Essential Fatty Acid Deficiency Profiles in Idiopathic Immunoglobulin A Nephropathy

Essential Fatty Acid Deficiency Profiles in Idiopathic Immunoglobulin A Nephropathy Ralph T. Holman, PhD, Susan B. Johnson, BS, Douglas Bibus, BS, Dorothy C. Spencer, and James V. Donadio, Jr, MD • The profiles of faHy acids (FAs) of plasma phospholipids (the compartment reflecting the essential FA status of tissue lipids), nonesterified FAs (the precursor pool for autacoid synthesis), urine protein excretion, and glomerular filtration rate were measured before and after supplementation with fish oil in 15 patients with immunoglobulin A nephropathy. In the FA profiles, there was deficient 18:3w3 (a-linolenic acid), the parent compound of w3 polyunsaturated FA, and deficient chain elongation products of both w3 and w6 polyunsaturated FAs with replacement by saturated and monounsaturated short-chain, odd-chain, and branched-chain FAs, producing significant loss of w3 FA. These alterations indicate nutritional or functional (w3) and metabolic (w6) deficiencies. Additionally, the mean melting point of the FAs was significantly increased, implying an inherent decrease in cell membrane fluidity. Enhancement of 20: 5w3 (eicosapentaenoic acid) and 22:6w3 (docosahexaenoic acid) and suppression of 20:4w6 (arachidonate) after supplementation with fish oil were accompanied by important decreases in proteinuria and improved glomerular filtration rate. Omega-3 polyunsaturated FAs may favorably influence immunoglobulin A nephropathy through a modulation of the pathologic actions of the w6 eicosanoids and other diverse actions on various mediators produced by an initial immune injury.

© 1994 by the National Kidney Foundation, Inc. INDEX WORDS: Polyunsaturated fatty acids; faHy acid profile; w3 faHy acids; mean melting point; immunoglobulin A nephropathy.

T

HE ESSENTIAL fatty acids (EFAs) are polyunsaturated fatty acids (PUFAs) that are ubiquitous in cellular membranes of animals, including humans. The PUFAs are nutrients required for normal function 1,2 and are diminished in the lipids of membranes in nutritional EFA deficiency. The content of PUFAs in lipids of vital organs, including the kidney,3 responds to changes in dietary EFAs of both the w6 and w3 families. The profiles offatty acids (FAs) of phospholipids (PLs) of plasma reflect the EFA status of tissue lipids in human EFA deficiency states,4 and many human diseases are accompanied by abnormalities, generally deficiencies, of the PUFA profile in plasma PLS. 5 From the Hormel Institute, University o/Minnesota, Austin, MN; and the Division 0/ Nephrology and Infernal Medicine, Mayo Clinic and Mayo Foundation, Rochester, MN. Received December 9, 1993; accepted in revised form December 14, 1993. Supported in part by Scotia Pharmaceuticals, Ltd 0/ England, the Mayo Foundation Research Fund, and the Hormel Foundation. Presented at the Third International Congress on Essential Fally Acids and Eicosanoids, Adelaide, Australia, March 1992, and at the 42nd Annual Meeting o/the National Kidney Foundation, Baltimore, MD, November 1992. Address reprint requests to James V. Donadio, Jr, MD, Mayo Clinic, 200 First St SW, Rochester, MN 55905. © 1994 by the National Kidney FOllndation, Inc. 0272-6386/94/2305-0003$3.00;0 648

Recently, the efficacy of fish oil supplementation has been tested in experimental nephritis and human renal diseases, including immunoglobulin A (lgA) nephropathy, in several laboratories, and the results have varied. 6 Therefore, the profiles ofFAs in plasma PLs (the compartment reflecting structural lipid components of membranes) and nonesterified fatty acids (NEFAs) (the precursor pool for the synthesis of autacoids) of a series of patients with IgA nephropathy were measured with our method to assess their EFA status. The patients were treated with w3 PUFA, and the effects of supplementation were assessed. This study was conducted during the formative period of the Mayo Nephrology Collaborative Group, a multicenter consortium of clinical nephrologists organized to conduct clinical trials in patients with chronic glomerular diseases for which no proven treatments are available. Observations made in a pilot study were encouraging. 7 We present the profiles and effects of fish oil supplement on the profiles of plasma PLs and NEFAs. PATIENTS AND METHODS Patients with idiopathic 19A nephropathy included in the study had a renal morphologic diagnosis ofigA nephropathy established at the Mayo Clinic between August 1987 and July 1988. Patients were excluded if there was clinical or serologic evidence of disease associated with 19A nephropathy, such as systemic lupus erythematosus. chronic liver disease. antiglo-

American Journal of Kidney Diseases, Vol 23, No 5 (May), 1994: pp 648-654

649

ESSENTIAL FATTY ACID DEFICIENCY PROFILES Table 1. Description of Patients With Idiopathic Immunoglobulin A Nephropathy Coo,

Patient No.

