokfulam, Hong Kong SAR, China 999077. E-mail: shenjg@hku.hk; fax: +852 28725476; or Zhiping Lv, School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, Guangdong, China. E-mail: lzp48241@126.com; fax: +020 61648244.

    Search for more papers by this author
    • These authors contributed equally to this work.


  • Potential conflict of interest: Nothing to report.

  • This work was supported by the Seed Funding Program for Basic Research (no. 201111159021), the Seed Funding Program for Applied Research (no. 201109160022), The University of Hong Kong, and the General Research Fund, Hong Kong SAR (HKU 777611M).

Abstract

Caveolin-1 (Cav-1) is known to participate in many diseases, but its roles in alcoholic liver injury remain unknown. In the present study, we aimed to explore the roles of Cav-1 in protecting hepatocytes from ethanol-mediated nitrosative injury. We hypothesized that Cav-1 could attenuate ethanol-mediated nitrosative stress and liver damage through regulating epidermal growth factor receptor/signal transducer and activator of transcription 3/inducible nitric oxide synthase (EGFR/STAT3/iNOS)-signaling cascades. Ethanol-fed mice had time- and dose-dependent increases of Cav-1 in serum and liver with peak increase at 12 hours. Compared to wild-type mice, Cav-1 deficiency mice revealed higher expression of iNOS, higher levels of nitrate/nitrite and peroxynitrite, and had more serious liver damage, accompanied with higher levels of cleaved caspase-3 and apoptotic cell death in liver, and higher levels of alanine aminotransferase and aspartate aminotransferase in serum. Furthermore, the results revealed that the ethanol-mediated Cav-1 increase was in an extracellular signal-regulated kinase–dependent manner, and Cav-1 protected hepatocytes from ethanol-mediated apoptosis by inhibiting iNOS activity and regulating EGFR- and STAT3-signaling cascades. In agreement with these findings, clinical trials in human subjects revealed that serum Cav-1 level was time dependently elevated and peak concentration was observed 12 hours after binge drinking. Alcohol-induced liver lesions were negatively correlated with Cav-1 level, but positively correlated with nitrate/nitrite level, in serum of binge drinkers. Conclusions: Cav-1 could be a cellular defense protein against alcoholic hepatic injury through inhibiting reactive nitrogen species and regulating EGFR/STAT3/iNOS-signaling cascades. (Hepatology 2014;60:687–699)

Abbreviations
1400W

N-(3-(aminomethyl)benzyl)acetamidine

ADH-1

alcohol dehydrogenases class 1

ALD

alcoholic liver diseases

ALDH2

aldehyde dehydrogenase class 2

ALT

alanine aminotransferase

AST

aspartate aminotransferase

BAC

blood-alcohol concentration

Bcl-2

B-cell lymphoma 2

b.w.

body weight

CAT

catalase

Cav-1

caveolin-1

EC

endothelial cell

EGFR

epidermal growth factor receptor

eNOS

endothelial nitric oxide synthase

ERK

extracellular signal-regulated kinase

GSH

glutathione

GSSG

glutathione disulfide

iNOS: inducible nitric oxide synthase; KO

knockout

mRNA

messenger RNA

NO

nitric oxide

3-NT

3-nitrotyrosine

P

phosphorylated

RNS

reactive nitrogen species

ROS

reactive oxygen species

RSNOs

S-nitrosothiol

SEM

standard error of the mean

SOD

superoxide dismutase

STAT3

signal transducer and activator of transcription 3

TUNEL

terminal deoxynucleotidyl transferase dUTP nick end labeling

WT

wild type

Binge drinking is defined as episodic excessive drinking of alcoholic beverages over a short period of time.[1] “Binge” commonly means consuming five or more standard drinks (male), or four or more drinks (female), on one occasion. Binge drinking becomes a major public health issue causing fatty liver, hypertension, neuronal damage, and so on.[2, 3] Many binge drinkers are susceptible to suffer from advanced alcoholic liver diseases (ALD), including steatohepatitis, hepatic fibrosis, and cirrhosis.[3, 4] Given that alcoholic liver injury has become a common health care problem, exploring underlying mechanisms of alcoholic liver injury is of great importance.

