N05016: Is folic acid considerably more effective than folates in raising folate status?

Study Duration: July 1998 to February 2002

Contractor: Institute of Food Research

Vitamin folate is a generic term for naturally originating compounds, especially found in green leafy vegetables, which have nutritional properties and chemical structures similar to those of folic acid, the synthetic form of the vitamin that is used extensively for food fortification purposes, and supplements. The two major naturally occurring folate forms in food are 5-methyl-tetrahydro-folates and 5-formyl-tetrahydro-folates.

Folates are crucial for the normal functioning of cells and are necessary for the interconversion of amino acids, particularly the re-methylation of homocysteine back to methionine, and the construction and repair of DNA (the ‘blueprint’ control for all cellular actions). The periconceptual supplementation of women with folic acid has been shown to significantly reduce the incidence and reoccurrence of neural tube defects such as Spina Bifida. Marginal vitamin folate deficiency is linked to risk for some cancers and is also associated with elevated plasma homocysteine, which is a risk marker (if not a risk factor) for vascular disease (particularly heart disease and stroke), Dementia and Alzheimer's.

Rational and Objectives

At the present time, our understanding of the influence of food composition and physiological variables on folate absorption is limited and it is not possible to predict folate bioavailability for a given food, diet, or even pure folate isolate. Isotopic methods that utilise 'stable-isotope'-labelled folates (rather than 'radio-isotope'-labelled folates), which can ethically be used for human studies, may permit the examination of folate absorption and metabolism with a high degree of specificity and sensitivity and, in conjunction with kinetic modelling techniques, estimate for absolute absorption (rather than comparative absorption relative to a folic acid 'reference' dose) and subsequent partitioning into other body pools. Stable-isotope studies directed at quantifying the bioavailability of natural folates in food have not been possible due to lack of suitably labelled food material and appropriate analytical procedures to quantify metabolites in biological samples.

The scientific objectives relate to answering the following questions:

Q1. Is there a difference in absorption and metabolism of folic acid versus a naturally-occurring folate such as 5-formyl-tetrahydrofolic acid?

Q2. Is folic acid more effectively absorbed and utilised than naturally-occurring folates (mainly 5-methyl-tetrahydro-folates and 5-formyl-tetrahydro-folates) in vegetables?

• ‘Stable-isotope’ (carbon-13; ¹³C-) labelled 5-formyl-tetrahydrofolic acid was prepared and its purity analysed.
• ‘Stable-isotope’ (initially carbon-13, ¹³C-, but later nitrogen 15,¹⁵N-) intrinsically labelled spinach was grown hydroponically and the degree of labelling of its natural folate content analysed.
• A combined ‘liquid chromatography-mass spectrometry' (LC-MS) analytical procedure for the determination of folic acid, 5-formyl-tetrahydrofolic acid, and 5-methyl-tetrahydrofolic acid (which would be the expected labelled-folate eventually appearing in plasma; original test doses having undergone metabolic transformation), and stable-isotope labels (¹³C, ¹⁵N & deuterated hydrogen internal standard, ²H) in the plasma from blood samples (which would be drawn at intervals after an oral test dose) was further developed and validated.
• Ethical permission was obtained to give test doses of stable-isotope-labelled folic acid, 5-formyl-tetrahydrofolic acid & natural spinach folates (a mixture of 5-methyl-tetrahydro-folates and 5-formyl-tetrahydro-folates), and to collect 12 blood samples at intervals over an 8 hour period following ingestion.
• A 1-compartment (comprising a balance between liver and plasma pools) mathematical model of folate absorption was developed and applied to the labelled plasma folate response over the 8-hour period following ingestion.

(1). There is a marked difference in absorption and metabolism of folic acid versus a 'natural' folate, such as 5-formyl-tetrahydrofolic acid. The 'rate of appearance' and 'time for maximum appearance' of labelled 5-methyl-tetrahydrofolic acid in the plasma from an initial liquid oral dose of labelled-folic acid is slower, and at least 1 hour later, respectively, than for the 'rate of appearance' and 'time for maximum appearance' of labelled 5-methyl-tetrahydrofolic acid from an initial liquid oral dose of labelled-5-formyl-tetrahydrofolic acid.

(2). Unexpectedly, the total plasma folate response to test folates is significantly greater than the plasma labelled-folate responses; suggesting a large tissue displacement of unlabelled folate not directly originating from the labelled test-doses. Furthermore, also unexpectedly, an initial straight forward ‘relative’ comparison of the plasma labelled-folate ‘positive area-under-the-curve’ (AUC; the perturbation above baseline concentration recorded at time zero) would indicate that folic acid is absorbed to a significantly lower extent than the ‘natural’ 5-formyl-tetrahydrofolic acid or spinach folates (mainly 5-methyl-tetrahydro-folates and 5-formyl-tetrahydro-folates).

