Diets could prevent many diseases

Abstract

The 2002 ISSFAL Meeting arranged a special evening discussion with professional dietitians about diet-tissue-disease relationships involving essential fatty acids and eicosanoids. The balance of eicosanoid precursors in human tissues differs widely, reflecting voluntary dietary choices among different groups worldwide. An empirical quantitative diet-tissue relationship fits these diverse values as well as other research reports on essential fatty acid metabolism. Information for dietitians and nutritionists about essential fatty acids and eicosanoids is also given in two distance learning web sites, http://ods.od.nih.gov/eicosanoids/ and http:// efaeducation.nih.gov/, which facilitate dietitian education and diet counseling. These sites also have an innovative, interactive diet planning software program with the empirical equation embedded in it to help evaluate personal food choices in the context of the diet-tissue-disease relationship and other widely recommended dietary advice.

Full text

ABSTRACT: The 2002 ISSFAL Meeting arranged a special evening
discussion with professional dietitians about diet-tissue-disease re-
lationships involving essential fatty acids and eicosanoids. The bal-
ance of eicosanoid precursors in human tissues differs widely, re-
flecting voluntary dietary choices among different groups world-
wide. An empirical quantitative diet-tissue relationship fits these
diverse values as well as other research reports on essential fatty
acid metabolism. Information for dietitians and nutritionists about
essential fatty acids and eicosanoids is also given in two distance
learning web sites, http://ods.od.nih.gov/eicosanoids/ and http://
efaeducation.nih.gov/, which facilitate dietitian education and diet
counseling. These sites also have an innovative, interactive diet
planning software program with the empirical equation embedded
in it to help evaluate personal food choices in the context of the
diet-tissue-disease relationship and other widely recommended di-
etary advice.
Paper L9161 in Lipids 38, 317–321 (April 2003).
BENEFITS FROM DECREASING
EXCESSIVE AUTACOID SIGNALING
Non-steroidal anti-inflammatory drugs (NSAID) are some of the
most widely used medications worldwide. The >70–100 million
NSAID prescriptions written annually offer a huge profitability
potential with annual sales of $13 billion. The 24.5 million Cele-
brex
®
prescriptions in 2001 in the United States were closely
matched by 23.7 million prescriptions for Vioxx
®
; these two
alone had worldwide sales of $5.7 billion. The drugs are taken
to decrease an overactive self-healing response by the body’s
autacoids (auto = self, akos = healing), the n-6 eicosanoids de-
rived from essential fatty acids obtained from the diet (1). This
review examines ways in which less costly dietary approaches
might also beneficially decrease those overactive responses. The
NSAID produce their beneficial anti-inflammatory, analgesic,
antipyretic, antithrombotic effects by inhibiting the cyclooxy-
genase enzyme that converts highly unsaturated fatty acid
(HUFA) precursors into active eicosanoids. The worldwide
success of NSAID provides ample “proof of principle” for the
desirability of preventing excessive n-6 eicosanoid formation
in many disorders such as thrombosis, the arrhythmia of heart
attacks, and inflammatory/immune problems in atherosclero-
sis, arthritis, asthma, and tumor angiogenesis. Moderating
those excesses brings an improved quality of life to millions
of people. Unfortunately, nearly all of the drugs have some
undesired side effects or limitations that are much discussed
in assessing benefits and risks of medical treatments.
Ironically, the self-healing eicosanoids that are so impor-
tant in the body’s normal adaptive processes are sometimes
overly mobilized, causing people to seek various self-admin-
istered medication to decrease the “friendly fire” of those
overresponses. The n-6 eicosanoid is formed more rapidly in
the case of prostaglandin derivatives (2,3), and acts more in-
tensely in receptor signaling (4) compared with its competing
n-3 eicosanoid analog (esp. leukotriene B chemotaxis). These
differences produce less intense actions when the mixture of
released HUFA precursors has a lower proportion of the n-6
analog to compete for the eicosanoid-forming enzymes. Phys-
iologic benefits depend on moderate transient actions of these
autacoids formed from local precursors present in nearly all
tissues of the body (1). Once formed, eicosanoids are rapidly
inactivated so that active autacoid usually appears only in
small amounts at nearby tissue receptors and travels little dis-
tance from its site of synthesis. As a result, a small increase
or decrease in the rate of synthesis causes a very important
difference in the intensity of an autacoid’s action at local re-
ceptors (1). Inhibitory drugs are voluntarily used by people to
moderate an overresponse when eicosanoid action is more in-
tense or for a longer time than desired. However, the differ-
ence in potency between the n-3 and n-6 eicosanoids gives
another opportunity to decrease chronic excessive formation
and action of n-6 eicosanoids by increased dietary intakes of
naturally occurring competitive n-3 analogs rather than with
pharmaceutical antagonists.