Sex

Age (yr)

Duration of Disease (mo)*

Proteinuria (g/24 hr)

(mL/min/1.73 m2 )

Hypertension at Entry Into Studyt

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

M M M F M M M M M M M M F M M

47 33 34 24 61 36 55 45 44 39 36 41 38 55 20

2 72 14 4 7 36 24 21 4

2.90 2.04 5.05 1.25 2.60 2.45 1.38 3.14 1.97 3.31 3.28 1.15 5.72 4.33 5.50

85 26 58 59 30 51 65 76 105 72 83 43 21 45 56

Yes Yes Yes Yes Yes No No Yes No No Yes Yes No Yes No

7 17 5 4 9

* The interval between renal biopsy diagnosis and analysis of PUFA.

t

The patient was considered to have hypertension if blood pressure was> 140/85 mm Hg or if the patient was being treated for hypertension as defined in the text.

merular basement membrane glomerulonephritis, or HenochSchiinlein purpura. s The study was approved by the Institutional Review Board of the Mayo Clinic and Foundation. Blood was drawn from 15 patients at the time baseline renal function studies were performed and before fish oil supplementation to measure baseline profiles of FAs of the plasma PLs and NEFAs. The clearance of iothalamate (C io') was used to estimate glomerular filtration rate, the measure of renal function used in our study.9 The patients were then given 6 to 12 g of menhaden oil (MaxEPA) daily for I year. Blood samples were again drawn at 6 weeks, 6 months, and 12 months, and plasma PL and NEFA profiles were measured to assess the effects of fish oil supplement on those profiles. Cio, and 24-hour urine protein were also measured at these intervals. Control subjects were 100 omnivores recruited from the students and staff of the U ni versity of Minnesota. Each completed a health and diet history and were found to be healthy by a physician. IO The group included 37 men and 63 women with a mean age of 29.3 years. An earlier study found no significant correlation between FA composition of plasma PLs and either age or sex. II

Methods ofAnalysis Plasma samples were prepared from the blood samples at the Mayo Clinic and kept frozen until analyzed at the Hormel Institute. Lipids were extracted from 2-mL samples with 6 mL of chloroform/methanol (2: I vol/vol) and were centrifuged. The aqueous layer was drawn off, and the extract was filtered, dried under N 2 , and redissolved in 100 ILL of chloroform. For each sample, PLs and NEFAs were separated by thin-layer chromatography. and each lipid was converted to methyl esters and analyzed for FA composition by methods described previously. 12

Presentation of Data Lipids occur in vesicles and membranes. the metabolism of which occurs at the surfaces. The kinetics of lipid metab-

olism are governed by concentrations at two-dimensional surfaces rather than by three-dimensional concentrations in aqueous medium. Therefore. the percentage of a FA in a lipid class expresses the concentration of a lipid substrate better than the concentration in the aqueous medium does. Normalcy ratio is the ratio of the experimental value to the control value; it expresses the relative difference in concentration of a FA substrate in the diseased condition compared with normal values. The normalcy ratio was calculated from the mean values for the two groups. and significance was calculated for the difference between the two groups. When plotted on a logarithmic scale. equal shifts of increase or decrease are equalfold changes in concentration. Probabilities were calculated with Student's (-test. Because the 25 FAs that were measured were interdependent. association could occur randomly at the probability level of P < 0.04. For this reason. single significance values at this level were not considered meaningful. The triene to tetraene ratio. the first-used index ofEFA status. is 20:3w9 to 20:4w6. I3 The double-bond index is the average number of double bonds per FA. The mean melting point (MMP) of the FA of a lipid is calculated as the sum of the products of (mole fraction) X (melting point) for each FA present. A similar calculation for mean chain length of the FA of a lipid is expressed in carbon atoms. In our experience. the entire FA profile is more revealing than a single index of EFA status. Moreover, the triene to tetraene ratio is unchanged in many instances of disease in which there are profound changes in profile. In our opinion. the MMP is the best single index of fluidity. because it is based directly on a physical property related to fluidity and on all the FAs present in a lipid. 14

RESULTS

The descriptions of the patients with 19A nephropathy enrolled in our study are summarized in Table 1. There were 13 men and two women (median age, 38 years; age range, 20 to 61 years).

