Original Paper

Protection from age-related increase in lipid biomarkers and inflammation contributes to cardiovascular protection in Gilbert's syndrome
Marlies Wallner, Rodrig Marculescu, Daniel Doberer, Michael Wolzt, Oswald Wagner, Libor Vitek, Andrew C. Bulmer, Karl-Heinz Wagner
Clinical Science May 10, 2013, 125 (5) 257-264; DOI: 10.1042/CS20120661

Abstract

Recent epidemiological and clinical data show protection from CVD (cardiovascular disease), all-cause mortality and cancer in subjects with GS (Gilbert's syndrome), which is characterized by a mildly elevated blood bilirubin concentration. The established antioxidant effect of bilirubin, however, contributes only in part to this protection. Therefore we investigated whether mildly elevated circulating UCB (unconjugated bilirubin) is associated with altered lipid metabolism. The study was performed on GS and age- and gender-matched healthy subjects (n=59 per group). Full lipoprotein profile, TAG (triacylglycerols), Apo (apolipoprotein)-A1, Apo-B, lipoprotein(a), the subfractions of LDL (low-density lipoprotein) and selected pro-inflammatory mediators were analysed. A hyperbilirubinaemic rodent model (Gunn rats, n=40) was investigated to further support the presented human data. GS subjects had significantly (P<0.05) improved lipid profile with reduced total cholesterol, LDL-C (LDL-cholesterol), TAG, low- and pro-atherogenic LDL subfractions (LDL-1+LDL-2), Apo-B, Apo-B/Apo-A1 ratio and lower IL-6 (interleukin 6) and SAA (serum amyloid A) concentrations (P=0.094). When the control and GS groups were subdivided into younger and older cohorts, older GS subjects demonstrated reduced lipid variables (total cholesterol and LDL-C, TAG and LDL-C subfractions, Apo-B/Apo-A1 ratio; P<0.05; Apo-B: P<0.1) compared with controls. These data were supported by lipid analyses in the rodent model showing that Gunn rat serum had lower total cholesterol (2.29±0.38 compared with 1.27±0.72 mM; P<0.001) and TAG (1.66±0.67 compared with 0.99±0.52 mM; P<0.001) concentrations compared with controls. These findings indicate that the altered lipid profile and the reduced pro-inflammatory status in hyperbilirubinaemic subjects, particularly in the older individuals, probably contribute additionally to the commonly accepted beneficial antioxidant effects of bilirubin in humans.

  • apolipoprotein
  • cardiovascular disease (CVD)
  • Gilbert’s syndrome
  • low-density lipoprotein (LDL)-subfractions
  • lipid metabolism
  • unconjugated bilirubin

CLINICAL PERSPECTIVES

  • • Epidemiological evidence indicates that circulating blood bilirubin is strongly associated with protection from chronic disease; however, a mechanistic explanation for this protection remains unknown.

  • • The results in the present study reveal a novel link between bilirubin and lipid metabolism, which is clearly perturbed in individuals with GS and the hyperbilirubinaemic Gunn rat. A clear trend towards prevention of the age-related increase in lipid biomarkers and inflammation in GS individuals has revealed an important clinical observation in hyperbilirubinaemic individuals.

  • • This finding significantly contributes to a topic that has so far been neglected in clinical research and practice and suggests a new direction of discussing the disease prevention using a novel approach of studying persons with a benign condition that affects millions of people worldwide.

INTRODUCTION

Strong epidemiological evidence indicates that circulating blood bilirubin protects from NCDs (non-communicable diseases) such as CVD (cardiovascular disease) [1] and cancer [2,3]. Furthermore, all-cause mortality in the general population [4] and in chronic haemodialysis patients [5] is negatively associated with plasma bilirubin levels. GS (Gilbert's syndrome) is a benign condition characterized by moderately elevated UCB (unconjugated bilirubin) levels (≥17.1 μM) because of additional TA repeats in the gene promoter for bilirubin UGT (uridine diphosphate glucuronosyltransferase 1 family, polypeptide A1). Bilirubin conjugation and excretion into the bile is decreased thereby increasing the circulating UCB concentration. This polymorphism affects approximately 3–13% of the general population [6,7].

Thus far, a full explanation for the protective effects of bilirubin in large epidemiological studies is lacking and is mainly based on bilirubin's in vitro antioxidant capacity [8]. Investigations in GS subjects, and experiments utilizing the rat model of hyperbilirubinaemia (Gunn rat), show protection from ex vivo [9,10] and in vivo [11] oxidation, coronary atherosclerosis [12], protection from thiol and protein oxidation [13] as well as anti-inflammatory [14] or anti-proliferative properties [15,16] of bilirubin.

However, evidence from an earlier meta-analysis indicates that circulating bilirubin alone could not account for all of the cardiovascular protection seen in hyperbilirubinaemic subjects [1]. These observations suggest other variables are interacting with serum bilirubin, resulting in cardiovascular protection. When exploring possible variables, lipid metabolism must be considered since it possesses great relevance to CVD protection. Very limited, but remarkable recent data support this link showing an impact of bilirubin on susceptibility of sd-LDL (small dense LDL) and circulating lipids to oxidation in individuals with GS [8,17]. Lower concentrations of the VLDL (very-low-density lipoprotein), IDL (intermediate-density lipoprotein) and LDL sub-fractions, in addition to reduced TAG (triacylglycerol) concentrations [18] and reduced total serum cholesterol [13] in GS subjects are reported. Particularly, the focused study of LDL subfractions is important because they can predict atherogenic risk of individuals better than LDL-C (LDL-cholesterol) [19]. Further evidence of the influence of bilirubin on lipid metabolism is shown in the Gunn rat model (equivalent to human Crigler-Najjar syndrome), where hepatic bilirubin glucuronidation is minimal resulting in high unconjugated hyperbilirubinaemia [10,20].