PROPORTIONS OF PRECURSORS RELATE
TO INDIVIDUAL HUFA AND TO CORONARY
HEART DISEASE (CHD) DEATH RATES
Dietary intake is the only route of entry for eicosanoid precur-
sors, and food choices have an important effect on tissue HUFA
levels and tissue autacoid responses. Dietary 18-carbon essen-
Copyright © 2003 by AOCS Press 317 Lipids, Vol. 38, no. 4 (2003)
*Address correspondence at 6100 Westchester Park Dr., Apt. 1219, College
Park, MD 20740. E-mail: wemlands@att.net
Abbreviations: AA, arachidonic acid; CHD, coronary heart disease; DGLA,
di-homo-γ-linolenic acid; HUFA, highly unsaturated fatty acids; MRFIT,
multiple risk factor intervention trial; NSAID, non-steroidal anti-inflamma-
tory drugs.
Diets Could Prevent Many Diseases
William E.M. Lands*
College Park, Maryland 20740

tial fatty acids are metabolized to different chain lengths [such
as the 20-carbon arachidonic acid, AA (20:4n-6), and EPA
(20:5n-3)] and stored as HUFA precursors esterified to tissue
phospholipids from which they are mobilized by phospholipase-
catalyzed hydrolysis (5). The closely related n-3 and n-6 essen-
tial fatty acids compete with each other for accumulation in tis-
sue phospholipids, a process long recognized since the descrip-
tion of competitive hyperbolic interactions for these two types
of nutrient by Mohrhauer and Holman (6,7). The voluntary food
choices that people make day by day provide diverse propor-
tions of n-6 and n-3 essential fatty acids associated with very
different proportions of n-6 HUFA among the total HUFA of
plasma phospholipids, which vary from 25 to 85% (Fig. 1).
Competition between the n-6 and n-3 fatty acids is clearly evi-
dent in the decreased proportions of n-3 HUFA (DHA and EPA)
associated with increased proportions of n-6 HUFA [AA and di-
homo-γ-linolenic acid (DGLA)] that are stored in the tissue
phospholipid HUFA.
High proportions of the n-6 precursor in the tissue HUFA that
will be released during a stimulus will give high rates of forma-
tion of n-6 eicosanoids, whereas low proportions will give low
rates of formation. In this way, the balance of n-3 and n-6 acids
in the diet influences the balance of n-3 and n-6 HUFA in tissues
and therefore the eventual balance of n-3 and n-6 eicosanoid ac-
tions in self-healing processes. CHD involves excessive n-6
eicosanoid actions in chronic and acute inflammatory processes
in vascular walls that predispose people to fatal heart attacks as
well as in the thrombosis and arrhythmia of the acute event. Be-
cause CHD is a major cause of death, many drug treatments are
marketed vigorously to meet the need to treat people and reduce
an imminent risk. Figure 2 shows that the age-adjusted risk of
CHD mortality is less when the proportion of n-6 eicosanoid
precursors in people’s tissue HUFA is lower. The wide diversity
in abscissa values for Figures 1 and 2 raises the question, “What
proportion of n-6 eicosanoid precursor is stored on the shelves
of your body’s medicine chest?” It also prompts a closer, more
quantitative look at the association between tissue HUFA and
CHD mortality.
