Jean-Pierre Knapp

Genetic defects as important factors for moderate hyperhomocysteinemia

Abstract The genes for the enzymes methylenetetrahydrofolate reductase (MTHFR), methionine synthase (MS), methionine synthase reductase (MSR) and cytathionine-Β-synthase (CBS) play an important role in homocysteine metabolism. Rare mutations in these genes cause severe hyperhomocysteinemia and clinical symptoms. Growing interest has focused on common mutations with moderate effects on homocysteine levels. We studied 280 subjects of different age groups for the following mutations: MTHFR677C→T and 1298A→C, MS2756A→G, MSR66A→G and the 68 bp insertion in the CBS gene. The median value for homocysteine increased significantly with age (median homocysteine levels: 7.5, 12.4 and 16.5 μmol/l in the age groups 20–43, 65–75 and 85–96 years, respectively). The genotypes of the MTHFR677C→T mutation were associated with differences in plasma homocysteine levels, but without reaching significance. Individuals homozygous for the MTHFR677C→T mutation had a 2.3 μmol/l higher median homocysteine level compared to individuals with the wild-type allele. This effect was pronounced in combination with low folate levels and abolished with higher folate in plasma. For the other three mutations no association with homocysteine values could be determined. The analysis of homocysteine metabolite cystathionine by backward regression analysis revealed a significant correlation of the MS2756A→G mutation with cystathionine level. This increase could indicate a disturbed remethylation. In summary, larger and homogeneous study populations are necessary to quantify the small effects of common mutations on homocysteine levels. This may also be the reason that no effects of genetic interactions between two genotypes were observed.

Simultaneous quantification of S-adenosyl methionine and S-adenosyl homocysteine in human plasma by stable-isotope dilution ultra performance liquid chromatography tandem mass spectrometry

S-adenosyl methionine (SAM) is an important methyl group donor that is formed from methionine. S-adenosyl homocysteine (SAH) is formed after demethylation of SAM and represents a potent inhibitor of many methyltransferases. We developed an improved stable-isotope dilution ultra performance liquid chromatography tandem mass spectrometry (UPLC-MS/MS) method for the simultaneous quantification of SAM and SAH in biological samples. The method comprises a phenylboronic acid-containing solid-phase extraction procedure, serving for binding and clean-up of SAM and SAH. After extraction, samples were separated and detected using either a HPLC SymmetryShield RP(18) or an Acquity UPLC BEH C(18) column with a HPLC-MS/MS or an UPLC-MS/MS system. The best results were obtained by Acquity UPLC BEH C(18) column. In plasma samples, the estimated intraassay coefficients of variation (CVs) for SAM and SAH were 3.3% and 3.9%, respectively, the interassay CVs were 10.1% for SAM and 8.3% for SAH. Mean recovery of SAM and SAH at two different concentrations was 100.0% for SAM and 101.7% for SAH. The quantification limits were 0.5 and 0.7nmol/L for SAM and SAH, respectively. In 31 plasma samples, the mean concentrations (SD) were 85.5 (11.1)nmol/L for SAM and 13.3 (5.0)nmol/L for SAH with a SAM/SAH ratio of 7.0 (1.8). The new UPLC-MS/MS method showed very high sensitivity and selectivity for SAM and SAH, low CVs and fast sample preparation (40 samples in 60min) and analysis time (3min). This new assay can be used for large-scale clinical studies.

The vegetarian lifestyle and DNA methylation.