For other lipid-related biomarkers such as Apo (apolipoprotein)-A1, Apo-B, lipoprotein (a) and pro-inflammatory cytokines, reports in GS are lacking. The Apo-B/Apo-A1 ratio is an independent predictor of the vascular disease risk [21] and metabolic syndrome. IL-6 (interleukin 6), a pro-inflammatory cytokine, induces the hepatic synthesis of pro-inflammatory CRP (C-reactive protein), which is a known independent risk factor for CVD and Type 2 diabetes [22–23]. A novel pro-inflammatory lipolytic adipokine, expressed in human adipocytes is SAA (serum amyloid A). SAA is an acute-phase protein that is increased in obese subjects [24] and correlates with the presence of CVD [25].

Descriptive studies investigating GS, lack either a specific focus on lipid metabolism and inflammation and/or an appropriate number/matching of subjects. Therefore we investigated well-accepted biomarkers of lipid metabolism and adiposity and also novel lipid-associated CVD risk markers including 12 lipoprotein subfractions in a larger cohort of individuals with mildly increased UCB levels and compared them with age- and gender-matched controls. These human data are strongly supported by additional results from the hyperbilirubinaemic Gunn rat model.

MATERIALS AND METHODS

Subject recruitment

A total of 118 healthy subjects were recruited from the general Austrian population and met the inclusion criteria of normal liver function, absence of any underlying disease and an age range between 20 and 80 years. Reticulocytes, γ-GT (γ-glutamyl transferase), ALT (alanine transferase), AST (aspartate aminotransferase), LDH (lactate dehydrogenase), ALP (alkaline phosphatase), haemoglobin and haematocrit were measured at an initial screening examination. Subjects with liver, heart or kidney disease, haemolysis, diabetes, cholelithiasis, organ transplants, history of CVD, cancer, smoking (>1 cigarettes /day), alcohol consumption (>7 standard drinks/week), excessive physical activity (>10 h/week) and vitamin supplementation (4 weeks before the first blood sampling) or other prescribed medication (other than the contraceptive pill) were excluded. Allocation to the GS group was based on a fasting serum UCB concentration of ≥17.1 μM measured by HPLC (see below). Subjects were age- and gender- (38 men and 21 women) matched in each group (total 118 subjects). Written informed consent was obtained from each patient, and the study was approved by the ethical committee of the Medical University of Vienna and the General Hospital of Vienna (#274/2010) and was performed according to the Declaration of Helsinki.

Sample preparation

An overnight fasting blood sample was collected into lithium heparin and serum vacutainers. Samples were stored on ice in the dark until further analysis. Blood collection tubes were centrifuged (15 min at 1350 g at 4°C) and serum or plasma was divided into aliquots, used fresh or stored at −80°C for further analysis. Erythrocytes from lithium heparin Vacutainers were washed three times with isotonic phosphate buffer, divided into aliquots and stored at −80°C.

Animal experiments

Hyperbilirubinaemic Gunn rats [homozygous for a mutation in UGT1A1 (uridine diphosphate glucuronosyltransferase 1 family), jj] and respective controls (normobilirubinaemic Gunn rats, heterozygous for a mutation in UGT1A1, jJ) from the same genetic background were obtained from Charles University in Prague (Prague, Czech Republic) and acclimatized in the breeding facility of the Medical University of Vienna (Himberg, Austria) 1 week prior to experiments. The animals were housed in plastic cages (Macrolon type IV; Techniplast), under standard conditions (24±1°C, humidity 50±5°C, 12 h light/dark cycle) and fed with a standard diet (ssniff R/M-H Extrudat; ssniff Spezialdiäten) and ad libitum access to fresh water. All experiments were carried out with 7–8-week-old rodents (18 males and 22 females; a total number of 40 rodents). The study was approved by the ethical committee of the Austrian Federal Ministry of Science and Research (BMF-66.006/0008-II/3b/2011).

Sample preparation

Animals were killed by decapitation. Blood was collected into sodium-heparinized tubes (Ebewe) and immediately transferred into ficoll containing tubes (Greiner Bio-One) for PBMC (peripheral blood mononuclear cell) isolation. Plasma was obtained after centrifugation, then divided into aliquots and stored at −80°C until further analysis.

Blood biochemistry

Liver function enzymes (γ-GT, AST, ALT, LDH and ALP), Apo-A1, Apo-B and lipoprotein(a), serum TAG (triacylglycerols), total cholesterol and HDL-C [HDL (high-density lipoprotein)-cholesterol] were analysed using routine diagnostic tests on Olympus 5400 clinical chemistry analysers (Beckman Coulter) and measured on the day of blood sampling. The quantitative analysis of lipoprotein families and the lipoprotein subfractions, VLDL, IDL1-3, LDL1, LDL2, LDL3-7, HDL the low-atherogenic lipoprotein profile (IDL3 and LDL1) and the pro-atherogenic lipoprotein profile (VLDL, IDL1-2 and LDL2-7) in serum from a subsample of 36 age- and gender-matched samples, was performed by the Lipoprint LDL system (Quantimetrix) [26]. In the same subsample, IL-6 (high-sensitive ELISA; eBioscience) and SAA (N Latex SAA®; DADE Behring) were determined.

Anthropometric measurements

All anthropometric measurements were obtained from participants who were lightly dressed without wearing shoes. Body height (m) was measured with a stadiometer (Seca, Modell 214) and body mass (kg) using standard analogue scales (Seca Selecta 791). BMI (body mass index) was calculated as kg/m2.

