N02007: Effect of diet on chain elongation and desaturation of alpha-linolenic acid in man

Study Duration: April 1998 to May 2001

Contractor: University of Southampton

The proportion of n-3 long chain polyunsaturated fatty acids (PUFA), principally eicosapaentaenoic acid (EPA) and docosahexaenoic acid (DHA) in cell membrane phospholipids, affects both the structure and function of cells in the body and is believed to play an important role in the maintenance of health and the development of disease. Since these fatty acids cannot be synthesised de novo in humans they must either be acquired pre-formed from the diet or produced within the body from carbon chain elongation and desaturation of their precursor, alpha-linolenic acid (ALNA). Despite increasing evidence for beneficial effects of n-3 polyunsaturated fatty acids (PUFA) within the diet, recent UK reports on diet and health have been unable to make firm recommendations for population changes in this class of fatty acids. There is a substantive lack of information on the type and amount of n-3 PUFA required to maintain optimal health. Whilst the cholesterol-lowering properties of PUFAs, particularly the n-6 PUFA, are well recognised, there is an increasing body of evidence which suggest that the balance of n-6 to n-3 PUFA may be important in determining the optimal function of specific cells and tissues, with implications for a number of chronic diseases of humans such as thrombosis, inflammatory disease states, hypertension and hypertriglyceridaemia. Although there appears to be general agreement that n-3 PUFA intakes should increase, there is insufficient evidence to determine the form in which n-3 PUFA should be supplied in the diet.

Given the relatively low EPA and DHA content of the UK diet, attention has focussed on the extent to which the conversion of the more abundant form, ALNA, can meet the metabolic demands of the body for EPA and DHA. The extent of conversion is not clear and the processes that regulate synthesis of EPA and DHA from ALNA are largely unknown in humans. Studies using isotopically labelled forms of the precursor ALNA to trace the conversion of ALNA to EPA, DPA and DHA have shown the pathway to be active in isolated cell preparations as well as in rats and non-human primates. Although there have been several tracer studies reporting conversion of the n-6 PUFA in adults, and n-3 PUFA in newborn infants, there are relatively few studies which have directly measured the conversion of ALNA to EPA and DHA in adults. Collectively, these studies have demonstrated that the capacity for ALNA conversion may be constrained (particularly in terms of the synthesis of DHA), that conversion may take place both in the cells of the gastrointestinal and the liver and that the conversion may be influenced by the fatty acid composition of the preceding diet or environment in which the cells are grown.

There were however, several key issues that remain unresolved: the extent to which these n-3 fatty acids are absorbed across the gastrointestinal tract (their availability); the relative contributions made by the cells of the gut and the liver in converting ALNA; how n-3 PUFA are handled within the body in terms of whether they are stored or used for energy; the extent to which conversion differs between individuals and the factors that may influence this conversion; whether altering the amount and balance of n-3 and n-6 PUFA in the test meal or background diet will alter the extent of conversion. Without this information on n-3 PUFA metabolism, our ability to formulate future nutrition policy relating to the consumption of n-3 and n-6 PUFA and for guiding food industry manufacturers on the optimal composition of spreads and oils for domestic use and in food is severely limited.

Rationale And Objectives

This project applied stable isotope tracer methodologies to determine whether the conversion of dietary ALNA to EPA and DHA in humans is a viable alternative to dietary sources of these longer chain n-3 PUFA and the extent to which this conversion is influenced by the n-3 and n-6 PUFA composition within a given test meal or the background diet. By incorporating [U-13C] ALNA acid into a test meal, we were able to determine i) the availability of ALNA across the gut (by measuring faecal losses), ii) the extent to which there was conversion of ALNA to EPA and DHA within the cells of the gut or the liver by measuring the appearance of labelled EPA and DHA derived from labelled ALNA in different lipid fractions which reflect the transport pools from the gut into the body (the chylomicrons) and from the liver to other tissues (the phospholipids of VLDL) and iii) the extent to which ALNA was either oxidised for energy or retained within the body by measuring the recovery of labelled CO₂ on the breath. Having characterised n-3 PUFA metabolism using this approach, we then sought to determine whether providing more precursor (high ALNA) stimulates conversion of ALNA to EPA and DHA and whether more product (high EPA and DHA) or linoleic acid inhibits ALNA conversion by altering the amounts of ALNA or EPA/DHA either within the test meal (Phase I) or background diet (Phase II).