Both reactive oxygen species (ROS) and reactive nitrogen species (RNS) are important free radicals involved in ethanol-mediated liver injury.[5, 6] Redox homeostasis is regulated by multiple antioxidant enzymes, including superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase, glutathione reductase, and glutathione (GSH). During ethanol challenge, overproduction of free radicals, which exceed the antioxidant capacity, induces oxidation of proteins, lipids, and DNA, impair cellular functions, and trigger cell death in all types of liver cells. Free radicals play key roles in inflammation, ischemia, fibrosis, necrosis, apoptosis, and malignant transformation.[7, 8] Nitric oxide (NO) and peroxynitrite (ONOO), the representative RNS, are critical players in ethanol-mediated hepatotoxicity. Endothelial nitric oxide synthase (eNOS) and inducible NOS (iNOS) are two regulatory enzymes during ethanol consumption.[9, 10] Although a low dose of ethanol stimulates eNOS activation to produce a low concentration of NO and induce vasodilatation, a high dose can activate iNOS to generate a high concentration of NO and impair endothelial functions.[9] iNOS-mediated alcoholic liver injury has been confirmed by using iNOS knockout (KO) mice as well as pharmacological interventions, showing that both iNOS deficiency and inhibition of iNOS with N-(3-(aminomethyl)benzyl)acetamidine (1400W) had remarkably reduced ethanol-induced hepatotoxicity.[11, 12] Ethanol stimulation can mediate production of NO and superoxide anions (O2•−). When NO and O2•− are simultaneously produced, their reaction is extremely rapid, with a rate constant that is nearly diffusion controlled (k2 = 4.7 × 109 M−1 s−1) to form peroxynitrite and subsequently mediate liver injury.[13, 14] However, how cellular defense systems in response to ethanol stimulation for preventing hepatotoxicity occur remains largely unknown.

Two previous studies suggest a role of caveolin-1 (Cav-1), a plasma membrane-signaling protein, in preventing hepatotoxicity during alcohol consumption. Increased hepatic Cav-1 expression has been associated with reduced eNOS activity in chronic alcohol-treated rats.[15] Alcoholic hepatitis samples had significantly higher hepatic Cav-1 messenger RNA (mRNA) levels than cirrhosis and normal biopsy specimens.[16] As a scaffold protein, Cav-1 regulates multiple cellular events, including cell growth, apoptosis, and cholesterol trafficking.[17-19] Cav-1 can also bind all types of NOS and inhibit NO production[20-24] and modulates epidermal growth factor receptor (EGFR)-signaling activity by receptor sequestration and trafficking.[25] Signal transducer and activator of transcription 3 (STAT3), a DNA-binding transcription factor, can regulate iNOS activity through nuclear interaction with EGFR.[26] STAT3 is activated by cell-surface EGFR following its association with the C-terminus of EGFR.[26, 27] However, the underlying mechanisms of Cav-1 in regulating ethanol-induced liver injury remain unknown. In the present study, we conducted a series of cellular animal experiments and clinical trials to test the hypothesis that Cav-1 plays critical roles in preventing alcoholic liver injury through inhibiting RNS production and regulating EGFR/STAT3/iNOS signal cascades.

Materials and Methods

Animal Study

Experimental protocol was approved by the University of Hong Kong Institutional Animal Care and Ethical Committee. Male Cav-1 KO mice (Cav-1−/− mice, strain Cav-1tm1Mls/J; Jackson Laboratory, Bar Harbor, ME), Cav-1 heterozygous littermates (Cav-1+/−), and wild-type (WT) littermates (Cav-1+/+), between 8 and 12 weeks of age, were genotyped and subjected to binge drinking, as previously described.[28] After 6 hours of fasting, mice were given 50% (v/v) ethanol at an accumulative dosage of 5 g/kg body weight (b.w.) by intragastric administration. Vehicle mice received the same volume of water. In parallel studies, mice were treated with N-(3-(aminomethyl)benzyl)acetamidine (1400W; iNOS inhibitor, 5 mg/kg, intraperitoneally; Calbiochem, San Diego, CA) at 30 minutes before ethanol feeding, as previously reported.[11, 29]

Human Binge Drinking Project

Study Population

Serum samples were obtained from 57 healthy adult volunteers before and after binge drinking. All of the volunteers were healthy without long-term drinking or alcohol dependence. The study was registered at the Chinese Clinical Trial Registry (ChiCTR-TRC-13003344), guided by the ethical guidelines of the 1975 Declaration of Helsinki and approved by the appropriate institutional review committee with written informed consent from all participants.

