If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
A competitive, hyperbolic relationship predicts the proportions of n-3 and n-6 HUFA accumulated from n-3 and n-6 nutrients.
•
The %n-3 in HUFA is the inverse of %n-6 in HUFA that predicts the likely intensity of formation and action of n-6 eicosanoids.
•
Blood biomarker values for %n-3 in HUFA have less noise than values for EPA + DHA in total fatty acids.
•
Omega 3-6 balance scores give likely impacts of food items on blood HUFA balance to guide effective nutritional interventions.
•
Eating less n-6 nutrients allows competing n-3 nutrients to raise more effectively the %n-3 in HUFA.
Abstract
Careful handling of data on fatty acid composition is needed when interpreting evidence for the influence of dietary n-3 and n-6 essential fatty acids on brain function and health conditions. The relative dietary supplies of competing n-3 and n-6 nutrients determine the balance of 20- and 22-carbon n-3 and n-6 highly unsaturated fatty acids (HUFA) which accumulate competitively at the 2-position of tissue phospholipids. In turn, the HUFA balance expressed as the %n-6 in HUFA affects the likely intensity of n-6 eicosanoid actions in diverse health conditions. As a result, measures of HUFA balance are important, valid biomarkers for designing and monitoring successful preventive nutrition interventions. Successful interventions must also consider the ability of fatty acid ligands to saturate binding sites of enzymes and receptors and give paradoxical dose-response results.
The flow of fatty acids into and out of various tissue storage and transport forms was the focus of lipid biochemistry during the mid-20th century. Most dietary lipids are hydrolyzed in the gut and re-esterified into lipoproteins that are transported to all tissues via blood (see Fig. 1). Adipose tissue predominantly accumulates dietary essential fatty acids (EFA) in triacylglycerols, for which the weight percent (wt%) content tends to reflect directly dietary EFA intakes expressed as a percent of daily food energy (en%) [
Validity of the fatty acid composition of subcutaneous fat tissue microbiopsies as an estimate of the long-term average fatty acid composition of the diet of separate individuals.
]. The average American adult may have about 30% of their body mass as triacylglycerols that may contain over one in six acyl groups as linoleate, 18:2n-6. Perhaps half of the triacylglycerols contain linoleate. Liver tissue predominantly converts dietary 18-carbon polyunsaturated fatty acids (PUFA) into 20- and 22- carbon highly unsaturated fatty acids (HUFA) and secretes them as phospholipids that are transported in blood lipoproteins to all tissues.
Fig. 1Flow of dietary fatty acids through diverse metabolites and tissues. Three stages are involved: PUFA precursors of HUFA; HUFA precursors of eicosanoids; eicosanoid receptor actions.
Interpretations of how dietary n-3 and n-6 EFA influence brain function and human health can be improved by recognizing the multiple stages through which the nutrients pass on their way to modulating physiological events. Many steps in absorption, transport and storage in tissues occur with little discrimination between the n-3 or n-6 structures, although some steps (e.g., acyltransferases and phospholipases) favor acids with greater chain length and number of double bonds [
]. The 18-carbon EFA, 18:2n-6 and 18:3n-3, are abundant (and imbalanced) in daily foods, and they compete almost equally in forming and accumulating the 20- and 22-carbon highly unsaturated fatty acids (HUFA) [
]. Selectivity also occurs in some synthase conversions of PGH into bioactive eicosanoids (PGD, PGE, PGF, PGI, TXA) which selective cellular receptors bind and transduce into intracellular signaling events [
]. Nearly every tissue has one or more eicosanoid receptors (e.g., DP1, DP2, EP1, EP2, EP3, EP4, FP, IP, TP), for which the diverse expression and action provides a rich array of ways in which an imbalance in n-3 and n-6 nutrients can cause healthy physiology to drift into pathophysiology [
The arachidonic acid cascade includes oxidative alterations of HUFA by cyclooxygenases, lipoxygenases and cytochrome P450 enzymes, and it has more examples of harmful actions by n-6 than n-3 mediators [
]. Clinical research has recognized many health conditions in which excessive n-6 mediator actions are therapeutic targets for new drug development to treat the unwanted pathophysiology. Alternately, nutritionists can design preventive nutrition strategies that decrease the proportion of n-6 in HUFA and thereby decrease the need for treatments [
]. Successful preventive nutrition will need careful handling of fatty acid composition data at three stages: PUFA precursors of HUFA; HUFA precursors of eicosanoids; eicosanoid-mediated health conditions.
