Long-chain n-3 PUFA: nomenclature, natural origin and role in the food chain

The important LC n-3 PUFA are 20:5n-3 (EPA), 22:5n-3 (docosapentaenoic acid) and 22:6n-3 (DHA). EPA and DHA are naturally found in substantial amounts in marine-derived foods such as oil-rich fish and to a lesser extent in white fish and shellfish. This natural occurrence is a direct result of phytoplankton, a diverse range of microscopic organisms found in water that are the abundant natural producers of LC n-3 PUFA at the base of the food chain. Algae (a type of phytoplankton) are important constituents of a range of ecosystems and are particularly known for their role as primary producers of LC n-3 PUFA⁽Reference Guschina and Harwood³⁾. In the marine food chain microalgae (a subgroup of algae) are consumed by zooplankton, which in turn are consumed by planktiverous fish. Thus, EPA- and DHA-rich lipids originally synthesised by microalgae are transferred into the lipid stores of planktiverous fish. Larger pisciverous fish feed on planktiverous fish, therefore obtaining and depositing LC n-3 PUFA in their lipid stores, and both planktiverous and pisciverous fish are consumed by the human population, at which point these fatty acids enter the human food chain. The concentrations of LC n-3 PUFA in edible fish differ subtly according to species, geographical location and season of catch and are also different in wild and farmed species, with the lipid profiles of farmed species resembling those of the fish oil or fishmeal on which they are fed⁽Reference Bandara, Batista and Nunes^4, Reference Aidos, van der Padt and Luten⁵⁾.

Another important and dietary essential n-3 PUFA is α-linolenic acid (ALA; 18:3n-3). This fatty acid is not marine derived but is found in plant oils, with linseed oil being one of the richest sources (approximately 50–60% total fatty acids)⁽Reference Kralik, Škrtić and Suchý⁶⁾. The role of ALA is reviewed elsewhere⁽Reference Ruxton, Calder and Reed^7–Reference Arterburn, Hall and Oken⁹⁾, but it should be noted that the health effects of ALA (particularly its cardiovascular effects) are not the same as those associated with marine-derived PUFA. Investigations have also shown that the conversion efficiency of ALA to EPA and DHA in human subjects not only proceeds with low efficiency but is influenced by factors such as gender, age and genotype. Efficiency of conversion, particularly along the n-3 pathway, has been shown to be variable at 5–10% ALA being converted to EPA and 2–5% ALA converted to DHA⁽Reference Emken, Adlof and Gulley^10–Reference Burdge¹⁵⁾. The most likely limiting factor in the conversion is the Δ6 desaturase step, as expression and activity of this enzyme is influenced by numerous factors including competition by other fatty acids⁽Reference Cleland, Gibson and Pedler¹⁶⁾.

In conclusion, preformed EPA and DHA are therefore needed in the human diet to satisfy requirements for optimum health. Dietary habits have changed over the years, resulting in marked increases in the consumption of n-6 PUFA. It has been suggested that n-6:n-3 should ideally be approximately 1:1, but currently it is more likely to be 15:1⁽Reference Simpoulous¹⁷⁾. It has also been noted that linoleic acid intake has increased substantially in Western societies⁽Reference Ailhaud, Guesnet and Cunnane¹⁸⁾, which is of concern because of the shared metabolic pathways of both the n-3 and n-6 PUFA and the resulting competition for metabolite production⁽Reference Brenner and Peluffo¹⁹⁾, which has had an impact on the efficiency of conversion of ALA to EPA. It has been concluded that the concept of n-6:n-3 is not as useful as considering the impact of actual low consumption of LC n-3 PUFA⁽Reference Stark, Crawford and Reifen²⁰⁾. In either case, the concurrent increase in n-6 consumption and decrease in n-3 consumption deserves attention, including consideration of the options for increased supply of LC n-3 PUFA.

