- Email: [email protected]
Differential regulation of the formation of prostaglandins and related substances from arachidonic acid and from dihomogammalinolenic acid. I. effects of ethanol
Prostaglandins
and
Medicine
3:
118-128,
1979
DIFFERENTIAL REGULATION OF THE FORMATION OF PROSTAGLANDINS AND RELATED SUBSTANCES FROM ARACHIDONIC ACID AND FROM DIHOMOGAMMALINOLENIC ACID. I. EFFECTS OF ETHANOL. M.S. Manku, M. Oka and D.F. Horrobin, Clinical 110 Pine Avenue West, Montreal H2W lR7, Canada.
Research
Institute,
ABSTRACT Ethanol, over the concentration range 33 to 300 mg/lOO ml (7.2-65.2 x low3 M) caused a dose dependent and highly significant enhancement of conversion of 14C-dihomogammalinolenic acid (DGLA) to prostaglandin (PG) El and to thromboxane (TX) Bl by human platelets. Ethanol had no significant effect on conversion of 14C-arachidonic acid to PGE2 and TxB2. This concentration range is the one involved in human alcoholic intoxication. The effect could be related to enhanced transport of DGLA to the active site of the cycle-oxygenase enzyme complex, to a modification of the enzyme complex which changes the way it metabolizes DGLA but not arachidonic acid or to two different enzymes. Whatever the mechanism it seems that it is possible to regulate selectively the formation of 1 and 2 series PGs probably at the cycle-oxygenase level. The physiological and clinical implications of this are discussed. INTRODUCTION It is commonly assumed that the same cycle-oxygenase enzyme system converts arachidonic acid (AA) and dihomogammalinolenic acid (DGLA) to their respective endoperoxides. As a consequence, it is also commonly assumed that factors which modulate cycle-oxygenase activity will affect AA and DGLA metabolism in ways which are qualitatively and quantitatively similar. These assumptions have never been specifically and directly experimentally tested. This is partly because they are so persuasive and partly because radioactive AA is readily available while radioactive DGLA has not been until recently. It seemed to us possible that the theoretical assumptions might not be valid and we have therefore experimentally tested the effects of various agents on the conversion by human platelets of 14C-AA and 14CDGLA to their respective products. Some agents have similar effects on both reactions but others affect the two substances in quite different ways. Possible explanations for this observation are discussed.
119
MATER I ALS AND METHODS [1-14Cl arachidonic acid and Cl -14C1 dlhomogammal inolenic acid were purchased from New England Nuclear. They were diluted with hexane to specific activities of about 5 uCi/vmol. PGs and thromboxanes were kindly suppl ied by Dr. John Pike, Upjohn Co., Kalamazoo, Michigan. All organic solvents were reagent grade and were obtained from BDH, Montreal, except for ethanol which was obtained from Fisher, Montreal. One day expi red (2 days old) human platelets were obtained from the Canadian Red Cross, Montreal. Silica gel thin layer chromatography (TLC) plates were obtained from Analtech Inc. The platelet concentrates were always used within 48 hours of expiration. One unit was centrifuged at 1000 g for 15 minutes and the supernatant drawn off. The platelet pellet was resuspended in Tris-NaCl-EDTA buffer as described by Hamberg et al (1). The buffer was made up of 0.15 M M NaEDTA (90:8:2 v/v/v). The NaCl, 0.15 M Tris HCl at pH 7.4 and 0.077 platelets were recentrifuged, the supernatant removed and the pellet resuspended in Krebs-Henseleit buffer (without calcium) at pH 7.4. The washed platelet suspension contained about l-2% red blood cells. Al 1 glassware used in the preparation of the platelets were si liconized using “Prosi 1” (Canlab, Montreal). Four equal sized 1 ml aliquots of the platelet suspension, containing 109 platelets/ml were incubated with 0.5 PCi 14C-DGLA for five minutes. At the beginning of the incubation ethanol in concentrations of 0, 33, 100 and 300 mg/lOO ml (7.2, 21.7 and 65.2 x 10-3 M) was added to the suspensions The reaction was stopped after five minutes by addition of l/10 volume The suspension was then extracted three times with of 10% formic acid. The three ethyl acetate fractions obtained from each ethyl acetate. The extract was alcohol concentration were pooled and dried under vacuum. then taken up with 5 ml chloroform/methanol (2/l, v/v). Recovery of radioactive material in the extract was checked by taking 50 ~1 of the chloroform/methanol and counting by liquid scintillation. Recovery was in the range 80-95% in most experiments. The chloroform/methanol extract was then reduced in volume to 1 ml under Thin layer chromatography was carried out on dry prepurif ied nitrogen. 500 pg precoated, prescored silica gel G Uniplates (Analtech). Plates were activated by heating to llO°C for 1 hour immediately prior to use. methanol:acetic acid:water (90:8:1: The solvent system was chloroform: Fla, and thromboxane Bl) were run 0.8). Reference compounds (PGs El, at the same time and visualized by phosphomolybdic acid spray followed by brief heating. The bands on the plates corresponding to the reference PGEl, PGFla and TxBl were scraped off and eluted with 20 ml acetone. Each elution was then evaporated to dryness and counted by 1 iquid scintillation (Beckman 100 LS counter). Using the same batch of iments were carried out compounds.