The first phase of studies investigated the effect of meal n-3 fatty acid content upon ALNA conversion. Six men aged between 27 - 40 years of normal weight consumed a test meal on each of three occasions which was either modelled on the UK n-3 fatty acid intake (Control, n-3/n-6 ratio 0.15), had a 13-fold greater ALNA content (High ALNA) or an increased (19-fold) EPA+DHA content (High EPA/DHA). Having initially established that nearly all of the labelled ALNA in the test meal was absorbed (faecal losses <1%), a finding consistent with that which we and others have observed for n-6 PUFA, we agreed with the Agency that there was little further to gain by pursuing this line of investigation and facael collecions were discontinued.

The second phase of studies examined the effect of altering the dietary n-3 fatty acid content of the background diet on ALNA conversion. These studies were conducted in association with Prof Williams as part of the BBSRC-LINK project (AFQ111) after the first two cohorts of subjects had completed the dietary intervention. These results demonstrated that whilst increasing the amount of ALNA or EPA/DHA in the background altered the n-3 status of the subjects, there was no demonstrable difference between the moderate and high ALNA or EPA/DHA arms and that the principle physiological changes were evident after 8 weeks which then remained stable for the remaining 4 month intervention period. Therefore, with the approval of the Agency and those responsible for the LINK project, it was agreed that the study design should be altered - only three groups would be studied (Control, High ALNA and High EPA/DHA), that the studies should be made before and after the dietary intervention (as opposed to a single study after the intervention as this would increase the power of the study for the size of the study group) and that the dietary intervention would last for 8 weeks after a 4 week control diet. Fourteen men (26-69 years) maintained their habitual diet but substituted their usual margarine for a spread representative of the typical UK fatty acid intake for 4 weeks (Control) when the extent of conversion of ALNA to longer-chain fatty acids was determined over 48 hours using the control test meal described above. The subjects were then assigned to either continuing the Control intervention (n=5), increased ALNA consumption (10 grams per day) (n=4) or increased EPA+DHA consumption (1.5 grams per day, n=5) for 8 weeks before reassessing ALNA conversion.

Outcome and key results

The first set of key results relate to n-3 PUFA metabolism following a test meal containing amounts of ALNA, EPA and DHA which reflect that usually observed in the UK diet on the subject's habitual diet (Phase I). We were able to follow the appearance and disappearance of label within the different lipid pools in the circulation over the next 21 days. Labelled ALNA was mobilised as triglyceride within chylomicrons from the gut, whilst labelled ALNA incorporation into plasma phospholipids occurred later, probably by the liver. The time scale of appearance of labelled ALNA into EPA and docosapentaenoic acid (DPA - the final step before conversion to DHA) in plasma phospholipids over the next 14 days suggested that the liver was the primary site of conversion, although there was some indication that some limited conversion had taken place in the enterocyte. There was however, no detectable appearance of labelled DHA in any of the plasma fractions or cell membranes at any time point up to 21 days. This suggests that whilst some inter-conversion occurs, there is a constraint of DHA formation down-stream of DPA synthesis. Finally, about one third of the labelled ALNA was oxidised over the study period which is comparable to that seen for other fatty acids studied within our group. This would suggest that dietary n-3 PUFA are neither preferentially oxidised or stored in any way differently to that seen for other fatty acids.

The second set of key results relate to the effect of altering the n-3 PUFA composition of the test meal (Phase I). There was no difference in the conversion or oxidation of the labelled ALNA when test meals containing either more ALNA or more EPA/DHA were consumed. Both these test meals had more total n-3 PUFA than in the control test meal and this resulted in a down-regulation of EPA and DPA synthesis. However, once again, no conversion to DHA was detected following any of the test meals.

