A F Howie

Genetic polymorphisms in the human selenoprotein P gene determine the response of selenoprotein markers to selenium supplementation in a gender-specific manner (the SELGEN study)

Selenium (Se), a micronutrient essential for human health, is incorporated into at least 25 selenoproteins including selenoprotein P (SePP)′ which transports Se within the body. This research identified two single nucleotide polymorphisms (SNPs) in the SePP gene, one in the coding region (position 24731, causing an Ala to Thr change) and one in the 3′untranslated region (position 25191). Their frequency was similar in Caucasian, Chinese, and South Asian populations. Prospectively geno‐typed volunteers were supplemented for 6 wk with 100 μg sodium selenite/day. Blood samples were analyzed for plasma Se and selenoprotein biomark‐ers at baseline, after supplementation, and during a washout period. Plasma Se, SePP, and glutathione peroxidase 3 (GPx3) levels increased with supplementation. Baseline plasma Se content depended on both SePP genotypes and body mass index. Pre‐supplementation SePP concentration was associated with gender and genotype at SNP 24731 and post‐supplementation concentration with SNP 25191. Both SNPs and gender were associated with differences in GPx3 activity, plasma, and erythrocyte thioredoxin reductase 1 concentrations and lymphocyte glutathione peroxidase 1 and 4 activities and concentrations. In conclusion, the data reveal two common functional SNPs within the human SePP gene that may predict behavior of biomarkers of Se status and response to supplementation and thus susceptibility to disease.—Méplan, C., Crosley, L. K., Nicol, F., Beckett, G. J., Howie, A. F., Hill, K. E., Horgan, G., Mathers, J. C., Arthur, J. R., Hesketh, J. E. Genetic polymorphisms in the human selenoprotein P gene determine the response of selenoprotein markers to selenium supplementation in a gender‐specific manner (the SELGEN study) FASEB J. 21, 3063–3074 (2007)

The developmental expression of alpha-, mu- and pi-class glutathione S-transferases in human liver

The developmental expression of the alpha, mu and pi class glutathione S-transferases has been defined in human liver using radioimmunoassay and immunohistochemistry. Expression of alpha and mu class isoenzymes increased significantly at birth, while that of the pi isoenzyme declined during the first trimester. Mu-class isoenzymes (GST1 1, GST1 2, GST1 2-1) were expressed in hepatocytes but not in other liver cell types.

Thioredoxin reductase is the major selenoprotein expressed in human umbilical-vein endothelial cells and is regulated by protein kinase C.

Damage to the endothelium by reactive oxygen species favours atherogenesis. Such damage can be prevented by selenium, which is thought to exert its actions through the expression of selenoproteins. The family of glutathione peroxidases (GPXs) may have antioxidant roles in the endothelium but other intracellular and extracellular selenoproteins with antioxidant actions may also be important. The selenoproteins expressed by cultured human umbilical-vein endothelial cells (HUVECs) were labelled with [(75)Se]selenite and separated using SDS/PAGE. HUVECs secreted no extracellular selenoproteins. There were distinct differences between the intracellular selenoprotein profile of (75)Se-labelled HUVECs and those of other tissues. A single selenoprotein with a molecular mass of 58 kDa accounted for approx. 43% of the intracellular (75)Se-labelled proteins in HUVECs. This protein was identified by Western blotting as the redox-active lipid-hydroperoxide-detoxifying selenoprotein, thioredoxin reductase (TR). TR expression in HUVECs was down-regulated by transiently exposing cells to the phorbol ester PMA for periods as short as 1 min. However, there was a delay of 48 h after PMA exposure before maximal down-regulation of TR was observed. The protein kinase C (PKC) inhibitor bisindolylmaleimide I hydrochloride had no effect on TR expression when added alone, but the agent prevented the down-regulation of TR expression seen with PMA. The calcium ionophore A23187 increased TR expression in HUVECs after a 12-h exposure, but the maximal effect was only observed after a 35-h exposure. These findings suggest that TR may be an important factor in the known ability of Se to protect HUVECs from peroxidative damage. Furthermore, the results also suggest that TR expression can be negatively regulated through PKC. It is possible that TR expression may be positively regulated by the calcium-signalling cascade, although TR induction by A23187 may be due to toxicity.