platelets at the same time exactly similar with 14C-AA and PGEZ, PGF2a, and TxB2 as
120
experreference
TxB2
30 r 0
f m E
2520 -
iz g 158 Q
lo-
1 s
5OL
0
33
100
300
0
33
loo
300
0
33
100
Fig. 1. The effects of ethanol on arachidonic acid metabolism, Each column shows the mean and SEM for six platelet incubations. Results are expressed as the number of counts appearing in the bands corresponding to authentic PGEZ, PGF2 alpha and TxB2 standards.
121
300
TXBl
0
33
100
300
0
33
100
xxx
T
300
0
33
loo
300
ETHANOL CONCENTRATION (mg/lOOml)
Fig. 2. The effects of ethanol on dihomogammalinolenic Each column shows the mean and SEM for six metabol i sm. Statistical analysis was by paired t test: incubations. values for differences from incubations without ethanol Results are expressed as xx - p
122
acid platelet p are: the authentic
RESULTS The effects of the concentrations of ethanol used on the conversion of AA There was no ef feet on the formation to its products are shown in Fig. 1. of either PGE2 or PGF2a. TxB2, formed in much higher amounts than the evidence of a non-significant inhibtwo primary PGs, showed some slight ition of its formation at the two higher ethanol concentrations. Figure 2 shows the results with the DGLA incubations. Under basal conditions approximately equal amounts of PGEl and TxBl were produced, with only a slightly smaller amount of PGFla. Ethanol caused a dose dependent and statistically significant rise in the formation of both PGEl and TxBl and a dose dependent but not significant rise in PGFlcr formation. The absolute counts in each band are shown in figs. 1 and 2. Figure 3 is derived from these and shows the percentage changes in the sums of PGE, PGF and TxB formed from either AA or DGLA. The dramatic effect of ethanol on DGLA metabolism is clearly contrasted with the lack of any effect on AA metabolism. DISCUSSION Differences in the products formed by human platelets from DGLA and AA, similar to those reported here, have been previously described (2,3). In general with AA as substrate, TxB2 is the dominant product, whereas with DGLA the products are more evenly distributed. These differences in the pattern of products are usually be1 ieved to be due to differences in metabolism at stages beyond the cycle-oxygenase enzyme. The observations reported in the present paper, while confirming the previous reveal in addition a new phenomenon. work, While ethanol has no clear effect on the formation of any of the three products of AA metabolism analysed, it substantially enhances the formation of the three DGLA products. The similarity of the effect on ail three DGLA products suggests that the main action of ethanol is at the cycle-oxygenase level or earlier. Small differences between the ethanol effects on PGEl, TxBl and PGFlc( probably represent experimental variation but small effects beyond the cycle-oxygenase stage cannot be ruled out completely. There are a number of possible explanations for these results. The first is that ethanol did not affect PG synthesis directly but enhanced conversion of DGLA to AA via the desaturase enzyme, leading to formation of 2 series products from DGLA incubation. This can be ruled out for two reasons. First the extra material formed from DGLA under the influence of alcohol should have reflected the pattern of AA metabolism with far more TxB being formed. In fact at the highest concentration of alcohol used, the extra material formed was almost equally divided between TxBl and PGEl. Second the PGEl and PGE2 spots resulting from DGLA and AA incubations respectively were run using authentic PGEl and PGE2 standards on silver nitrate plates which separate these two PGs. The product from DGLA was indeed PGEl and that from AA was PGEE, ruling out a major effect of ethanol on the desaturase as an explanation of these results.
123
240 220 a
&
200
-
s 0 0
180 -
ik a@
160 -
z
100 80 I
I
I
0
33
100
ETHANOL
CONCENTRATION
I
300 (mg/lOOml)
Fig. 3. Summary of the effects of ethanol on arachidonic acid and dihomogammalinolenic acid metabolism. For each substrate the sum of the counts appearing in the PGE, PGF and TxB bands in the absence of ethanol is expressed as 100%. The sums of the counts at the various ethanol concentrations are then shown as percentages of this control value. Absolute counts are shown in figs 1 and 2.