(3). With mathematical modelling, but without introducing any adjustment for a liver ‘first-pass’ effect (whereby a defined fraction of newly absorbed folate, which is initially delivered from the gut into the body-proper via the hepatic portal vein, is deemed to have been taken out by the liver leaving only the residual folate fraction to eventually pass into the systemic blood system), the ‘absolute’ fractional absorption of test dose into the plasma is ca. 0.25 for folic acid but 0.38 and 0.44 for 5-formyl-tetrahydrofolic acid and natural spinach folates respectively. By introducing a 70% liver ‘first-pass’ or ‘clearance’ effect, the fractional ‘absolute’ absorption of folic acid is recalculated as 0.89 – a figure in keeping with literature values from radioisotope work in rats and in one human. However, applying this 70% liver ‘first-pass’ factor to either the 5-formyl-tetrahydrofolic acid or spinach test doses results in impossible ‘absolute’ fractional absorptions of approximately 1.7. Only by the introduction of a much lower liver ‘first-pass’ effect (50%) is the ‘absolute’ fractional absorption of 5-formyl-tetrahydrofolic acid or spinach folates reduced down to 0.84 and 0.89, respectively.

Differences in absorption and metabolism of folic acid versus a ‘natural’ folate, such as 5-formyl-tetrahydrofolic acid, indicate not only that previously acceptable methods for estimating unlabelled folate absorption ‘relative’ to an unlabelled folic acid ‘reference’ dose could be flawed but also that unadjusted ‘absolute’ absorptions from mathematically-modelled plasma labelled-folate responses may also give flawed results.

When mathematically modelling an estimate of ‘absolute' absorption, the apparent need to use a much higher liver 'first-pass' effect (70%) for folic acid than the maximum term used for either 5-formyl-tetrahydrofolic acid or spinach folates (50%) indicates that a significant amount of the folic acid dose absorbed by mucosal cells of the small intestine must have been passed into the hepatic portal vein unchanged as 'folic acid'. This suggests that the main site for reduction and methylation of folic acid is the liver, and not the mucosal absorptive cells. Furthermore, the calculated 'absolute' fractional absorptions of the 5-formyl-tetrahydrofolic acid (0.84) or spinach folates (0.89) will be notably reduced if there is actually a significantly lower liver 'first-pass’ effect (than 50%) for these natural folates than that finally used in the mathematical modelling. So, because of problems generated by differing metabolism, we still do not know if folic acid is better absorbed that 'natural' folates.

It is concluded that a large tissue displacement of folate, not of test-dose origin, may negate use of either measurement of the rate of increase, or the maximum increase of plasma 5-methyl-tetrahydrofolic acid, or the plasma 5-methyl-tetrahydrofolic acid 'positive area-under-the-curve' (AUC; the perturbation above baseline concentration recorded at time zero), when attempting to estimate the comparative response from oral doses of unlabelled folates with an unlabelled folic acid 'reference’ dose, as has hitherto historically been the case. Furthermore, comparison of labelled folate plasma AUC from a labeled folate test dose ‘relative’ to that from a labeled folic acid ‘reference’ would also be flawed.

These results have implications for the future funding of folate absorption work by the FSA. It is suggested that future priority be given to elucidating the fundamental underlying mechanisms of folate absorption in human volunteers from pure folate isolates. Such an understanding is essential before any assessment of folic acid/folate absorption from food sources (with their additional problems of variable digestion, folate release and matrix effects) can even be attempted.

Sanderson P, McNulty H, Mastroiacovo P, McDowell IF, Melse-Boonstra A, Finglas PM, Gregory JF 3rd (2003) Folate bioavailability: UK Food Standards Agency workshop report. Br J Nutr. 90, 473-479.

McKillop DJ, Pentieva KD, Scott JM, Strain JJ, McCreedy R, Alexander J, Patterson K, Hughes J, McNulty H. (2003) Protocol for the production of concentrated extracts of food folate for use in human bioavailability studies. J Agric Food Chem. 51, 4382-4388.

McKillop DJ, Pentieva K, Daly D, McPartlin JM, Hughes J, Strain JJ, Scott JM, McNulty H. (2002) The effect of different cooking methods on folate retention in various foods that are amongst the major contributors to folate intake in the UK diet. Br J Nutr. 88, 681-688.

Contact: Dr Alison Tedstone
Tel: 020 7276 8929
Email: alison.tedstone@foodstandards.gsi.gov.uk