Abstract Vegetarians have a lower intake of vitamin B12 than omnivores do. Vitamin B12 deficiency (holotranscobalamin II <35pmol/L or methylmalonic acid >271nmol/L) was found in 58% of 71 vegetarians studied. Higher homocysteine levels (>12μmol/L) found in 45% indicate disturbed remethylation of homocysteine to methionine. The methylation of DNA is strongly linked to homocysteine metabolism. Since DNA methylation is an important epigenetic factor in the regulation of gene expression, alteration of the methylation pattern has been associated with aging, cancer, atherosclerosis and other diseases. Three observations indicate that DNA methylation could be diminished by a vegetarian lifestyle. The vegetarian diet has a low content of methionine, remethylation of homocysteine is reduced by vitamin B12 deficiency and elevated homocysteine levels can induce the generation of S-adenosylhomocysteine (SAH), a potent inhibitor of methyltransferases. In our study we observed a significant correlation between SAH and whole-genome methylation (r=−0.36, p<0.01). This observation underlines the role of SAH as a potent inhibitor of methyltransferases. The methylation status was not correlated with homocysteine or S-adenosylemethionine (SAM). These results indicate that the degree of methylation does not depend on the supply of methyl groups and that the reverse generation of SAH has no influence. In addition to whole-genome methylation, the specific promoter methylation of the p66Shc gene was studied. However, the latter did not correlate with SAH, SAM or homocysteine. Obviously, the promoter methylation of the p66Shc gene is controlled in a specific way, without following the general regulating influence of SAH. In conclusion, an inhibitory effect of SAH on whole-genome methylation was found, but from our data no interaction between vegetarian lifestyle and DNA methylation could be determined.

Quantification of key folate forms in serum using stable-isotope dilution ultra performance liquid chromatography-tandem mass spectrometry.

Folates act as essential coenzymes in many biological pathways. Alteration in folate form distribution might have biological significance, especially in relation to certain genetic polymorphisms. We developed a stable-isotope dilution ultra performance liquid chromatography-mass spectrometry (UPLC-MS/MS) method for quantification of the folate forms 5-methyltetrahydrofolate (5-methylTHF), 5-formylTHF, 5,10-methenylTHF, THF, and folic acid in serum. After extraction using an ion exchange and mixed mode solid-phase, samples were separated and detected using an UPLC-MS/MS system. The quantification limits were between 0.17nmol/L (5-formylTHF) and 1.79nmol/L (THF), and the assay was linear up to 100nmol/L (5-methylTHF) and 10nmol/L (5-formylTHF, 5,10-methenylTHF, THF, and folic acid). The intraassay CVs for 5-methylTHF and 5-formylTHF were 2.0% and 7.2%, respectively. Mean recoveries were between 82.3% for THF and 110.8% for 5,10-methenylTHF. Concentrations of total folate measured by the new method showed a strong correlation with those measured by an immunologic assay (r=0.939; p<0.001). The mean total folate from 32 apparently healthy subjects was 18.09nmol/L, of which 87.23% was 5-methylTHF. Concentrations of homocysteine showed a better correlation to the total folate measured by the new method compared to that obtained by an immunologic assay. We also confirmed that MTHFR polymorphism has a significant effect on folate distribution in this small population of non-supplemented subjects.

The role of genetic factors in the development of hyperhomocysteinemia.

Abstract Moderate hyperhomocysteinemia has been identified as a new independent risk factor for cardiovascular and neurodegenerative diseases. This fact has produced interest in the study of genetic variants involved in homocysteine metabolism and its relationship to pathogenesis. Recently, more than 15 different genes were studied for their relationship to plasma homocysteine levels. We determined the influence of genetic variants in five genes (5,10-methylenetetrahydrofolate reductase (MTHFR) 677C→T, serine hydroxymethyltransferase (SHMT) 1420C→T, thymidylate synthase (TS) 2R→3R, catechol-O-methyltransferase (COMT) 1947G→A and transcobalamin (TC) 776C→G) on plasma homocysteine, folic acid and parameters of vitamin B12 metabolism in 111 vegetarians (mean age: 46±15 years) and 118 healthy seniors (mean age: 82±6.5 years). Median homocysteine concentration in plasma was significantly influenced by the MTHFR genotypes in both populations. In the vegetarians the median homocysteine level was increased by 8 μmol/l in individuals homozygous for the mutation as compared to wild-type or heterozygous genotypes (20.4 μmol/l vs. 12.9 and 12.7 μmol/l, respectively). This unexpected increase was observed although the folate levels were in medium to elevated ranges. Our results suggest that vegetarians have a higher demand for folate to neutralize the genotype effect. Preclinical vitamin B12 deficiency in vegetarians may be the cause for disturbed remethylation and folate trap. Plasma homocysteine was not significantly influenced by the SHMT, TS, COMT and TC mutations. In addition, for the TC mutation a trend toward cellular vitamin B12 deficiency was observed. The methylmalonic acid (MMA) levels were slightly elevated and the holotranscobalamin-II (holoTC-II) levels decreased. In the vegetarian group a significant relationship between the COMT genotype and holoTC-II concentration in plasma was determined, whereas the high activity COMT genotype (G/G) resulted in increased levels (35 μmol/l vs. 21 μmol/l for heterozygous and low activity genotypes). The MMA levels were inversely correlated to holoTC-II concentrations. In conclusion, the study on vegetarians and seniors documents interesting lifestyle-genotype interactions. Although the TC and COMT mutations influence cellular vitamin B12 metabolism, this effect did not result in overt homocysteine elevation.

