PLATELET-MEMBRANE FATTY ACIDS, PLATELET AGGREGATION, AND THROMBOXANE FORMATION DURING A MACKEREL DIET

441

PLATELET-MEMBRANE FATTY ACIDS, PLATELET AGGREGATION, AND THROMBOXANE FORMATION DURING A MACKEREL DIET P. ROTH I. KURZMANN P. C. WEBER

W. SIESS

B. SCHERER B. BÖHLIG

Medizinische Klinik Innenstadt der Universität, Ziemssenstrasse 1, 8000 München 2, West Germany

Eicosapentaenoic acid (C20:5&ohgr;3), which is present in high concentration in certain salt-water fish, may reduce the incidence of cardiovascular disease in Greenland Eskimos by reducing platelet aggregation and adhesion.1 Changes in platelet and plasma fatty acids, platelet aggregation, and thromboxane (TX) formation were studied in 7 healthy White men who had been on a mackerel diet for 1 week. Platelet aggregation and TX synthesis after low-dose collagen stimulation in platelet-rich plasma were reduced. This could be due to the marked change of the ratio of C20:5 to arachidonic acid (C20:4&ohgr;6) in platelet membranes. When the men were on the mackerel diet, their C20:5 level in platelet membranes and in plasma increased and that of C20:4 dropped. Both fatty acids were released during platelet aggregation. The reduction in the amount of C20:4 liberated and/or a diminished conversion of C20:4 to TXA2 by competitive inhibition of the platelet cyclo-oxygenase by the released C20:5 could be responsible for the decreased platelet aggregation.

Changing the dietary fatty acids to those taken by Greenland Eskimos may prevent certain cardiovascular disorders in Western communities. Therefore, we studied the effect of a fish diet on plasma and platelet fatty-acid composition, thromboxane (TX) formation, and platelet aggregation in healthy White men. We chose a mackerel diet, because (1) this fish has a high content of C20:5, and (2) a change in the fatty-acid composition of human plasma lipids occurs after a mackerel diet.5

Summary

Introduction GREENLAND Eskimos have a low incidence of cardiovascular disorders, favourable lipid and lipoprotein levels in their plasma, and a bleeding tendency which may be the result of reduced platelet aggregation.’ Their intake of monounsaturated and co-3 polyunsaturated fatty acids is high and that of linoleic and arachidonic acids is low. Both their plasma and platelet lipids have 1 a similar fatty-acid pattern to that present in their diet.’ Eicosapentaenoic acid (C20:5w3) seems to be the most important of the dietary fatty acids because of its anti-aggregatory effect,2,3 which results most probably from competitive inhibition of formation of pro-aggregatory arachidonate (C20:4M6) metabolites.4 In addition, C20:5 may be converted to a prostacyclin (PGI) of the 3-series whose biological properties are similar to those

ofPGIz.3,4 16. Siegel S.

Nonparametric

Subjects

and Methods

7 healthy White men, who had received no medication in the preceding 3 weeks, were put on a mackerel diet. The diet consisted almost exclusively of mackerel in two forms (stewed and smoked). The daily mackerel intake was 500-800 g, which resulted in a consumption of 7-11 g ofC20:3. Practically all other forms of proteins and lipids were eliminated from the diet, but carbohydrate and fluid intakes were not restricted. A pilot study with 3 persons over 2 weeks had shown that considerable changes of plasma and platelet fatty acids occurred within 3 to 6 days after starting the diet. Therefore, studies on platelet aggregation, TX synthesis, and platelet fatty acids were done 3 days and 1 day before, and on the third and sixth day after starting, the mackerel diet. Citrated blood was taken from fasting subjects and immediately centrifuged at 150-200 g for 5 min to get platelet-rich plasma (PRP). To obtain a constant good platelet yield from blood (50-60%) the g numbers were varied for each sample and were lowered during the mackerel diet. Platelet-poor plasma (PPP) was prepared by centrifugation of resting blood at 2000 g for 15 min. For platelet aggregation studies and TX measurements platelets in PRP were adjusted to 250 000/pl PRP with autologous PPP. Platelet aggregation was measured according to the method of Born6 by the use of a ’LaborAggregometer’ (Fresenius, Bad Homburg, FRG). The change in light transmission after addition of aggregating agents was recorded as a percentage, 100% corresponding to a change of 0-14 optical density. The following aggregating agents were tested at pre-determined times between 40 and 90 min after blood sampling: collagen-1, 2, and lOf1-g/ml PRP (HormChemie, Munich, FRG); arachidonic acid-1-8mmol/l; adenosine diphospate (ADP)-l and 2-3mol/1; and 1-adrenaline-1-500 mol/1 (all from Serva, Heidelberg FRG). TX

