Perla C. Schmidt

Docosahexaenoic acid ingestion inhibits natural killer cell activity and production of inflammatory mediators in young healthy men

The purpose of this study was to examine the effects of feeding docosahexaenoic acid (DHA) as triacylglycerol on the fatty acid composition, eicosanoid production, and select activities of human peripheral blood mononuclear cells (PBMNC). A 120-d study with 11 healthy men was conducted at the Metabolic Research Unit of Western Human Nutrition Reach Center. Four subjects (control group) were fed the stabilization diet throughout the study; the remaining seven subjects were fed the basal diet for the first 30 d, followed by 6 g DHA/d for the next 90 d. DHA replaced an equivalent amount of linoleic acid; the two diets were comparable in their total fat and all other nutrients. Both diets were supplemented with 20 mg d α-tocopherol acetate per day. PBMNC fatty acid composition and eicosanoid production were examined on day 30 and 113; immune cell functions were tested on day 22, 30, 78, 85, 106, and 113. DHA feeding increased its concentration from 2.3 to 7.4 wt% in the PBMNC total lipids, and decreased arachidonic acid concentration from 19.8 to 10.7 wt%. It also lowered prostaglandin E2 (PGE2) and leukotriene B4 (LTB4) production, in response to lipopolysaccharide, by 60–75%. Natural killer cell activity and in vitro secretion of interleukin-1β and tumor necrosis factor α were significantly reduced by DHA feeding. These parameters remained unchanged in the subjects fed the control diet. B-cell functions as reported here and T-cell functions that we reported previously were not altered by DHA feeding. Our results show that inhibitory effects of DHA on immune cell functions varied with the cell type, and that the inhibitory effects are not mediated through increased production of PGE2 and LTB4.

The effect of dietary docosahexaenoic acid on plasma lipoproteins and tissue fatty acid composition in humans

Normal, healthy male volunteers (n=6) were fed diets [high docosahexaenoic acid-DHA] containing 6 g/d of DHA for 90 d. The stabilization (low-DHA) diet contained less than 50 mg/d of DHA. A control group (n=4) remained on the low-DHA diet for the duration of the study (120 d). Blood samples were drawn on study days 30 (end of the stabilization period), 75 (midpoint of the intervention period), and 120 (end of the intervention period). Adipose tissue (AT) samples were taken on days 30 and 120. The plasma cholesterol (C), low density lipoprotein (LDL)-C and apolipoproteins (apo) [Al, B, and lipoprotein (a)] were unchanged after 90 d, but the triglycerides (TAG) were reduced from a mean value of 76.67±24.32 to 63.83±16.99 mg/dL (n=6, P<0.007 using a paired t-test) and the high density lipoprotein (HDL)-C increased from 34.83±4.38 mg/dL to 37.83±3.32 mg/dL (n=6, P<0.017 using a paired t-test). The control group showed no significant reduction in plasma TAG levels. Apo-E, however, showed a marked increase in the volunteers’ plasma after 90 d on the high-DHA diet, from 7.06±4.47 mg/dL on study day 30 to 12.01±4.96 mg/dL on study day 120 (P<0.002 using a paired t-test). The control subjects showed no significant change in the apo-E in their plasma (8.46±2.90 on day 30 vs. 8.59±2.97 on day 120). The weight percentage of plasma DHA rose from 1.83±0.22 to 8.12±0.76 after 90 d on the high-DHA diet. Although these volunteers were eating a diet free of eicosapentaenoic acid (EPA), plasma EPA levels rose from 0.38±0.05 to 3.39±0.52 (wt%) after consuming the high-DHA diet. The fatty acid composition of plasma lipid fractions—cholesterol esters, TAG, and phospholipid—showed marked similarity in the enrichment of DHA, about 10%, after the subjects consumed the high-DHA diet. The DHA content of these plasma lipid fractions varied from less than 1% (TAG) to 3.5% (phospholipids) at baseline, study day 30. EPA also increased in all plasma lipid fractions after the subjects consumed the high-DHA diet. There were no changes in the plasma DHA or EPA levels in the control group. Consumption of DHA also caused an increase in AT levels of DHA, from 0.10±0.02 to 0.31±0.07 (wt%) (n=6, P<0.001 using a paired t-test), but the amount of EPA in their AT did not change. Thus, dietary DHA will lower plasma TAG without EPA, and DHA is retroconverted to EPA in significant amounts. Dietary DHA appears to enhance apo-E synthesis in the liver. It appears that DHA can be a safe and perhaps beneficial supplement to human diets.