Selenium supplementation acting through the induction of thioredoxin reductase and glutathione peroxidase protects the human endothelial cell line EAhy926 from damage by lipid hydroperoxides

The human endothelial cell line EAhy926 was used to determine the importance of selenium in preventing oxidative damage induced by tert-butyl hydroperoxide (tert-BuOOH) or oxidised low density lipoprotein (LDLox). In cells grown in a low selenium medium, tert-BuOOH and LDLox killed cells in a dose-dependent manner. At 555 mg/l LDLox or 300 microM tert-BuOOH, >80% of cells were killed after 20 h. No significant cell kill was achieved by these agents if cells were pre-incubated for 48 h with 40 nM sodium selenite, a concentration that maximally induced the activities of cytoplasmic glutathione peroxidase (cyGPX; 5.1-fold), phospholipid hydroperoxide glutathione peroxidase (PHGPX;1.9-fold) and thioredoxin reductase (TR; 3.1-fold). Selenium-deficient cells pre-treated with 1 microM gold thioglucose (GTG) (a concentration that inhibited 25% of TR activity but had no inhibitory effect on cyGPX or PHGPX activity) were significantly (P<0.05) more susceptible to tert-BuOOH toxicity (LC(50) 110 microM) than selenium-deficient cells (LC(50) 175 microM). This was also the case for LDLox. In contrast, cells pre-treated with 40 nM selenite prior to exposure to GTG were significantly more resistant to damage from tert-BuOOH and LDLox than Se-deficient cells. Treatment with GTG or selenite had no significant effect on intracellular total glutathione concentrations. These results suggest that selenium supplementation, acting through induction of TR and GPX, has the potential to protect the human endothelium from oxidative damage.

Comparison of Fructosamine With Glycosylated Hemoglobin and Plasma Proteins as Measures of Glycemic Control

The relative value of fructosamine as an alternative to glycosylated hemoglobin (HbA1) and other measures of glycemic control was assessed in 100 insulin-dependent (IDDM) and 104 non-insulin-dependent (NIDDM) diabetic patients. We measured HbA1 (by electrophoretic and affinity methods), plasma glucose, glycosylated plasma proteins, and fructosamine in blood taken at a single clinic visit. The values were compared both by correlation analysis and by considering whether the various indices of glycemic control placed the patients in the same clinical decision categories as they were in by the HbA1 (affinity) result. Fructosamine correlated moderately well with HbA1 (affinity; r = .8) and placed 71% of IDDM and 72% of NIDDM patients in the same clinical category of good, moderate, or poor control. Differences can probably be partly attributed to the different periods over which HbA1 and fructosamine reflect average glycemia and partly to imprecision.

Decreased maternal hypothalamic-pituitary-adrenal axis activity in very severely obese pregnancy: Associations with birthweight and gestation at delivery