318 W.E.M. LANDS
Lipids, Vol. 38, no. 4 (2003)
FIG. 1. Different tissue highly unsaturated fatty acid (HUFA) proportions among humans. The phospholipid fatty acids from 380 different plasma sam-
ples described in other studies from the United States [n = 293 (8)] and Japan [n = 87 (9)] were analyzed by gas chromatography. The proportions of in-
dividual n-3 HUFA decrease (A: EPA, 20:5n-3, ●●; B: DHA, 22:6n-3, ●●) as the n-6 HUFA increase [A: arachidonic acid (AA), 20:4n-6, ▲▲; B: di-homo-γ-
linolenic acid (DGLA), 20:3n-6, ▲▲].
FIG. 2. Coronary heart disease (CHD) mortality rates associated with
tissue HUFA proportions. Results from the United States, Japan, and
Greenland were discussed earlier (10,11,12) as were quintile results
from the Multiple Risk Factor Intervention Trial (MRFIT) study [■ (13)]
and those from Quebec Inuit (14), Quebec Cree (15), and Quebec over-
all (16). For abbreviation see Figure 1.

The Multiple Risk Factor Intervention Trial (MRFIT) was a
multimillion-dollar clinical trial in which the “Usual Care” co-
hort provided prospective longitudinal evidence that people con-
suming greater levels of n-3 HUFA had a lower relative risk of
CHD than those eating less (13). The voluntary intake of n-3
HUFA for 6258 men (expressed as percentage of food energy,
en%) had quintile means of 0.001, 0.004, 0.019, 0.063, and
0.272 en%. These relatively low intakes corresponded to pro-
portions of 83, 82, 81, 77, and 62% n-6 HUFA in overall tissue
HUFA, respectively. The 1251 men with the most n-3 HUFA in-
take had a significantly lower relative risk of CHD, 0.6 com-
pared with 1.0, 1.08, 0.91, 0.88. However, Table 1 shows that
the limited diversity among Americans in voluntary intake of
n-3 HUFA causes this significant effect to be less apparent when
all five quintiles are combined together. Predicting the likely
proportion of n-6 HUFA in tissue HUFA for each of the quin-
tiles (in column 6) and combining them with the known fraction
of deaths in each quintile (column 7) predicts values of 78% for
cases and 75% for surviving controls (column 8). Direct gas
chromatographic analysis of the HUFA in plasma phospholipids
gave good agreement with the values predicted from the dietary
data, with an average of 79% n-6 HUFA in tissue HUFA for 94
cases and 76% for 94 controls (17). The clear benefit seen for
one quintile and emphasized by Dolecek and Grandits (13)
seems less visible when combining all quintiles together as re-
ported by Simon et al. (17). Nevertheless, the latter authors con-
cluded that a 1 SD increase in n-6 DGLA (20:3n-6) was associ-
ated with an increase in CHD risk of 40%, whereas a 1 SD in-
crease in DHA (22:6n-3) was associated with a decrease in CHD
risk of 33% (17). They concluded that the decreased CHD inci-
dence may reflect decreased platelet aggregability after in-
creased dietary intake of n-3 polyunsaturated fatty acids.
Recent results from Quebec (see Fig. 2) fit closely to the
relationship observed for the MRFIT cohort as well as for the
United States, Japan, and Greenland populations:
CHD mortality = 3 × (% n-6 HUFA in tissue HUFA) – 75 [1]
Although cross-national analyses have been regarded with cau-
tion, the Quebec population of groups from the same province
of the same country (14–16) follows the same trend for tissue
HUFA influence on the likelihood of fatal CHD events. Knowl-
edge of the molecular mechanisms for n-6 eicosanoid actions in
the disease processes of inflammation, thrombosis, and arrhyth-
mia and the success of many NSAID in moderating those
processes leads inevitably to considering dietary steps to pre-
vent excessive n-6 eicosanoid actions before they occur rather
than relying only on treatment after disease processes become
evident. The strong correlation coefficient of 0.99 (r
2
= 0.986)
for the results in Figure 2 suggests that making dietary choices
that decrease the proportion of tissue HUFA that is n-6 HUFA
can be an effective primary prevention strategy to decrease the
risk of fatal CHD events in a population. Figure 1 shows
that many people already do. There are undoubtedly other
eicosanoid-mediated diseases for which better nutrition may pre-
vent what must otherwise be treated with drugs.