A lack of interest in details of the competitive dynamics of n-3 and n-6 nutrients forming HUFA has delayed development of effective prevention of a current imbalance in n-3 and n-6 nutrients that shifts healthy physiology toward pathophysiology [
]. We examine here some quantitative dynamic aspects of essential fatty acid supply and metabolism to strengthen the biomedical community's ability to design an effective decrease in the severity of many unwanted food-related chronic health conditions.
2. Materials and experimental methods
2.1 Sources of gas chromatographic data
Gas chromatographic data on HUFA balance in 1,015 finger-tip blood-spot samples from Americans were obtained from archived historic de-identified records as described earlier [
]. Data from a recently published systematic review examining the EPA and DHA of plasma, plasma phospholipid, erythrocytes and whole blood from healthy adults across the globe was collated and screened for comprehensive fatty acid profiles to allow for further calculations [
An important aspect of enzyme catalytic sites and receptor binding sites is the manner in which they become saturated with ligands and give hyperbolic patterns for ligand dose-response actions (Fig. 2). As ligand concentrations rise, the receptor-mediated action also rises in a linear manner until the site becomes saturated, and an unchanging maximal response continues with further ligand increases. This paradoxical behavior is predicted by the 100-year old Michaelis-Menten equation. It is often a first lesson in beginning biochemistry, but neglected in many research reports.
(1)
Fig. 2Hyperbolic accumulation of highly unsaturated fatty acid (HUFA) from dietary precursors. Curve A shows the fractional maximal response to dietary intake of either 18:2n-6 (LA) or 18:3n-3 (ALA) alone has a sensitive linear response from 0% to 0.5% of food energy 0–0.5 en%), with a half-maximal response near 0.1 en% (curve A). In the presence of the USA average 0.7 en% ALA (curve B), the circle indicates a response near maximal for 6 en% LA and near half-maximal when both nutrients are at 0.7 en%. Curve C has a circle indicating that intake of ALA in the presence of the USA average 6 en% LA gives little accumulated n-3 HUFA and would give a half-maximal response when both nutrients are at 6 en%.
Fig. 2 uses the half-maximal response near 0.1 en% to predict little change in accumulated arachidonic acid (20:4n-6) levels when dietary linoleic acid (18:2n-6) changes between 3% of food energy (3 en%) and 8 en%. Under these conditions, conversion of dietary 18:2n-6 into 20:4n-6 is operating near maximal rather than not at all. Intakes between 0 and 0.5 en% show linear dose-response behavior. Detailed quantitative displays of sensitive biological responses to small dietary doses of EFA were reported in 1963 for rats [
]. Physiological responses to small amounts of n-3 and n-6 nutrients have concomitant accumulation of 20- and 22-carbon n-3 and n-6 HUFA displacing n-9 HUFA from the 2-position of tissue phospholipids.
The competitive, saturable dose-response dynamic for tissue HUFA accumulation [
] and summarized in the form of a data-based empirical equation using values for EFA as the percent of food energy (en%) to describe diet-tissue dynamics [
]. The similar dynamics for competing n-3 and n-6 substrates allow their relative abundance in diets to influence greatly the relative proportions of n-3 and n-6 HUFA accumulated in tissues. Curve C in Fig. 2 illustrates how the current high intakes of n-6 LA create a competitive environment in which tissue n-3 HUFA levels are determined more by eating pre-formed EPA and DHA than by ALA intake.
