Modified method for determination of sulfur metabolites in plant tissues by stable isotope dilution-based liquid chromatography–electrospray ionization–tandem mass spectrometry

A wide variety of sulfur metabolites play important roles in plant functions. We have developed a precise and sensitive method for the simultaneous measurement of several sulfur metabolites based on liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS) and ³⁴S metabolic labeling of sulfur-containing metabolites in Arabidopsis thaliana seedlings. However, some sulfur metabolites were unstable during the extraction procedure. Our proposed method does not allow for the detection of the important sulfur metabolite homocysteine because of its instability during sample extraction. Stable isotope-labeled sulfur metabolites of A. thaliana shoot were extracted and utilized as internal standards for quantification of sulfur metabolites with LC–MS/MS using S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), methionine (Met), glutathione (GSH), and glutathione disulfide (GSSG) as example metabolites. These metabolites were detected using electrospray ionization in positive mode. Standard curves were linear (r² > 0.99) over a range of concentrations (SAM 0.01–2.0 μM, SAH 0.002–0.10 μM, Met 0.05–4.0 μM, GSH 0.17–20.0 μM, GSSG 0.07–20.0 μM), with limits of detection for SAM, SAH, Met, GSH, and GSSG of 0.83, 0.67, 10, 0.56, and 1.1 nM, respectively; and the within-run and between-run coefficients of variation based on quality control samples were less than 8%.     

Evidence for altered methionine methyl-group utilization in the diabetic rat's brain

The methionine (MET) derivative, S-adenosylmethionine (SAM), provides methyl-groups for methylation reactions in many neural processes. In rats made diabetic with streptozotocin (SZ), brain SAM levels were generally lower (10–20%) than in controls, with a constant decrease being observed five weeks after onset of diabetes. This decrease in SAM levels may be due to reduced precursor (MET) availability because greatly elevating plasma MET concentrations in SZ diabetic rats by dietary manipulation increased their neural SAM concentrations to be approximately or even greater than (5–20%) those of controls. In contrast, neural levels of SAM's demethylated product, S-adenosylhomocysteine (SAH), were reduced to a greater extent (17–44%) than SAM levels in all groups of SZ diabetic rats independent of their plasma MET concentrations or brain SAM levels. This indicates that the decrease in SAH levels is not simply due to substrate (SAM) restriction. These changes in MET metabolites appear to be a general effect of diabetes rather than a non-pancreatic side-effect of SZ, because genetically diabetic BB Wistar rats also exhibited reduced brain SAM (25%) and brain SAH (46%) levels. These results indicate that methyl-groups from MET are handled differently in the brain of the diabetic rat, which considering the variety and importance of neural methylation reactions, could have important consequences for the diabetic.Abbreviations MET methionine - SAM S-adenosylmethionine - SAH S-adenosylhomocysteine - SZ streptozotocin - BBW BB Wistar - LNAA large neutral amino acids - BCAA branchedchain amino acids - MET:BCAA methionine to branched-chain amino acid ratio - MET:LNAA methionine to large neutral amino acid ratio     

The Regional Concentrations of S-Adenosyl-l-Methionine, S-Adenosyl-l-Homocysteine, and Adenosine in Rat Brain

Abstract: The concentrations of S -adenosyl- l -methionine (SAM), S -adenosyl- l -homocysteine (SAH), and adenosine (Ado) were determined in whole brain and rat brain regions by HPLC. The whole brain contains, respectively, 22 nmol, 1 nmol, and 64 nmol of SAM, SAH, and Ado per g of wet tissue. Their distribution indicated that SAM and SAH levels are highest in brainstem, whereas the Ado level is highest in cortex. With aging the SAM concentrations decrease in whole brain, brainstem, and hypothalamus (–25%) and SAH levels increase by 90% in striatum and by 160% in cerebellum, while Ado levels are increased in all regions by 100–180%.     