measurements were

made in connection with the

plate-

let-aggregation studies. TXB2, the stable hydrolysis product of TXA2, was measured by radioimmunoassay, after acidification and organic solvent extraction of the samples. We used a specific TXB2-antibody (a gift from L. Levine, Brandeis University, Waltham, Mass., USA), standard TXB, (a gift from Dr J. Pike, Upjohn Company, Kalamazoo, USA), and 3H-TXB2

(New England Nuclear, Boston, USA).

statistics for the behavioural sciences. New York:

McGraw Hill, 1956. 17. Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. Analysis and examples. Br J Cancer 1977; 35: 1-39. 18. Ledger WJ, Sweet RL, Headington JT. Prophylactic cephaloridine in the prevention of postoperative pelvic infections in premenopausal women undergoing vaginal hysterectomy. Am J Obstet Gynecol 1973; 115: 766-74. 19 Bivens MD, Neufeld J, McCarty WD. The prophylactic use of keflex and keflin in vaginal hysterectomy. Am J Obstet Gynecol 1975; 122: 169-72. 20 Ohm MJ, Galask RP. The effect of antibiotic prophylaxis on patients undergoing vaginal operations. I. The effect on morbidity. Am J Obstet Gynecol

1975, 123: 590-96. 21 Freiman JA, Chalmers TC, Smith H, et al. The importance of beta, the type II error and sample size in the design and interpretation of the randomized

control trial. N Engl J Med 1978; 299: 690-94. 22. Chodak GW, Plaut ME. Systemic antibiotics for prophylaxis in gery: a critical review. J Urol 1979; 121: 695-99.

urologic sur-

23. Cruse P. Infection surveillance: Identifying the problems and the high-risk patient. South Med J 1977; 70 (suppl. 1): 4-7. 24. Mathews DD, Agarwal V, Ross H. A randomized controlled trial of a short course of cephaloridine in the prevention of infection after abdominal hysterectomy. Br J Obstet Gynæcol 1978; 85: 381-85. 25. Gilmore OJA, Martin TDM, Fletcher BN. Prevention of wound infection after appendicectomy. Lancet 1973; i: 220-22. 26. Bates T, Down RHL, Houghton MCV, et al. Topical ampicillin in the prevention of wound infection after appendicectomy. Br J Surg 1974; 61: 489-92. 27. Donovan IA, Ellis D, Gatehouse D, et al. One-dose antibiotic prophylaxis against wound infection after appendicectomy: a randomized trial of clindamycin, cefazolin sodium and a placebo. Br J Surg 1979; 66: 193-96. 28. Burke J. Effective period of preventive antibiotic action in experimental incisions and dermal lesions. Surgery 1961; 50: 161-68. 29. Ledger WJ, Gee C, Lewis WP. Guidelines for antibiotic prophylaxis in gynecology. Am J Obstet Gynecol 1975; 121: 1038-45. 30. Chodak GW, Plaut ME. Wound infections and systemic antibiotic prophylaxis in gynecologic surgery. Obstet Gynecol 1978; 51: 123-27.

442 TABLE I-FATTY ACID DISTRIBUTION

*p<001; Washed

(%)

IN PLATELET MICROSOMES BEFORE

(CONTROL) AND AFTER A 6-DAY MACKEREL DIET

SE in parentheses.

platelet microsomes

were

prepared

from washed platelet

suspensions

after

Platelet fatty-acid analyses were made on washed platelet suspensions after extraction and separation.7 Free fatty acids were methylated with methanolic boron trifluoride.’ Phospho-

lipid fatty acids were transmethylated with methanolic sulphuric acid.8 The fatty acids were analysed on a 419 Packard gas chromatograph according to previously reported methods.9 For quantitative determination of free and phospholipid fatty acids, heptadecanoic acid and di-heptadecanoic-lecithin (Nattermann, Cologne, FRG), respectively, were added as internal standards before extraction. Plasma triglycerides and cholesterol were measured by standard techniques. Protein determinations were made according to Lowry et a1.1o Statistical evaluation of the results was done by Student’s t test. Results