Dietary α-linolenic acid alters tissue fatty acid composition, but not blood lipids, lipoproteins or coagulation status in humans

We examined the effect of dietary α-linolenic acid (ALA) on the indices of lipid and coagulation status and on the fatty acid composition of serum and peripheral blood mononuclear cell (PBMNC) lipids in ten healthy men (age 21–37 yr) who consumed all their meals at the Western Human Nutrition Research Center for 126 d. There was a stabilization period of 14 d at the start when all 10 subjects consumed the basal diet (BD) containing 23.4 energy percent (en%) fat and two intervention periods of 56 d each. During the first intervention period, 5 subjects consumed the BD containing 23.4 en% fat, and 5 subjects consumed a diet providing 6.3% calories from α-linolenic acid [flaxseed oil (FSO) diet containing 28.8 en% fat]. Diets were crossed over between the two groups during the second intervention period. Feeding the FSO diet did not nignificantly alter serum triglycerides, cholesterol, highdensity lipoproteins, low-density lipoproteins, apoprotein A-I and apoprotein B when compared to the corresponding values in the subjects fed the BD, nor was there any effect of the FSO diet on the bleeding time, prothrombin time and partial prothrombin time for these subjects. Feeding the ALA-containing diet did cause a significant increase in ALA concentration in serum (P<0.001) and PBMNC lipids (P<0.05). It also caused a significant increase (P<0.05) in the eicosapentaenoic and docosapentaenoic acid contents of PBMNC lipids, and a decrease (P<0.01) in linoleic and eicosatrienoic acid contents of serum lipids. Thus, dietary ALA, fed for 56 d at 6.3% of calories, had no effect on plasma triglyceride or very low density lipoprotein levels or the common risk factors associated with atherosclerosis, although these parameters have been reported by others to be influenced by fatty acids, such as palmitic or linoleic acids, in the diet. Dietary ALA did significantly alter the fatty acid composition of plasma and PBMNC.

The effect of dietary arachidonic acid on plasma lipoprotein distributions, apoproteins, blood lipid levels, and tissue fatty acid composition in humans.

Normal healthy male volunteers (n=10) were fed diets (high-AA) containing 1.7 g/d of arachidonic acid (AA) for 50 d. The control (low-AA) diet contained 210 mg/d of AA. Dietary AA had no statistically significant effect on the blood cholesterol levels, lipoprotein distribution, or apoprotein levels. Adipose tissue fatty acid composition was not influenced by AA feeding. The plasma total fatty acid composition was markedly enriched in AA after 50 d (P<0.005). The fatty acid composition of plasma lipid fractions, cholesterol esters, triglycerides, free fatty acids, and phospholipid (PL) showed marked differences in the degree of enrichment in AA. The PL plasma fraction from the subjects consuming the low-AA diet contained 10.3% AA while the subjects who consumed the high-AA diet had plasma PL fractions containing 19.0% AA. The level of 22:4n-6 also was different (0.67 to 1.06%) in the plasma PL fraction after 50 d of AA feeding. After consuming the high-AA diet, the total red blood cell fatty acid composition was significantly enriched in AA which mainly replaced linoleic acid. These results indicate that dietary AA is incorporated into tissue lipids, but selectively into different tissues and lipid classes. Perhaps more importantly, the results demonstrate that dietary AA does not alter blood lipids or lipoprotein levels or have obvious adverse health effects at this level and duration of feeding.