Participants

Participants (n = 65) in the binge drinking project were young male adult volunteers 21-35 years of age. Subjects were nonalcoholic social drinkers that met the criteria for either binge drinkers (n = 40; consumed 5+ drinks in 2 hours) or vehicle control drinkers as placebo (n = 25; consumed the same volume of water in 2 hours). A binge was operationally defined as consuming 5 or more drinks per occasion for men, according to the National Institute on Alcohol Abuse and Alcoholism.

Statistical Analysis

Results were expressed as means ± standard error of the mean (SEM). Statistical analysis was performed using an unpaired t test or one-way analysis of variance, followed by Tukey's multiple comparison test on dependent experimental designs. Correlation tests between continuous variables were performed by Pearson's coefficient for linear regression. Spearman's rank correlation test was used if the regression was nonlinear. In the experiments, difference was considered to be statistically significance when P < 0.05.

See the Materials and Methods section in the Supporting Information for detailed information on experimental designs and methods.

Results

Ethanol Induced a Time- and Dose-Dependent Increase of Cav-1 in Mouse Liver and Serum

We first investigated the dynamic changes of Cav-1 in liver and serum after WT mice were treated with ethanol (5 g/kg b.w.). Samples were collected for Cav-1 detection at 0, 2, 6, 9, 12, and 24 hours after binge. Levels of Cav-1 in both serum and liver were increased and simultaneously reached maximum at 12 hours and dropped to normal at 24 hours after binge (Fig. 1A,B). There was a strong positive correlation between serum Cav-1 level and hepatic Cav-1 expression (r = 0.808; Fig. 1C). We then selected the time point of 12 hours for the following experiments, and mice were treated with ethanol with doses of 1, 3, 5, and 7 g/kg. Ethanol stimulation within a concentration of 5 g/kg led to a dose-dependent increase of Cav-1 in the liver (Fig. 1D). Notably, as indicated by staining with hepatocyte marker arginase-1[30] and endothelial cell (EC) marker CD31, increased Cav-1 expression was mainly located at hepatocytes (Fig. 1E) and ECs (Fig. 1F) around the central veins.

Figure 1.

Ethanol induced a time- and dose-dependent increase of Cav-1 in mouse liver and serum. (A) ELISA analysis for detecting time dependent response in serum Cav-1 level in WT mice after binge (n = 8/group). (B) Western blot analysis for detecting time-dependent response in hepatic Cav-1 expression in WT mice after binge (n = 12/group). (C) Spearman rank correlation test on the correlation between the serum and hepatic Cav-1 levels. (D) Western blot analysis for detecting dose-dependent responses in hepatic Cav-1 expression in WT mice after binge (n = 6/group). (E) Immunofluorescent analysis for costaining Cav-1 expression with arginase-1 marked for hepatocyte distribution. (F) Immunofluorescent analysis for costaining Cav-1 expression with CD31 marked for endothelial cells distribution. rs: Spearman's correlation coefficient. Data were represented as mean ±SEM (*P < 0.05, **P < 0.01, ***P < 0.001).

Cav-1 Deficiency Increased Susceptibility to Alcoholic Hepatic Injury in Mice

We then used WT, Cav-1+/−, and Cav-1−/− mice at the age of 8-12 weeks with matched body weights (Supporting Fig. 1A,B) and investigated the roles of Cav-1 in regulating alcoholic liver injury. Ethanol stimulation up-regulated the expressions of Cav-1 mRNA and protein in WT mice (Fig. 2A,B). Interestingly, both Cav-1+/− and Cav-1−/− mice had lower B-cell lymphoma 2 (Bcl-2) expression and higher cleaved caspase-3 in the liver tissues, compared to WT littermates, at 12 hours after ethanol treatment (Fig. 2A,C). Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay revealed that Cav-1 deficiency mice had higher rates of apoptotic cell death in liver tissues than WT littermates (Fig. 2D). Consistently, Cav-1−/− mice had higher levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in serum than WT littermates (Fig. 2E). These results indicate that Cav-1 has hepatic protective effects against alcoholic liver injury and cell death after binge drinking.

Figure 2.