The final set of key results relates to the effect of altering the n-3 PUFA composition of the background diet (Phase II). Once again, there was evidence of ALNA conversion to EPA and DPA, but unlike the studies in Phase I, we were now able to detect some conversion to DHA. However, the apparent fractional conversions were low: ALNA 95.9%, EPA 2.8%, DPA 1.2 % and 0.04% DHA. There was no significant change in plasma PC composition or ALNA inter-conversion between trials in the control group as one would expect given that the diet was held constant. Increasing ALNA intake increased the amount of EPA in the plasma PC, but not DHA, whilst the conversion of ALNA to EPA, DPA or DHA was unaltered. In contrast, increasing EPA/DHA intake increased the amount of EPA and DHA in plasma PC, but this was associated with decreased EPA and DPA synthesis. The oxidation of ALNA was not affected by the n-3 PUFA content of the background diet.

Conclusions

Taken together, it is possible to draw the following conclusions which we believe make an important contribution to our understanding of n-3 PUFA metabolism in man. Firstly, although ALNA may be readily converted to EPA and DPA, DHA synthesis from dietary ALNA appears to be extremely modest (<1% over 21 days) and seems to be independently regulated from the synthesis of EPA and DPA. Secondly, although the total n-3 PUFA content of the test meal appears to acutely reduce ALNA inter-conversion, changing the relative proportions of ALNA to EPA/DHA within the test meal did not alter the extent of conversion. Thirdly, it does not appear to be possible to force the conversion of ALNA to EPA by increasing the supply of the dietary precursor (ALNA) in the background diet, although it may be decreased by lowering the demand for EPA and DHA following a period of raised EPA and DHA in the background diet. Finally, it appears that the extent to which ALNA is oxidised after consumption appears to be independent of the amount or relative proportions of n-3 PUFA in the test meal or background diet.

The ability of adult men to convert ALNA to DHA may be determined more by their intrinsic metabolic competence to make DHA or metabolic requirement for DHA than be governed by dietary supply of ALNA. In other words, if the primary objective is to increase the amount of DHA within body pools, then increasing the ALNA content of the diet will not stimulate DHA synthesis although the body pools of EPA and DPA may increase. These observations both support and explain the principle findings of the LINK study in that increasing ALNA in the diet increases tissue and circulating EPA but not DHA levels and at feasible dietary levels ALNA does not appear to mimic the effects of the long chain omega-3 PUFA on a blood lipids, haemostatic factors or immune and inflammatory function. If future research establishes DHA as an important component of the cardioprotective actions of fish oil fats, these findings have implications for potential dietary sources of DHA in adult humans for optimal heart health. Thus the only effective means of increasing DHA status would be to consume preformed DHA either by increased seafood consumption, fortification of food with DHA or the use of DHA dietary supplements. If research establishes EPA as the cardioprotective component of fish oils, then this may also explain the epidemiological evidence for cardioprotective effects of high dietary ALNA intakes.

Sanderson, P., Finnegan, Y. E., Williams, C. M., Calder, P. C., Burdge, G. C., Wootton, S. A., Griffin, B. A., Millward, D. J., Pegge, N. C. and Bemelmans W. J. E. (2002) UK Food Standards Agency alpha-linolenic acid workshop report. Br J Nutr 88, 573-579

Burdge GC, Jones AE & Wootton (2002) 'Eicosapentaenoic and docosapentaenoic acids are the principal products of alpha-linolenic acid metabolism in young men' Br J Nutr 88, 355-363

Burdge GC & Wootton SA (2002) 'Conversion of alpha-linolenic acid to eicosapentaenoic, docosapentaenoic and docosahexaenoic acids in young women' Br J Nutr 88, 411-420

Burdge GC, Finnegan YE, Minihane AM, Williams CM & Wootton SA (2003) 'Effect of n-3 fatty acid intake upon plasma lipid fatty acid composition, conversion of [13C] alpha-linolenic acid to longer-chain fatty acids and partitioning towards beta-oxidation in older men' Br J Nutr 90, 311-321.

Burdge GC, Wootton SA (2003) Conversion of alpha-linolenic acid to palmitic, palmitoleic, stearic and oleic acids in men and women. Prostaglandins Leukot Essent Fatty Acids. 69, 283-290.

Contact: Peter Sanderson
Tel: 020 7276 8920
Email: peter.sanderson@foodstandards.gsi.gov.uk