650

HOLMAN ET AL

Included in the study were II patients who recei ved fish oil in a pilot program 7 and the first four patients who received fish oil in an ongoing clinical trial to determine the effects of fish oil in IgA nephropathy in patients at risk for developing progressive renal failure . IS-I S The 24-hour proteinuria was 1.15 to 5.50 g (mean, 3.07 g) and the 24-hour Cio! was 21 to 105 mL/min/1.73 m 2 (mean, 58 mL/min/1.73 m 2). The nine patients who had hypertension all received the angiotensin-converting enzyme inhibitor enalapril maleate to achieve blood pressure control (defined as lowering the average systemic blood pressure to :::;140/85 mm Hg). The w6 PUFA profile of plasma PLs, the FA pool representing structural lipid, is shown in the left column of Fig I. The profile was significantly deficient in w6 PUFAs beyond the first chain elongation steps in the metabolic cascade. Chain elongation of 18 :2w6 to 20:2w6 , a minor product that is not further desaturated, was suppressed to 63% of its normal level (0.47%; P < 0.001). The precursor 18:2w6 was normal, and the first desaturation product, 18:3w6, was not significantly high. However, the chain elongation product of 18:3w6, 20:3w6 , was suppressed to 82% of its normal value (3.41 %; P < 0.01). Moreover, 20: 4w6 was not significantly different from normal, but its elongation product, 22:4w6, was strongly suppressed to 53% of normal (0.76%; P < 0.001) and its subsequent desaturation product, 22:5w6, was suppressed to 64% of normal (0.60%; P < 0.00 I). These four suppressions of minor constituents suggested a diminished capacity for chain elongation. Total w6 acids were slightly but significantly low at 96% of normal content

PUFA

(42.1 %; P < 0.001). Thus, by this measure, the concentration of w6 acids collectively were only slightly diminished. The w3 PUFA profile showed that 18:3w3, the precursor member of the w3 family, was low at 77% of normal (0.21 %; P < 0.05) of total PL FAs. In addition, 20:5w3 was slightly but not significantly increased to 113% of normal (0.59%). Its elongation product, 22:5w3, was suppressed to 77% of normal (1.13%; P < 0.0 I), and 22:6w3, the most abundant w3 PUFA, was suppressed to 75% of normal (3.59%; P < 0.01). Thus, total w3 PUFAs were suppressed to 80% of their normal content (5 .53%; P < 0.01). This is a 20% loss of w3 acids compared with a 4 %loss of w6 acids. The profile of nonessential FA, shown in the right column of Fig I, revealed progressive discrimination against saturated F As of increasing chain length. This phenomenon occurs in multiple scierosis,I2 during pregnancy,14 and in several diseases studied in our laboratory (not published). The same phenomenon was seen among the monoenoic acids in our study and contributed to a shortening of the mean chain length of the composite FAs of PLs. In IgA nephropathy, the mean chain length of plasma PLs was 17.33 ± 0.03 carbon atoms (normal, 17.50 ± 0.04 carbon atoms). This seemingly small change was highly significant (P < 0.001). The MMP was 19.1°C ± 2.3°C compared with 14.8° ± 2.5°C in controls, an increase of 4.3 ° (P < 0.001). The serial PUFA profile for plasma PLs is shown in Fig 2, in which the more revealing items of the PUFA profile are shown for baseline, 6 weeks, 6 months, and 12 months in a pseudothree-dimensional plot. Supplementation with

Non-PUFA

18:2W6 18:3W6 20:2W6 20:3W6 20:4W6 22 :4W6 22:5W6

14:0 16:0 18:0 20:0 22:0 24:0 l: Sat.

l:W6

16:1W7

18:3W3 20:5W3 22:5W3 22 :6W3

18:1W9 20:1W9 22:1W9 24:1W9 l: Mono Branch Odd

l:W3 20:3W9

l: PUFA 0.1

1.0

10.0

0.1

1.0

10.0

Fig 1. Essential PUFA and nonessential PUFA (NonPUFA) profiles of plasma PLs of 15 patients with IgA nephropathy compared with those of 100 control subjects measured at the beginning of the study. Normalcy ratio is the ratio of the experimental value to the control value; it expresses the relative difference in concentration of a FA substrate in the IgA nephropathy group compared with normal values. The vertical axis is normal. The open bars = nonsignificant change; the stippled bars = P < 0.05; the oblique striations = P < 0.01; and the solid bars = P < 0.001.