Cav-1 deficiency increased susceptibility to alcoholic hepatic injury in mice. (A) Western blot analysis for the expressions of Cav-1, cleaved caspase3 and Bcl-2 in the livers of WT, Cav-1+/− and Cav-1−/− mice after binge (n = 6/group, two experiments). (B) Quantitative real-time PCR analysis for detecting Cav-1 mRNA in liver tissues of WT and Cav-1+/− mice after binge (n = 8/group). (C) Immunohistochemical studies on cleaved caspase3 in liver tissues of WT and Cav-1−/− mice after binge. (D) TUNEL assay for detecting apoptotic cell death in liver tissues. (E) The liver lesion was assessed by measuring serum ALT and AST in WT and Cav-1−/− mice after binge. Data were represented as mean ±SEM (*P < 0.05, **P < 0.01, ***P < 0.001).

Cav-1 Deficiency Promoted iNOS Expression and RNS Production in Mouse Liver After Binge Drinking

We subsequently addressed the underlying mechanisms of Cav-1 in regulation of alcoholic liver injury. First, we compared expression or activities of antioxidant and alcoholic metabolizing enzymes in WT and Cav-1−/− mice. Compared to WT littermates, Cav-1−/− mice had no significant difference in expressions of antioxidant enzymes in livers, including SOD-1, SOD-2, CAT, alcohol dehydrogenases class 1 (ADH-1) and aldehyde dehydrogenase class 2 (ALDH2; Fig. 3A and Supporting Fig. 2A), as well as activities of alcoholic metabolizing enzymes in serum, including ADH-1, total SOD and CAT, GSH, and glutathione disulfide (GSSG; (Supporting Fig. 2B,C). Interestingly, Cav-1 deficiency resulted in significant changes in expression of iNOS and production of RNS in mice livers after binge. Cav-1−/− mice had a higher expression level of iNOS than WT mice (Fig. 3A). We subsequently detected serum nitrite/nitrate levels, the oxidized derivatives of NO. Compared to WT mice, Cav-1−/− mice had significantly increased levels of nitrate/nitrite in serum (Fig. 3B). Moreover, we used HKGreen-3A, a newly developed highly selective ONOO probe,[31, 32] to detect ethanol-induced ONOO production. Cav-1−/− mice had significantly higher ONOO levels in livers than WT mice (Fig. 3C). Collectively, those results suggest that Cav-1 could inhibit iNOS activity and RNS production during alcoholic liver injury.

Figure 3.

Cav-1 deficiency promoted iNOS expression and RNS production in mouse liver after binge drinking. (A) Western blot analysis for the expressions of iNOS, eNOS and SOD1/2 in WT and Cav-1−/− mice after binge (n = 12/group). (B) Colorimetric detections of serum nitrite/nitrate levels in WT and Cav-1−/− mice after binge. (C) In situ detection of ONOO production with HKGreen-3A, a novel fluorescence probe for ONOO, in livers of WT and Cav-1−/− mice after binge (n = 6/group). Data were represented as mean ±SEM (*P < 0.05, **P < 0.01).

1400W, a Selective iNOS Inhibitor, Alleviated Alcoholic Liver Injury in WT and Cav-1−/− Mice

We further confirmed the involvement of iNOS-mediated nitration in alcoholic liver injury by using 1400W. After ethanol treatment, Cav-1−/− mice had higher expression levels of 3-nitrotyrosine (3-NT) and higher rates of apoptotic cell death in liver tissues than WT mice (Fig. 4A,B). As expected, 1400W did not affect hepatic Cav-1 level in WT mice (Supporting Fig. 3A,B). However, pretreatment of 1400W abolished ethanol-induced iNOS and 3-NT formation in both WT and Cav-1−/− mice (Fig. 4A-C) and inhibited serum iNOS activity and S-nitrosothiol (RSNOs) level in Cav-1−/− mice (Fig. 4D and Supporting Fig. 4). Treatment of 1400W alleviated hepatic apoptotic cell death (Fig. 4E) and decreased serum ALT and AST levels in WT and Cav-1−/− mice after binge (Fig. 4F). These results indicate that RNS-mediated nitration contributes to ethanol-mediated hepatic toxicity.

Figure 4.

