The effect of Viagra (sildenafil citrate) on liver injury caused by chronic ethanol intragastric feeding in rats

https://doi.org/10.1016/j.yexmp.2004.09.004Get rights and content

Abstract

Rats fed with ethanol and a nutritious diet intragastrically develop liver pathologic changes associated with cyclic elevation of blood and urinary ethanol levels (BAL and UAL cycle). At the peaks of the UAL cycle, the livers are hypoxic. When the liver portal hepatic blood flow is temporarily clamped for 2 min and then released, the livers at the peak UAL fail to recover completely compared to the control livers and the livers at the UAL cycle troughs. Viagra was fed to the ethanol-fed rats to enhance the effects of nitric oxide. Since nitric oxide is known to increase hepatic blood flow, it was anticipated that Viagra would prevent the liver hypoxia at the UAL cycle peaks and also improve the post-clamp recovery from the post-clamp ischemia challenge. Viagra tended to improve the post-clamp recovery of the liver surface pO2 levels of the ethanol-fed rats probably by slowing O2 consumption as result of NO inhibition of mitochondrial cytochrome c oxidase activity. However, Viagra increased the pathology score when fed with ethanol. For this reason, Viagra is a two-edged sword. On the one hand, it tended to be protective in the post-ischemic injury in the ethanol-fed rats and on the other hand, it enhanced the liver injury caused by ethanol. Viagra did not affect the UAL cycle.

Introduction

To enhance the effects of nitric oxide (NO) in improving liver blood flow, rats were fed Viagra (sildenafil citrate) with ethanol and a liquid diet intragastrically for 1 month. To measure the improvement in liver, blood flow pO2 and perfusion units were measured on the surface of the liver in rats where the blood alcohol levels (BAL) and urinary ethanol levels (UAL) were high during the UAL cycle.

Viagra selectively inhibits cyclic guanosine monophosphate (cGMP)-specific phosphodiesterase type 5 (PDE5) resulting in enhanced smooth muscle relaxation initiated by nitric oxide (NO). Endothelial NO derived from NO synthase enters smooth muscle cells and activates guanylate cyclase, resulting in an increase in cGMP levels. This is responsible for the vasodilator effects of NO (Moncada and Higgs, 1993). In the case of the liver, liver sinusoidal endothelial cells are responsible for nitric oxide modulation of resistance to blood flow in the sinusoids (Shah et al., 1997). The half-life of cGMP is short, since it is rapidly degraded by PDE5 (Gibson, 2001). In the present study, the PDE5 inhibitor sildenafil (Viagra) was fed with ethanol to improve liver blood flow and prevent the liver hypoxia that develops during recovery from a brief ischemia episode induced by clamping the portal blood flow for 2 min.

There is evidence that an acute high does of ethanol causes an increase in NO levels (Baraona et al., 2002) as a response to adenosine release cause by ethanol (Orrego et al., 1988). Nitric oxide modulates hepatic vascular tone in normal rat liver (Mittal et al., 1994). A similar increase in NO occurs in binge drinking in man (Oekonomaki et al., 2004). In rats fed with ethanol intragastrically, nitrites increase in the liver to the same extent at high and low blood alcohol levels indicating that NO production is chronically increased in the liver regardless of the blood alcohol level (Bardag-Gorce et al., 2002). The question addressed in the present study is: will enhancement of NO effects change the circulation dynamics in the liver to alter the alcohol liver damage in chronic ethanol-fed rats? Nanji et al. (2001) showed that l-arginine supplement to enhance NO synthesis reversed the pathologic changes induced by intragastric ethanol feeding.

The intragastric ethanol feeding rat model used here is characterized by a cyclic change in blood alcohol levels (BAL) and urinary alcohol levels over a span of 7–10 days (Tsukamoto et al., 1985). Liver hypoxia occurs only when the BAL and UAL are very high during the cycle (Arteeel et al., 1997, French, 2004, Li et al., 2004a). Arteeel et al. (1997) likened this model to ischemic–reperfusion injury because liver hypoxia occurs when the BAL is high and reoxygenation occurs when the BAL is low during the cycle. Sinusoidal endothelial-derived NO has a protective effect during cold ischemia (Oishi et al., 2001). NO prevents hypoxia/reoxygenation injury in mouse liver (Taniai et al., 2004). NO protects rat hepatocytes against reperfusion injury by preserving the mitochondrial permeability transition (Kim et al., 2004). The role of NO as a cytoprotective agent during I/R injury, however, is complicated by the cytotoxic effects NO (Shah and Kamath, 2003). Thus, it is difficult to predict how NO enhancement will effect liver hypoxia and alcohol-induced liver damage. The results of this study indicate that NO enhancement did not effect the UAL cycle but improved the post-ischemic recovery of liver pO2. Paradoxically, liver injury was worse when Viagra was fed with ethanol.