Statistical analysis

All statistical analyses were conducted using SPSS (IBM statistics, Version 17.0). Normal distribution within the data set was tested by the Kolmogorov–Smirnoff test (KS). Therefore some variables were log10- (e.g. UCB) transformed and an inverse transformation (1/x) for age was conducted. To determine differences between two groups, an independent sample Student's t test (parametric data) or Mann–Whitney U test (non-parametric data) was conducted. Depending on the homogeneity of variance and normality of data within groups, one-way ANOVA (parametric data) or Kruskal–Wallis H test (non-parametric data) was used for multiple group comparisons. Student–Newman–Keuls post-hoc tests were performed after one-way ANOVA to determine group differences. Pearson coefficient (parametric data) or Spearman rho correlation (non-parametric data) tested the bivariate relationships between the independent and dependent variables. Forward stepwise regression analysis was used to predict the effect of multiple-independent variables on lipid variables. Data are expressed as means±S.D., and a P value <0.05 was considered significant.

RESULTS

General subject characteristics

UCB concentrations were significantly higher in GS subjects (P<0.001) as compared with controls. No differences in liver enzyme activities existed between the groups. After age and gender matching, GS subjects showed a trend towards lower BMI than the respective controls (P=0.075).

Human and animal lipid data

Whole group comparisons indicated that GS subjects exhibited reduced total cholesterol, TAG and LDL-C concentrations compared with controls (P<0.05; Table 1). LDL subfractions were significantly (LDL-1 and LDL-2; P<0.05) or tended (LDL-3) to be lower in GS (Table 1). The total scores of pro-atherogenic markers (sum of VLDL, IDL-1, IDL-2 and LDL2–7) were not significantly lower in GS; however, the low-atherogenic parameters (IDL-3+LDL-1) were significantly reduced in GS compared with controls (P<0.05) (Table 1). Significantly reduced Apo-B and the Apo-B/Apo-A1 ratio (P<0.05) (Table 1) in GS was also demonstrated.

View this table:
Table 1 Demographic features, biochemical parameters, biomarker for fat metabolism and inflammation of individuals with GS and controls

Values are means (SD). *Significantly different from the GS group; †difference between GS and control group for trend (P<0.10). Low-atherogenic score, IDL3 and LDL1; pro-atherogenic score, VLDL, IDL1–2 and LDL2–7.

Lipid status in Gunn rats and littermate controls strongly supported the different lipid biochemistry data in GS subjects. Hyperbilirubinaemic rodents had significantly reduced total cholesterol, TAG and HDL-C concentrations (P<0.05). Hyperbilirubinaemia and hypocholesterolaemia in Gunn rats were associated with significantly reduced body mass in female rats only (P<0.001; Table 2).

View this table:
Table 2 Body mass, UCB and lipid parameters in combined, male and female littermate-matched Gunn (n=20) and Wistar (n=20) rats

Values are means (S.D.).

When analysing the entire cohort significant negative correlations between UCB and BMI (r=−0.211, P<0.05), total cholesterol (r=−0.245, P<0.01), LDL-C (r=−0.243, P<0.01), TAG (r=−0.248, P<0.01), LDL-1 (r=−0.247, P<0.05), LDL-2 (r=−0.272, P<0.05) and the low-atherogenic subfractions (r=−0.276, P<0.05) were found. These data were corroborated by analyses of Gunn rat sera. In female Gunn rats, strong negative correlations between serum UCB and total cholesterol (r=−0.882, P<0.001), TAG (r=−0.502, P<0.05) and HDL-C (r=−0.919, P<0.001) existed. In male Gunn rats, UCB was negatively correlated with TAG (r=−0.649, P<0.05).

Forward stepwise regression analysis revealed that bilirubin explained 21% of the variance in LDL-C, 15% of total cholesterol, 4.5% of LDL-1 and 4% of LDL-C in GS subjects. However, bilirubin explained 44.6% of the variance in total cholesterol, 42.4% of HDL-C and 28.3% of TAG in rodents.

Since lipid parameters, as risk factors for CVD, rise with age, more specific analyses were conducted by dividing the human cohort into two age subgroups (age group 1: <30 years; age group 2: ≥30 years with a range from 30 to 72 years) as shown in Table 3.

View this table:
Table 3 Effect of age on biomarkers of lipid metabolism and inflammation in GS and control subjects

Values are means (S.D.). aP<0.05 compared with the control of the age group (young or old); bP for trend < 0.1 to control of the age group (young or old); cP<0.05 to younger GS and control. *Pro-atherogenic score, VLDL, IDL1-2 and LDL2–7; †low-atherogenic score: IDL3 and LDL1. n.s., not significant.

For almost all lipid parameters the older controls showed either higher values than older GS subjects (i.e. significantly greater for total cholesterol, TAG, LDL-C, LDL-1, the low-atherogenic subfraction and the Apo-B/Apo-A1 ratio, P<0.05; Table 3) or tended to be greater for VLDL-C and Apo-B. The older control group had significantly higher total cholesterol, TAG, LDL-C, LDL-subfractions (except LDL-3), Apo-B and the Apo-B/Apo-A1 ratio than the younger control group (P<0.05). Interestingly, this age-related increase in these parameters was not observed between the younger and the older GS group. In order to determine whether GS individuals were protected from age-related dyslipidaemia, the age-associated increase in total cholesterol, LDL-C, TAG and HDL-C concentrations in addition to BMI were plotted against age in both groups (see Supplementary Figure S1 at http://www.clinsci.org/cs/125/cs1250257add.htm). These data indicate that significant age-dependent increases in these variables existed in both groups. However, the relationship between age and lipid variables remained offset, below the relationship in controls. Furthermore, despite remaining significant in both groups, the age-associated increase in BMI remained clearly lower in GS subjects (r2=0.29, P<0.05), compared with controls (r2=0.44, P<0.01).