1400W, a selective iNOS inhibitor, alleviated alcoholic liver injury in WT and Cav-1−/− Mice. (A) and (B), Immunofluorescent analysis for detecting the expressions of iNOS (A, left) and 3-NT (B, left) in livers of WT and Cav-1−/− mice fed with ethanol, and co-immunofluorescent staining of iNOS (A, right) or 3-NT (B, right) with TUNEL assay for detecting apoptosis in the livers of Cav-1−/− mice fed with ethanol (n = 6/group). (C) Western blot analysis for iNOS in Cav-1−/− mice after 1400W treatment (n = 12/group). (D) Biochemical analysis for serum iNOS activity in WT and Cav-1−/− mice after binge (n = 12/group). (E) TUNEL assay. (F) Effects of 1400W on serum ALT and AST levels in WT and Cav-1−/− mice after binge. Data were represented as mean ±SEM (*P < 0.05, **P < 0.01, ***P < 0.001).

Ethanol-Induced Cav-1 Expression Was in an ERK-Dependent Manner and Involved in the Regulations of EGFR and STAT3 During Alcoholic Liver Injury

The extracellular signal-regulated kinase (ERK) pathway may regulate Cav-1 expression in hepatocytes,[33] and the nuclear interaction of EGFR/STAT3 could activate iNOS to induce NO production.[26] Herein, we examined the interactions between Cav-1 and EGFR/ERK/STAT3-signaling cascades. After ethanol treatment, WT mice had increased Cav-1, phosphorylated (P)-ERK1/2, and P-STAT3 Tyr705, but unchanged P-EGFR in liver tissues. Cav-1−/− mice displayed much higher P-EGFR and P-STAT3 Tyr705 levels, but lower P-ERK1/2 levels, than WT mice (Fig. 5A). Results herein suggest that Cav-1 could inhibit P-STAT3 Tyr705 and P-EGFR, but promote P-ERK1/2 during alcoholic liver injury. Moreover, treatment of 1400W had no effect on Cav-1 and P-STAT3 Tyr705 (Supporting Fig. 3A,C), indicating that regulation of EGFR/STAT3/ERK cascades mediated by Cav-1 could be NO independent.

Figure 5.

Ethanol-induced Cav-1 expression was in an ERK-dependent manner and involved in the regulations of EGFR and STAT3 during alcoholic liver injury. (A) Western blot analysis for the expression of P-STAT3/STAT3, P-EGFR/EGFR, P-ERK/ERK and Cav-1 in WT and Cav-1-/- mice liver after binge (n = 12/group). (B) Western blot analysis for the expression of P-STAT3/STAT3, P-EGFR/EGFR, P-ERK/ERK and Cav-1 in L02 cells treated with 100 mM or 200 mM ethanol for 12 hrs. *P < 0.05, ethanol 200 mM Vs control in Cav-1; #P < 0.05 for ethanol 200 mM Vs control in P-STAT3 Tyr705/STAT3; ΔΔP < 0.01 for ethanol 200 mM Vs control in P-ERK/ERK. (C) Blocking ERK activation with PD98059 (20 μM, 40 μM) showing dose-dependent inhibition of Cav-1 expression in L02 cells. (D) Inhibiting STAT3 activation by SD-1029 (10 μM) showing no significant effect on Cav-1 expression in L02 cells. (E) Western blot analysis of P-ERK, ERK, P-STAT3, STAT3 and Cav-1 expressions in L02 cells treated with 20 nM Cav-1 siRNA or scrambled control siRNA for 6 hrs. Data were represented as mean ±SEM (*P < 0.05, **P < 0.01, ***P < 0.001). siRNA, short interfering RNA.

The in vitro cultured human L02 hepatocyte experiments revealed that ethanol dose dependently up-regulated expressions of Cav-1 and P-ERK1/2, but down-regulated P-STAT3 Tyr705 (Fig. 5B). Treatment with PD98059, an ERK selective inhibitor, down-regulated expression of Cav-1 (Fig. 5C), whereas incubation of SD-1029, a STAT3 inhibitor, had no effect (Fig. 5D). Knockdown of Cav-1 with RNA interference up-regulated expression of P-STAT3 Tyr705, but had no effect on P-ERK1/2, verifying the roles of Cav-1 in regulating ERK/STAT3-signaling cascades (Fig. 5E). Collectively, these in vivo and in vitro experiments provide direct evidence indicating that ethanol-mediated Cav-1 expression could be ERK dependent and that Cav-1 might have protective effects against alcoholic liver injury through promoting P-ERK, but inhibiting P-EGFR, P-STAT3, and iNOS signaling cascades.