Decreased p66Shc promoter methylation in patients with end-stage renal disease.

Abstract Background: p66Shc is a stress response protein and partially regulated by epigenetic modifications. Mice lacking p66Shc have reduced atherosclerosis, increased resistance to oxidative stress and a prolonged life time. The aim of the present study was to compare promoter methylation of the p66Shc gene between healthy controls and patients with end-stage renal disease (ESRD). There are two reasons for studying patients with ESRD. First, patients with ESRD have a disturbed homocysteine metabolism, and second an increased risk of morbidity and mortality from cardiovascular disease is a constant finding in these patients. Methods: In our study, we measured fasting levels of homocysteine, S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH) and 8-isoprostane in 22 patients and in 26 healthy, age- and sex-matched controls. The methylation of the p66Shc promoter and Line-1, as surrogate marker of whole genome methylation was quantified in peripheral blood mononuclear cells. Results: In comparison to the control group, homocysteine, SAM, SAH, 8-isoprostane and whole genome methylation were significantly elevated in ESRD patients, while the p66Shc promoter methylation was significantly reduced. A significant correlation was found between SAH and p66Shc promoter methylation in the patient group. This observation underlines the role of SAH as a potent inhibitor of methyltransferases. Using backward regression analysis, we demonstrated that 8-isoprostane has a significant influence on p66Shc promoter methylation. In the control group and in patients with ESRD, increasing 8-isoprostane levels were linked to an elevated promoter methylation. Conclusions: Under physiological conditions, based on the results of the control group, the p66Shc expression is more silenced through epigenetic modifications. The atherosclerotic risk is dramatically increased in ESRD patients; therefore, our experimental results of methylation are in accordance with the clinical situation. Clin Chem Lab Med 2007;45:1764–70.

Fluorescence-based single-strand conformation polymorphism analysis of mutations by capillary electrophoresis.

Capillary electrophoresis in combination with fluorescence‐based single‐strand conformation polymorphism (SSCP) analysis was used to screen for known mutations as well as for unknown mutations. The mutations causing hemochromatosis and thrombogenetic diseases (factor V Leiden mutation and prothrombin mutation) are well defined. Familial hypercholesterolemia is caused by mutations in the low density lipoprotein (LDL) receptor gene. Because the mutations are heterogeneously localized in all 18 exons of the LDL receptor gene, effective screening procedures are necessary. The three well known mutations and 59 of 61 previously characterized mutations in the LDL receptor gene were detected by a distinct abnormal fragment pattern in capillary electrophoresis. The remaining two mutations in the LDL receptor gene showed only slight abnormalities under standard electrophoresis conditions (13 kV, 30°C, 30 min). However, the abnormal pattern could be amplified by increasing the electrophoresis temperature. In all cases, heterozygous and homozygous mutations could clearly be differentiated from wild‐type alleles. Because of the high efficiency of mutation detection, capillary electrophoresis in combination with fluorescence‐based SSCP analysis would be attractive for the detection of well‐defined mutations as well as for the screening of unknown mutations. The accuracy and the degree of automation make this technique well suited for routine genetic diagnosis.