S-Adenosylmethionine metabolism in HL-60 cells: effect of cell cycle and differentiation

The effect of the cell cycle and differentiation on S-adenosylmethionine (SAM) metabolism in HL-60 cells has been investigated. Synthesis and pool sizes of SAM and S-adenosylhomocysteine (SAH) were cell-cycle-independent (SAM, 315, μM; SAH, 4.6 μM). The SAM-synthase (ATP: l-methionine S-adenosyltransferase) of HL-60 cells has a K_m for methionine of 12.8±2.0 μM and thus appears to be of the intermediate K_m type found in other malignant tissues. The enzyme does not show cell-cycle regulation. Treatment of cells with DMSO resulted in a rapid and marked decrease of SAM and SAH levels without affecting pool turnover or the SAM/SAH ratio. A decrease in SAM concentration could also be observed in a variant cell line resistant to differentiation with DMSO. DMSO inhibited SAM-synthase in cell-free extracts. This inhibition was noncompetitive with respect to l-methionine. Inhibition of SAM-synthase by cycloleucine lowered SAM levels in intact cells, but resulted in differentiation of only a minor percentage of cells. These data indicate that changes in SAM and SAH levels in HL-60 cells seem to be a consequence rather than a cause of differentiation.     

The effects of dietary supplementation of methionine on genomic stability and p53 gene promoter methylation in rats

Methionine is a component of one-carbon metabolism and a precursor of S-adenosylmethionine (SAM), the methyl donor for DNA methylation. When methionine intake is high, an increase of S-adenosylmethionine (SAM) is expected. DNA methyltransferases convert SAM to S-adenosylhomocysteine (SAH). A high intracellular SAH concentration could inhibit the activity of DNA methyltransferases. Therefore, high methionine ingestion could induce DNA damage and change the methylation pattern of tumor suppressor genes. This study investigated the genotoxicity of a methionine-supplemented diet. It also investigated the diet's effects on glutathione levels, SAM and SAH concentrations and the gene methylation pattern of p53. Wistar rats received either a methionine-supplemented diet (2% methionine) or a control diet (0.3% methionine) for six weeks. The methionine-supplemented diet was neither genotoxic nor antigenotoxic to kidney cells, as assessed by the comet assay. However, the methionine-supplemented diet restored the renal glutathione depletion induced by doxorubicin. This fact may be explained by the transsulfuration pathway, which converts methionine to glutathione in the kidney. Methionine supplementation increased the renal concentration of SAH without changing the SAM/SAH ratio. This unchanged profile was also observed for DNA methylation at the promoter region of the p53 gene. Further studies are necessary to elucidate this diet's effects on genomic stability and DNA methylation.     

Regulation of S-adenosyl methionine synthesis in the mouse embryo

In early embryos, methylation is involved in "gamete imprinting" and inactivation of artificially introduced foreign genes. We studied the biosynthesis of the universal methylation cofactor: S-Adenosyl methionine (SAM). In the mouse, SAM conversion from methionine is limited by saturation of the methionine endogenous pool. SAM is present at a practically unchanged level from the unfertilized oocyte to early morula. SAM synthesis is increased at the time of compaction. In blastocysts, although methionine uptake is increased, the conversion rate from methionine is lowered. We observed no differences between C57 Black and Swiss albino random bred strains. In few experiments with human unfertilized oocytes and spared embryos, we observed higher methionine incorporation, and higher conversion to SAM. Next, the effect of two methylation inhibitors was tested, on early mouse embryonic development, at the one-cell and the two-cell stage. We found that ethionine is very toxic, even at the lowest tested concentration of 25 microM. Homocysteine is more potent at the one-cell stage than at the 2-cell stage, and it only partially blocks blastocyst formation from the 2-cell stage even at a concentration of 500 microM. It clearly acts as a methylation inhibitor; it lowers the SAM pool and the methylation index, SAH/SAM ratio (SAH: S-Adenosyl Homocysteine). We also found that homocysteine is an unexpected competitor for methionine influx and efflux.     