ESSENTIAL FATTY ACID DEFICIENCY PROFILES

651

Fig 2. Serial PUFA profiles, monounsaturated (MONO) and saturated (SAT) acids of plasma PLs in 15 patients with IgA nephropathy presented as mean percentages of total FAs, double-bond index (OBI) presented as the average number of double bonds per FA, and MMP (DC) at baseline (day 0) and after 6 weeks, 6 months, and 12 months of fish oil supplementation. The value for the control population is shown for each parameter by the oblique line through its profile.

fish oil containing w3 PUFA dramatically enhanced the content of eicosapentaenoic acid (EPA; 20:5w3), docosahexaenoic acid (OHA; 22: 6w3), and total w3 PUFA of plasma PLs. The content of 18:3w3, the precursor of the series, was consistently less than the normal value (0.21 %) of total FAs ofPLs, perhaps because fish oil is not rich in this member of the family and because incorporation of the long-chain members of the w3 family into PLs is favored over that of 18:3w3. The content oflinoleic acid (l8:2w6) was normal at day 0, and supplementation with fish oil suppressed it slightly at 6 weeks and at 6 months, but by 12 months it had returned to normal. Arachidonic acid (20:4w6) was very close to normal at day 0 and was suppressed by the long-chain w3 supplement, as expected from the known suppression of w6 metabolism by w3 supplements. 19 The total w6 PUFA reacted in a similar fashion. Monounsaturated acids were slightly less than the control value at day 0 and diminished with w3 supplementation, as expected. Saturated acids in PLs at day 0 were somewhat increased and remained so during w3 supplementation. The double-bond index, less than normal at day 0, increased to normal with w3 supplementation. The elevated MMP remained high at each study interval after supplementation. The fish oil supplement did not restore this index to normal. The profile for plasma NEFA, the FA pool for eicosanoid synthesis, showed no significant changes among the w3 PUFA. However, 18:2w6 was suppressed to 81% of normal (18.4%; P < 0.001), 20:3w6 to 32% of normal (0.52%; P < 0.01), and 20:4w6 to 34% of normal (2.42%; P

< 0.001). These losses of PUFA were compensated by increases of 14:0 (P < 0.001), 16:0 (P < 0.001), 20:0 (P < 0.01), and 16: lw7 (P < 0.01), respectively. The MMP increased to 29.2°C from a norm of 24.1 0c. Fish oil had little effect on total w6 PUFA. For the w3 PUFA, the greatest responses to fish oil supplementation were in the 20:5w3 and 22:6w3 levels. Measurements of protein excretion and Cot are shown in Figs 3 and 4. Linear regression analysis of these data indicated that the decrease in protein excretion and the increase in glomerular filtration rate were significant within 1 year after treatment, and the magnitude of change in both variables was similar in both normotensive and hyperten-

Fig 3. Twenty-four-hour urine protein values for 15 patients with IgA nephropathy at baseline and after supplementation with fish oil and antihypertensive treatment with enalapril maleate (solid lines) and fish oil alone (broken lines). For each person, the regression of proteinuria (y) on the natural log of time was an approximately linear relationship, and in all 15 patients the changes were negative. The mean slope (short horizontal lines) decreased 2.15 ± 0.38 (SEM) g/24 hour (range, -5.0 to -2.0 g/24 hour) during the 12-month treatment period (P < 0.0001) and the slopes were similar for both hypertensive and normotensive subjects.