Pro-inflammatory parameters

GS subjects also experienced significantly (P<0.05) reduced concentrations of the pro-inflammatory cytokine IL-6 compared with controls. A similar trend was shown for SAA (P=0.094; Table 1).

When considering age (threshold 30 years; Table 3) the older control group had the highest SAA, CRP and IL-6 concentrations when compared with the other age-subgroups (older/younger GS subjects and the younger control group; P<0.05).

DISCUSSION

The results of the present study provide three lines of evidence to suggest protection from age-related dyslipidaemia/inflammation in GS. First, BMI, total cholesterol, TAG, LDL-C, LDL sub-fractions, Apo-B, the Apo-B/Apo-A1 ratio, SAA and IL-6 were significantly lower in aged GS compared with matched controls (Table 3). Secondly, the significant age-associated increase in total cholesterol, TAG, LDL-C, LDL sub-fractions, pro- and low-atherogenic lipid scores, Apo-B, SAA, IL-6 and CRP occurred in controls but not in GS (Table 3). Finally, the highly significant differences in lipid profile are presented in both male and female hyperbilirubinaemic Gunn rats (Table 2), providing a clear mechanistic link between bilirubin/UGT1A1 function and perturbation of lipid status. Together, these data provide the translational evidence that bilirubin or its metabolism modulates lipid metabolism in humans, which could contribute to reduced CVD risk in GS.

BMI

The GS and control groups in this study were matched for age and gender. Matched GS subjects tended to possess lower BMI, which can mainly be explained by the significant differences between the older subgroups (≥30 years; Table 3 and Supplementary Figure S1). In general, the BMI was in the range of the Austrian population [27]. The older GS group did not differ significantly from the younger GS group in BMI; however, the older control group had a significantly higher BMI than the older GS subgroup and could be defined as being overweight (>25 kg/m2), whereas the older GS subgroup was, by definition, normal (<25 kg/m2). This observation is supported by other large epidemiological observations showing lower BMI in hyperbilirubinaemic subjects ([4,28]; see [30] for review). Interestingly, in Gunn rats, a difference in body mass was only found in female rats, which is supported by recent studies in a smaller group of aged female Gunn rats [13] and cross-sectional clinical studies [29] as recently reviewed [30]. A recent short-term weight loss study was effective in significantly increasing bilirubin levels back to the normal reference range. In this study, a linear relationship was observed between increase in total bilirubin concentrations and amount of weight loss [31].

These observations indicate that bilirubin is linked to weight or weight loss and that GS subjects have a lower BMI which, particularly at a greater age, could significantly contribute to prevention of CVD. A low BMI is an independent protector against NCDs, since overweight and obesity is correlated to higher mortality from NCDs [32].

Lipid metabolism

Focused reports on the relationship between elevated bilirubin, the metabolic state of GS and lipid data have not been published in the literature thus far, although scattered findings have been presented, particularly recently. For example, reduced circulating total cholesterol [17], LDL-C [18], TAG [33] and elevated HDL/LDL ratio [9] have been documented in GS, although findings are not always consistent between published studies [34], because of differences in the subject matching and cohort characteristics [30].

The results of the present study demonstrate that many pro-atherogenic risk markers of lipid metabolism are significantly reduced in GS subjects (Table 1). These observations were more pronounced in Gunn rats, which also showed significantly lower HDL-C (Table 2). An interesting observation was the specific effect in female Gunn rats, which were also lighter than littermate controls. This effect is comparable with a recently published study of homozygote animals reporting total cholesterol values one-third of heterozygote and wild-type animals [10]. These data could reflect a dose response of circulating bilirubin on lipid metabolism that is species independent. Mechanistically, altered lipid metabolism could reflect a role for either UGT1A1 or unconjugated bilirubin in modifying lipid metabolism and cholesterol excretion [30]. This observation probably explains the reduction in body mass in littermate-matched homozygote animals for which, until now no explanation has been found. The pronounced effect of the Gunn phenotype in female animals could suggest an additional role for oestrogen in enhancing the bilirubin or UGT1A1 effect. This could be mediated by competition of oestrogen and bilirubin for glucuronidation [35] by UGT1A1, which would also explain the slightly elevated bilirubin levels compared with male Gunn rats (Table 2). In humans there were no significantly stronger effects in women compared with men (results not shown).

The greater reduction in HDL (compared with LDL) in the Gunn rats is reflected by a greater contribution of HDL to total cholesterol in rats and suggests underlying perturbation of cholesterol metabolism/excretion/absorption within dysfunctional UGT1A1 syndromes (human and animals) as reported [11] and recently comprehensively reviewed [30]. One limitation is the higher UCB concentration in Gunn rats compared with GS; however, the general effect on lipid metabolism (hypocholesterolaemia) was similar to humans and other adult hyperbilirubinaemic animal models are not available. Evidence for the impact of bilirubin metabolism on lipoprotein assembly is described by an effect on LDL subfraction concentrations. Small dense LDL particles appear to confer a higher level of CVD risk than the larger less dense LDL particles and to be more susceptible to oxidative modification [36]. We used a new, recently published electrophoretic method, which enables the analysis of up to 12 lipoprotein subfractions [18]. Atherogenic lipoprotein profile is characterized by the presence of VLDL, IDL and the presence of small dense LDL lipoproteins. The sum of pro-atherogenic subfractions that consist of VLDL, IDL1–2 and LDL2–7 was not significantly lower in GS (Table 1). Closer evaluation of the data reveals a significant increase in pro-atherogenic subfraction values in older controls (18.4% higher than that of the young control group), which did not increase significantly in older GS subjects (compared with younger GS subjects; Table 3). Interestingly, this finding emphasizes the protection from age-related dyslipidaemia in GS. The concentrations of the low-atherogenic fractions in the young groups compared with the older GS group were similar; but, however, were significantly reduced compared with the older control group (P<0.05; Table 3).