During the mackerel diet, the pattern of microsomal and total phospholipid fatty acid composition in platelets became similar to that of mackerel lipid composition-the proportion of monounsaturated and &ohgr;-3 polyunsaturated fatty acids increased, whereas the proportions of stearic acid, linoleic acid, arachidonic acid, and docosanoic acid decreased (table i). The changes in plasma free and phospholipid fatty acids were similar (data not shown). There were also changes in the absolute amounts of these fatty acids in total platelet phospholipids : C20:5 and C22:6 increased from 1-8±0-6 (SD) and 1.9±0.5 to 6-2±1.4 and 5-6±1-0 lJ-g/mg platelet protein, respectively (p<001); and C18:2 and C20:4 decreased from 5-6±1-1 and 25-3±2.3 to 3-3±0-8 and 18-3±1-3 µg/mg platelet protein, respectively (p<0.05). Therefore, the ratio of C20:5 to C20:4 in plasma and platelet membrane phospholipids changed considerably while the men were on the mackerel diet, and the change depended on the amount of mackerel ingested (fig. 1). The mean C20:5/C20:4 ratio was highest in plasma free fatty acids (3-6±0’4) and exceeded that

homogenisation

and differential

ultracentrifugation."

found in plasma phospholipids and platelet phospholipids by twofold and tenfold, respectively (fig. 1). In addition, there was a direct relation between the change in C20:5/C20:4 ratio in plasma free fatty acids with that in platelet membrane fatty acids (y=88x-021; The release of both C20:4 and r =0 .98; p<0.001). C20:5 from platelet membranes after thrombin-induced aggregation (table n) indicated a predominant role of these fatty acids in platelet function. When the volunteers were on the mackerel diet invitro platelet aggregation and TXB2 formation induced by low doses of collagen were significantly reduced (fig. 2, table III). The reduction of platelet aggregation correlated with diminished TXB2 formation (y=2-1x—13.3, r=0-83; p<0-01) (fig. 3) and was dependent on the C20:5/C20:4 ratio in platelet phospholipids (fig. 4). Platelet aggregation after 1-adrenaline, and the concurrent TXB2 formation as well as the TXB2 formation after high-dose collagen, decreased (table n), however, these differences did not reach significance. ADPinduced platelet aggregation tended to decrease but to a significant level only in some subjects. Platelet aggrega-

2-PRP (250x 109/1) stimulated (arrows) by ADP (25 µM). arachidonic acid (AA, 1-8 mM), and collagen (Coll., 1 and 10 µg/m1) under control conditions (left) and after a week ona

Fig.

Fig. 1--Change in C20:5/C20:4 ratio in plasma (closed circles) and platelet (open triangles) phospholipids produced by the mackerel diet. Values given are mean + SE.

mackerel diet (right).

The

amounts

ng/ml PRP.

of

TXB2 produced during aggregation

are

given =

443 TABLE II-RELEASE OF

C20:4 AND C20:5 UPON PLATELET 6-DAY MACKEREL DIET

AGGREGATION AFTER A

*p<0.05. SE in parentheses. The release was measured after stimulation of platelet suspensions with thrombm (150 NIH/ml, Roche, Switzerland). To block oxygenation of the released fatty acids, the platelet suspensions were preincubated with 30 µmol/1 eicosatetraynoic acid." Detection limit: 0.04 µg/mg protein.

tion and TXBz-synthesis induced by exogenous arachidonic acid were not changed (table in, fig. 2), which indicated that platelet cyclo-oxygenase activity itself was not altered by the mackerel diet. There was also a reduction in both plasma cholesterol (from 193±20[SD] mg/dl to 170±23) and plasma triglycerides (from 63±27 [SD] mg/dl to 33±13); however, these changes did not reach

Fig. 3-Correlation between TXB2 formation and platelet aggregation upon stimulation of PRP with 1 µg/ml collagen when subjects were on a mackerel diet. -

significance. Discussion A one-week mackerel diet produced considerable changes in the fatty-acid pattern in plasma and platelets from healthy White men. The fatty-acid composition in platelet membranes (that is microsomal lipids and total phospholipids) after 6 days of this fish diet was very similar to that of total platelet lipids of Greenland Eskimos.’ Platelet aggregation and TX synthesis induced by low-dose collagen stimulation were significantly reduced by the mackerel diet. The mechanisms by which the changes in plasma and platelet fatty acids reduce platelet aggregation induced by physiological stimuli are unknown. Since prostaglandins influence platelet aggregation, the changes in C20:4 and C20:5, the precursor fatty acids of prostaglandins of the two and three series, seem to be the most important ones: C20:4 promotes platelet aggregation by forming the pro-aggregatory TXA2, and C20:5 inhibits platelet aggregation in vitro,2,3 most probably by inhibiting platelet cyclo-oxygenase.4 We found a decrease of C20:4 and an increase of C20:5 in plasma and platelet fatty acids, which pro-

