• transplant
• liver
• gamma-glutamyltransferase

The aim of this study is test the diagnostic power of GGT fraction for hepatic diseases in comparison with that of total GGT and its usefulness in the setting of liver transplantation. Cirrhosis and chronic liver failure, and HCC are leading causes of morbidity and mortality worldwide. The diagnosis of cirrhosis and the determination of the etiology remain complex, in fact, no serologic test or radiological study can accurately diagnose cirrhosis. Besides, assays in most standard liver panels do not reflect the function of the liver correctly. With appropriately selected patients, liver transplantation is a definitive curative therapy for long-term survival and good quality of life for patients with end stage liver disease facing death. Accurate diagnosis and prognosis are essential for patients management pre and post-transplant, and for patients prioritization for organ allocation for liver transplantation. This requires the choice of biomarkers that provide adequate diagnostic information, at the minimum cost to more accurately select candidates for liver transplantation, to monitor post-transplant outcome and provide an optimal treatment regimen.
Serum gamma-glutamyltransferase (GGT) activity is a sensitive marker of liver dysfunction, but its specificity is modest, in fact, its value increases in all liver dysfunctions. GGT has been already included in diagnostic algorithm (i.e.: the Fatty Liver Index, FLI), its specificity was high if considered with other markers, but low if considered alone. The currently used laboratory GGT assays do not allow discriminating among the different causes of GGT increase, thus reducing the clinical value and specificity of this otherwise sensitive disease biomarker. A new method based on molecular-size exclusion chromatography, followed by a GGT-specific post-column reaction, allowed to identify and quantify, in healthy subjects, 4 plasma GGT fractions with high sensitivity, specificity and reproducibility. These fractions, named big-GGT (b-GGT), medium-GGT (m-GGT), small-GGT (s-GGT), and free-GGT (f-GGT) showed different molecular weight (MW), i.e. 2000, 1000, 250 and 70 kDa, respectively. It has been previously shown that in healthy subjects f-GGT is the most abundant fraction, while b-GGT showed the highest degree of correlation with established cardiovascular risk factors. Interestingly b-GGT has been found in atherosclerotic plaques together with products deriving from the pro-oxidant reactions catalysed by the enzyme, The liver is one of the main organs that generates free radicals, one of the mechanisms of hepatocyte injury in response to diverse insults occurring in different pathological conditions. For example, oxidative stress has been demonstrated to be implicated as a cause of hepatic fibrosis. Besides, liver damage is characterized by increased iron storage which elicits a free-radical mediated peroxidation.

In the period between February 2008 and April 2011, 264 patients during evaluation for liver transplant [215 men; median (25th – 75th percentile); age 54.5 (50-60 years)] were enrolled at the Department of Surgery, Liver Transplantation Unit of the University Hospital of Pisa. At the visit, attendees underwent anamnestic-physical examination and blood sampling for the laboratory assessment of liver function. In this cohort: 39 patients were diagnosed with metabolic cirrhosis (MC), 96 with viral cirrhosis (VC) 129 with viral cirrhosis and hepatocellular carcinoma (HCC). As control 200 blood donors were selected and studied for the determination of fractional GGT reference values. Blood samples were also collected from 14 LT recipients preoperatively before native liver hepatectomy (T0), and for 10 consecutive days post-transplant. Bile samples were collected intra-operatively during duct anastomosis (T0) and 10 days following the surgical procedure of transplantation through Kehr-tube. Standard assay of all blood tests were simultaneously performed according to the standard clinical laboratory procedures by automated analysers at the Clinical Laboratories of the University Hospital of Pisa.
Analysis of total and fractional GGT was performed using an FPLC (fast protein liquid chromatography) system. Separation of fractional GGT was obtained by gel filtration chromatography and the enzymatic activity was quantified by post-column injection of the fluorescent substrate for GGT. The area under chromatogram peak is proportional to fractional GGT activity. Total area and fractional GGT area was calculated by a MatLab program. Localization of GGT protein in liver biopsies was performed by automated indirect immunohistochemical analysis, using a polyclonal antibody directed against the C-terminal 20 amino acids of GGT heavy chain. Histological sections were analysed using the image software MetaAnalisys.