124
There are tinguished 1.
at least on the
three basis
DGLA and AA are ethanol enhances
other possible explanations of currently available data.
metabolised the transport
by
2.
DGLA and AA are metabolised by ethanol alters the effectiveness without changing it for AA.
3.
There
are
different
which cannot These are:
the same cycle-oxygenase of DGLA but not of AA to the same of this
cycle-oxygenases
cycle-oxygenase system with
involved
in
be dis-
system but the active
system regard to
DGLA and
site.
but DGLA
AA metabol
i sm.
There have been several previous reports of effects of ethanol possibly Co11 ier et al (4) reported that etharelated to changes in PG synthesis. nol enhanced the tone of rat stomach strips in a dose related manner. The half maximal effect was produced by a concentration of ethanol of The effect could be blocked by aspirin. about 3.3 g/loo ml (7.2 x 10-l M). that the inhibition of gastric secretion by Karppanen et al (5) reported 1.8 M) in anesthetized rats could be 8 g/l00 ml, 10% ethanol (about reduced by pre-treatment of the animals with either indomethacin or mefeBoth observations are consistent with enhanced formation of namic acid. PGs under the influence of ethanol. However, since PGs of both series have been reported to cause contractions of rat stomach strips and to either 1 or 2 series PGs could be involved. inhibit gastric secretion, Bul 1 Two reports have investigated formation of 2 series PGs from AA. seminal vesicles increased PG production by about 50% when incubated Guinea pig with 2.5-5.0 g/l00 ml ethanol (in the 10” M range) (6). platelets failed to show changes in conversion of AA to TxB2 and PGs either when the platelets were incubated in vitro with ethanol up to 135 were taken from animals chronically mg/lOO ml (2.9 x 10 -4 M) or when they Much higher concentrations of ethanol (threshold treated with ethanol (7). 0.27 M) inhibited formation of TxB2 in human platelets (71, an effect consistent with the small non-significant inhibition of TxB2 formation observed in the present study. Whatever the mechanism, the results leave 1 i ttle doubt that clinically relevant concentrations of alcohol significantly enhance conversion of DGLA to PGs and related products without much affecting AA conversion. This observation is likely to prove of physiological and clinical significance. In human platelets the products of DGLA metabolism tend to inhibit, whereas those of AA metabolism tend to enhance aggregation (3). Human platelets produce PGEl and even under basal conditions more PGEl than PGE2 is formed (2,8). Attempts have been made with some success to enhance platelet PGEl formation and inhibit aggregation by dietary means, using DGLA or its immediate precursor, garnnalinolenic acid (9,lO). A whole new field is opened up by the possibility that dietary manipulation could be reinforced by agents which selectively enhance PGEl formation from DGLA, without increasing metabolism of AA. PGEl not only inhibits platelet aggregation. It is a potent dilator of coronary and other blood vessels. It is able to reduce infarct size following coronary occlusion in conscious dogs (11)) to inhibit platelet aggregation and improve clinical
125
status in humans with pre-infarction angina and acute myocardial infarction (12), to reduce vasospasm and improve ulceration in Raynaud’s disease (13) and dramatically to relieve pain and promote ulcer healing in patients with advanced 1 imb arteriosclerosis (14). There are obvious 1 imits to the usefulness of exogenous PGEl but the possibility of selectively enhanced endogenous PGEl production is promising. There may indeed al ready be evidence of its value in ischaemic heart disease. A recent epidemiological study of ischaemic heart disease in countries with a basically Western life style showed strong negative correlations with both alcohol and polyunsaturated fat consumption (15). If alcohol is in part exerting its effect by converting DGLA to PGEl, the combination of alcohol and polyunsaturates would seem particularly desirable. Other possible consequences of selective enhancement of PGEl formation relate to control of 2 series PG synthesis and to immunological function. Feinstein et a.1 (16) have shown in platelets that PGEl is able to inhibit mobilisation of free arachidonate and formation of its products. It is unknown at present whether this is a general mechanism but if it is, selective enhancement of PGEl formation may be of value in the control of situations where overproduction of 2 series PG is important. Such a mechanism could perhaps explain the inhibition of arachidonate mobilisation and 2 series PG formation by an ethanol-containing vehicle following UV irradiation of the skin (17). There is a good deal of evidence that PGEl may have a specific role in regulation of thymus and T lymphocyte function and this may be worth exploring (18,19). The control of infections by alcohol consumption may prove to have a scientific basis! Further investigation is required before the importance, if any, of enhanced PG 1 series formation in the actions of ethanol can be elucidated. The fact that the thresholds of the effects of ethanol on PG production Platelets from and on human behaviour are so close is of interest. a disease in which there is behaviour similar to that patients with mania, seen in mild alcoholic intoxication, have an enhanced abi 1 ity to convert DGLA to PGEl in the presence of ADP (2). Other evidence that PGEl may be important in brain function relates to its effects on nerve conduction (21) and to the evidence that low levels may be involved in schizophrenia (22) * In conclusion we have formation of 1 series physiology and clinical
demonstrated PGs. This medicine.