Further support of the protective effects of GS on age-related dyslipidaemia was revealed on correlational analysis. In control and GS subjects total cholesterol, LDL-C and TAG increased in an age-dependent manner; however, the relationship in GS subjects remained consistently offset below that of controls (Figure S1). Indeed, when observing the absolute total cholesterol and LDL-C concentrations in aged GS and controls, aged controls were hypercholesterolaemic (≥5.92 mM total cholesterol; ≥3.68 LDL-C), as defined by the American Heart Association [37], however, age-matched GS subjects were not.

Generally, these results are in agreement with a recently published study, showing decreased small dense lipoprotein fractions in GS [18]. However, this study investigated only men and did not age match the subjects. Regardless, these data and those of Ocadlik et al. [18] support the impact of bilirubin metabolism on TAG and VLDL synthesis, potentially by down-regulating their synthesis/assembly.

Further biomarkers to describe the CVD risk include the structural proteins of lipoproteins. Apo-1 and -2 are the major structural proteins of HDL particles, whereas Apo-B is a major protein of every other lipoprotein particle but HDL. The Apo-B/Apo-A1 ratio is considered as one of the strongest plasma lipid-associated predictors of CVD risk [38], which indicates the balance between potentially atherogenic and anti-atherogenic particles [39]. GS subjects showed similar Apo-A1 and Apo-B concentrations to control subjects (Table 1). Similar to the above-mentioned observations, the older GS subjects were not significantly different to the younger cohorts in their Apo-B values, whereas the older control group showed significantly higher Apo-B than the younger subjects (P<0.05). These data indicate protection from the age-related increase in plasma Apo-B concentration in GS. The Apo-B/Apo-A1 ratio was also lower in GS subjects (P<0.05 compared with control group), and was shown to be highest in the older control group with a mean of 0.69±0.15 compared with 0.61±0.21 in the older GS group. This shows that the older control cohort was almost classified as being at medium risk of myocardial infarction, the criterion for which is determined by a ratio of 0.7 or higher [40], but the older GS subgroup was clearly at a lower risk level.

Pro-inflammatory markers

SAA is a pro-inflammatory adipokine in humans linked to obesity and is a predictor of CVD [24,41]. Adipose tissue is inflamed in obese individuals with increased secretion of pro-inflammatory and decreased expression of anti-inflammatory adipokines [24,25]. We report a trend towards reduced SAA levels in GS subjects (Table 1), which was significantly lower in the older GS compared with control subjects (Table 3). The same effect was evident for IL-6, with the older GS group having lower plasma levels than the older control group (Table 3). CRP showed a similar trend, but due to considerable variation, the results failed to show significance.

These data show that in older subjects that are traditionally assumed to possess a higher risk of CVD, GS individuals have lower levels of pro-inflammatory cytokines and do not experience the age-related induction of dyslipidaemia and inflammation. Elevated bilirubin has previously been associated with low CRP levels [42], which is a widely used diagnostic criterion to assess the inflammation status. However, this is the first study to report lower IL-6 concentrations in GS, which induces hepatic liberation of CRP into the circulation. Lower levels of CRP and also IL-6 are related to a lower risk of CVD [22] and diabetes [23] and further emphasize the importance of these biomarkers in contributing to cardiovascular protection in individuals with mildly elevated unconjugated bilirubin levels.

Conclusions

The results of the present study show that elevated circulating bilirubin, particularly the benign condition of GS, is associated with reduced concentrations of lipid and inflammation biomarkers and a trend to decreased BMI. Furthermore, it is shown that older subjects, who are generally at greater disease risk, are likely to benefit more from a mild congenital hyperbilirubinaemia. These data are strongly supported by serum lipid analysis in Gunn rats, with more pronounced effects in the female animals. The observations of the present study could contribute significantly to revealing a new mechanistic explanation for the previously reported protective effects of hyperbilirubinaemia against NCDs in many published epidemiological studies. Such efforts could assist in developing novel strategies to mildly increase circulating bilirubin concentrations for protection against NCDs.

AUTHOR CONTRIBUTION

Marlies Wallner performed the human study, assisted in the animal study and performed statistical analysis; Rodrig Marculescu and Daniel Doberer wrote and submitted the human Ethics application, oversaw patient recruitment, assisted in the planning of the study and performed some of the lipid analysis; Michael Wolzt provided clinical staff and material for the clinical trial and assisted in subject recruitment together with Marlies Wallner and Karl-Heinz Wagner; Oswald Wagner provided clinical staff and material for the clinical trial; Libor Vitek provided the Gunn rats and proofread the paper before submission; Andrew Bulmer planned the studies together with Karl-Heinz Wagner, obtained the funding with Karl-Heinz Wagner, gave advice on the animal work, assisted with the statistical analyses and contributed to the paper; Karl-Heinz Wagner obtained the funding, planned the studies together with Andrew Cameron Bulmer, supported Marlies Wallner and wrote the paper.

FUNDING

This work was supported by the Austrian Science Fund [grant number P21162-B11] and by the Vienna Science and Technology Fund [grant number LS07-031].

Acknowledgments

We would like to gratefully acknowledge the clinical staff of the Department of Clinical Pharmacology for subject care and performance of blood analysis.

Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; Apo, apolipoprotein; AST, aspartate aminotransferase; BMI, body mass index; CRP, C-reactive protein; CVD, cardiovascular disease; γ-GT, γ-glutamyl transferase; GS, Gilbert’s syndrome; HDL, high-density lipoprotein; HDL-C, HDL-cholesterol; IDL, intermediate-density lipoprotein; IL, interleukin; LDH, lactate dehydrogenase; LDL, low-density lipoprotein, LDL-C, LDL-cholesterol; NCD, non-communicable disease; SAA, serum amyloid A; TAG, triacylglycerol(s); UCB, unconjugated bilirubin; UGT, uridine diphosphate glucuronosyltransferase; UGT1A1, uridine diphosphate glucuronosyltransferase 1 family, polypeptide A1; VLDL, very-low-density lipoprotein

References

  1. ↵
    1. Novotny L.,
    2. Vitek L.
    (2003) Inverse relationship between serum bilirubin and atherosclerosis in men: a meta-analysis of published studies. Exp. Biol. Med. 228:568–571.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    1. Temme E. H.,
    2. Zhang J.,
    3. Schouten E. G.,
    4. Kesteloot H.
    (2001) Serum bilirubin and 10-year mortality risk in a Belgian population. Cancer Causes Control 12:887–894.
    OpenUrlCrossRefPubMedWeb of Science
  3. ↵
    1. Zucker S. D.,
    2. Horn P. S.,
    3. Sherman K. E.
    (2004) Serum bilirubin levels in the U.S. population: gender effect and inverse correlation with colorectal cancer. Hepatology 40:827–835.
    OpenUrlCrossRefPubMedWeb of Science
  4. ↵
    1. Horsfall L. J.,
    2. Rait G.,
    3. Walters K.,
    4. Swallow D. M.,
    5. Pereira S. P.,
    6. Nazareth I.,
    7. Petersen I.
    (2011) Serum bilirubin and risk of respiratory disease and death. JAMA, J. Am. Med. Assoc. 305:691–697.
    OpenUrlCrossRefPubMedWeb of Science
  5. ↵
    1. Chen Y. H.,
    2. Hung S. C.,
    3. Tarng D. C.
    (2011) Serum bilirubin links UGT1A1*28 polymorphism and predicts long-term cardiovascular events and mortality in chronic hemodialysis patients. Clin. J. Am. Soc. Nephrol. 6:567–574.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Bosma P. J.,
    2. Chowdhury J. R.,
    3. Bakker C.,
    4. Gantla S.,
    5. de Boer A.,
    6. Oostra B. A.,
    7. Lindhout D.,
    8. Tytgat G. N.,
    9. Jansen P. L.,
    10. Oude Elferink R. P.,
    (1995) The genetic basis of the reduced expression of bilirubin UDP-glucuronosyltransferase 1 in Gilbert's syndrome. N. Engl. J. Med. 333:1171–1175.
    OpenUrlCrossRefPubMedWeb of Science
  7. ↵
    1. Fertrin K. Y.,
    2. Goncalves M. S.,
    3. Saad S. T.,
    4. Costa F. F.
    (2002) Frequencies of UDP-glucuronosyltransferase 1 (UGT1A1) gene promoter polymorphisms among distinct ethnic groups from Brazil. Am. J. Med. Genet. 108:117–119.
    OpenUrlCrossRefPubMedWeb of Science
  8. ↵
    1. Stocker R.
    (2004) Antioxidant activities of bile pigments. Antioxid. Redox Signaling 6:841–849.
    OpenUrlCrossRefPubMedWeb of Science
  9. ↵
    1. Bulmer A. C.,
    2. Blanchfield J. T.,
    3. Toth I.,
    4. Fassett R. G.,
    5. Coombes J. S.
    (2008) Improved resistance to serum oxidation in Gilbert's syndrome: a mechanism for cardiovascular protection. Atherosclerosis 199:390–396.
    OpenUrlCrossRefPubMedWeb of Science
  10. ↵
    1. Muchova L.,
    2. Vanova K.,
    3. Zelenka J.,
    4. Lenicek M.,
    5. Petr T.,
    6. Vejrazka M.,
    7. Sticova E.,
    8. Vreman H. J.,
    9. Wong R. J.,
    10. Vitek L.
    (2011) Bile acids decrease intracellular bilirubin levels in the cholestatic liver: implications for bile acid-mediated oxidative stress. J. Cell. Mol. Med. 15:1156–1165.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Dennery P. A.,
    2. McDonagh A. F.,
    3. Spitz D. R.,
    4. Rodgers P. A.,
    5. Stevenson D. K.
    (1995) Hyperbilirubinemia results in reduced oxidative injury in neonatal Gunn rats exposed to hyperoxia. Free Radical Biol. Med. 19:395–404.
    OpenUrlCrossRefPubMedWeb of Science
  12. ↵
    1. Schwertner H. A.,
    2. Jackson W. G.,
    3. Tolan G.
    (1994) Association of low serum concentration of bilirubin with increased risk of coronary artery disease. Clin. Chem. 40:18–23.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Boon A. C.,
    2. Hawkins C. L.,
    3. Bisht K.,
    4. Coombes J. S.,
    5. Bakrania B.,
    6. Wagner K. H.,
    7. Bulmer A. C.
    (2012) Reduced circulating oxidized LDL is associated with hypocholesterolemia and enhanced thiol status in Gilbert syndrome. Free Radical Biol. Med. 52:2120–2127.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Yoshino S.,
    2. Hamasaki S.,
    3. Ishida S.,
    4. Kataoka T.,
    5. Yoshikawa A.,
    6. Oketani N.,
    7. Saihara K.,
    8. Okui H.,
    9. Shinsato T.,
    10. Ichiki H.,