Liver Lesions Negatively Correlated With Cav-1, But Positively Correlated With Nitrate/Nitrite Levels, in Binge Drinkers

We finally conducted clinical trials on volunteer healthy human subjects to further confirm the roles of Cav-1 against alcoholic liver injury after binge drinking. Baseline characteristics of the participants are presented in Table 1. Both binge drinking and placebo groups were comparable at entry regarding demographics, anthropomorphic measurements, and serological features on liver function. Fifty-seven participants completed the binge drinking study, whereas 8 dropped out (Fig. 6A). After binge, blood-alcohol concentration (BAC) reached maximum at 2 hours and back to normal within 12 hours (Fig. 6B). The increased serum ALT and AST levels were within the normal range, ensuring the safety of this clinical trial (Fig. 6C and Supporting Fig. 5). In a parallel study, no significant change in serum Cav-1 level was found in the placebo group at all time points, excluding a potential diet influence on results (Supporting Fig. 6A). Interestingly, both Cav-1 and nitrate/nitrite levels were increased in serum after binge, with maximum levels at 12 hours (Fig. 6D and Supporting 6B). With Pearson's linear correlation tests, we found a negative correlation between serum Cav-1 and ALT levels in the overall samples (n = 160; r = −0.244; Fig. 6E) and a positive correlation between serum nitrate/nitrite and AST levels (r = 0.3491), as well as the correlation between serum nitrate/nitrite and ALT levels (r = 0.1968; Fig. 6F). Besides, BAC reached maximum at 2 hours after binge, and BAC was negatively correlated with serum Cav-1 level (r = −0.4255; Supporting Fig. 7A). We then evaluated the ultimate protective effect of Cav-1 on liver function against binge and received a negative correlation between serum Cav-1 and ALT at 24 hours (r = −0.4004) and a positive correlation between serum nitrate/nitrite and AST at 24 hours (r = 0.4936; Supporting Fig. 7B,C). Conclusively, this clinical study indicates the correlations of serum Cav-1 and nitrate/nitrite levels with liver lesions and supports the protective roles of Cav-1 in inhibiting RNS production during alcohol liver injury.

Table 1. Biological Characteristics of Healthy Participants
CharacteristicsBinge DrinkingPlaceboTotal
  1. Data are presented as mean ± SD.

  2. Abbreviations: BMI, body mass index; A/G, albumin/globumin ratio; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

No.402565
SexMaleMaleMale
Age, years23.2 ± 1.423.5 ± 1.923.3 ± 1.6
Body weight, kg66.4 ± 8.3664.42 ± 9.065.7 ± 8.6
Height, cm173.5 ± 5.3172.3 ± 5.7173.1 ± 5.4
Total body water, L41.3 ± 3.040.5 ± 3.441.0 ± 3.2
BMI, kg/m222.1 ± 2.721.6 ± 2.421.9 ± 2.6
Consumption of ethanol, g75.9 ± 5.50 
Placebo, g074.3 ± 6.3 
Water, mL/h184.7 ± 13.3181.0 ± 15.4183.3 ± 14.2
Serum ALT level, U/L15.5 ± 6.014.1 ± 4.415.0 ± 5.4
Serum AST level, U/L19.9 ± 3.518.3 ± 3.819.3 ± 3.7
Total protein, g/L74.6 ± 4.3578.8 ± 4.4276.3 ± 4.75
Albumin, g/L45.9 ± 2.647.0 ± 2.546.3 ± 2.6
Globulin, g/L30.1 ± 8.431.8 ± 4.730.7 ± 7.3
A/G1.59 ± 0.451.52 ± 0.321.57 ± 0.40
Total bilirubin, µmol/L15.5 ± 5.1514.0 ± 3.6615.0 ± 4.68
Direct bilirubin, µmol/L3.69 ± 2.283.73 ± 1.623.71 ± 2.04
Indirect bilirubin, µmol/L11.9 ± 3.210.3 ± 2.211.3 ± 3.0
Total cholesterol, mmol/L4.29 ± 0.774.16 ± 0.694.24 ± 0.74
HDL, mmol/L1.3 ± 0.151.3 ± 0.211.3 ± 0.17
LDL, mmol/L2.42 ± 0.582.42 ± 0.512.42 ± 0.56
Figure 6.