FOOD CHOICES AFFECT PROPORTIONS
OF PRECURSORS IN TISSUE HUFA
Diet choices influence the composition of HUFA in tissue
phospholipids, whose composition affects the likelihood of
phospholipase releasing n-6 eicosanoid precursors; that in
turn affects the probable intensity of n-6 eicosanoid synthesis
and action in tissues. In maintaining health and preventing
chronic diseases, the tissue is the issue. Extending the work
of Mohrhauer and Holman, we examined quantitative meta-
bolic relationships between dietary supply and the propor-
tions of n-6 and n-3 eicosanoid precursors stored in the HUFA
of tissue phospholipids. The proportion of n-6 in tissue HUFA
can be estimated by an empirical relationship (12) for the
quantitative metabolic competitions among four separate
types of essential fatty acids in the diet:
1. 18-carbon n-3 PUFA; 18:3n-3
2. 18-carbon n-6 PUFA; 18:2n-6
3. 20- and 22-carbon n-3 HUFA; 20:5; 22:5; 22:6n-3
4. 20- and 22-carbon n-6 HUFA; 20:3; 20:4; 22:4; 22:5n-6
Each of the four types contributes differently to the overall tis-
sue balance. Discussing the four types combined into a single
DIETS COULD PREVENT MANY DISEASES 319
Lipids, Vol. 38, no. 4 (2003)
TABLE 1
Combined Quintile Results from the MRFIT Study
a,b
12345 6 7 8
Quintile of n-3 HUFA n Deaths RR Dietary en% Estimated % n-6 HUFA 3 × 6 Cases (2–3) × 6 Controls
MRFIT #5 1251 24 0.60 0.272 62.3 1495. 76,434.
MRFIT #4 1252 35 0.88 0.063 76.8 2687. 93,428.
MRFIT #3 1251 35 0.91 0.019 80.9 2830. 98,335.
MRFIT #2 1197 39 1.08 0.004 82.4 3215. 95,447.
MRFIT #1 1307 42 1.08 0.001 82.8 3476. 104,692.
Total 6258 175 Dolecek data predict 78.3. 74.8.
Cases Controls
Simon data observed 78.7. 76.5
a
Results in the first five columns are from Dolecek and Grandits (13), and the proportion of n-6 HUFA in total HUFA was estimated by an empirical quantitative
metabolic relationship (12). Column 7 combines columns 3 and 6 to obtain the average estimated value for coronary heart disease cases, whereas column 8 gives
the average estimated value for survivors. Analytical results of Simon et al. (17) are also in columns 7 and 8.
b
Abbreviations: MRFIT, Multiple Risk Factor Intervention Trial; HUFA, highly unsaturated fatty acids; RR, relative risk; en%, percentage of food energy.

ratio of total dietary n-6/total dietary n-3, the ratio has no
clear meaning because tissue proportions of n-6 HUFA in
total HUFA result from the four separate contributions ex-
pressed as a percentage of total food energy (en%). To make
estimates with all four variables, a simple calculator is acces-
sible at http://efaeducation.nih.gov/sig/dietbalance.html to
help design general features of dietary intervention protocols.
To facilitate estimates of how different combinations of spe-
cific foods can give very different proportions in body tissues,
we developed an interactive personalized computer program that
manages U.S. Department of Agriculture information about the
essential fatty acids in 9214 different servings of food. The soft-
ware combines information concerning a person’s daily food
choices and predicts the resulting tissue HUFA proportions,
which are biomarkers for essential fatty acid intake (as noted in
Ref. 12) as well as surrogate clinical markers (as seen in Fig. 2)
that are also metabolically related to the probable intensity of an
eicosanoid response as noted above. It can be downloaded with-
out cost from http://ods.od.nih.gov/eicosanoids/. Some ethnic
food choices achieve healthy tissue proportions and moderate
eicosanoid responses by containing large amounts of n-3 HUFA
(Japanese or Greenlanders), whereas some populations eat lower
amounts of the n-6 18-carbon fatty acids (Mediterranean peo-
ple). Although intake of linoleate is similar in the United States
and Japan (quintile means from 3 to 9 en%), the mean intake for
the lowest quintile of dietary n-3 HUFA in Japan (0.31, 0.46,
0.56, 0.71, 0.92 en%) is greater than the highest for the United
States (0.001, 0.003, 0.017, 0.057, 0.249), making tissue eico-
sanoid responses appreciably different for the two populations.