Effects of methionine synthase and methylenetetrahydrofolate reductase gene polymorphisms on markers of one-carbon metabolism

Genetic and nutritional factors play a role in determining the functionality of the one-carbon (1C) metabolism cycle, a network of biochemical reactions critical to intracellular processes. Genes encoding enzymes for methylenetetrahydrofolate reductase (MTHFR) and methionine synthase (MTR) may determine biomarkers of the cycle including homocysteine (HCY), S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH). MTHFR C677T is an established genetic determinant of HCY but less is known of its effect on SAM and SAH. Conversely, the relationship between MTR A2756G and HCY remains inconclusive, and its effect on SAM and SAH has only been previously investigated in a female-specific population. Folate and vitamin B₁₂ are essential substrate and cofactor of 1C metabolism; thus, consideration of gene–nutrient interactions may clarify the role of genetic determinants of HCY, SAM and SAH. This cross-sectional study included 570 healthy volunteers from Kingston, Ontario, Ottawa, Ontario and Halifax, Nova Scotia, Canada. Least squares regression was used to examine the effects of MTR and MTHFR polymorphisms on plasma HCY, SAM and SAH concentrations; gene–gene and gene–nutrient interactions were considered with the inclusion of cross-products in the model. Main effects of MTR and MTHFR polymorphisms on HCY concentrations were observed; however, no gene–gene or gene–nutrient interactions were found. No association was observed for SAM. For SAH, interactions between MTR and MTHFR polymorphisms, and MTHFR polymorphism and serum folate were found. The findings of this research provide evidence that HCY and SAH, biomarkers of 1C metabolism, are influenced by genetic and nutritional factors and their interactions.     

Cigarette smoke extract increases S-adenosylmethionine and cystathionine in human lung epithelial-like (A549) cells

The effect of cigarette smoke extract (CSE) on S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), and sulfur amino acid metabolism was examined in human lung epithelial-like (A549) cells exposed to various CSE concentrations (2.5-100%) for 24 or 48 h. Intracellular SAM and SAM/SAH ratio were elevated after exposure to CSE for 48 h. Cell SAH content decreased, but the effect was not consistent. Cellular cystathionine, cysteine, and methionine levels were increased after CSE exposure for 48h. Sub-acute exposure to CSE induced increases in cellular SAM and SAM/SAH ratio. The transsulfuration pathway was likely activated by CSE since cystathionine increased, potentially contributing to the increased total intracellular GSH content.     

Variations in S-adenosylmethionine, S-adenosylhomocysteine and adenosine concentrations in rat liver.

The hepatic concentrations of S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH) and adenosine (Ado) in the rat were examined diurnally and as a function of fasting. Ado concentrations increased continuously throughout the fasting period; concentrations after 2 days of fasting were 7.5-fold higher than control values. Diurnally, the concentration of Ado was highest during the light hours. SAM and the ratio of SAM/SAH were reduced greater than 50% due to fasting and exhibited a significant daily rhythm which appeared to be related to dietary methionine availability. Hepatic SAM concentrations decreased continuously during the light hours and increased during the dark period to levels 7.3-fold greater than the lowest light values. The concentration of SAH was altered in a similar fashion yet to a much lesser degree such that the ratio of SAM/SAH paralleled the changes in the concentration of SAM. The SAM/SAH ratio exhibited a 4.5-fold difference between the peak and nadir values.     

Homocysteine alterations in experimental cholestasis and its subsequent cirrhosis

Homocysteine (Hcy), an intermediate in methionine metabolism, has been proposed to be involved in hepatic fibrogenesis. Impaired liver function can alter Hcy metabolism. The aim of the present study was to determine plasma Hcy alterations in acute obstructive cholestasis and the subsequent biliary cirrhosis. Cholestasis was induced by bile duct ligation and sham-operated and unoperated rats were used as controls. The animals were studied on the days 7th, 14th, 21st and 28th after the operation. Plasma Hcy, cysteine, methionine, nitric oxide (NO) and liver S-adenosyl-methionine (SAM), S-adenosyl-homocysteine (SAH), SAM to SAH ratio and glutathione were measured. Chronic L-NAME treatment was also included in the study. Plasma Hcy concentrations were transiently elevated by the day 14th after bile duct ligation (P < 0.01) and subsequently returned to control levels. Similar relative fluctuations in plasma Hcy were observed in BDL rats after intraperitoneal methionine overload. Plasma methionine, cysteine and nitrite and nitrate were significantly increased after bile duct ligation. SAM to SAH ratio was diminished by the 1st week of cholestasis and remained significantly decreased throughout the study. These events were accompanied by a decrease in GSH to GSSG ratio in the liver. Chronic L-NAME treatment improved SAM to SAH ratio and prevented the elevation of plasma Hcy and methionine (P < 0.05) while couldn't influence the other parameters. In conclusion, this study demonstrates alterations in plasma Hcy and liver SAM and SAH contents in precirrhotic stages and in secondary biliary cirrhosis, for the first time. In addition, we observed that plasma Hcy concentrations in BDL rats follow a distinct pattern of alteration from what has been previously reported in other models of cirrhosis. NO overproduction may contribute to plasma Hcy elevation and liver SAM depletion after cholestasis.     