To estimate the impact of individual food items in an overall daily intake of EFA upon the likely resulting blood HUFA biomarkers, the EFA contents of each food item can be expressed as mg/kcal in the form of an omega 3–6 balance score [
], and they allow estimates of the likely resulting blood HUFA balance that is associated with each individual's typical daily pattern of food intake. For example, removing ten items with the most negative score from the 100 most frequently eaten American foods would shift the average score of the USA diet (near −6.5) to that for a traditional Mediterranean diet (near −3) [
]. In the following equations, letters A, B, C, D represent n-6 HUFA (20:3n-6, 20:4 n-6, 22:4n-6, 22:5n-6) and F, G, H are n-3 HUFA (20:5n-3, 22:5n-3, 22:6n-3). The seldom seen 20:3n-9 is E, and other letters are non-HUFA acids that often add up to 80% of total fatty acids.
Traditional GC analytical reports express each acid's content as a percent of the total fatty acids (% in totFA) detected in the sample. The denominator thus contains many diverse fatty acids (from many diverse lipids in whole blood samples). Also, depending on laboratory handling conditions, corrections may be needed to adjust for evaporative losses of more volatile acids and oxidative losses of highly unsaturated acids. Eqs. ((2), (3), (4), (5)) describe how standard gas chromatographic data are used to calculate biomarker values.
(2)
(3)
(4)
(5)
In contrast, HUFA-oriented biomarkers express HUFA content as a percent of the total HUFA detected. In this way, each HUFA serves as an internal recovery control for the other chemically related HUFA.
(6)
(7)
An inherent difference between the two types of biomarkers is evident when examining their reciprocal forms. Putting Eqs. ((2), (5)) into a reciprocal format (recip) gives Eqs. ((8), (9)), respectively. Each contains the recip[%n–3 in HUFA] (recip[Eq. (6)]) plus a non-HUFA component:
By adding the appropriate residual non-HUFA component, we can convert the HUFA Balance biomarker, %n-3 in HUFA, into values for either the % n-3 in totFA or the %EPA+DHA in totFA (called the “Omega-3 Index” when applied to RBC). However, doing so creates a loss of signal clarity with no gain in interpretable information. The relative “noise” for these biomarkers is evident in Fig. 3 which compares values for recip[Eq.6] with values obtained by adding appropriate non-HUFA residual values to give comparable displays for recip[% n-3 in totFA] and recip[%EPA+DHA in totFA].
Fig. 3Recognizing signal and noise in different biomarkers Data are from 1,015 assays described in detail elsewhere
. Data for recip[%n-3 in HUFA] have less “noise” than data for recip[%n-3 in HUFA] plus residual-8 (which include DPA) and data for recip[%n-3 in HUFA] plus residual-9 (which have DPA deleted).
Because most people have negligible amounts of 20:3n-9, the measured value for %n-6 in HUFA equals 100 - %n-3 in HUFA. As a result, any relationship of disease severity with %n-3 in HUFA has a similar, but inverse, relationship with the %n-6 in HUFA. The %n-6 in HUFA biomarker value measured in blood lipids also relates directly to the predicted calorie-weighted average daily balance of n-3 and n-6 nutrients expressed as mg/calorie [
], approaches to designing low-cost high-throughput assays were discussed at the 2004 ISSFAL workshop. Discussants regarded the finger-tip blood spot on filter paper to be less labor-intensive than centrifuging to obtain red blood cell (RBC) membranes [
]. This reduced gas chromatography time then led to a re-examination of sample handling and fatty acid methyl ester preparation that resulted in rapid direct transesterification protocols [
]. All of the above aspects of discipline for sample handling in gas chromatographic analyses impact credibility when interpreting how the collected compositional data relate to human health. The data can be expressed in different ways to meet different purposes.