TABLE III-PLATELET AGGREGATION AND

TXB2-FORMATION

4-Relation between platelet aggregation induced by 1 and the C20:5/C20:4 ratio in platelet phospholipids before (open circles) and after a mackerel diet (closed circles) in 6 subjects.

Fig.

µg/ml collagen

duced a raised C20:5/C20:4 ratio. The value of this ratio depended on the duration of the mackerel diet and the amount of mackerel ingested. The linear positive correlation for this ratio between plasma free fatty acids and platelet membrane fatty acids supported previous work that plasma free fatty acids influence directly the fatty-acid composition of platelet membranes.12 C20:4

IN PRP BEFORE

(CONTROL) AND 3 AND 6 DAYS AFTER STARTING THE

MACKEREL DIET

SE in parentheses. *p<001, p<0.02 compared with control. Platelet aggregation is measured as % change of light transmission; and TXB2-formation in ng/ml. PRP contained 250x 109 platelets/1.

444

and C20:5 in platelets were found predominantly as esters in membrane phospholipids; only trace amounts were detected as free fatty acids13 (table n). The immediate effects of stimulation of platelet aggregation by physiological aggregating agents such as collagen or thrombin are the activation of a platelet phospholipase and the selective release of C20:4 from membrane phospolipids;14 but with the mackerel diet C20:5, obviously incorporated into the platelet membranes while the subjects were on that regimen, was released as well. The liberated C20:5 could competitively inhibit platelet cyclo-oxygenase,2,3 as has been shown in vitro.4 Free C20:5 in plasma may also have a direct effect on platelet cyclo-oxygenase. Such a mechanism seems, however, to be less likely: aggregation and TX synthesis after exogenous arachidonic acid were not diminished during the mackerel diet, indicating that endogenous C20:5 levels in plasma did not inhibit platelet cyclo-oxygenase directly. In addition, in studies of control platelets resuspended in PPP obtained when patients were on the mackerel diet, we found no inhibition of platelet aggregation. Further, the concentration of C20:5 necessary for inhibiting platelet aggregation in vitro has been reported to be 5-10 times higher2,3 than the free C20:5 plasma level observed in subjects on the mackerel diet. Thus, the reduction in platelet aggregation induced by low-dose collagen stimulation during the mackerel diet may result from an increase of C20:5 and a reduction of C20:4 in platelet membranes, which lead to diminished formation of proaggregatory TXA2. This interpretation is supported by our observation that the decrease in platelet aggregation correlated with both diminished TXB2 formation and with an increase in the C20:5/C20:4 ratio in platelet phospholipids (figs. 3 and 4). The highest C20:5/C20:4 ratio in platelet phospholipids reached in 1 subject during this study was 0-52. In Eskimos who have an even higher C20:5/C20:4 ratio (approx. 1 - 0) platelet aggregation is reduced to a greater extent. The latter, however, could be also partly due to a reduced platelet count.1 In addition to these suggestions of how platelet aggregation is reduced, one should keep in mind that other fatty acids are also much changed by the mackerel diet (i.e., C20:1; C22:6; C24:1). An increased proportion of unsaturated fatty acids in microsomes could influence platelet function by other means-e.g., by changing physicochemical characteristics, such as viscosity and fluidity of the platelet membranes. IS Furthermore, a reduction in plasma cholesterol by the mackerel diet (although not significant in this study) could contribute to the decrease in platelet aggregation by causing diminished cholesterol incorporation into platelet membranes. The opposite effect, an increased platelet aggregability in cholesterol-loaded platelets, has been shown.16 In conclusion, our findings, taken in conjunction with previous epidemiological studies in Eskimos, suggest that a change of fatty-acid consumption such that the diet contained fat with a high C20:5/C20:4 ratio could reduce platelet aggregability and hence exert a beneficial influence on certain cardiovascular disorders. This study was supported by the Deutsche Forschungsgemeinschaft We 681. We thank S. Fischer, H. Witzgall, Th. Dorbic, and A. Wiehl

for cooperation.