    (2011) Relationship between bilirubin concentration, coronary endothelial function, and inflammatory stress in overweight patients. J. Atheroscler. Thromb. 18:403–412.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Ollinger R.,
    2. Bilban M.,
    3. Erat A.,
    4. Froio A.,
    5. McDaid J.,
    6. Tyagi S.,
    7. Csizmadia E.,
    8. Graça-Souza A. V.,
    9. Liloia A.,
    10. Soares M. P.,
    (2005) Bilirubin: a natural inhibitor of vascular smooth muscle cell proliferation. Circulation 112:1030–1039.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Ryter S. W.,
    2. Choi A. M.
    (2009) Heme oxygenase-1/carbon monoxide: from metabolism to molecular therapy. Am. J. Respir. Cell Mol. Biol. 41:251–260.
    OpenUrlCrossRefPubMedWeb of Science
  17. ↵
    1. Tapan S.,
    2. Karadurmus N.,
    3. Dogru T.,
    4. Ercin C. N.,
    5. Tasci I.,
    6. Bilgi C.,
    7. Kurt I.,
    8. Erbil M. K.
    (2011) Decreased small dense LDL levels in Gilbert's syndrome. Clin. Biochem. 44:300–303.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Ocadlik I.,
    2. Hlinstakova S.,
    3. Oravec S.
    (2011) Relationship between unconjugated hyperbilirubinemia and lipoprotein spectrum. Neuro. Endocrinol. Lett. 32:360–364.
    OpenUrlPubMed
  19. ↵
    1. Otvos J. D.,
    2. Mora S.,
    3. Shalaurova I.,
    4. Greenland P.,
    5. Mackey R. H. Jr.,
    6. Goff D. C.
    (2011) Clinical implications of discordance between low-density lipoprotein cholesterol and particle number. J. Clin. Lipidol. 5:105–113.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Chowdhury J. R.,
    2. Kondapalli R.,
    3. Chowdhury N. R.
    (1993) Gunn rat: a model for inherited deficiency of bilirubin glucuronidation. Adv. Vet. Sci. Comp. Med. 37:149–173.
    OpenUrlPubMed
  21. ↵
    1. Emerging Risk Factors Collaboration,
    2. Di Angelantonio E.,
    3. Sarwar N.,
    4. Perry P.,
    5. Kaptoge S.,
    6. Ray K. K.,
    7. Thompson A.,
    8. Wood A. M.,
    9. Lewington S.,
    10. Sattar N.,
    11. Packard C. J.,
    (2009) Major lipids, apolipoproteins, and risk of vascular disease. JAMA, J. Am. Med. Assoc. 302:1993–2000.
    OpenUrlCrossRefPubMedWeb of Science
  22. ↵
    1. Van Gaal L. F.,
    2. Mertens I. L.,
    3. De Block C. E.
    (2006) Mechanisms linking obesity with cardiovascular disease. Nature 444:875–880.
    OpenUrlCrossRefPubMedWeb of Science
  23. ↵
    1. Pradhan A. D.,
    2. Manson J. E.,
    3. Rifai N.,
    4. Buring J. E.,
    5. Ridker P. M.
    (2001) C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA, J. Am. Med. Assoc. 286:327–334.
    OpenUrlCrossRefPubMedWeb of Science
  24. ↵
    1. Yang R. Z.,
    2. Lee M. J.,
    3. Hu H.,
    4. Pollin T. I.,
    5. Ryan A. S.,
    6. Nicklas B. J.,
    7. Snitker S.,
    8. Horenstein R. B.,
    9. Hull K.,
    10. Goldberg N. H.,
    (2006) Acute-phase serum amyloid A: an inflammatory adipokine and potential link between obesity and its metabolic complications. PLoS Med. 3:e287.
    OpenUrlCrossRefPubMed
  25. ↵
    1. King V. L.,
    2. Thompson J.,
    3. Tannock L. R.
    (2011) Serum amyloid A in atherosclerosis. Curr. Opin. Lipidol. 22:302–307.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Hoefner D. M.,
    2. Hodel S. D.,
    3. O’Brien J. F.,
    4. Branum E. L.,
    5. Sun D.,
    6. Meissner I.,
    7. McConnell J. P.
    (2001) Development of a rapid, quantitative method for LDL subfractionation with use of the Quantimetrix Lipoprint LDL System. Clin. Chem. 47:266–274.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Elmadfa I.,
    2. Freisling H.,
    3. Nowak V.,
    4. Hofstädter D.
    (2009) Austrian Nutrition Report 2008 (Federal Ministry of Health, Vienna).
  28. ↵
    1. Choi S. H.,
    2. Yun K. E.,
    3. Choi H. J.
    (2013) Relationships between serum total bilirubin levels and metabolic syndrome in Korean adults. Nutr. Metab. Cardiovasc. Dis. 23:31–37.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Kwon K. M.,
    2. Kam J. H.,
    3. Kim M. Y.,
    4. Chung C. H.,
    5. Kim J. K.,
    6. Linton J. A.,
    7. Eom A.,
    8. Koh S. B.,
    9. Kang H. T.
    (2011) Inverse association between total bilirubin and metabolic syndrome in rural Korean women. J. Womens Health 20:963–969.
    OpenUrlCrossRef
  30. ↵
    1. Bulmer A. C.,
    2. Verkade H. J.,
    3. Wagner K. H.
    (2012) Bilirubin and beyond: a review of lipid status in Gilbert's syndrome and its relevance to cardiovascular disease protection. Prog. Lipid Res. 52:193–205.
    OpenUrlPubMed
  31. ↵
    1. Andersson C.,
    2. Weeke P.,
    3. Fosbol E. L.,
    4. Brendorp B.,
    5. Kober L.,
    6. Coutinho W.,
    7. Sharma A. M.,
    8. Van Gaal L.,
    9. Finer N.,
    10. James W. P.,