The increased analytical throughput has enabled research comparing fatty acid biomarkers across various blood fractions such as whole blood, plasma and red blood cells (RBC) and allowed the development of large datasets of fatty acid composition. These large datasets allow for thorough investigations into the utility of different fatty acid blood biomarkers. Two recent reports on biomarkers [
], the % n-3 in HUFA relates to the % n-6 in HUFA in a clear linear relationship over a wide range, and the distinct slope contrasts with that for the sum of the % of EPA + DHA in total fatty acids (Fig. 4).
Fig. 4Comparing biomarkers for 1,015 Americans. Biomarker values are from de-identified records described earlier
] and they were relatively consistent across RBC (100 studies), plasma phospholipid (91 studies) plasma total lipid (89 studies), and whole blood (8 studies) (Fig. 5). The HUFA-based approach resulted in consistent linear relationships across all blood fractions, whereas the relationship was weak and varied across blood fractions with the total fatty acid approach. The differences in the biomarker properties are largely driven by the exclusion of unrelated non-HUFA information and not omitting relevant DPAn-3 information. The HUFA-based biomarker has a recognized quantitative association with competing nutrient intakes and the predicted intensity of n-6 eicosanoid actions. In contrast, the sum of the % of EPA + DHA in total fatty acids and the risk of cardiovascular disease has been defined previously [
Advantages of a HUFA-based approach have been reported in the literature previously. Stronger correlations between different blood fractions were first reported in 2005 [
]. The HUFA-based approach was better at predicting omega-3 status of various tissues from blood measures as compared with EPA+DHA in total fatty acids and the ratio of total n-3/n-6 PUFA [
]. These observations are largely driven by the tendency for HUFA to be in the sn-2 position of glycerophospholipids, so focusing on the HUFA pool eliminates variation across tissues created by different mixtures of complex lipids such as triacylglycerols and cholesteryl esters. Recently, values of EPA+DHA in total fatty acids were shown to be translated across different blood pools, but the process can be quite complex [
The HUFA-based approach also accommodates methodological variation in fatty acid analysis as the HUFA have similar chemical properties and also similar location within phospholipids. Variations in preparation of fatty acid methyl esters for gas chromatography tend to affect similarly the fatty acids in the HUFA pool while other classes of fatty acids such as saturates may respond differently. Specifically, the HUFA pool seems free from the incomplete derivatization seen for fatty acids in the sphingolipid pool and for transesterification efficiency for nonpolar and polar lipid pools [
The HUFA-based biomarker has also been shown to be somewhat resistant to changes in status due to oxidation during less than optimal sample storage as HUFA oxidize at a similar rate as compared with less saturated fatty acids [
]. While the HUFA-based approach results in little change in estimated omega-3 status, values for the % EPA+DHA in total fatty acids drop dramatically as measured values for EPA and DHA drop in comparison with saturates and monunsaturates. Correcting for oxidation during storage at −20 °C is possible for the EPA+DHA biomarker approach, but it requires calibration equations with multiple imputations [
3.2 Interpreting linoleic acid levels in foods and blood
Blood levels of 18:2n-6 range from 12% to 39% of total fatty acids in Fig. 4, Fig. 5, indicating no current shortage of 18:2n-6 in American or global diets. Fig. 5 shows that 18:2n-6 content in blood varies among blood fractions, with plasma tending to have higher 18:2n-6 levels as the cholesteryl ester component in plasma tends to have a high content of 18:2n-6 [
]. Despite this, the 18:2n-6 levels are similarly high and independent of the %n-6 in HUFA across all blood fractions. An absence of appreciable 20:3n-9 in these databases confirms that average daily intakes of PUFA are much higher than the half-maximal value near 0.1 en% [
The different n-3 and n-6 EFA do not enter the body in an independently random manner, but have different covariances determined by the diverse foods eaten. Characteristic proportions of n-3 and n-6 nutrients in individual food items [
] may combine in daily diet patterns to give counter-intuitive impacts on HUFA balance. For example, food frequency data paradoxically showed that people in the highest quintile of eating 18:3n-3 had the highest 18:2n-6 intake [
Predictions of health risk and benefit associated with dietary linoleate intakes have been made by combining observed food intake associations with arithmetical manipulations regarded as “replacing” or “substituting” dietary saturated fatty acid with linoleic acid as if a change in diet had actually been made and its outcome observed. For example, Wang et al. [
] concluded that “Replacing 5%of energy from saturated fats with equivalent energy from PUFA and MUFA was associated with reductions in total mortality of 27%and 13%, respectively.” However, such predictions from manipulated associations are still hypotheses to be tested by the reality of an actual controlled nutritional intervention.