Ethanol exposure modulates hepatic S-adenosylmethionine and S-adenosylhomocysteine levels in the isolated perfused rat liver through changes in the redox state of the NADH/NAD system

Methionine metabolism is disrupted in patients with alcoholic liver disease, resulting in altered hepatic concentrations of S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), and other metabolites. The present study tested the hypothesis that reductive stress mediates the effects of ethanol on liver methionine metabolism. Isolated rat livers were perfused with ethanol or propanol to induce a reductive stress by increasing the NADH/NAD⁺ ratio, and the concentrations of SAM and SAH in the liver tissue were determined by high-performance liquid chromatography. The increase in the NADH/NAD⁺ ratio induced by ethanol or propanol was associated with a marked decrease in SAM and an increase in SAH liver content. 4-Methylpyrazole, an inhibitor the NAD⁺-dependent enzyme alcohol dehydrogenase, blocked the increase in the NADH/NAD⁺ ratio and prevented the alterations in SAM and SAH. Similarly, co-infusion of pyruvate, which is metabolized by the NADH-dependent enzyme lactate dehydrogenase, restored the NADH/NAD⁺ ratio and normalized SAM and SAH levels. The data establish an initial link between the effects of ethanol on the NADH/NAD⁺ redox couple and the effects of ethanol on methionine metabolism in the liver.     

Ethanol exposure modulates hepatic S-adenosylmethionine and S-adenosylhomocysteine levels in the isolated perfused rat liver through changes in the redox state of the NADH/NAD(+) system

Methionine metabolism is disrupted in patients with alcoholic liver disease, resulting in altered hepatic concentrations of S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), and other metabolites. The present study tested the hypothesis that reductive stress mediates the effects of ethanol on liver methionine metabolism. Isolated rat livers were perfused with ethanol or propanol to induce a reductive stress by increasing the NADH/NAD(+) ratio, and the concentrations of SAM and SAH in the liver tissue were determined by high-performance liquid chromatography. The increase in the NADH/NAD(+) ratio induced by ethanol or propanol was associated with a marked decrease in SAM and an increase in SAH liver content. 4-Methylpyrazole, an inhibitor the NAD(+)-dependent enzyme alcohol dehydrogenase, blocked the increase in the NADH/NAD(+) ratio and prevented the alterations in SAM and SAH. Similarly, co-infusion of pyruvate, which is metabolized by the NADH-dependent enzyme lactate dehydrogenase, restored the NADH/NAD(+) ratio and normalized SAM and SAH levels. The data establish an initial link between the effects of ethanol on the NADH/NAD(+) redox couple and the effects of ethanol on methionine metabolism in the liver.     

Methionine and homocysteine modulate the rate of ROS generation of isolated mitochondria in vitro

Dietary methionine restriction and supplementation in mammals have beneficial (antiaging) and detrimental effects respectively, which have been related to chronic modifications in the rate of mitochondrial ROS generation. However it is not known if methionine or its metabolites can have, in addition, direct effects on the rate of mitochondrial ROS production. This is studied here for the methionine cycle metabolites S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), homocysteine and methionine itself in isolated rat liver, kidney, heart, and brain mitochondria. The results show that methionine increases ROS production in liver and kidney mitochondria, homocysteine increases it in kidney and decreases it in the other three organs, and SAM and SAH have no effects. The variations in ROS production are localized at complexes I or III. These changes add to previously described chronic effects of methionine restriction and supplementation in vivo.     