652

~ '. ' 20

:;;;:.:;;--------:

!:e~-g ro

~-~----------~ 6 wk 6 mo 12 mo Time

Baseline

Fig 4. Clearance of iothalamate (Clot) values for 15 patients with IgA nephropathy. The mean clearance slope (short horizontal lines) increased 9.6 ± 2.3 (SEM) mL/ min/1.73 m2 (range, -2 to 31 mL/min/1.73 m2 ) during the 12-month period, indicating improved renal function (P < 0.001). The changes were similar in both hypertensive (solid lines) and normotensive (broken lines) subjects.

sive subjects. Note that proteinuria changes showed a linear response and continued to decrease with time, although at a slower rate after 6 weeks, whereas Ciot remained relatively unchanged from 6 weeks to 12 months. Changes over 12 months in Ciot and proteinuria did not directly correlate significantly with changes individually in 20:5w3, 22:6w3, or 20: 4w6, or with changes in total w3 PUFA or in estimated MMP of plasma PL using the Pearson and Spearman correlation tests. DISCUSSION

At baseline, decreased levels of a-linolenic acid (18 :3w3), the parent compound of w3 PUFA; the suppressed chain elongation products of EPA (20: 5w3), 22:5w3, and DRA (22:6w3); and th~ low total w3 PUFA were undoubtedly related to a nutritional or functional deficiency of w3 PUFA in our patients who were from the midwest United States and who were not accustomed to a diet rich in marine lipids. Normal levels of linoleic acid (18:2w6), the parent compound of w6 PUFA, and the suppression of w6 elongation products (20:2w6, 20:3w6, 22:4w6, and 22:5w6) and total w6 PUFA with replacement by saturated and monounsaturated FAs of shorter chain lengths (CWC 1S) and odd-chain and branched F As appear to be of metabolic origin. When long-chain PUFAs are lost through metabolic or nutritional deficiencies, the components providing the most unsaturation and the lowest melting point (most "fluidity") to the composite F As of structural lipid are lost. Replacement by saturated and monounsaturated

HOLMAN ET AL

acids of lesser chain length (CWC 1S) and by small proportions of odd-chain and branched-chain F As contributes to shortening the mean chain length and decreases the MMP. However, in these PL profiles, the enhanced proportions of shortchain, branched-chain, and odd-chain F As were insufficient to maintain a low MMP and normal fluidity. In our patients, MMP was significantly increased above normal in both of the lipid compartments tested. The MMP was comparably increased to levels found in various other unrelated disorders, including insulin-dependent and noninsulin-dependent diabetes mellitus, rheumatoid arthritis, pre-eclampsia, and coronary occlusion, and in renal transplant recipients receiving cyclosporine.1 4 The MMP of component FA of PLs is now considered the single best index of membrane fluidity. "Membrane fluidity" refers to the physical state of all acyl chains comprising the cell membrane bilayer structure. The behavior of the acyl chains is also influenced by other components of the cell membrane, including proteins and cholesterol. Increased MMP, such as in our patients, implies inherent decreased membrane fluidity, a diminished temperature range below body temperature at which membranes remain fluid. The relationship of altered membrane fluidity to the pathophysiology of IgA nephropathy and other nutritional, physiologic, ethnic, and disease states (some listed above) studied in our laboratoryl4 are unidentified but may be part of the immunologic, thrombotic, and inflammatory responses in many of these conditions. In the serial measurements of PUFA profiles after fish oil supplementation, there was a remarkable and expected enhancement of EPA (20: 5w3), DRA (22:6w3), and total w3 PUFA, with suppression of arachidonic acid (20:4w6) and total w6 PUFA. The MMP was unaffected. The enhancement of plasma PLs and NEFAs with EPA and DRA was accompanied by a significant decrease in 24-hour urine protein excretion and improvement in glomerular filtration rate that were similar in patients who were normotensive and treated with fish oil only as well as in those with hypertension who were treated with both fish oil and enalapril maleate. However, the improvement in renal characteristics did not directly correlate significantly with changes in 20:5w3, 22: 6w3 , 20:4w6, total w3 PUFA, or MMP in plasma PLs. Despite the lack of significant correlations