The validity of any predicted estimate for an associative risk or benefit depends on the dynamics of EFA flow from diet to tissue to health outcome. More certain causal evidence for harm or benefit comes after actually changing people's diet and measuring the resultant outcomes relative to those for control subjects. There is a risk of error when discussing a predicted association as if it were an experimentally measured event. Such risk may be high when treating the EFA diet-tissue relationship as if it were linear when it includes hyperbolic components (Fig. 2).
When clinical interventions need many people over many years to have sufficient statistical power, researchers often use a surrogate biomarker to monitor intermediate conditions to save time and money [
]. HUFA balance can be a valuable surrogate since it has a recognized predictable relationship to dietary nutrients and also to physiological outcomes [
]. Fig. 6 shows that in individuals with less than 50% n-6 in HUFA, activated cytosolic phospholipase A2 is likely to encounter more n-3 HUFA than arachidonate. Average HUFA balance for Americans is near 75–80% n-6 in HUFA [
Essential HUFA form eicosanoids that regulate many healthy physiological functions. However, excessive actions by n-6 eicosanoids formed by the arachidonic acid cascade [
] can shift healthy physiology toward pathophysiology. This causal role for arachidonate makes measures of its availability potentially valid surrogate biomarkers for those conditions. Validity is an important aspect since not all associated biomarkers have causal roles. Epidemiological evidence of cardiovascular disease (CVD) associations with surrogate biomarkers led to hypotheses that the disease may be caused by elevated blood cholesterol [
]. However, large, expensive interventional trials testing the hypotheses showed that lowered biomarkers levels were not accompanied by lowered CVD [reviewed in [
]]. Thus, elevated blood cholesterol, obesity, and type 2 diabetes are associated comorbidities, whereas HUFA balance is a rational causal mediator that awaits substantial validation.
Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet heart Study and updated meta-analysis.
] clinical interventions showed that having patients eat more 18:2n-6 gave the expected lowering of blood cholesterol while it also caused more death. Such evidence does not support the hypothesis that lowering blood cholesterol lowers cardiovascular death. Rather, it fits an alternate concept that blood cholesterol levels predict death only to the extent that n-6 HUFA exceed n-3 HUFA [
Omega-6 fatty acids and risk for cardiovascular disease: a science advisory from the american heart association nutrition subcommittee of the council on nutrition, physical activity, and metabolism; council on cardiovascular nursing; and council on epidemiology and prevention.
]. On the basis of observational associative evidence, the AHA advised people to eat “an omega-6 PUFA intake of at least 5–10% of energy” and warned that “To reduce omega-6 PUFA intakes from their current levels would be more likely to increase than to decrease risk for CHD.” [
Omega-6 fatty acids and risk for cardiovascular disease: a science advisory from the american heart association nutrition subcommittee of the council on nutrition, physical activity, and metabolism; council on cardiovascular nursing; and council on epidemiology and prevention.
], another hypothesis not yet tested by controlled interventional data. An alternate view is that “The widespread consumption of diets with more than 2% energy as LA should be recognized for what it is — a massive uncontrolled human experiment without adequate rationales or proven mechanisms.” [
Significant associations of more disease with higher proportions of n-6 HUFA have been reported from observational data of populations which include a wide range of HUFA balance [
]. Future nutritional interventions for diminishing these diseases may develop clearer insight on EFA actions by including measurement of a valid surrogate biomarker like the %n-6 in HUFA [
] to design a nutritional intervention that appreciably shifts the HUFA balance.
