Abstract (eng)
Unconjugated Bilirubin (UCB), the product of heme catabolism, is known as a potent antioxidant. A mild elevation of circulating plasma UCB, also known as Gilbert`s syndrome (GS) is associated with a reduced risk of non communicable diseases (NCDs) such as cardiovascular diseases and cancer.
This study aimed to find explanations for the reduced disease risk observed in GS by exploring whether serum bilirubin concentrations are linked to changes in DNA damage, to altered lipid metabolism and inflammation status in human and animal models.
Therefore, a variety of biomarkers were investigated in a cross sectional case control study comparing healthy subjects with Gilbert`s syndrome to age- and gender matched controls (n= 38 per group, total n= 76). Changes in DNA oxidation were evaluated using the cytokinesis-block micronucleus assay (CBMN), buccal micronucleus cytome assay (BMcyt) and the single cell gel electrophoresis (SCGE) assay. In addition, urinary 8-oxo-2`-deoxyguanosine (8oxodGuo) for estimation of DNA oxidation and urinary 8-oxo-guanosine (8oxoGuo) that reflects RNA oxidation were measured. Moreover, the lipid profile (lipoproteins, triglycerides (TG), cholesterols, including subfractions of LDL) and biomarkers of inflammation (interleukin 6 (IL-6), IL-1β, tumor necrosis factor α (TNF-α), serum amyloid A (SAA), C-reactive protein (CRP)) as well as heme metabolites (UCB, carboxy hemoglobin (CO-Hb) and iron) were analysed.
In addition, a hyperbilirubinemic rodent model (Gunn rats vs. Wistar rats, n= 20 per group, total n= 40) was selected to support the data observed in the human study. Within this trial, changes in DNA damage were determined in different tissues by using the SCGE assay, together with assessing the lipid profile.
The GS group and the Gunn rats had significantly higher UCB concentrations and a lower body mass index (BMI; human data) or lower body mass (female Gunn rats only) compared to their respective controls (p<0.05). With reference to DNA damage no significant difference in GS subjects versus controls was observed. Surprisingly, significantly higher DNA strand breaks (measured in white blood cells, PBMCs) were identified in Gunn rats when compared to controls.
Relative to their controls GS subjects had a significantly improved lipid profile (reduced total cholesterol, LDL-cholesterol, TG, LDL subfractions (LDL-1+LDL-2), lipoprotein (Apo-B), Apo-B/Apo-A1 ratio (p<0.05). Moreover, lower IL-6 and SAA concentrations and higher IL-1β concentrations were found when compared to controls (p<0.05; p=0.09; p<0.01). These data were confirmed by lipid analyses in the rodent model, showing that Gunn rat serum contained lower total cholesterol and TG concentrations as compared to that of normobilirubinemic rats (p<0.001). Furthermore, individuals with GS had significantly higher levels of CO-Hb and iron (p<0.01; p<0.001). When lifestyle effects (age and BMI) were considered statistically, lower formation of buccal anomalies were found in the GS subgroup of ≥30 years compared to the GS subgroup of <30 years of age (p<0.05). GS subjects of ≥30 years of age had reduced lipid variables (total cholesterol and LDL-C, TG and LDL-C subfractions, Apo-B/Apo-A1 ratio, Apo-B) when compared to older controls (p<0.05). The group with a BMI ≥25 kg/m2 had lower 8oxodG concentrations and tended to have higher DNA strand breaks. Simultaneously, these subjects had lower UCB levels than normal weight subjects (p<0.05; p=0.064; p<0.05).
These findings suggest DNA protection in older GS individuals` epithelia, however, a disease preventing effect which has been reported previously in a series of epidemiological studies, does not seem to be based on an immediate UCB-derived DNA protection. Consequently, the effects observed in PBMCs utilising the SCGE assay were not as pronounced. The link to a better health status (e.g. reduced risk of NCDs) observed in GS individuals, is probably rather predicted on the lower BMI, altered lipid profile and reduced inflammation status that were noticed in hyperbilirubinemic subjects, particularly in the older individuals. Therefore, it is likely that UCB, in addition to its commonly known antioxidant behavior, also exceeds its beneficial health effects via alternate mechanisms which are yet to be explored.
In this respect a newly hypothesised relationship between UCB, heme, iron and CO-Hb possibly indicates a positive feedback loop that includes UCB as a direct inducer of HO-1.
The results of the present study that are mainly based on observations, should be pursued in future, to follow-up on this and other hypotheses.