Long-chain n-3 PUFA and human health

Cardiovascular health

Metabolic syndrome is a complex web of conditions that are the consequence of a number of ‘dysregulated’ metabolic pathways⁽Reference Roche, Philips and Gibney²¹⁾. It gives rise to markedly increased risk of serious morbidity or mortality from a cardiovascular-related illness⁽Reference Shaw, Hall and Williams^22, Reference Nugent²³⁾, which according to the British Heart Foundation is the cause of four in ten deaths in the UK each year⁽²⁴⁾. Fish, fish oil and thus LC n-3 PUFA have been indicated in the reduction of risk factors associated with metabolic disease and mortality since the work of Hugh Sinclair in the 1950s⁽Reference Sinclair²⁵⁾ and in later research showing a 10-fold lower incidence of CVD for Greenland Inuits than for age- and gender-matched Danes despite a higher total fat intake⁽Reference Dyerberg²⁶⁾. These findings were attributed to regular consumption of seal and whale blubber in Greenland, which had previously been observed to be a concentrated source of LC n-3 PUFA. Other epidemiological studies of populations such as those of Japan and Alaska whose diets are rich in fish and LC n-3 PUFA have indicated a lower prevalence of CHD⁽Reference Lands, Hamazaki and Yamazaki^27, Reference Newman, Middaugh and Guzman²⁸⁾. A selection of studies investigating the link between fish and LC n-3 PUFA consumption and CVD and all-cause mortality are shown in Table 1. In addition, one of the largest randomised controlled trials that has been conducted in this area of research is the GISSI Prevenzione Study⁽²⁹⁾, a placebo-controlled study of >11 000 Italian patients post myocardial infarction who were given 885 mg EPA+DHA/d or 300 mg vitamin E/d or both. The greatest reduction in risk was found for sudden death (44%) after the 3·5-year follow up, along with 30, 35 and 32% reduction in risk of cardiovascular, cardiac and coronary deaths respectively; the reduction was found in both the EPA+DHA and EPA+DHA+vitamin E groups, although the magnitude of risk reduction was not found to differ significantly between the two groups, indicating no additional benefit of vitamin E. These outcomes are impressive and indicative of a marked benefit of a modest daily intake of EPA+DHA.

Table 1. Examples of studies reporting an inverse association between fish and long-chain (LC) n-3 PUFA consumption and CVD and mortality

M, males; F, females; MI, myocardial infarction.

The large body of evidence to support the inverse relationship between CVD and consumption of fish has been extensively reviewed⁽Reference Wang, Harris and Chung^30–Reference von Schacky³³⁾. A meta-analysis of eleven key studies has concluded that for a 20 g/d increase in fish consumption there is an associated 7% reduction in CHD mortality risk and that consumption of fish once per week confers reduced CHD mortality rates⁽Reference He, Song and Daviglus³¹⁾. A large meta-analysis encompassing forty-six randomised controlled trials, prospective cohort studies and case–control studies has concluded that data from both secondary and primary prevention studies support the association between increased EPA and DHA consumption and reduced risk of all-cause mortality, cardiac death, sudden death and stroke but this outcome is not apparent for ALA⁽Reference Wang, Harris and Chung³⁰⁾. The analysis emphasises that the strongest evidence available at the time was for secondary prevention rather than for primary prevention, but this finding may have been a result of fewer comprehensive studies in this area⁽Reference Wang, Harris and Chung³⁰⁾. The variety of mechanisms that potentially contribute to the reduction in CVD have also been well studied and are summarised in Table 2; they include decreased arrhythmias, decreased TAG, increased plaque stability and modest reduction in blood pressure.

Table 2. Effects of long-chain (LC) n-3 PUFA on cardiovascular risk factors (adapted from Calder⁽Reference Calder¹²⁴⁾ and Minihane & Lovegrove⁽Reference Minihane, Lovegrove, Williams and Buttriss¹²⁵⁾)

↓ , Decreased; ↑, Increased.