Structural basis for recognition of S-adenosylhomocysteine by riboswitches

S-adenosyl-(L)-homocysteine (SAH) riboswitches are regulatory elements found in bacterial mRNAs that up-regulate genes involved in the S-adenosyl-(L)-methionine (SAM) regeneration cycle. To understand the structural basis of SAH-dependent regulation by RNA, we have solved the structure of its metabolite-binding domain in complex with SAH. This structure reveals an unusual pseudoknot topology that creates a shallow groove on the surface of the RNA that binds SAH primarily through interactions with the adenine ring and methionine main chain atoms and discriminates against SAM through a steric mechanism. Chemical probing and calorimetric analysis indicate that the unliganded RNA can access bound-like conformations that are significantly stabilized by SAH to direct folding of the downstream regulatory switch. Strikingly, we find that metabolites bearing an adenine ring, including ATP, bind this aptamer with sufficiently high affinity such that normal intracellular concentrations of these compounds may influence regulation of the riboswitch.     

Immunoassay of <Emphasis Type="Italic">S</Emphasis>-adenosylmethionine and <Emphasis Type="Italic">S</Emphasis>-adenosylhomocysteine: the methylation index as a biomarker for disease and health status

Background

S-Adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH) are relevant to a variety of diseases. Previous reports that quantified SAM and SAH were based on HPLC or LC–MS/MS. No antibody against SAM has been generated, and the antibody against SAH cannot be used with blood samples. Immunoassays have not been used to measure SAM and SAH. In this study, ELISA was used to measure blood SAM and SAH levels.

Results

Specific antibodies against SAM were produced for the first time using a stable analog as the antigen. The monoclonal antibodies against SAM and SAH were characterized. No cross-reactivity was detected for the analyzed analogs. For the anti-SAM antibodies, the ELISA sensitivity was ~2 nM, and the affinity was 7.29 × 10¹⁰ L/mol. For the anti-SAH antibodies, the sensitivity was ~15 nM, and the affinity was 2.79 × 10⁸ L/mol. Using high-quality antibodies against SAM and SAH, immunoassays for the detection of SAM and SAH levels in blood and tissue samples were developed. Clinical investigations using immunoassays to measure SAM, SAH and the methylation index (MI) in normal and diseased samples indicated that (1) the SAM level is age and gender dependent; (2) the SAM level is associated with the severity of liver diseases, inflammatory reactions and other diseases; and (3) the methylation index (MI) is significantly reduced in many diseases and may serve as a screening biomarker to identify potentially unfavorable health conditions.

Conclusion

It is possible to generate antibodies against active small biomolecules with weak immunogenicity, such as SAM and SAH, using traditional hybridoma technology. The antigens and antibodies described here will contribute to the development of immunoassays to measure SAM, SAH and related molecules. These assays enable the MI to be measured specifically, accurately, easily and quickly without costly equipment. This preliminary study indicates that the MI could be an effective indicator of general health, except under conditions that may alter the value of the MI, such as special diets and medications.

     

S‐adenosyl‐l‐methionine usage during climacteric ripening of tomato in relation to ethylene and polyamine biosynthesis and transmethylation capacity

S‐adenosyl‐l ‐methionine (SAM) is the major methyl donor in cells and it is also used for the biosynthesis of polyamines and the plant hormone ethylene. During climacteric ripening of tomato (Solanum lycopersicum ‘Bonaparte’), ethylene production rises considerably which makes it an ideal object to study SAM involvement. We examined in ripening fruit how a 1‐MCP treatment affects SAM usage by the three major SAM‐associated pathways. The 1‐MCP treatment inhibited autocatalytic ethylene production but did not affect SAM levels. We also observed that 1‐(malonylamino)cyclopropane‐1‐carboxylic acid formation during ripening is ethylene dependent. SAM decarboxylase expression was also found to be upregulated by ethylene. Nonetheless polyamine content was higher in 1‐MCP‐treated fruit. This leads to the conclusion that the ethylene and polyamine pathway can operate simultaneously. We also observed a higher methylation capacity in 1‐MCP‐treated fruit. During fruit ripening substantial methylation reactions occur which are gradually inhibited by the methylation product S‐adenosyl‐l ‐homocysteine (SAH). SAH accumulation is caused by a drop in adenosine kinase expression, which is not observed in 1‐MCP‐treated fruit. We can conclude that tomato fruit possesses the capability to simultaneously consume SAM during ripening to ensure a high rate of ethylene and polyamine production and transmethylation reactions. SAM usage during ripening requires a complex cellular regulation mechanism in order to control SAM levels.     