that ethanol may selectively observation may prove important
enhance in both
ACKNOWLEDGMENTS We thank the Fisher Family this study.
Muscular Dystrophy Association Foundation and Mrs. Sydney
REFERENCES
126
of Duder
Canada, Dr. for financial
David Roy, the support for
1.
J, Wakabayashi Hamberg M, Svensson structure of two prostaglandin Proc Nat Acad aggregation.
2.
Different Lagarde M, Gharib A, Dechavanne M. donic acid and dihomogammalinolenic acids prostaglandin synthetase. Biochimie 59:
3.
Falardeau P, Hamberg M, Samuelsson eicosatrienoic acid in human Acta 441: 193-200, 1976.
4.
Collier
HOJ, glandin affecting
McDonald-Gibson biosynthesis gut function.
5.
Karppanen H, Puurunen Eur in the rat.
6.
Collier
HOJ, glandin 1: 702,
Smith blood
by
CP. The platelets.
8,11,14Biophys
Stimulation of prostain vitro of some drugs 58: 193-9, 1976.
indomethacin 35: 221-3,
WJ, Saeed capsaicin,
and platelet
utilization of arachiby human platelet 935-7, 1977.
B. Metabolism of Biochim platelets.
WJ, Saeed SA. effects drugs: Br J Pharmacol
J. Ethanol, J Pharmacol
McDonald-Gibson biosynthesis 1975.
SN, by
by
T, Samuelsson B. Isolation endoperoxides that cause Sci USA 71: 345-9, 1974.
and 1976.
gastric
secretion
SA. Stimulation of prostaethanol and tyramine. Lancet
7.
Pennington thesis
effect of ethanol Prostaglandins
on thromboxane Med 2: 43-50,
8.
Lagarde M, Dechavanne M, Rigaud platelet prostaglandins: international Prostaglandin Abstract Book, p 65.
9.
Sim
10.
Kernoff
PBA, Will is AL, Stone KJ, Davies botic potential of dihomo-gamma-linoleic Med J 2: 1441-4, 1977.
11.
Jugdutt
B I, Hutchins GM, Bulkley BH, Becker LC. of PGEl and PGE2 on myocardial infarct size occlusion in conscious dogs. international Conference, Washington, May 1979, Abstract
12.
Bierenbaum ML, Oudhof JH. Platelet hyperaggregabil nary disease managed with PGEl. International Conference, Washington, May 1979, Abstract
13.
Data
M, Durand J. Basal PGEl is more elevated Meeting, Washington,
syn1979.
level of human than PGE2. May 1979,
AK, McGraw AP. The activity of y-linolenate and dihomo-ylinolenate methyl esters in vitro and in vivo on blood platelet function in non-human primates and in man. Thrombosis Res lo: 385-7, 1977.
JL, Martin WB, DePree HE, Molony BA, El for Raynaud’s disease. International ence, Washington, May 1979, Abstract
127
JA,
McNicol GP. Antithromacid in man. Brit
Stern
Dissimilar effects after coronary Prostaglandin Book p 57. i ty in acute Prostaglandin Book, p 10.
coro-
PM. Prostaglandin Prostaglandin ConferBook, p 23.
14.
Olsson
AG, Carlson LA. Prostaglandin El in ischemic peripheral artery disease. International Prostaglandin Conference, Washington May 1979, Abstract Book, p 89.
15.
St.Leger AS, Cochrane AL, Moore F. Factors associated mortality in developed countries with particular consumption of wine. Lancet 1: 1017-20, 1979.
16.
Feinstein MB, Becker EL, Fraser C. Thrombin, collagen and A23187 stimulated platelet arachidonate metabolism: differential inhibition by PGEl, local anesthetics and a serine-protease inhibitor. Prostaglandins 14: 1075-94, 1977.
17.
Black
AK, Greaves MW, Hensby CN, Plummer of indomethacin on arachidonic acid F2a levels in human skin 24 h after Br J Clin Pharmacol 6: 261-6, 1978.
18.
Zurier
RB, Sayadoff DM, Torrey treatment in NZB/NZW mice.