Observational data led to the hypothesis that combining a lower intake of n-6 nutrients with higher intake of n-3 nutrients can lower the %n-6 in HUFA and give health benefits [
] in which two modes of dietary intervention reduced the need for medications in three months, and health benefits were related quantitatively to the extent of lowering of the %n-6 in HUFA. Patients reducing 18:2n-6 intake from 7.4 en% to 2.4 en% shifted HUFA balance from 78% to 75% n-6 in HUFA and had a small, but significant, reduction in pain and need for medication. Fig. 2 predicts that a small decrease in HUFA balance is likely at this high range of linoleate intake. The patients who included a higher intake of n-3 nutrients with the lowered n-6 nutrient intake shifted their HUFA balance from 77% to 61% n-6 in HUFA in three months, and the pain and need for medication lowered to nearly half the baseline value [
], but attention continues to go more to treating the conditions than preventing the need to treat them. The design of effective primary prevention has been delayed by a lack of funds and a low interest by traditional governmental and pharmaceutical research organizations for translating existing knowledge into public health practice. Fortunately, no prescriptions are needed for people to eat less n-6 EFA and more n-3 EFA as an approach to “eating healthy” [
American employers already spend millions for wellness programs to lower health-related financial losses. Those programs may be enhanced by helping employees learn their personal HUFA balance and its association with diverse health conditions [
]. Explicit information from finger-tip blood-spot assays plus explicit information on foods that either raise or lower the blood biomarker empowers participants of wellness programs to choose voluntarily a HUFA balance that fits their own personal priorities. Observational data from the longitudinal MRFIT study [
] show the quintile with an estimated HUFA balance near 63% n-6 in HUFA had nearly half the cardiovascular deaths compared to those with an estimated HUFA balance near 80% n-6 in HUFA [
] suggests that a longitudinal observational study of employees in wellness programs that monitor HUFA balance may show lower annual healthcare costs as employees lower their HUFA balance from values near 80% toward values near 60% n-6 in HUFA [
]. Organizations facilitating such a wellness intervention may find a useful return on investment that can sustain the program and provide useful information to the broader public [
The balance of n-3 and n-6 HUFA that accumulate in the 2-position of tissue phospholipids affects the intensity of n-6 eicosanoid mobilization and action during diverse chronic health conditions. As a result, measures of HUFA balance are important, valid biomarkers for designing and monitoring successful preventive nutrition interventions. Paradoxical dose-responses occur because HUFA balance depends in a competitive non-linear way upon the relative amounts of n-3 and n-6 EFA nutrients eaten. Monitoring either type of EFA alone gives an incomplete and imprecise view of the likely physiological outcomes from food choices that affect the HUFA balance. Reducing the intake of dietary n-6 nutrients allows n-3 nutrients to lower more effectively the %n-6 in HUFA and thereby lower the risk of excessive n-6 eicosanoid-mediated events.
Sources of support
K.D.S. is supported through a Canada Research Chair in Nutritional Lipidomics (950-228125).
References
van Staveren W.A.
Deurenberg P.
Katan N.B.
Burema J.
de Groot L.C.
Hoffmans M.D.
Validity of the fatty acid composition of subcutaneous fat tissue microbiopsies as an estimate of the long-term average fatty acid composition of the diet of separate individuals.
Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet heart Study and updated meta-analysis.
Omega-6 fatty acids and risk for cardiovascular disease: a science advisory from the american heart association nutrition subcommittee of the council on nutrition, physical activity, and metabolism; council on cardiovascular nursing; and council on epidemiology and prevention.
30152-1&id=gr1.jpg){kind=link}
30152-1&id=gr2.jpg){kind=link}
30152-1&id=gr3.jpg){kind=link}
30152-1&id=gr4.jpg){kind=link}
30152-1&id=gr5.jpg){kind=link}
30152-1&id=gr6.jpg){kind=link}