The feeding experiment was performed using the intragastric tube-feeding model in which ethanol was fed at a constant rate for 1 month. In this model, the BAL cycles up and down over a 7- to 10-day period as previously described (Tsukamoto et al., 1985). At the end of the experiment, pO2 was monitor on the surface of the liver to measure ischemia–reperfusion injury after 2 min clamping of the porta hepatis.

Eight Wistar male rats, weighing about 275 g, were fed with ethanol at 13 g/kg/day for 6 weeks. Four of the rats were fed with Viagra (10 mg/kg/day) in the ethanol diet. Eight pair-fed control rats were fed a dextrose solution isocaloric to the ethanol. Four of the pair-fed control rats had Viagra (10 mg/kg/day) added to the diet and were pair-fed with ethanol + Viagra-fed rats. The diet provided 271 kcal/kg/day in all four groups. The diet used included a salt and vitamin mix (Dyets, Bethlehem, PA) as described by the American Institute of Nutrition Manual Mix and Vitamin Mix for Optimum Growth of Rats (French et al., 1997). The source of protein was lactalbumin. The diet was supplemented with 500 mg of choline and 1 g of methionine per liter of diet. The dietary calories were derived as follows: 33% from fat, 25% from protein, 6.7% from dextrose, and 34% from ethanol.

Rats fed with ethanol were monitored for blood ethanol levels (BAL) at the termination of the experiment. Urinary alcohol levels (UAL) were determined daily on 24 h urines collected under toluene to prevent evaporation. The UAL was measured daily to identify the peak and trough UALs in order to document the UAL cycle. UALs were measured using the QUED Saliva Alcohol Text Kit A150 (STC Technologies, Bethlehem, PA). Blood levels of serum alanine aminotransferase were measured by using a clinical analyzer (A kinetic rate method on Synch Ronex Systems; Beckman Instruments) at the beginning and the end of the experimental period. Body weights were measured weekly and at termination. Liver weights were measured at termination and the liver weight/body weight ratio calculated. The procedures were approved by the Research and Education Institute Animal Care Committee in accordance with guidelines for animal care as described by the National Academy of Sciences (1996).

Liver histology was examined after fixation of the liver in Zinc formalin and stained with hematoxylin and eosin. The liver pathology score was determined blinded to the experimental treatments of the rats using a previously reported scoring method (Bardag-Gorce et al., 2002). PCNA-positive hepatocytic nuclei were quantitated morphometrically using Nikon metamorph software. The liver sections were stained with a mouse monoclonal antibody to rat PCNA (clone PC10) (Signet Pathology Systems Inc., Dedham, MA).

This experiment was designed to test whether Viagra affected pO2 and perfusion unit levels before, during or after clamping of the portal hepatis. pO2 (mm Hg) was measured using an O2 probe on the surface of the liver during laparotomy under isofluorane anesthesia. The pO2 probe was calibrated to room air (159–160 mm Hg) using an automatic internal barometer for transcutaneous oxygen pO2 (tcp O2) before every use. The pO2 probe module used was PE 50 ys. (Perimed, North Royalton, OH). Perfusion arbitrary units (PU) were simultaneously measured as an indication of blood flow using a laser Doppler perfusion instrument (Laser Doppler System PR407-1 master probe). The laser Doppler module used was PF5010. (Perimed, North Royalton, OH). The Doppler probe was calibrated from zero PU and 250 PU using a motility standard which simulates random movement times velocity of concentrated moving blood cells. The data were analyzed using Data Acquisition Software PSW 325 (L) with the assistance of Robert Kostelny (North Royalton, OH).