Effects of dietary arachidonic acid on human immune response

Arachidonic acid (AA) is a precursor of eicosanoids, which influence human health and the in vitro activity of immune cells. We therefore examined the effects of dietary AA on the immune response (IR) of 10 healthy men living at our metabolic suite for 130 d. All subjects were fed a basal diet containing 27 energy percentage (en%) fat, 57 en% carbohydrate, and 16 en% protein (AA, 200 mg/d) for the first and last 15 d of the study. Additional AA (1.5 g/d) was incorporated into the diet of six men from day 16 to 65 while the remaining four subjects continued to eat the basal diet. The diets of the two groups were crossed-over from day 66 to 115. In vitro indexes of IR were examined using the blood samples drawn on days 15, 58, 65, 108, 115, and 127. The subjects were immunized with the measles/mumps/rubella vaccine on day 35 and with the influenza vaccine on day 92. Dietary AA did not influence many indexes of IR (peripheral blood mononuclear cell proliferation in response to phytohemagglutinin, Concanavalin A, pokeweed, measles/mumps/rubella, and influenza vaccines prior to immunization, and natural killer cell activity). The postimmunization proliferation in response to influenza vaccine was about fourfold higher in the group receiving high-AA diet compared to the group receiving low-AA diet (P=0.02). Analysis of variance of the data pooled from both groups showed that the number of circulating granulocytes was significantly (P=0.03) more when the subjects were fed the high-AA diet than when they were fed the low-AA diet. The small increases in granulocyte count and the in vitro proliferation in response to influenza vaccine caused by dietary AA may not be of clinical significance. However, the lack of any adverse effects on IR indicates that supplementation with AA may be done safely when needed for other health reasons.

The effect of a salmon diet on blood clotting, platelet aggregation and fatty acids in normal adult men.

This study was designed to measure the effect of dietary n−3 fatty acids (FA) on platelets and blood lipids. Healthy men (n=9), ages 31 to 65, were fed diets in which salmon was the source of n−3 fatty acids. They were confined in a nutrition suite at this Center for 100 days. Food intake and exercise levels were rigidly controlled. Initially they were placed on a stabilization diet for 20 days, then six men were fed the salmon diet for 40 days. The others remained on the stabilization diet. The two groups switched diets for the last 40 days of the study. Both diets were isocaloric [16% protein, 54% carbohydrate, and 30% fat by energy-% (En%)]. The salmon diet contained 7.5% of calories from n−6 FA and 2% from n−3 FA, primarily eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in a 40∶60 ratio, while the stabilization diet contained 7.5% of calories from n−6 FA and less than 0.3% n−3 FA, mainly 18∶3n−3. The bleeding time was unaffected by the diets in this study. The prothrombin time was shortened (11.6 sec.vs. 12.6 sec., p<0.01) for the subjects consuming the salmon diet as compared to that measured after 20 days of the stabilization diet. Mean platelet volume increased significantly during the period in which the volunteers consumed the salmon diet compared to the baseline diet (p<0.01), while the mean platelet levels decreased. Platelet aggregation (PA) was measured in platelet rich plasma before, during, and after the salmon diet using collagen, ADP, arachidonic acid (AA), and thrombin agonists. The PA threshold for ADP was significantly increased for the subjects on the salmon diet (p<0.05). No change in the PA threshold was detected for collagen or thrombin. The PA threshold for AA was unchanged also, but the platelets in subjects consuming the salmon diet had a prolonged time to maximum aggregation (p<0.01) with this reagent compared to platelets from men on the stabilization diet. Plasma, red cell, and platelet total FA composition was determined by capillary GLC. While the men consumed the salmon diets, there were marked increases (3 to 10-fold) in the EPA and DHA levels in all blood components with concomitant decreases in linoleic acid and AA levels. Thus, a salmon diet, high in n−3 FA, did not influence the bleeding times, but it decreased the sensitivity of platelets to ADP and AA, increased the mean platelet size, decreased the platelet count, and changed the FA composition of the plasma, RBC and platelet membrane lipids.