653

ESSENTIAL FATTY ACID DEFICIENCY PROFILES

between these measures of kidney function and these indicators of EFA deficiency with supplementation with menhaden oil, trends that may be significant with testing greater numbers bfpatients were apparent. Nevertheless, the FA compositional status of the plasma PLs of patients with IgA nephropathy indicated pronounced deficiencies of both w3 and w6 EFAs, indicating a nutritional problem that must be addressed. Omega-3 PUFA may favorably influence IgA nephropathy by exerting diverse actions on many potential mediators that stem from an initial glomerular immunologic injury.6 These mediators produce altered glomerular hemodynamics, mesangial contraction, and inflammation. Interruption of these pathways by 20:5w3 and 22:6w3 may ameliorate glomerular injury. Eicosapentaenoic acid and DHA undergo biologic transformation and produce trienoic eicosanoids, including prostacyclin (prostaglandin h), which has vasodilatory and platelet antiaggregating properties, and thromboxane A3, which has no vasoconstricting or platelet aggregating properties. Thromboxane A2 induces renal vasoconstriction and mesangial contraction. 20,21 Prostaglandin 13 could ameliorate both of these renal effects in the presence of decreased amounts of active thromboxane A2 replaced by inactive thromboxane A3. The net result of w3 FA on eicosanoid production is to change the hemostatic balance toward a more vasodilatory state with less platelet aggregation, Production of inactive leukotriene B5 from EPA, the preferred substrate for the lipoxygenase pathway,22,23 could reduce leukocyte and sensitized monocyte involvement in inflammation, such as occurs in immune-mediated glomerular and interstitial injury.23 Decreased production of interleukin-l and tumor necrosis factor in stimulated peripheral blood monocytes also may be related to decreased synthesis ofleukotriene B4.24 This effect may further alleviate glomerular inflammation through inhibition of cytokine production by mesangial and endothelial cells. 25 Furthermore, vascular damage in the kidney may be modified by the influence of 20:5w3 and 22: 6w3 on blood rheology,26 inhibition of platelet aggregation,27 and decrease of elevated levels of plasma lipids.28 Thus, after EPA and DHA are incorporated into plasma and cell membrane PL pools, they may modify various potential mediators and affect important biologic changes that,

in turn, may alleviate the glomerular injury and its consequences, The data presented here indicate an association of IgA nephropathy with significantly altered profiles of FA of plasma PLs. The patterns characteristic of EFA deficiency in tissue PLs in animals and humans are also apparent in plasma PLS.4 Thus, the analysis of plasma PLs is an indicator of tissue EFA status. The level of EFA deficiency, as indicated by MMP based on all FAs measurable in PLs, is on par with that found in various disorders.14 In IgA nephropathy, the deficiencies in the PUFA pattern occur for several members of the w6 and w3 families of PUFAs, both required for normal composition and function of tissue membranes. Currently, it is not known whether chronic low-level nutritional deficiencies of these essential nutrients contribute to the pathogenesis of nephropathy or whether the disease induces the deficiencies by altered metabolism, Altered metabolism seems to be present in nephropathy, with a reduced ability to chain elongate the dietary EFAs, Chronic lowlevel nutritional deficiencies of w3 may likewise be associated with the continued injury of nephropathy. Our controls showed a broad range of content of individual EFAs in their plasma lipids, 10, II some having low contents of w6 and w3 EFA that may nutritionally predispose them to a distorted or faulty metabolism of EFAs. Treatment of IgA nephropathy should include enhanced and balanced intake of EFAs as well as correction of the more obvious features associated with the disease, such as hypertension,

REFERENCES 1. Holman RT: Essential fatty acid deficiency. Prog Chern Fats Other Lipids 9:275-348, 1971 2. Holman RT: Biological activities and requirements for polyunsaturated fatty acids. Prog Chern Fats Other Lipids 9: 607-682, 1971 3. RieckehoffI, Holman RT, Burr GO: Polyethenoid fatty acid metabolism. Effect of dietary fat on polyethenoid fatty acids of rat tissues. Arch Biochem 20:331-340, 1949 4. Paulsrud JR, Pensler L, Whitten CF, Stewart S, Holman RT: Essential fatty acid deficiency in infants induced by fatfree intravenous feeding. Am J Clin Nutr 25:897-904, 1972 5. Holman RT: Control of polyunsaturated fatty acids in tissue lipids. JAm Coll Nutr 5:236-265, 1986 6. Donadio JV, Jr: Omega 3 polyunsaturated fatty acids: A potential new treatment of immune renal disease. Mayo Clin Proc 66:1018-1028, 1991 7. Donadio JV, Jr, Holman RT, Holub RJ, Bergstralh EJ: Effects of omega 3 polyunsaturated fatty acids in mesangial 19A nephropathy. Kidney Int 37:255, 1990 (abstr)