BACKGROUND The maternal hypothalamic-pituitary-adrenal-axis (HPAA) undergoes dramatic activation during pregnancy. Increased cortisol and corticotrophin-releasing-hormone (CRH) associate with low birthweight and preterm labor. In non-pregnant obesity, the HPAA is activated but circulating cortisol levels are normal or lower than in lean women. We hypothesized that maternal cortisol levels would be lower in obese pregnancy, and would associate with increased fetal size and length of gestation. METHOD Fasting serum cortisol was measured at 16, 28 and 36 weeks gestation and at 3-6 months postpartum in 276 severely obese and 135 lean women. In a subset of obese (n=20) and lean (n=20) we measured CRH, hormones that regulate bioavailable cortisol (corticosteroid-binding-globulin, estradiol, estriol, and progesterone). Urinary glucocorticoid metabolites were measured in pregnant (obese n=6, lean n=5) and non-pregnant (obese n=7, lean n=7) subjects. RESULTS Maternal cortisol and HPAA hormones were lower in obese pregnancy. Total urinary glucocorticoid metabolites increased significantly in lean pregnancy, but not in obese. Lower maternal cortisol in obese tended to be associated with increased birthweight (r=-0.13, p=0.066). In obese, CRH at 28 weeks correlated inversely with gestational length (r=-0.49, p=0.04), and independently predicted gestational length after adjustment for confounding factors (mean decrease in CRH of -0.25 pmol/L (95% CI -0.45 to -0.043 pmol/L) per/day increase in gestation). CONCLUSION In obese pregnancy, lower maternal cortisol without an increase in urinary glucocorticoid clearance may indicate a lesser activation of the HPAA than in lean pregnancy. This may offer a novel mechanism underlying increased birthweight and longer gestation in obese pregnancy.

Purification of acidic glutathione S-transferases from human lung, placenta and erythrocyte and the development of a specific radioimmunoassay for their measurement

Human acidic glutathione S-transferases (GST) have been purified from placenta, lung and erythrocytes. The purification protocol resulted in a high yield of pure protein for each tissue, when compared to previous procedures. An apparent subunit Mr of 24,800 was calculated for each of the acidic GST and each enzymes had a pI of 4.75. No immunochemical differences were detected between the acidic GST isolated from the three tissues. A specific and sensitive radioimmunoassay suitable for the measurement of human acidic GST in plasma or tissues is described.

Selenium supplementation attenuates procollagen-1 and interleukin-8 production in fat-loaded human C3A hepatoblastoma cells treated with TGFβ1

BACKGROUND Non-alcoholic fatty liver disease (NAFLD) is associated with obesity, insulin resistance and hepatic steatosis. Non-alcoholic steatohepatitis (NASH) is a serious consequence of NAFLD where chronic tissue damage and inflammation result in fibrosis which may progress to cirrhosis. Transforming growth factor beta1 (TGFbeta1), proinflammatory cytokines and oxidative stress are thought to play crucial roles in the pathogenesis of these conditions. The contributions of individual liver cell types to fibrogenesis remain controversial and the influence of selenium status has not been investigated. METHODS In this study we have used a cell culture model of fat-loading using oleate-treated human hepatoblastoma (C3A) cells to investigate how fat-loading and selenium status might influence the production of collagen in response to TGFbeta1. The secretion of inflammatory cytokines was also investigated, together with the epithelial character of the treated cells. RESULTS We found that in response to treatment with TGFbeta1, C3A cells produced mRNA encoding the pro-alphaI chain of procollagen I, secreted procollagen I peptide, up-regulated production of the proinflammatory cytokine interleukin-8 (IL-8) and the mesenchymal marker vimentin, and down-regulated albumin production. Most of these responses were considerably enhanced when cells were fat-loaded with oleate and were attenuated by selenium addition at a dose that optimised the expression of thioredoxin reductase and glutathione peroxidase. CONCLUSIONS Our data establish that both fat-loading and suboptimal selenium status enhance collagen and IL-8 production by C3A hepatocytes in response to TGFbeta1, possibly as part of an epithelial to mesenchymal transition. GENERAL SIGNIFICANCE These findings suggest that the hepatocyte may be an important contributor to the pathogenesis of fibrosis associated with NAFLD.