The pO2 and perfusion units (PU) were recorded simultaneously using combined laser Doppler and transcutaneous oxygen probes. The perfusion unit reflects the movement of blood, which includes Brownian movement and movement caused by respiration and the excursion of the diaphragm, causing micromotion of the blood in the liver. When the diaphragm contracts downward, the pressure in the thorax becomes negative which increases hepatic venous outflow. When the diaphragm relaxes, the intrathoracic pressure increases and hepatic venous flow decreases. This is why the flow oscillates in synchrony with respiration. Because of this mechanism, when the portal blood flow is clamped, the baseline measurements still record movement of blood in synchrony with respiration.

Recording of pO2 and the laser Doppler PU levels were taken over a 4-min baseline. Then, the hepatic portal vessels at the porta hepatis were clamped for 2 min. The clamp was then released and the recovery of pO2 and blood flow were continuously monitored for 4 min more. The rats were then terminated by exsanguination. Data for each point were calculated as the Mean ± SEM in the 4 rats/group, during the preclamp, clamp, and post-clamp recovery.

P values were determined by t test or one-way ANOVA, and all pair-wise multiple comparison procedures such as Bonferroni, or the Student–Newman–Keuls method (Sigma-Stat Software, San Francisco, CA).

Section snippets

Results

The rats in all four groups gained the same amounts of weight (Fig. 1). The weight of the liver was increased in both ethanol-fed and ethanol-plus Viagra-fed groups compared to their dextrose pair-fed controls (P < 0.001) (Fig. 2). The liver weights were not different when the ethanol-fed and the ethanol-plus Viagra-fed groups were compared. Likewise, the liver weight/body weight ratio was increased in the ethanol-fed and ethanol-plus Viagra-fed groups when compared to their controls (P <

Discussion

There is evidence that the UAL cycle causes ischemic/reperfusion (I/R) injury in the liver due hypoxia at the peaks of the cycle and reoxygenation at the troughs of the cycle (Arteeel et al., 1997, Bardag-Gorce et al., 2002, French, 2004). In I/R injury of the liver, post-ischemic oxidant stress increases the expression of inducible nitric oxide synthase (iNOS) among other factors (Jaeschke, 2003).

NO is potent vasodilator which acts intracellularly by activation of guanylate cyclase which

Acknowledgments

The authors would like to thank Adriana Flores for typing the manuscript and Fawzia Gorce-Bardag, PhD for digitizing the figures. Supported by NIH/NIAAA grants 8116 and P50-011999 Alcohol Center Grant on Alcoholic Liver and Pancreatic Diseases Morphology Core.

References (35)

  • V. Shah et al.

    Nitric oxide in liver transplantation: pathobiology and clinical implications

    Liver Transplant

    (2003)
  • A. Venkatraman et al.

    Chronic alcohol consumption increases the sensitivity of rat liver mitochondrial respiration to inhibition by nitric oxide

    Hepatology

    (2003)
  • Y. Watanabe et al.

    Expression of type I plasminogen activator inhibitor and nitric oxide generation in ischemic preconditioning of the rat liver

    Hepatol. Res.

    (2003)
  • R. Wiest et al.

    The paradox of nitric oxide in cirrhosis and portal hypertension: too much or too little

    Hepatology

    (2002)
  • F.H. Agani et al.

    Role of nitric oxide in the regulation of HIF-1α expression during hypoxia

    Am. J. Physiol.: Cell Physiol.

    (2002)
  • G.F. Arteeel et al.

    Chronic ethanol treatment causes hypoxia in rat liver in vivo

    Hepatology

    (1997)
  • T.M. Buetler et al.

    Role of superoxide as a signaling molecule

    News Physiol. Sci.

    (2004)
  • Cited by (19)

    • Protective effect of sildenafil on liver injury induced by intestinal ischemia/reperfusion

      2013, Journal of Pediatric Surgery
      Citation Excerpt :

      According to Ji et al. [8], sildenafil citrate did not induce significant changes in hepatic function and regenerative activity after partial hepatectomy in alcohol-fed rats. However, Li et al. [30] used PCNA to quantify the mitotic index in an experimental study designed to investigate the effects of sildenafil on liver injury caused by chronic ethanol feeding. They found an increase in the numbers of liver cells in the S phase in rats fed ethanol plus sildenafil.

    View all citing articles on Scopus
    View full text