The effects of dietary methionine and folate on plasma tHcy in 8-week old apolipoprotein E (apoE) deficient mice were investigated. It was first determined whether dietary methionine, during a 4-week period, affected fasting plasma tHcy levels in chow-diet fed male apolipoprotein E-deficient mice. The chow diet was analyzed to contain 2.5 mg/kg folate and 4.2 g/kg methionine, resulting in an approximate daily intake of 6.6 μg folate and 11 mg of methionine per mouse. Three increasing amounts of methionine were fed chronically, by methionine addition to the drinking water. The mean fasting plasma tHcy in mice without methionine supplementation was 7.3 ± 1.5 μM, and increasing methionine in the diet resulted in increasing concentrations of fasting plasma tHcy, with estimated 132 mg/day methionine intake leading to fasting tHcy of 22 ± 2.5 μM (). Nonlinear regression of this data yields a Bmax of 28 μM tHcy (r2=0.90), suggesting that this is the theoretical limit of fasting level of tHcy that could be obtained in these mice on a normal folate diet by methionine supplementation in the drinking water.
Effect of varying dietary methionine levels on fasting plasma Hcy levels in apolipoprotein E-deficient mice
To investigate the effects of both folate and methionine, we used a low folate diet that was analyzed to contain 0.065 mg/kg folate and 4.3 g/kg methionine (methionine content similar to the level in the chow diet). This diet was supplemented with drinking water containing either: 1) no additions to yield the low folate/low methionine diet (LoF/LoM); 2) 4% (w/v) methionine to yield the low folate/high methionine diet (LoF/HiM); 3) 10 mg/l folate to yield the high folate/low methionine diet (HiF/LoM); or 4) 10 mg/ml folate plus 4% methionine to yield the high folate/high methionine diet (HiF/HiM). The estimated daily intake of folate and methionine for these four diets are shown in . Mice were fed these four diets chronically from 9 weeks of age until 16 weeks of age (8–10 per group), and their tHcy levels, after an overnight fast and after replacing the folate/methionine supplemented drinking water with plain water, are shown in . The two groups on low methionine diets, HiF/LoM and LoF/LoM, had mean fasting tHcy levels of 9.3 and 12.6 μM, respectively. The LoF/HiM diet dramatically increased mean fasting plasma tHcy to 22.8 μM. However, the fasting hyperhomocysteinemia caused by the LoF/HiM diet was ameliorated by feeding high amounts of folate in the HiF/HiM diet, lowering the mean fasting tHcy levels to 10.0 μM.
Approximate daily folate and methionine intakes per mouse on different diets.
Effects of the test diets on tHcy levels and other parameters.
Fasting tHcy levels may not representative of the average tHcy levels in mice with free access to food and water; thus, plasma tHcy levels were monitored every 6 h over one day (). Mice are nocturnal and thus we observed a large diurnal variation in non-fasted plasma tHcy levels, with higher values at night and lower values at day. Interestingly, the fed tHcy values were greater than the fasting values for both high methionine diets, whereas the fed tHcy values were less than the fasting tHcy levels for both low methionine diets (compare to ). By averaging the diurnal tHcy values at midnight, 6AM, noon, and 6 PM, the mean fed tHcy values were calculated for the four experimental diets, which we have called the effective tHcy levels and which are a better reflection of the chronic tHcy levels than the fasted tHcy values (). The effective tHcy levels were 4.5±0.9 μM (HiF/LoM), 5.7±1.2 μM (LoF/LoM), 37±10 μM (LoF/HiM), and 15±4.0 μM (HiF/HiM). The most striking difference between the fasted and effective tHcy values was observed in the HiF/HiM fed group, where the fasting values were roughly similar to the values for the LoF/LoM and HiF/LoM groups, while the effective values were about three-fold higher than the values for the LoF/LoM and HiF/LoM groups. The lack of significance between the effective tHcy of the HiF/HiM group and the LoF/LoM and HiF/LoM groups is only due to the small sample size and the conservative nature of the ANOVA Newman-Keuls multiple comparison posttest, as a similar test comparing data from just these 3 diets found the effective tHcy levels of mice on the HiF/HiM diet different from both the LoF/LoM and HiF/LoM groups (p<0.01), as do t-tests for HiF/HiM vs. both LoF/LoM and HiF/LoM (p<0.02).
Diurnal variation in non-fasted plasma Hcy levels
After obtaining the fasting blood samples, the mice on the four test diets were subjected to oral gavage with 320 mg/kg methionine and bled 2.5 hours later at the time of sacrifice. These post-gavage bloods were used to obtain plasma tHcy values after oral methionine loading. There were dramatic increases in plasma tHcy after oral methionine loading for mice on all four test diets, with values > 100 μM detected for both folate-deficient diets (). The fold increase in tHcy levels comparing the methionine loaded to fasted samples is shown in . Regardless of the folate content of the diet, the fold-increase in tHcy levels after methionine loading was almost twice as great (~8-fold) for mice fed the two low methionine diets than for mice fed the two high methionine diets (~4.5-fold). Therefore, chronic methionine intake resulted in an adaptive response which limited the induction-fold in tHcy levels after an acute methionine load.