Recommended intakes of long-chain n-3 PUFA

In 1994 the Department of Health recommended that the population average consumption of LC n-3 PUFA should be increased from the estimated intake of approximately 100 mg/d to 200 mg/d (1·5 g/week)⁽⁵⁰⁾. In 2004 the report Advice on Fish Consumption: Benefits and Risks was published jointly by the Scientific Advisory Committee on Nutrition and the Committee on Toxicity⁽⁵¹⁾, having been commissioned specifically to examine the nutritional and toxicological evidence relating to fish consumption and its benefits and potential risks. The report acknowledges the Department of Health's earlier recommendation⁽⁵⁰⁾, but notes that the UK population is at high risk of CHD and therefore recommends a review of intakes. It also considers new evidence for health benefits that could substantiate an increase in recommended intakes of fish, whilst also considering the potential adverse affects of consumption because of the presence of toxins. Based on available evidence for CVD, but not neurological effects, the report recommends consumption of 450 mg LC n-3 PUFA/d, which is still easily achievable by consuming two portions of fish per week, of which one should be oil-rich fish.

The report also highlights the low consumption of oil-rich fish in the UK (approximately 27% of adults are consumers) but also the lack of sufficient data to formulate recommendations for the sector of the population who consume little or no oil-rich fish⁽⁵¹⁾. It also acknowledges that there is evidence to demonstrate the need to consume ≥1·5 g LC n-3 PUFA/d in order to have any noticeable effects on many CVD risk factors such as reducing TAG and inflammatory response, but that making a generalised recommendation of >450 mg/d is not ‘practical’. The lack of recommendations for intakes >450 mg/d, despite convincing evidence to the contrary, has in recent years been subject to much debate⁽Reference Winkler⁵²⁾. However, it will become clear in the following discussions that average intakes are <450 mg/d at present and thus more emphasis should be placed on increasing intakes in general rather than attempting to provide exact guidelines for intakes in different population sectors.

Recommended v. current intakes: effects of age and gender

Estimates of intakes of specific fatty acids can be derived from food consumption data, chemical analysis of diets or from blood or tissue lipid samples used as biomarkers of intake and good-quality food-consumption data⁽Reference Sanders⁵³⁾. Despite the availability of good-quality data from the UK National Diet and Nutrition Survey⁽Reference Henderson, Gregory, Irving and Swan⁵⁴⁾ there have been no detailed reports of LC n-3 PUFA intake across the UK population showing the relative contributions of different foods to total LC n-3 PUFA intake until a recent report of work that couples food intakes with composition data for fish, meat, milk and milk products⁽Reference Givens and Gibbs⁵⁵⁾. This report estimates mean current intakes at 244 mg/d for UK adults, but highlights the possible effect of only 27% of the population being consumers of oil-rich fish and suggests that intakes amongst non-consumers could be as low as 100 mg/d⁽Reference Givens and Gibbs⁵⁵⁾. It should be noted that whilst animal-derived foods currently supply only minor amounts of LC n-3 PUFA to UK adults (mg EPA+DHA per adult per d; oil-rich fish 131, chicken meat 26, beef 4, lamb 2), these foods are the only source for the majority of the population⁽Reference Givens and Gibbs⁵⁵⁾.