To help clinical investigators design diets that produce the
desired tissue HUFA balance, a simple spreadsheet was
arranged to combine the four separate dietary influences with
the empirical relationship described earlier (12) to predict the
likely long-term outcome. Illustrations of three typical ethnic
dietary combinations of essential fatty acid intakes (expressed
as en%) are in the columns at the right of Table 2. In a 2400
kcal/d diet, 266 mg of fat is 0.10 en%, and a standard 1-g fish
oil supplement is equivalent to 180 mg 22:6 + 120 mg 20:5,
i.e., ~0.11 en% n-3 HUFA per capsule.
Dietitians who know the importance of voluntary food
choices can use their familiarity with nutrient tables and the
new interactive software on essential fatty acids to help peo-
ple choose healthy combinations of palatable foods that will
fit their personal tastes and meet their personal target for the
balance of tissue HUFA. The interactive software combines
the selected foods into daily meal plans, tracking the calories
and estimating the probable proportions of n-6 HUFA in the
total tissue HUFA, which can vary from 25 to 85%. The
American Heart Association has recommended at least two
meals per week of fatty fish to help provide better tissue bal-
ance. Careful attention to raising the relative amount of n-3
to n-6 contents of salad and cooking oils is another easy step
in shifting tissue HUFA to a healthier balance. Also, discov-
ering and eating the vegetables and legumes with relatively
higher proportions of n-3 acids is made easier with a comput-
erized “sort” command. All of the above aspects open the way
for dietitians to lead in developing primary prevention strate-
gies that decrease the frequency and severity of eicosanoid-
mediated disorders for the whole population.
ACKNOWLEDGMENTS
Support for developing the empirical quantitative metabolic diet-
tissue relationship was from a Pfizer Biomedical Research Award;
the distance learning web site (http://ods.od.nih.gov/eicosanoids/)
hosted by the Office of Dietary Supplements provides the free down-
loaded interactive food choice software that was developed with
support from the National Institute on Alcohol Abuse and Alco-
holism. I thank Drs. W.S. Harris and M. Kobayashi for sharing their
detailed analytical results.
REFERENCES
1. Lands, W.E.M. (1979) The Biosynthesis and Metabolism of
Prostaglandins, Annu. Rev. Physiol. 41, 633–652.
2. Kulmacz, R.J., Pendleton, R.B., and Lands, W.E.M. (1994) Inter-
action Between Peroxidase and Cyclooxygenase Activities in
Prostaglandin-Endoperoxide Synthase, J. Biol. Chem. 269, 5527–
5536.
3. Malkowski, M.G., Thuresson, E.D., Lakkides, K.M., Rieke, C.J.,
Micielli, R., Smith, W.L., and Garavito, R.M. (2001) Structure
of Eicosapentaenoic and Linoleic Acids in the Cyclooxygenase
Site of Prostaglandin Endoperoxide H Synthase-1, J. Biol. Chem.
276, 37547–37555.
4. Needleman, P., Raz, A., Minkes, M.S., Ferrendelli, J.A., and
Sprecher, H. (1979) Triene Prostaglandins: Prostacyclin and
Thromboxane Biosynthesis and Unique Biological Properties,
Proc. Natl. Acad. Sci. USA 76, 944–948.
5. Lands, W.E.M., and Samuelsson, B. (1968) Phospholipid Pre-
cursors of Prostaglandin, Biochim. Biophys. Acta 164, 426–429.
6. Mohrhauer, H., and Holman, R.T. (1963) Effect of Linolenic
Acid upon the Metabolism of Linoleic Acid, J. Nutr. 81, 67–74.