Liver lesions negatively correlated with Cav-1, but positively correlated with nitrate/nitrite levels in binge drinkers. (A) Participants flow diagram illustrating the two study groups. (B) The BAC was assessed by measuring blood alcohol concentration and breath alcohol concentration using gas chromatography/mass spectrometry (GC-MS) and alcohol breath testing instrument. (C) The levels of ALT&AST in serum of binge drinkers detected by automatic biochemical analyzer. (D) ELISA analysis for Cav-1 level and colorimetric detection for serum nitrite/nitrate levels in the serum of binge drinkers (n = 32 per time point). (E) Pearson linear correlation tests for Cav-1 and ALT&AST levels in the serum of binge drinkers. (F) The correlation analysis for nitrite/nitrate and ALT&AST levels. rp: Pearson's correlation coefficient. Data were represented as mean ±SEM (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 7.

A model depicting the key role of Cav-1 in the pathogenesis of binge drinking-induced liver injury. Binge drinking induces the up-regulation of serum Cav-1 correlated with liver Cav-1 level due to the dispensability and dynamic changes of Cav-1. As the most important organ in ethanol consumption, the liver reveals more significant pathophysiologic change. First, for the acute phase reaction, the up-regulation of Cav-1 has a crosstalk with ERK activation. Excess ethanol stimulates the activation of EGFR/STAT3 pathway and subsequent iNOS transcription, whereas Cav-1 inhibits EGFR/STAT3 cascades and suppresses the over-expression of iNOS. On the other hand, Cav-1 directly inhibits iNOS activity via scaffold domain. Consequently, Cav-1 suppresses NO/ONOO releasing and protects against alcoholic liver injury as a master regulator of RNS stress.

Discussion

To our knowledge, this is the first report about the role of Cav-1 in regulating alcoholic liver injury. We provide direct evidence to demonstrate the dynamic changes of Cav-1 and its functions in preventing alcoholic hepatic injury as a master regulator for RNS production.

Cav-1 exists in plasma membrane, lipid droplets, the endoplasmic reticulum, Golgi, and/or mitochondria.[17, 34, 35] The dispensability of Cav-1 in liver allows redistribution of this protein from caveolae-enriched domains to the noncaveolar fractions,[36] which may account for its appearance in serum. The essential role of Cav-1 in regulating liver function is evident from the low survival of Cav-1−/− mice and impaired liver regeneration after partial hepatectomy.[37] A prominent increase of hepatic Cav-1 was found in cirrhosis and chronic alcoholic fatty liver,[15, 38] resulting from the roles of Cav-1 in regulating intracellular cholesterol trafficking[39] and lipid homeostasis.[34] Cav-1 appears to be a hepatocyte fate determinant for resistance to apoptosis.[40] Herein, we demonstrated that Cav-1 may play a critical role in regulating alcoholic liver injury. Both serum and hepatic Cav-1 levels were time and dose dependently increased in mice fed with ethanol. Cav-1 deficiency mice had aggravated alcoholic liver injury, as evidenced by increased ALT, AST, and cleavage caspase-3 and reduced Bcl-2 expression, along with increased hepatocyte apoptosis, in comparison to WT mice.

Cav-1 can inhibit iNOS through the caveolin-binding motif.[20-22] Caveolin-NOS interaction participates in cirrhosis, inflammation, and carcinogenesis.[22, 23, 38] Overexpression of iNOS increases the vulnerability to ethanol, and iNOS deficiency alleviates ethanol-induced hepatic injury.[10, 11] Besides, chronic alcohol consumption affects interaction of Cav-1 and eNOS as well.[15] In the present study, after binge, Cav-1 deficiency mice had higher levels of iNOS, 3-NT, NO, and ONOO in liver than WT mice, and 1400W treatment ameliorated alcoholic liver injury in both WT and Cav-1−/− mice. Those results support the notion that loss of Cav-1 could accelerate alcoholic liver injury through activating hepatic iNOS and inducing NO/ONOO production. Notably, we used a newly developed ONOO-specific fluorogenic probe, HKGreen-3A, which was designed on the basis of the rhodol scaffold and an ONOO-specific oxidation reaction. Previous studies demonstrated the specificity and sensitivity of the probe for detecting ONOO.[31, 32] After ethanol treatment, Cav-1−/− mice had higher fluorescence of HKGreen-3A and 3-NT than WT mice, directly showing the effects of Cav-1 on ONOO formation and tyrosine nitration. Consistently, Cav-1−/− mice had higher levels of nitrate/nitrite, RSNOs and iNOS than WT mice after binge, indicating the effects of Cav-1 on NO production. Those results imply that Cav-1 could inhibit the ethanol-induced RNS production in liver tissues.