In terms of global intakes of EPA+DHA, the USA is reported to have intakes of 100–200 mg/d⁽Reference Kris-Etherton, Taylor and Yu-Poth⁵⁶⁾ whilst Australians (all ages) are estimated to consume approximately 175 mg/d⁽Reference Meyer, Mann and Lewis⁵⁷⁾. Furthermore, there is the potential for considerable variation between genders and age-groups⁽Reference Givens and Gibbs⁵⁵⁾ but because the UK National Diet and Nutrition Survey reports oil-rich fish intakes that include canned tuna, which should not be categorised as an oily fish (although this factor has been corrected for by the Scientific Advisory Committee on Nutrition and the Committee on Toxicity in their calculations of mean fish intake for all ages and genders⁽⁵¹⁾), it has not been possible to study this variation in detail. Consequently, further studies have been carried out to correct for canned tuna and to calculate current intakes of EPA+DHA by gender and across age-groups⁽Reference Gibbs, Givens and Rymer⁵⁸⁾. The results of these investigations show clear age-related trends emerging in both males and females. EPA+DHA intakes show an increase with increasing age, with intakes in young adults (19–24 years) the lowest at 97 mg/d and those of 50–64 year olds the highest at 330 mg/d⁽Reference Gibbs, Givens and Rymer⁵⁸⁾. It is not clear, however, whether these trends will continue in future generations as these data are only cross-sectional. A summary of studies that have reported intakes by age-group is shown in Table 3, which also identifies age-related trends in other countries. Estimates of LC n-3 PUFA intakes for the low-income subsection of the UK population have also been made using data from the Low Income Diet and Nutrition Survey⁽Reference Nelson, Erens and Bates⁵⁹⁾ and food fatty acid concentrations⁽Reference Gibbs, Givens and Rymer⁵⁸⁾. Overall, EPA+DHA intakes are approximately 50 mg/d lower than those for the national population (RA Gibbs, C Rymer and DI Givens, unpublished results) but the percentage of consumers of oil-rich fish in this group (15) is lower than that in the national population, suggesting that intakes in non-consumers would be lower than the mean.

Table 3. Studies reporting EPA and DHA intakes in different age-groups and genders

* Median values.

It is worth noting that the countries that have been discussed so far are Westernised and that intakes in Japan, for example, a nation for which fish is a substantial part of the diet, are higher, approximately 1·2 g LC n-3 PUFA per capita per d⁽Reference Sugano and Hirahara⁶⁰⁾. In conclusion, current intakes of EPA+DHA in UK adults in relation to recommendations are substantially <450 mg/d for males and females across all age-groups, with greater disparity in the youngest age-groups. Furthermore, widespread low consumption of oil-rich fish (67% non-consumers according to the National Diet and Nutrition Survey⁽Reference Henderson, Gregory, Irving and Swan⁵⁴⁾ and 85% non-consumers according to the Low Income Diet and Nutrition Survey⁽Reference Nelson, Erens and Bates⁵⁹⁾) coupled with evidence for poor in vivo conversion of ALA to EPA and DHA⁽Reference Burdge^15, Reference Burge and Calder⁶¹⁾ would suggest that LC n-3 PUFA status in at least two-thirds of the population is very low indeed, with the main contribution currently being from animal sources. Thus, strategies to increase intakes across the whole population are urgently needed, although strategies that specifically target different population subgroups may be appropriate.

The potential of animal-derived foods to increase intakes

Animal-derived foods have been highlighted as foods that can potentially be altered to accommodate optimum nutrient requirements for health⁽Reference Kris-Etherton, Taylor and Yu-Poth^56, Reference Givens and Shingfield⁶²⁾, including manipulation of their LC n-3 PUFA content. Animal nutrition plays a key role in determining animal fat composition and therefore is the central tool for changes in fat composition. The lipid fraction of animal-derived foods has particular potential for manipulation through dietary adjustment⁽Reference Leskanich and Noble⁶³⁾, although responses to such manipulations are species dependent. Meats and dairy foods currently contribute substantially to the human diet⁽Reference Henderson, Gregory, Irving and Swan^54, Reference Givens and Gibbs⁵⁵⁾. As animal fat is a major source of fat in the diet, exploiting the wide consumption of animal fats is potentially an important route to increased intakes of particular fatty acids⁽Reference Givens and Shingfield^62, Reference Leskanich and Noble⁶³⁾. The overall success of strategies to increase the supply of LC n-3 PUFA in the food chain through enriched foods is likely to depend on the following key factors⁽Reference Givens and Gibbs^55, Reference Givens and Shingfield⁶²–Reference De Henauw, Van Camp and Sturtewagen⁶⁶⁾:

• the extent to which the fatty acid composition of the food is amenable to manipulation;

• the wideness of consumption of the product across the general population;

• the quantities in which the product is consumed;

• taste and other quality characteristics;

• consumer acceptance and willingness to buy.