Different GGT fraction patterns were observed in cirrhotic patients and within the three sub cohorts (VC, MC, HC). s-GGT showed a broader and double profile not seen in controls, defined as s1-GGT and s2-GGT. The b/s ratio was lower in patients than controls. The diagnostic value of the b/s ratio was independent of the absolute values of total GGT and from the aetiology of the cirrhosis and the presence of liver cancer. Variations of the GGT fractions reflect different aspects of the liver cirrhosis: b-GGT behaves as a positive index of liver function, and reflects the progression of portal hypertension and splenomegaly; s2-GGT fraction reflects hepatocellular damage.
GGT activity in human bile is higher than that found in plasma, showing only two peaks corresponding to plasma b-GGT and f-GGT fractions, while m- and s-GGT fractions were not detectable. Regarding the nature and characteristics of biliary complex corresponding to the plasma b-GGT, the part of b-GGT fraction insensitive to the direct action of papain can be released into the bile associated with membrane vesicles such as exosomes. Immunolocalization of GGT in patients and control biopsy demonstrated different abundance and tissue distribution all over the section and quantification of GGT in liver tissue suggest that there is not a direct relationship between tissue and circulating GGT enzyme levels.
The post-operative course of the selected 14 patients was uneventful and there were no events of acute rejection. Soon after transplantation (24h), a sharp decline in total plasma GGT is observed and reflected on all fractions, in particular b-GGT. In 5-6 days after there has been a gradual increase in total plasma GGT. Plasma f-GGT fraction shows minor alterations, while other fractions have a similar trend as total GGT. In bile sample T0 GGT is present mainly as b-GGT and in less extent as f-GGT. The first 24 h post-transplant bile b-GGT activity is decreased followed by a sudden increase in its activity with a peak observed in the fourth day while bile f-GGT fraction shows minimal changes. An increase of bile GGT activity and an apparent peak of f-GGT preceded by an abrupt drop in b-GGT activity a day before is observed at days 6 and 10 in two patients: and a reversal of the proportions between the bile b-and f-GGT fractions in favour of b-GGT fraction has been observed in one of these patient at day 10 (T10) and in another patient on days 7 (T7) and 8 (T8). All fractions behave as positive index of cholestasis and liver function. Interestingly all fractions showed a positive correlation with direct bilirubin apart from s1-GGT, which showed a strict negative correlation. Unexpectedly, all fractions were negative associated with LDH, and b-GGT and m-GGT showed a negative correlation also with transaminases AST and ALT. Thus plasma GGT fractions, in particular b-GGT and m-GGT, were primarily related to ischemic-type biliary lesions following liver transplantation.

In conclusion the main findings of this study are:

1) patients with NAFLD and CHC display different GGT fraction patterns, despite similar total GGT activity values.
2) Collected data showed that the b/s ratio, independently of the absolute values of total GGT and its fractions, displays a high sensitivity and specificity for liver cirrhosis, and the values of the b/s ratio were lower than controls independently of the cause of the cirrhosis (viral or cryptogenetic) or the presence of associated liver cancer. This suggests that the b/s ratio is a specific biomarker of architectural and functional damage of the liver.
3) the elution profile of bile GGT activity showed the presence of only two forms corresponding to plasma fractions b-GGT and f-GGT, respectively. Similar to that found in plasma GGT fractions, the biliary f-GGT fraction consists of soluble protein and b-GGT fraction of exosomes. But, unlike plasma b-GGT, biliary b-GGT fraction is in part sensitive ti papain action; likely, the portion of biliary b-GGT sensitive to the proteolytic action might be consistuted of bile acids micelles.
4) Plasma b-GGT and m-GGT levels, in the first 10 days after liver transplant, were primarily related to ischemic-type biliary lesions following liver transplantation

The precise nature of GGT fractions has not yet been established, and at present it is not possible to speculate on the possible reasons conducting to different GGT fraction patterns in NAFLD and CHC and cirrhosis. Data collected suggest that GGT fraction pattern specificity might depend on its ability to reflect the different extents of inflammatory, structural and functional derangement in liver disease.
Further study on the nature and biological significance of plasma GGT fractions in health and disease might allow to improve the use of this sensitive but otherwise poorly specific biomarker in the numerous contexts in which it is employed, including multimarker algorithms comprising plasma GGT for the assessment of liver steatosis and fibrosis. Extensive investigation on the diagnostic value of GGT fractions might provide a novel diagnostic tool for liver diseases; understanding the nature, properties, and pathophysiological variations of GGT fraction pattern might allow a better understanding of the pathogenesis of the diseases associated with increased GGT.