Using S-adenosyl-l-homocysteine capture compounds to characterize S-adenosyl-l-methionine and S-adenosyl-l-homocysteine binding proteins

S-Adenosyl-l-methionine (SAM) is recognized as an important cofactor in a variety of biochemical reactions. As more proteins and pathways that require SAM are discovered, it is important to establish a method to quickly identify and characterize SAM binding proteins. The affinity of S-adenosyl-l-homocysteine (SAH) for SAM binding proteins was used to design two SAH-derived capture compounds (CCs). We demonstrate interactions of the proteins COMT and SAHH with SAH–CC with biotin used in conjunction with streptavidin–horseradish peroxidase. After demonstrating SAH-dependent photo-crosslinking of the CC to these proteins, we used a CC labeled with a fluorescein tag to measure binding affinity via fluorescence anisotropy. We then used this approach to show and characterize binding of SAM to the PR domain of PRDM2, a lysine methyltransferase with putative tumor suppressor activity. We calculated the K_d values for COMT, SAHH, and PRDM2 (24.1 ± 2.2 μM, 6.0 ± 2.9 μM, and 10.06 ± 2.87 μM, respectively) and found them to be close to previously established K_d values of other SAM binding proteins. Here, we present new methods to discover and characterize SAM and SAH binding proteins using fluorescent CCs.     

In Vitro Effect of Amphotericin B on Candida albicans,Candida glabrata and Candida parapsilosis Biofilm Formation

Candida spp. biofilm is considered highly resistant to conventional antifungals. The aim of this study was to investigate the in vitro effect of amphotericin B on Candida spp. biofilms at different stages of maturation. We investigated the activity of amphotericin B against 78 clinical isolates of Candida spp., representing three species, growing as planktonic and sessile cells, by a widely accepted broth microdilution method. The in vitro effect on sessile cell viability was evaluated by MTT reduction assay. All examined strains were susceptible to amphotericin B when grown as free-living cells. At the early stages of biofilm maturation 96.7–100.0 % strains, depending on species, displayed amphotericin B sessile minimal inhibitory concentration (SMIC) ≤1 μg/mL. Mature Candida spp. biofilm of 32.1–90.0 % strains displayed amphotericin B SMIC ≤1 μg/mL. Based on these results, amphotericin B displays species- and strain-depending activity against Candida spp. biofilms.     

Methylation potential associated with diet,genotype, protein,and metabolite levels in the Delta Obesity Vitamin Study

Micronutrient research typically focuses on analyzing the effects of single or a few nutrients on health by analyzing a limited number of biomarkers. The observational study described here analyzed micronutrients, plasma proteins, dietary intakes, and genotype using a systems approach. Participants attended a community-based summer day program for 6–14 year old in 2 years. Genetic makeup, blood metabolite and protein levels, and dietary differences were measured in each individual. Twenty-four-hour dietary intakes, eight micronutrients (vitamins A, D, E, thiamin, folic acid, riboflavin, pyridoxal, and pyridoxine) and 3 one-carbon metabolites [homocysteine (Hcy), S-adenosylmethionine (SAM), and S-adenosylhomocysteine (SAH)], and 1,129 plasma proteins were analyzed as a function of diet at metabolite level, plasma protein level, age, and sex. Cluster analysis identified two groups differing in SAM/SAH and differing in dietary intake patterns indicating that SAM/SAH was a potential marker of nutritional status. The approach used to analyze genetic association with the SAM/SAH metabolites is called middle-out: SNPs in 275 genes involved in the one-carbon pathway (folate, pyridoxal/pyridoxine, thiamin) or were correlated with SAM/SAH (vitamin A, E, Hcy) were analyzed instead of the entire 1M SNP data set. This procedure identified 46 SNPs in 25 genes associated with SAM/SAH demonstrating a genetic contribution to the methylation potential. Individual plasma metabolites correlated with 99 plasma proteins. Fourteen proteins correlated with body mass index, 49 with group age, and 30 with sex. The analytical strategy described here identified subgroups for targeted nutritional interventions.

Electronic supplementary material

The online version of this article (doi:10.1007/s12263-014-0403-9) contains supplementary material, which is available to authorized users.