Poultry meat as a prototype food to increase intakes

Poultry meat has been identified as meeting the criteria outlined, chiefly because of the amenability of the lipid fraction to accommodate PUFA⁽Reference Leskanich and Noble^63, Reference Rymer and Givens^65, Reference Marion and Woodruff^67–Reference Hulan, Ackman and Ratnayake⁷⁰⁾. Poultry meat is a popular food, consumed by approximately 80% of UK adults⁽Reference Henderson, Gregory, Irving and Swan⁵⁴⁾ and is also highly versatile. It is available in many forms at retail and the range is wide enough to meet the preferences of a wide range of consumers. According to TNS™ (London, UK; personal communication) 34% of the poultry meat purchased in retail outlets during 2006–8 was as meat portions, whilst 11% was purchased as whole birds. A further 14% of poultry meat was found in ready-prepared meals containing meat and carbohydrate and 14% as processed meat portions. Added together these sources account for approximately 75% of poultry-meat purchases, with the remaining 25% comprising hot rotisserie birds, the meat in poultry pies, cold and ready-to-eat poultry etc. There are no data available for quantities of poultry meat consumed as fast food, in restaurants and other kinds of catering or in takeaway sandwiches etc., but there is also likely to be a substantial amount consumed in this way.

The fatty acid composition of breast, leg and skin of broilers (chickens reared for meat) has been shown to reflect that of the diet⁽Reference Marion and Woodruff⁶⁷⁾. Targeted enrichment of tissues was first achieved in poultry in the late 1960s when it was demonstrated that feeding fish oil to turkeys increases the EPA and DHA content of depot fat and muscle lipids⁽Reference Neudoerffer and Lea⁶⁹⁾. Work began in the 1980s to study the effects of feeding ALA to poultry, but it was soon noted that the response of skinless breast, in terms of EPA and DHA deposition, is poor although the response in meat with skin is better. This finding is understandable because skin has a high lipid content and is also composed of a larger proportion of TAG⁽Reference Leskanich and Noble⁶³⁾, in which ALA accumulates. It was found that ALA itself is readily accumulated in tissues. A review of work on feeding ALA to poultry⁽Reference Rymer and Givens⁶⁵⁾ has concluded that feeding ALA in order to increase EPA and DHA deposition in edible tissues (breast and white meat particularly) is not at all efficient and therefore a preformed source of EPA and DHA is needed if poultry are to accumulate meaningful amounts in their edible tissues. No differences between gender or breeds of chicken have been found in terms of their response to dietary n-3 PUFA⁽Reference Leskanich and Noble⁶³⁾.

An increase in the EPA and DHA content of tissues as a result of feeding fish oil or algae has been noted by several authors⁽Reference Leskanich and Noble^63, Reference Leskanich and Noble^68, Reference Leskanich and Noble^69, Reference Leskanich and Noble^71–Reference Gonzalez-Esquerra and Leeson⁷⁵⁾. Over time, the emphasis has shifted to investigating the appropriate doses of dietary LC n-3 PUFA and the efficiency of responses that will achieve maximum potential for poultry meat to provide LC n-3 PUFA to the human diet. Studies have also investigated the potential drawbacks of such practices in terms of meat quality.