Requests for reprints should be addressed to P. C. W., Medizinische Klinik Innenstadt der Universitat, Ziemssenstrasse 1, 8000 Müncher. 2, West Germany. REFERENCES

Dyerberg J, Bang HO. Hæmostatic function and platelet polyunsaturated fatty acids in Eskimos. Lancet 1979; ii: 433-35. 2. Dyerberg J, Bang HO. Dietary fat and thrombosis. Lancet 1978; i: 152. 3. Dyerberg J, Bang HO, Stoffersen E, Moncada S, Vane JR. Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis? Lancet 1978, ii 1.

117-19. 4. Needleman

P, Raz A, Minkes MS, Ferrendelli JA, Sprecher H Triene prostaglandins: prostacyclin and thromboxane biosynthesis and unique biological properties. Proc Nat Acad Sci. 1979; 76: 944-48. 5. Von Lossonczy TO, Ruiter A, Bronsgeest-Schoute HC, van Gent OM. Hermus RJJ. The effect of a fish diet on serum lipids in healthy human subjects. Am J Clin Nutr 1978; 31: 1340-46. 6. Born GVR. Aggregation of blood platelets by adenosine diphosphate and its

reversal. Nature 1962; 194: 927-29. 7. Bills TK, Smith JB, Silver MJ. Metabolism of (14C)arachidonic acid by human platelets. Biochim Biophys Acta 1976; 424: 303-14. 8. Lehmann J, Schoene NW, Church JP. Essential fatty acid deficiency and platelet fatty acids of normotensive and genetically hypertensive rats Prostaglandins 1977; 13: 583-86. 9. Oelz O, Seyberth HW, Knapp, Jr HR, Sweetman BJ, Oates JA. Effects of feeding ethyl-dihomo&ggr;-linolenate on prostaglandin biosynthesis and platelet aggregation in the rabbit. Biochim Biophys Acta 1976; 431: 268-77 10. Lowry OH, Rosebrough HJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193: 265-75. 11. Needleman P, Moncada S, Bunting S, Vane JR, Hamberg M, Samuelsson B. Identification of an enzyme in platelet microsomes which generates thromboxane A2 from prostaglandin endoperoxides. Nature 1976; 261: 558-60. 12. Cohen P, Derksen A, van den Bosch H. Pathways of fatty acid metabolism in human platelets. J Clin Invest 1970; 49: 128-39. 13. Marcus AJ, Ullman HL, Safier LB. Lipid composition of subcellular particles of human blood platelets. Lipid Res 1969; 10: 108-14. 14. Bills TK, Smith JB, Silver MJ. Selective release of arachidonic acid from the phospholipids of human platelets in response to thrombin. J Clin Invest

1977; 60: 1-6. 15. Schaeffer BE, Curtis ASG. Effects on cell adhesion and membrane fluidity of changes in plasmalemmal lipids in mouse L929 cells. J Cell Sci 1977. 26: 47-55. 16. Shattil SJ, Anaya-Galindo R, Bennett J, Colman RW, Cooper RA. Platelet hypersensitivity induced by cholesterol incorporation. J Clin Invest 1975, 55: 636-43.

ENDOSCOPIC RETROGRADE CHOLANGIOGRAPHY AND PANCREATOGRAPHY IN INVESTIGATION OF POST-CHOLECYSTECTOMY PATIENTS W. S. J. RUDDELL M. G. ASHTON

D. J. LINTOTT A. T. R. AXON

Departments of Gastroenterology and Diagnostic Radiology, General Infirmary, Leeds 102

severely symptomatic post-cholecystectomy patients were studied by endoscopic retrograde cholangiography (ERCP), and successful ampullary cannulation was achieved in 101. All 29 patients with jaundice were correctly classified into intrahepatic and extrahepatic causes. 8% (73) patients with abdominal pain but no history of jaundice had retained biliary calculi, and 25% had an abnormal pancreatogram suggesting pancreatitis. The measurement of bileduct calibre alone did not reliably distinguish between the presence or absence of retained stones. It is suggested that ERCP is the investigation of choice in the symptomatic post-cholecystectomy patient

Summary

Introduction As many as 40% of patients continue to have symptoms after cholecystectomy.Many have only mild dyspepsia, abdominal discomfort, or diarrhoea which ca