Cav-1 can directly interact with EGFR and modulates EGFR-mediated signaling cascades.[41, 42] The nuclear interaction of EGFR/STAT3 promotes iNOS transcriptional activation.[26] In the current study, Cav-1−/− mice fed with ethanol had significant up-regulation of P-EGFR and P-STAT3 in liver, and 1400W had no effect on P-STAT3. Similar results were obtained in subsequent in vitro experiments. Both Cav-1 and P-STAT3 were significantly elevated in response to ethanol exposure. STAT3 inhibitor had no effect on Cav-1 expression, but Cav-1 silencing strikingly increased P-STAT3, suggesting that Cav-1 is the upstream signaling of STAT3. Those results indicate that the EGFR/STAT3/iNOS-signaling pathway could be a critical target of Cav-1 in regulating alcoholic hepatic injury. In addition, we speculate on the involvement of ERK in ethanol-mediated Cav-1 expression. According to the literature, there is a cross-talk between Cav-1 and ERK,[43] and activation of ERK leads to up-regulation of Cav-1 in hepatocytes.[33] Consistently, in vivo experiments revealed that Cav-1 deficiency blocked ethanol-mediated ERK activation in mice after binge. The in vitro human L02 hepatic cell studies corroborated the Cav-1-dependent ERK activation in response to ethanol stimulation. On the other hand, PD98059, a selective ERK inhibitor, showed an inhibitory effect on Cav-1 in L02 cells, but Cav-1 silencing did not affect ERK activation, indicating the regulatory effect of ERK on Cav-1 expression. The cross-talk between ERK and Cav-1 could be an important signaling pathway in alcoholic liver injury.

The human binge drinking project further confirmed the conclusion derived from cellular and animal experiments. Considering the ethical aspects, a clinical trial was designed to induce moderate acute alcoholic liver injury with levels of ALT and AST at safe ranges. We observed a time-dependent change of liver lesions after binge, which was correlated with serum Cav-1 in a negative manner, but with nitrate/nitrite levels in a positive manner. We also received a negative correlation between serum Cav-1 and BAC at the peak value time of ethanol concentration, corresponding to the in vivo data that Cav-1 deficiency aggravated mice drunken behaviors. Behavior studies demonstrated that Cav-1−/− mice had lower alcohol tolerance than WT mice, indicating the roles of Cav-1 in alcohol tolerance (Supporting Fig. 1C). Whether there is a relationship between serum Cav-1 level and individual ethanol tolerance is an interesting topic and deserves further in-depth studies. Attractively, correlation analysis showed a correlation bias toward ALT and AST with serum Cav-1 and nitrate/nitrite levels individually, which may be attributed to the different distribution of those transaminases in hepatocytes. Cav-1 is mainly located in plasma membrane and cytoplasm, where ALT is predominantly produced and gathered.[44] Except for the cytoplasm, mitochondrion is a major site for AST in liver,[44] which is susceptible to the binding of NO and cytochrome c oxidase.[45] Intracellular distribution may account for the strong correlation of AST with nitrate/nitrite. On the other hand, alcohol metabolizing enzymes and antioxidant enzyme activities are important regulating systems for preventing alcoholic hepatic injury. However, no significant difference was found in all of the alcohol metabolizing enzymes and antioxidant enzyme activities in both WT and Cav-1−/− mice after binge. Those results, taken together, suggest that Cav-1 could protect liver cells from alcoholic nitrosative injury by inhibiting RNS-mediated nitrosative damage instead of alcohol metabolisms and ROS-mediated oxidative injury.

In conclusion, Cav-1 plays essential roles in protecting hepatic cells against ethanol-induced cell injury. The underlying mechanisms are related to inhibited iNOS activity and RNS production and regulating EGFR- and STAT3-signaling pathways. Therefore, we raise the possibility that Cav-1 could be an important therapeutic target against binge drinking-induced liver injury.

Acknowledgment

The authors thank Prof. Dan Yang from the University of Hong Kong for supporting the ONOO fluorescence probe, Dr. Michelle Dee for her good suggestions and analysis, Drs. Xi Chen and Tingting Yan for their technical support, and Ms. Xiaoxia Zhu and Dr. Shaohui Huang from Southern Medical University for their assistance on clinical trials.

Ancillary