Challenges with enrichment of poultry meat

The effect of cooking on enriched poultry meat composition

Since all chicken meat is consumed in the cooked state it is crucial that the effects of cooking on fatty acid composition are also known. Fatty acid-enrichment studies have predominantly reported the fatty acid composition of raw meat and very few studies have directly compared uncooked and cooked conventional and enriched broiler meat. The fatty acid composition of grilled samples of LC n-3 PUFA-enriched chicken has been compared with that of raw meat samples derived from broilers fed a control diet or a diet containing 50 or 150 g tuna fishmeal with an antioxidant complex/kg feed for 21 d⁽Reference Howe, Downing and Greyner¹⁰⁷⁾. Concentrations of total LC n-3 PUFA in grilled breast and thigh were found to be increased compared with those of raw breast and thigh and it was suggested that moisture loss during grilling causes a concentrating effect⁽Reference Howe, Downing and Greyner⁷⁴⁾. In a study in which fatty acid concentrations of cooked thigh meat with skin were also measured it was found that cooking for 30 min at 80°C in a water bath results in highly significant reductions (P⩽0·001) of 6·2%, 6·8% and 5·7% in SFA, MUFA and PUFA respectively, but SFA:MUFA:PUFA is unchanged and there are no significant differences between the levels of EPA and DHA in raw and cooked thigh meat⁽Reference Cortinas, Villaverde and Galobart¹⁰⁸⁾. The fish oil concentrations of the diets used in the study were relatively low (0, 5, 10 or 20 g/kg feed) compared with those used in recent enrichment studies (⩽40 or 60 g/kg feed). Thus, a greater extent of enrichment and thus polyunsaturation of the lipids in edible tissues may have an impact on losses. The findings of these two studies are insufficient to draw firm conclusions in relation to the fate of EPA+DHA as a consequence of cooking.

To address this issue a study was designed to examine the effect of cooking skinless breasts taken from a group of broilers reared on either a control diet (main fat source soya oil at 50 g/kg feed) or a fish oil-supplemented (50 g fish oil/kg feed) diet and slaughtered at approximately 42 d of age. Each breast was cut in half and half was roasted for 20 min in a fan-assisted oven at 180°C. Both the cooked and raw breast pieces were analysed for fat content and fatty acid composition whilst maintaining the identity of each raw and cooked breast tissue pair. The concentrations of major fatty acids in the breast tissue were not found to be significantly affected, indicating that no significant loss of any fatty acid occurs as a result of cooking (by roasting; RA Gibbs, C Rymer and DI Givens, unpublished results). Furthermore, to establish whether there are any differences between different methods of cooking skinless breast samples were taken from forty-eight birds that had received either control or fish oil-supplemented diets at (/kg feed) 40 g fish oil+100 mg vitamin E (as α-tocopheryl acetate), 60 g fish oil+100 mg vitamin E, 60 g fish oil+150 mg vitamin E, 80 g fish+200 mg vitamin E. At 42 d broilers were slaughtered and breasts were taken and cut longitudinally into three pieces. Breast pieces were taken and cooked by one of four methods; boiling, pan frying in olive oil, grilling and roasting. Procedures were standardised and closely resembled those used in the home. Once the cooked chicken pieces had cooled each piece was analysed for fat content and fatty acid composition. Significant differences (P<0·0001) in fatty acid concentrations of chicken breast cooked by different methods were found, with results suggesting that boiling chicken meat (as a representation of casseroling in water-based liquid) is the most favourable method (RA Gibbs, C Rymer and DI Givens, unpublished results). A 135 g breast portion derived from broilers fed 40 g fish oil/kg feed, when boiled, has been shown to contain approximately 450 mg EPA+docosapentaenoic acid+DHA or 320 mg EPA+DHA. These levels are different from those for fried breast, which contains 210 and 270 mg of EPA+DHA and EPA+docosapentaenoic acid+DHA respectively (RA Gibbs, C Rymer and DI Givens, unpublished results).

Potential impact of enriched chicken meat on intakes of long-chain n-3 PUFA

Chicken meat currently provides a small proportion of the LC n-3 PUFA consumed in the human diet but it is clear that there is great potential for this enriched prototype to contribute to increased intakes. Much research has recently focused on techniques for optimising the LC n-3 PUFA content of poultry meat to address the needs of the consumer. However, to predict the potential impacts on nutrient intakes there is a need to relate modified-product composition to product consumption trends and the overall diet⁽Reference Givens and Gibbs^55, Reference Marion and Woodruff⁶⁷⁾. This issue is currently being addressed and initial findings, some of which are described here, indicate that cooked chicken breast could supply ⩽148 mg EPA+DHA/d based on current consumption trends of chicken meat in UK adults. However, as with all food technologies, governmental, commercial and consumer interest is desperately needed to make LC n-3 PUFA-enriched chicken meat a viable and widely-available product.