The impact of hyperhomocysteinemia as a cardiovascular risk factor in the prediction of coronary heart disease.

Abstract Coronary heart disease often occurs in the absence of traditional risk factors. Consequently, epidemiological studies exploring novel risk factors are necessary to improve the prediction of coronary heart disease. This study evaluated five promising markers of cardiovascular risk: homocysteine, C-reactive protein, fibrinogen, lipoprotein(a) (Lp(a)), free apolipoprotein(a) (apo(a)) and Lp(a) phenotypes. The study included 135 patients with angiographically confirmed atherosclerosis. The control group consisted of 93 sex- and age-matched individuals. The Mann-Whitney U-test was used for group comparison. New risk factors were evaluated by binary logistic regression. The odds ratios were calculated continuously for homocysteine in dependence on C-reactive protein. Low density lipoprotein (LDL)-cholesterol was nearly identical in controls and patients. Homocysteine, C-reactive protein, fibrinogen, high density lipoprotein (HDL)-cholesterol and Lp(a) discriminated highly significantly between both groups. The continuously calculated odds ratio for homocysteine demonstrated a distinct influence of C-reactive protein. In the group with high C-reactive protein levels, homocysteine levels above 9.6 μmol/l resulted in a markedly elevated risk (odds ratio 12), in the group with C-reactive protein levels below 5 mg/dl, a comparable risk increase was observed at a homocysteine level of 16.6 μmol/l. This data strongly suggests that plasma homocysteine helps identify individuals at risk, especially among those with elevated C-reactive protein levels.

In vitro inhibition of fibrinolysis by apolipoprotein(a) and lipoprotein(a) is size- and concentration-dependent.

Abstract Lipoprotein(a) (Lp(a)) is considered an independent risk factor for atherosclerotic heart and circulatory diseases. The unique, polymorphic character of Lp(a) is based on its apolipoprotein(a) (apo(a)), which has remarkable structural analogies with plasminogen, an important protein for fibrinolysis. The formation of plasmin from plasminogen is a fundamental step in the dissolution of fibrin. Repression of this step may lead to a deceleration of fibrinolysis. It has been suggested that Lp(a) has antifibrinolytic properties through apo(a) and that the apo(a)-size polymorphism has a distinct influence on the prothrombotic properties of Lp(a). However, the results on this topic are controversial. Therefore we used a standardized in vitro fibrinolysis model to provide further information on the influence of Lp(a) on plasmin formation. Monitoring the time-course of plasmin formation, we investigated the inhibition of plasmin formation through dependence on Lp(a), respectively, free apo(a) concentration. Furthermore, we investigated the influence of three Lp(a)/apo(a) phenotypes (22KLp(a), 22 kringle-4 repeats; 30KLp(a), 30 kringle-4 repeats; 35KLp(a), 35 kringle-4 repeats). Adding varying amounts of Lp(a) to our model, we observed that the rate of plasmin formation was inversely related to the Lp(a) concentration. At 0.1 µmol/l 30KLp(a), for example, the plasmin formation was reduced by 12.7% and decreased further by 40.7% at 0.25 µmol/l Lp(a). A similar but more distinct effect was observed when free 30Kapo(a) was added to the model (25.3% at 0.1 µmol/l vs. 59.3% at 0.25 µmol/l). Comparing the antifibrinolytic influence of different apo(a) phenotypes we found that the reduction of plasmin generation advanced with the size of apo(a). At 0.1 µmol/l Lp(a) the reduction of the plasmin formation increased in the order 22KLp(a), 30KLp(a) and 35KLp(a) from 3.7% to 10.7% and 22.3%, respectively. Experiments with different phenotypes of free apo(a) showed similar results (0.5 µmol/l: 22Kapo(a), 56.4% vs. 30KLp(a), 80.4%). Summarizing these results, our study indicates a distinct interrelation of Lp(a)/apo(a) phenotype and concentration with the formation of plasmin. From the antifibrinolytic Lp(a)/apo(a) effect in vitro it may be hypothesized that Lp(a)/apo(a) also has an inhibitory influence on in vivo fibrinolysis.