Modulation of thioredoxin reductase-2 expression in EAhy926 cells: Implications for endothelial selenoprotein hierarchy

BACKGROUND We examined the expression of the mitochondrial selenoenzyme TrxR2 in the endothelial cell line EAhy926 under conditions known to modify its cytoplasmic counterpart TrxR1. METHODS Cells were cultured with varying concentrations of selenite, sulforaphane or the Ca2+ ionophore A23187 for 72-h, prior to assay of TrxR concentration and activity. Further cultures underwent prolonged (7-day) Se-depletion before selenoprotein measurement. RESULTS In Se-deficient cultures, neither Se, A23187 or sulforaphane affected TrxR2 concentration, while these treatments induced TrxR1 concentration (p<0.05). When co-incubated, optimal concentrations of Se (40 nM) and sulforaphane (4 microM) only modestly increased TrxR2 protein (approximately 1.3-fold), compared with TrxR1 (approximately 4-fold). In Se-deficient cells, TrxR activity was unaffected by sulforaphane or A23187. Prolonged Se-depletion caused a comparatively small reduction in TrxR2 (66% TrxR2 retained) against TrxR1 and glutathione peroxidase-1 activity (38% and 17% retained, respectively). CONCLUSIONS The relative resistance of TrxR2 to Se-deprivation and induction by sulforaphane and A23187 suggests TrxR2 lies near the top of the selenoprotein hierarchy in EAhy926 cells and exhibits near maximum expression under a range of culture conditions. In Se deficiency an inactive (possibly truncated) TrxR1 is produced in response to stimulus by sulforaphane and A23187. GENERAL SIGNIFICANCE These observations underpin a likely critical antioxidant role for TrxR2 and TrxR1 in the endothelium.

An improved ex vivo method of primary porcine hepatocyte isolation for use in bioartificial liver systems

Introduction Primary porcine hepatocytes are commonly used in bioartificial liver devices and for in vitro studies of hepatocyte function. Although in vivo isolation of porcine hepatocytes can give high yield and viability, such methods are time‐consuming and expensive, requiring specialist surgical facilities. Aim To develop a simple, low‐cost, high viability, high yield, reproducible ex vivo method for obtaining functional porcine hepatocytes for use in bioartificial liver systems. Methods Weanling piglets (12 kg) were killed with pentobarbitone sodium, the infra‐hepatic inferior vena cava was clamped and the supra‐hepatic inferior vena cava cannulated. The whole liver was retrogradely perfused in situ with cold saline and excised, followed by an ex vivo open‐loop and re‐circulating perfusion method (at 37°C) in five steps. The liver was disrupted, sequentially filtered in washing buffer, purified by centrifugation and resuspended in Williams E medium. Viability and cell number were assessed using trypan blue exclusion. The cells were subsequently cultured in serum‐free chemically‐defined medium and function was assessed. Results The time interval from when the animals were killed to the final cell wash was 105 ± 5 min (n = 20). Cell viability was 85 ± 6% with a yield of (2.4 ± 0.5) × 1010 from 12 ± 1 kg piglets using 0.03% (w/v) collagenase (n = 20). Hepatocytes from all isolations were successfully plated and grown in monolayer culture. In freshly isolated hepatocytes (day 0) total protein content (TP) was 1.2 ± 0.1 mg/106 cells (n = 5) and 1.2 ± 0.3 mg/106 cells (n = 5) for day 2 monolayer cultures, corresponding to approximately 9 × 106 hepatocytes per dish. The percentage of total LDH released into the medium was 13 ± 4% for day 0 and 8 ± 4% at day 2; conversely, intracellular LDH activities were 87 ± 4% and 92 ± 4% of the total, respectively. The urea synthesis rate was 196 ± 36 nmol/h/mg total protein at day 0 (n = 5) and 292 ± 62 nmol/h/mg protein (n = 9) at day 2. The total P450 content was 99 ± 11 pmol/mg total protein for fresh cells (n = 5) and maintained at 89 ± 35 pmol/mg total protein in day 2 cultures. Conclusions This ex vivo method provides a high viability, high yield, cost‐effective and rapid technique for isolating functional porcine hepatocytes with high plating efficiency, which compares favourably with results obtained using complex in vivo techniques. Eur J Gastroenterol Hepatol 12:923‐930