Dose response and time course studies on superoxide dismutase as a urinary biomarker of carbon tetrachloride-induced hepatic injury in the Hanover Wistar rat

The commonly used methods for identifying drug- and chemical-induced liver toxicity or liver disease are serum/plasma clinical chemistry and histopathology. However, the reliability of both techniques has been questioned (Travlos et al. 1996). Clinical chemistry measurements reflect the importance of serum/plasma enzymes as markers of organ toxicity. For example, high levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) generally reflect liver injury (Wroblewski 1959; Balazs et al. 1961). However, many serum enzymes are not organ specific and often it is necessary to measure a combination of enzymes. Furthermore, in humans, in the case of some liver diseases, for example, hepatic fibrosis, serum enzyme analysis may fail to detect the condition until the later stages of disease development (Friedman 1999). In addition, both serum clinical chemistry measurements and histopathological examination involve invasive methods of sample collection which may induce some distress in the patient. Consequently, there is a need for more reliable, specific and non-invasive biomarkers of liver injury and liver disease. An obvious non-invasive source of possible biomarkers is urine as samples can be collected with ease. Urine analysis is becoming a fast, effective means of testing for toxicity or disease, and a variety of techniques have been applied to urine analysis in the search for biomarkers, including gas chromatography, liquid chromatography, mass spectrometry and high performance liquid chromatography (HPLC) (Obata et al. 2000; Murali et al. 2004; Poor et al. 2004). However, the majority of urinary biomarkers identified to date have been relatively non-specific and, for example, include general markers of oxidative stress (Hermanns et al. 1998; Obata et al. 2000; Murali et al. 2004). Nevertheless, more recently, parvalbumin has been identified as a biomarker for skeletal muscle toxicity (Dare et al. 2002) and calreticulin has been recognized as a marker for urinary bladder cancer (Kageyama et al. 2004). We have recently identified copper/zinc superoxide dismutase (Cu/Zn SOD, also known as SOD-1) in the urine of rats treated with carbon tetrachloride (CCl4) following both single and repeat dosing (Smyth 2004; Smyth et al. 2007, 2008a,b). CCl4 induces fatty liver and hepatocellular hydropic degeneration, but the primary lesion is centrilobular necrosis (Luckey & Petersen 2001). Previously reported work (Smyth 2004; Smyth et al. 2008a,b;) demonstrated that hydropic degeneration and centrilobular necrosis were induced by CCl4, and urine samples from CCl4-treated animals were found to contain elevated levels of SOD-1 as identified by gel electrophoresis, in-gel digestion and mass spectrometry. Western blotting of tissue homogenates with an SOD-1 antibody in our investigations confirmed the liver as the source of the enzyme in the CCl4-treated Hanover Wistar rat (Smyth et al. 2008a). The current studies are concerned with an assessment of the usefulness of SOD-1 as a urinary biomarker following hepatocellular injury. It is considered that a useful urinary biomarker should reflect the relatively early changes induced by CCl4, that levels should be related to the degree of injury induced (i.e., the effect should be dose-related) and also, that the response of the biomarker should be easily measured (de Zwart et al. 1998). It is also important that the biomarker should not be increased in urine as a result of kidney injury. In our earlier studies, we demonstrated nephrotoxicity using histopathological evaluation in CCl4-treated rats at dose levels of 1.2 ml/kg and above, but below this dose level, the kidneys were histologically normal and no renal lesions were identified (Smyth et al. 2008a). For this reason, the highest dose level of CCl4 used in this investigation was 0.8 ml/kg. There are many reports demonstrating that CCl4 can cause liver injury at very low dose levels, and hepatotoxic effects have been reported in rats following the administration of doses as low as 0.10 ml/kg (Recknagel & Ghoshal 1966; Soni & Mehendale 1993; Rao et al. 1997). It has also been recorded that serum ALT and AST levels have been significantly increased in rats treated with a single dose of CCl4 as early as 6 h postdosing (Soni & Mehendale 1993). However, other investigators have reported only limited morphological liver injury, as determined by histopathological examination, but significant increases in serum ALT and AST (Murali et al. 2004). This last finding suggests that serum levels of both ALT and AST are sensitive indicators of toxicity caused by low doses of CCl4; however, a disadvantage is that both enzymes are not specific to injury of the liver, and also the in vivo collection of blood samples for the preparation of serum/plasma is an invasive procedure. Therefore, the objectives of the present investigations were to determine if urinary SOD is a useful and sensitive non-invasive marker for CCl4-induced liver injury when administered at low dose levels in the rat, and to compare the response of urinary SOD with the more conventional serum enzymes and other parameters routinely measured as markers of liver injury. We also investigated approximately when SOD first appeared in the urine following CCl4 administration.