Fold increase in plasma Hcy levels in mice after an acute oral methionine load compared to fasting Hcy levels
There was no significant influence of the different diets on spleen weight, total cholesterol (), hemoglobin, hematocrit, and complete or differential blood counts (latter values not shown) in these apoE-deficient mice. The LoF/HiM diet did, however, lead to a significant decrease in body weight compared to the other three diets (18% decrease vs. the LoF/LoM diet, ). Despite this effect on body weight, all mice were healthy looking and well-groomed.
Based upon these results, we designed two new diets in which either 2% methionine or glycine, as a control amino acid, was milled into a solid-food diet containing 2.0 mg/kg folate, which we designated as MET and GLY diets, respectively. These diets are considerably easier to use compared to methionine supplementation in the water, which requires weekly mixing. We fed these diets to groups of male wild type and apoE-deficient mice starting at 7 weeks of age for a period of 5 weeks. The mice on the GLY diets had mean fasted Hcy levels of ~ 5 μM. The MET diet led to significantly higher mean fasted Hcy levels of 90 and 68 μM for the in the wild type and apoE-deficient mice, respectively (). Body weights were reduced by > 20% in both strains of mice fed the MET diet (), similar to the body weight reduction observed in the mice supplemented with 4% methionine in their drinking water (). One possible explanation for this finding would be that the mice did not like the taste of the MET diet and ate less food. However, this was not the case as we found no significant difference in the amount of food consumed comparing mice fed the MET (3.58 + 0.57 g/d, N=5) and GLY diets (3.70 + 0.58 g/d, N=5), which was assessed over a 24 hr period in 12-week old mice housed in metabolic cages. As expected, the plasma total cholesterol levels were ~ 500 mg/dl in the apoE-deficient mice, and below 100 mg/dl in the wild type mice (, p<0.001 for the apoE-deficient mice vs. the wild type mice by ANOVA posttest). The wild type mice fed the MET diet had 37% lower total cholesterol levels than the mice on the GLY diet (p<0.001), and the same trend was observed in the apoE-deficient mice, albeit not statistically significant. Since most of the plasma cholesterol in wild type mice is carried on HDL, we assayed HDL-C levels in these samples as well. The MET diet led to a highly significant 33% reduction in HDL-C levels in wild type mice, and a 25% reduction in HDL-C levels in apoE-deficient mice, which was not statistically significant (). As previously observed [21
], HDL-C levels were reduced in apoE-deficient mice compared to wild type mice (p<0.01 for the apoE-deficient mice vs. the wild type mice by ANOVA posttest).
MET and GLY diet effects on wild type and apoE-deficient mice
Cholesterol absorption and HDL turnover studies were performed to follow up on the MET diet induced reduction in HDL-C in wild type mice. Cholesterol absorption was measured by a dual isotope gavage method using [14C]cholesterol and the poorly absorbed [3H]sitostanol in two independent experiments each with 5 wild type male mice on each diet. Pooling data from both experiments, the mice on the GLY and MET diets had 80.0 ± 5.1% and 82.8 ± 10.1% mean cholesterol absorptions (not significant). HDL turnover was assayed by i.v. injection of [3H]cholesteryl oleyl ether labeled HDL into five wild type male mice each on the GLY and MET diets. Since the ether linkage of cholesteryl oleyl ether cannot be cleaved, the label is not readily releasable from tissues, and thus the plasma disappearance of the label is reflective of HDL turnover (). The turnover of the labeled HDL in each mouse was fit by non-linear regression to a 2-phase exponential decay. The initial rapid phase had average fractional catabolic rates (FCR) of 9.0 and 6.1 pools per hour for the mice on the GLY and Met diets, respectively (NS). This initial rapid phase of decay is probably due to plasma HDL equilibration with the interstitial fluid, and thus we calculated the HDL-cholesterol FCR by use of the second and slower phase of HDL clearance. The slow phase HDL FCRs were 0.070 and 0.071 pools per hour for the mice on the GLY and Met diets, respectively (NS, ). As before, the MET diet led to a significant reduction in body weight (24% reduction, ). And, the mean HDL-C levels were reduced 24%, going from 94.4 mg/dl on the GLY diet to 71.8 mg/dl on the MET diet. Using the HDL-cholesterol values as well as the calculated plasma volumes and body weights, we calculated HDL-cholesterol production rates of 2.53 and 1.91 μg/hr/g body weight for mice on the GLY and Met diets, respectively (p=0.019, ). Thus, the 24% decrease in HDL-C levels was reflected by a significant 25% decrease in HDL production rate adjusted per gram of body weight, and a significant 45% reduction in HDL transport rate not adjusted by body weight ().
HDL turnover in wild type mice fed GLY or MET diet
Methionine effects on HDL turnover