Increased dietary arachidonic acid enhances the synthesis of vasoactive eicosanoids in humans

Data on the effect of dietary arachidonic acid (AA) (20∶4n-6) on the synthesis of thromboxane and prostacyclin (PGI2) in humans are lacking. We measured the effect of 1.5 g/d (ca. 0.5 en%) of 20∶4n-6 added isocalorically to a stabilization (low-AA) diet on the excretion of 11-dehydrothromboxane B2 (11-DTXB2) and 2,3-dinor-6-oxo-PGF1α (PGI2-M). In a crossover design, 10 healthy men, living in a metabolic unit, were fed a diet (low-AA) containing 210 mg/d of 20∶4n-6 for 65 d and an identical diet (high-AA) that contained 1.5 g/d of additional 20∶4n-6 for 50 d. Three-day urine pools were collected at the end of each dietary period and analyzed for eicosanoids by gas chromatography-electron capture negative ion-tandem mass spectrometry. Mean excretion of 11-dehydrothromboxane B2 was 515±76, 493±154, and 696±144 ng/d (SD; n=10) during the acclimation (15 d) low-AA diet and high-AA diet periods, respectively (41% increase from low-AA to high-AA diet, P=0.0037); mean excretion of PGI2-M was 125±40, 151±36, and 192±55 ng/d (SD; n=10) during acclimation (15 d) low-AA and high-AA diets, respectively (27% increase from low-AA to high-AA diets; P=0.0143). Thus, both the metabolites of thromboxane and PGI2 increase on the high-AA diet. Furthermore, both indicated changes in metabolite excretion may be associated with measurable effects on several physiologically significant cellular functions, such as platelet aggregation in vivo and inflammation in response to immune challenges.

The urinary excretion of thiobarbituric acid reactive substances and malondialdehyde by normal adult males after consuming a diet containing salmon

In this study we investigated the output of thiobarbituric acid reactive substances (TBARS) and malondialdehyde (MDA), as thiobarbituric acid (TBA)-MDA adduct, in the urine from subjects eating a diet in which the only source of n−3 long-chain, polyunsaturated fatty acids was fresh salmon. Nine healthy men, ages 30–65, were confined in the United States Department of Agriculture Western Human Nutrition Research Center, San Francisco, CA, for 100 d; food intake and exercise levels were controlled. All subjects were placed on a stabilization diet (StD) for 20 d, then six were fed the salmon diet for 40 d. The others remained on the StD. The groups switched diets for the last 40 d. Both diets were isocaloric (16% protein, 54% CHO and 30% fat by energy %). The salmon diet contained 7.5% of calories from n−6 fatty acids (FAs) and 2% from n−3 FAs, primarily eicosapentaenoic acid and docosahexaenoic acid in a 50∶60 ratio, while the StD contained 7.5% from n−6 FAs and <0.3% n−3 FAs (with presumably no significant amounts of C20 or C22 n−3 FAs). Twenty-four hour urinary output was collected, and 2% 3−d pool samples prepared for analysis of urinary TBARS and the TBA-MDA adduct. The total urinary output of each individual varied considerably, and on a daily basis the concentration of autoxidation products in an individuals urine varied also. However, the mean daily output (in μmoles TBA-MDA equivalents/day) at the end of the salmon diet feeding period was significantly greater (7.05±1.33 TBARS,P<0.05; and 7.07±1.73 TBA-MDA adduct,P<0.01) compared to when the subjects were eating the StD (5.65±1.09 TBARS and 4.65±0.76 TBA-MDA adduct). When the TBARS and TBA-MDA adduct values were normalized relative to creatinine output (in nmoles TBA-MDA equivalents/μmole creatinine), the data achieved even greater statistical significance. The mean output of the group eating the salmon diet was 0.478±0.076 for TBARS (P<0.01) and 0.476±0.082 for the TBA-MDA adduct (P<0.001)vs. 0.345±0.059 for TBARS and 0.283±0.041 for the TBA-MDA adduct when the subjects were consuming the StD. Thus, the consumption of cooked fish may increase ones exposure to MDA and other autoxidation products, compounds that may be carcinogenic or mutagenic.