    11. ,
    12. SCOUT Executive Steering Committee; SCOUT investigators
    (2009) Acute effect of weight loss on levels of total bilirubin in obese, cardiovascular high-risk patients: an analysis from the lead-in period of the Sibutramine Cardiovascular Outcome trial. Metab., Clin. Exp. 58:1109–1115.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Leitzmann M. F.,
    2. Moore S. C.,
    3. Koster A.,
    4. Harris T. B.,
    5. Park Y.,
    6. Hollenbeck A.,
    7. Schatzkin A.
    (2011) Waist circumference as compared with body-mass index in predicting mortality from specific causes. PLoS ONE 6:e18582.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Vitek L.,
    2. Kraslova I.,
    3. Muchova L.,
    4. Novotny L.,
    5. Yamaguchi T.
    (2007) Urinary excretion of oxidative metabolites of bilirubin in subjects with Gilbert syndrome. J. Gastroenterol. Hepatol. 22:841–845.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Lippi G.,
    2. Targher G.,
    3. Montagnana M.,
    4. Salvagno G. L.,
    5. Guidi G. C.
    (2009) Bilirubin concentration and cardiovascular risk profile. Liver Int. 29:315–316.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Zhou J.,
    2. Tracy T. S.,
    3. Remmel R. P.
    (2011) Correlation between bilirubin glucuronidation and estradiol-3-gluronidation in the presence of model UDP-glucuronosyltransferase 1A1 substrates/inhibitors. Drug Metab. Dispos. 39:322–329.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Berneis K. K.,
    2. Krauss R. M.
    (2002) Metabolic origins and clinical significance of LDL heterogeneity. J. Lipid Res. 43:1363–1379.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Smith S. C. Jr.,
    2. Benjamin E. J.,
    3. Bonow R. O.,
    4. Braun L. T.,
    5. Creager M. A.,
    6. Franklin B. A.,
    7. Gibbons R. J.,
    8. Grundy S. M.,
    9. Hiratzka L. F.,
    10. Jones D. W.,
    (2011) AHA/ACCF secondary prevention and risk reduction therapy for patients with coronary and other atherosclerotic vascular disease: 2011 update: a guideline from the American Heart Association and American College of Cardiology Foundation endorsed by the World Heart Federation and the Preventive Cardiovascular Nurses Association. Circulation 58:2432–2446.
    OpenUrl
  38. ↵
    1. Holme I.,
    2. Aastveit A. H.,
    3. Hammar N.,
    4. Jungner I.,
    5. Walldius G.
    (2009) Relationships between lipoprotein components and risk of ischaemic and haemorrhagic stroke in the Apolipoprotein MOrtality RISk study (AMORIS). J. Intern. Med. 265:275–287.
    OpenUrlCrossRefPubMedWeb of Science
  39. ↵
    1. Walldius G.,
    2. Jungner I.
    (2006) The apoB/apoA-I ratio: a strong, new risk factor for cardiovascular disease and a target for lipid-lowering therapy – a review of the evidence. J. Intern. Med. 259:493–519.
    OpenUrlCrossRefPubMedWeb of Science
  40. ↵
    1. Walldius G.,
    2. Jungner I.
    (2007) Is there a better marker of cardiovascular risk than LDL cholesterol? Apolipoproteins B and A – new risk factors and targets for therapy. Nutr. Metab. Cardiovasc. Dis. 17:565–571.
    OpenUrlCrossRefPubMedWeb of Science
  41. ↵
    1. Johnson B. D.,
    2. Kip K. E.,
    3. Marroquin O. C.,
    4. Ridker P. M.,
    5. Kelsey S. F.,
    6. Shaw L. J.,
    7. Pepine C. J.,
    8. Sharaf B.,
    9. Bairey Merz C. N.,
    10. Sopko G.,
    11. ,
    12. National Heart, Lung, and Blood Institute
    (2004) Serum amyloid A as a predictor of coronary artery disease and cardiovascular outcome in women: the National Heart, Lung, and Blood Institute-Sponsored Women's Ischemia Syndrome Evaluation (WISE). Circulation 109:726–732.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Hwang H. J.,
    2. Lee S. W.,
    3. Kim S. H.
    (2011) Relationship between bilirubin and C-reactive protein. Clin. Chem. Lab. Med. 49:1823–1828.
    OpenUrlPubMed
View Abstract
Back to top

September 2013

Volume: 125 Issue: 5

Clinical Science: 125 (5)

Actions

Email
Protection from age-related increase in lipid biomarkers and inflammation contributes to cardiovascular protection in Gilbert's syndrome
Marlies Wallner, Rodrig Marculescu, Daniel Doberer, Michael Wolzt, Oswald Wagner, Libor Vitek, Andrew C. Bulmer, Karl-Heinz Wagner
Clinical Science Sep 2013, 125 (5) 257-264; DOI: 10.1042/CS20120661
Permalink:
del.icio.us logo Digg logo Reddit logo Technorati logo Twitter logo CiteULike logo Facebook logo Mendeley logo
Citation Tools
Print
Alerts
Request Permissions
Save to my folders

View Full PDF

Jump To

Keywords

apolipoprotein
cardiovascular disease (CVD)
Gilbert’s syndrome
low-density lipoprotein (LDL)-subfractions
lipid metabolism
unconjugated bilirubin

Related Articles

Cited By...