654 8. Rodicio JL: Idiopathic IgA nephropathy. Kidney Int 25: 717-729,1984 9. Duarte CG, Elveback LR, Liedtke RR: Glomerular filtration rate and renal plasma flow, in Duarte CG (ed): Renal Function Tests. Boston, MA, Little, Brown, 1980, pp 29-47 10. Phinney SD, Odin RS, Johnson SB, Holman RT: Reduced arachidonate in serum phospholipids and cholesteryl esters associated with vegetarian diets in humans. Am J Clin Nutr 51 :385-392, 1990 II. Holman RT, Smythe L, Johnson S: Effect of sex and age on fatty acid composition of human serum lipids. Am J Clin Nutr 32:2390-2399, 1979 12. Holman RT, Johnson SB, Kokmen E: Deficiencies of polyunsaturated fatty acids and replacement by nonessential fatty acids in plasma lipids in multiple sclerosis. Proc Nat! Acad Sci USA 86:4720-4724, 1989 13. Holman RT: The ratio of trienoic:tetraenoic acids in tissue lipids as a measure of essential fatty acid requirement. J Nutr 70:411-417, 1960 14. Holman RT, Johnson SB, Ogburn PL: Deficiency of essential fatty acids and membrane fluidity during pregnancy and lactation. Proc Nat! Acad Sci USA 88:4835-4839, 1991 15. Hood SA, Velosa JA, Holley KE, Donadio JV: IgAIgG nephropathy: Predictive indices of progressive disease. Clin Nephrol 16:55-62, 1981 16. Nicholls KM, Fairley KF, Dowling JP, Kincaid-Smith P: The clinical course of mesangial IgA associated nephropathy in adults. Q J Med 53:227-250, 1984 17. D'Amico G: The commonest glomerulonephritis in the world: IgA nephropathy. Q J Med 64:707-727,1987 18. Kusumoto y, Takebayashi S, Taguchi T, Harada T, Naito S: Long-term prognosis and prognostic indices of IgA nephropathy in juvenile and in adult Japanese. Clin Nephrol 28:118-124,1987 19. Holman RT: Nutritional and metabolic interrelationships between fatty acids. Fed Proc 23:1062-1067, 1964 20. Sakr HM, Dunham EW: Mechanism of arachidonic acid-induced vasoconstriction in the intact rat kidney: Possible

HOLMAN ET AL

involvement ofthromboxane A2 . J Pharmacol Exp Ther 221: 614-622, 1982 21. Scharschmidt LA, Lianos E, Dunn MJ: Arachidonate metabolites and the control of glomerular function. Fed Proc 42:3058-3063, 1983 22. Lee TH, Mencia-Huerta J-M, Shih C, Corey EJ, Lewis RA, Austen KF: Effects of exogenous arachidonic, eicosapentaenoic, and docosahexaenoic acids on the generation of 5lipoxygenase pathway products by ionophore-activated human neutrophils. J Clin Invest 74:1922-1933, 1984 23. Weaver BJ, Holub BJ: Health effects and metabolism of dietary eicosapentaenoic acid. Prog Food Nutr Sci 12:111150, 1988 24. Endres S, Ghorbani R, Kelley VE, Georgilis K, Lonnemann G, van der Meer JWM, Cannon JG, Rogers TS, Klempner MS, Weber PC, Schaefer EJ, WolffSM, Dinarello CA: The effect of dietary supplementation with n-3 polyunsaturated fatty acids on the synthesis of interleukin-I and tumor necrosis factor by mononuclear cells. N Engl J Med 320: 265-271, 1989 25. Wiggins R, Downer G, Phan S: Inflammation to sclerosis: How long does it take? Implications for therapeutic approaches in inflammatory renal disease, in Davison AM (ed): Nephrology. Proceedings of the Xth International Congress of Nephrology. London, UK, Balliere Tindall, 1988, pp 532-537 26. Cartwright IF, Pockley AG, Galloway JH, Greaves M, Preston FE: The effects of dietary n-3 polyunsaturated fatty acids on erythrocyte membrane phospholipids, erythrocyte deforrnability and blood viscosity in healthy volunteers. Atherosclerosis 55:267-281, 1985 27. SkeaffCM, Holub BJ: The effect offish oil consumption on platelet aggregation responses in washed human platelet suspensions. Thromb Res 51: 105-115, 1988 28. Phillipson BE, Rothrock DW, Connor WE, Harris WS, Illingworth DR: Reduction of plasma lipids, lipoproteins and apoproteins by dietary fish oils in patients with hypertriglyceridemia. N Engl J Med 312:1210-1216,1985