Studies on the hydrogen belts of membranes: II. Non-electrolyte permeability of liposomes of diester, diether, and dialkyl phosphatidylcholine and cholesterol

We have postulated the existence of lipid-lipid and protein-lipid hydrogen bonding in the hydrogen belts of membranes, i.e., the regions of hydrogen bond acceptors (carbonyl oxygens of esters and amides) and hydrogen bond donors (hydroxyls of cholesterol, sphingosine, proteins, water). To assess the possible effects of modifications of the hydrogen belts on membrane permeability, we prepared a diester phosphatidylcholine and two analogs lacking carbonyl oxygens, a diether and a dialkyl phosphatidylcholine, care being taken to synthesize lipids of identical efficient hydrophobic chain length. Relative permeation rates for glycerol and urea were determined by osmotic swelling of liposomes containing the phospholipids alone or with an equimolar quantity of cholesterol, with 4 mole % of dioleylphosphate added. The permeation rates of both solutes were similar for all three lipids, with Arrhenius activation energies ΔE* around 16 kcal/mole. Cholesterol reduced the permeability of all three membranes. The activation energy ΔE* of permeation did not change for diester and dialkyl phosphatidylcholine with cholesterol, but was lower by about 5 kcal/mole for the diether lipid with cholesterol. This corresponds to a reduction in the entropy of activation ΔΔS*∼-16 cal/mole/degree. We interpret the results as supporting the hypothesis of interaction between cholesterol hydroxyl and phospholipid carbonyl.

Effect of a salmon diet on the distribution of plasma lipoproteins and apolipoproteins in normolipidemic adult men

The effects of n−3 fatty acids on plasma lipids, lipoproteins and apoproteins have usually been studied in humans after feeding of purified fish oil. This study describes the effect of a natural diet, containing salmon as the source of n−3 fatty acids, on these parameters as compared to a diet very low in n−3 fatty acids. The subjects were nine normolipidemic, healthy males who were confined to a nutrition suite for 100 days. During the first 20 days of the study the participants were given a stabilization diet consisting of 55% carbohydrates, 15% protein, and 30% fat. The n−3 content of this diet was less than 1%, and it contained no 20- or 22-carbon n−3 fatty acids. After the stabilization period the men were split into two groups, one group continued on the stabilization diet while the other received the salmon diet that contained approximately 2.1 energy percent (En%) of calories from 20- and 22-carbon n−3 fatty acids. Both diets contained equal amounts of n−6 fatty acids. This regime continued for 40 days, then the two groups switched diets for the remainder of the study. Plasma triglycerides were lowered significantly (p<0.01) and high density lipoprotein cholesterol (HDL-C) was significantly elevated (p<0.01) after the men consumed the salmon diet for 40 days. The very low density lipoproteins (VLDL) were lowered, but the trend did not reach statistical significance during the intervention period. The total plasma cholesterol, total low density lipoprotein (LDL) and the total high density lipoprotein (HDL) levels were not influenced by the salmon diet. Within the HDL fraction, however, the larger HDL2 subfractions were significantly elevated (p<0.002), and the smaller, more dense HDL3 was lowered (p<0.002) by the salmon diet. These significant changes were detected by analytic ultracentrifugation and confirmed by gradient gel electrophoresis. Analysis of the apolipoproteins (apo) AI, AII, B, and E, and Lp(a) indicated only significant lowering of apoAI, consistent with the increased HDL2, which is higher in cholesterol but lower in the major HDL apolipoprotein, apoAI. Thus, the purported beneficial cardiovascular effects of consumption of n−3 fatty acids by humans may, in part, be attributable to changes in the HDL distribution,i.e., the lowering of the more dense HDL3 and the elevation of the larger, less dense HDL2.