7. Mohrhauer, H., and Holman, R.T. (1963) The Effect of Dose
Level of Essential Fatty Acids upon Fatty Acid Composition of
the Rat Liver, J. Lipid Res. 4, 151–159.
8. Harris, W.S., Ginsberg, H., Arunakul, N., Shachter, N.S., Wind-
sor, S.L., Adams, M., Berglund, L., and Osmundsen, K. (1997)
Safety and Efficacy of Omacor in Severe Hypertriglyceridemia,
J. Cardiovasc. Risk 4, 385–391.
9. Kobayashi, M., Sasaki, S., Kawabata, T., Hasegawa, K., Akabane,
M., and Tsugane, S. (2001) Single Measurement of Serum Phos-
pholipid Fatty Acid as a Biomarker of Specific Fatty Acid Intake
in Middle-Aged Japanese Men, Eur. J. Clin. Nutr. 55, 643–650.
10. Lands, W.E.M., Hamazaki, T., Yamazaki, K., Okuyama, H.,
Sakai, K., Goto, Y., and Hubbard, V.S. (1990) Changing Di-
etary Patterns, Am. J. Clin. Nutr. 51, 991–993.
11. Lands, W.E.M. (1991) Biosynthesis of Prostaglandins, Annu.
Rev. Nutr. 11, 41–60.
320 W.E.M. LANDS
Lipids, Vol. 38, no. 4 (2003)
TABLE 2
Daily Dietary Intakes Affect Tissue HUFA
a
Type of essential Trial
Typical diets
fatty acid in the diet diet U.S. Mediterranean Japan
1 en% 18:3n-3 1.00 0.85 0.50 0.83
2 en% 18:2n-6 6.00 6.82 2.30 5.41
3 en% n-3 HUFA 0.10 0.03 0.09 0.71
4 en% n-6 HUFA 0.10 0.08 0.08 0.09
% n-6 in total HUFA 75.10 80.10 63.10 48
a
For abbreviations see Table 1.10

12. Lands, W.E.M., Libelt, B., Morris, A., Kramer, N.C., Prewitt, T.E.,
Bowen, P., Schmeisser, D., Davidson, M.H., and Burns, J.H.
(1992) Maintenance of Lower Proportions of n-6 Eicosanoid Pre-
cursors in Phospholipids of Human Plasma in Response to Added
Dietary n-3 Fatty Acids, Biochim. Biophys. Acta 1180, 147–162.
13. Dolecek, T.A., and Granditis, G. (1991) Dietary Polyunsatu-
rated Fatty Acids and Mortality in the Multiple Risk Factor In-
tervention Trial (MRFIT), World Rev. Nutr. Diet. 66, 205–216.
14. Dewailly, E., Blanchet, C., Lemieux, S., Sauve, L., Gingras, S.,
Ayotte, P., and Holub, B.J. (2002) n-3 Fatty Acids and Cardio-
vascular Disease Risk Factors Among the Inuit of Nunavik, Am.
J. Clin. Nutr. 76, 85–92.
15. Dewailly, E., Blanchet, C., Gingras, S., Lemieux, S., and Holub,
B.J. (2002) Cardiovascular Disease Risk Factors and n-3 Fatty
Acid Status in the Adult Population of James Bay Cree, Am. J.
Clin. Nutr. 76, 85–92.
16. Dewailly, E.E., Blanchet, C., Gingras, S., Lemieux, S., Sauve,
L., Bergeron, J., and Holub, B.J. (2001) Relations Between n-3
Fatty Acid Status and Cardiovascular Disease Risk Factors
Among Quebecers, Am. J. Clin. Nutr. 74, 603–611.
17. Simon, J.A., Hodgkins, M.L., Browner, W.S., Neuhaus, J.M.,
Bernert, J.T., and Hulley, S.B. (1995) Serum Fatty Acids and
the Risk of Coronary Heart Disease, Am. J. Epidemiol. 142,
469–476.
[Received September 23, 2002, and in revised form and accepted
April 8, 2003]
DIETS COULD PREVENT MANY DISEASES 321
Lipids, Vol. 38, no. 4 (2003)