19.
Horrobin DF, Manku MS, Oka M, Morgan Ghavur T, Schweitzer M, Karmali ation of T lymphocyte function.
with cardiac reference to
NA, Warin AP. The effects and prostaglandin E2 and UVB and UVC irradiation.
SB, Rothfield Arthritis
NF. Prostaglandi Rheum 20: 723-8,
n E 1977.
RO, Cunnane SC, Ally Al, RA. The nutritional regulMed Hypotheses 5: 969-85,
1979.
20.
Ef feet of ADP on PGE formation in blood Abdul la YH, Hamadah K. platelets from patients with depression, mania and schizophrenia. Br J Psychiat 127: 591-5, 1975.
21.
Horrobi n DF, conduct Prostag
22.
Reconciliation Schizophrenia: Horrobin DF. and opioid concepts and pros tag 1andin, Lancet 529-531, 1979. ( March 10)
Durand on and andins
Prostaglandin El LG, Manku MS. interferes with iocal anesthetic 14: 103-8, 1977.
128
of the
mod i f ies action.
nerve
the dopamine, role of the pineal.
and
Medicine
3:
118-128,
1979
DIFFERENTIAL REGULATION OF THE FORMATION OF PROSTAGLANDINS AND RELATED SUBSTANCES FROM ARACHIDONIC ACID AND FROM DIHOMOGAMMALINOLENIC ACID. I. EFFECTS OF ETHANOL. M.S. Manku, M. Oka and D.F. Horrobin, Clinical 110 Pine Avenue West, Montreal H2W lR7, Canada.
Research
Institute,
ABSTRACT Ethanol, over the concentration range 33 to 300 mg/lOO ml (7.2-65.2 x low3 M) caused a dose dependent and highly significant enhancement of conversion of 14C-dihomogammalinolenic acid (DGLA) to prostaglandin (PG) El and to thromboxane (TX) Bl by human platelets. Ethanol had no significant effect on conversion of 14C-arachidonic acid to PGE2 and TxB2. This concentration range is the one involved in human alcoholic intoxication. The effect could be related to enhanced transport of DGLA to the active site of the cycle-oxygenase enzyme complex, to a modification of the enzyme complex which changes the way it metabolizes DGLA but not arachidonic acid or to two different enzymes. Whatever the mechanism it seems that it is possible to regulate selectively the formation of 1 and 2 series PGs probably at the cycle-oxygenase level. The physiological and clinical implications of this are discussed. INTRODUCTION It is commonly assumed that the same cycle-oxygenase enzyme system converts arachidonic acid (AA) and dihomogammalinolenic acid (DGLA) to their respective endoperoxides. As a consequence, it is also commonly assumed that factors which modulate cycle-oxygenase activity will affect AA and DGLA metabolism in ways which are qualitatively and quantitatively similar. These assumptions have never been specifically and directly experimentally tested. This is partly because they are so persuasive and partly because radioactive AA is readily available while radioactive DGLA has not been until recently. It seemed to us possible that the theoretical assumptions might not be valid and we have therefore experimentally tested the effects of various agents on the conversion by human platelets of 14C-AA and 14CDGLA to their respective products. Some agents have similar effects on both reactions but others affect the two substances in quite different ways. Possible explanations for this observation are discussed.
119
MATER I ALS AND METHODS [1-14Cl arachidonic acid and Cl -14C1 dlhomogammal inolenic acid were purchased from New England Nuclear. They were diluted with hexane to specific activities of about 5 uCi/vmol. PGs and thromboxanes were kindly suppl ied by Dr. John Pike, Upjohn Co., Kalamazoo, Michigan. All organic solvents were reagent grade and were obtained from BDH, Montreal, except for ethanol which was obtained from Fisher, Montreal. One day expi red (2 days old) human platelets were obtained from the Canadian Red Cross, Montreal. Silica gel thin layer chromatography (TLC) plates were obtained from Analtech Inc. The platelet concentrates were always used within 48 hours of expiration. One unit was centrifuged at 1000 g for 15 minutes and the supernatant drawn off. The platelet pellet was resuspended in Tris-NaCl-EDTA buffer as described by Hamberg et al (1). The buffer was made up of 0.15 M M NaEDTA (90:8:2 v/v/v). The NaCl, 0.15 M Tris HCl at pH 7.4 and 0.077 platelets were recentrifuged, the supernatant removed and the pellet resuspended in Krebs-Henseleit buffer (without calcium) at pH 7.4. The washed platelet suspension contained about l-2% red blood cells. Al 1 glassware used in the preparation of the platelets were si liconized using “Prosi 1” (Canlab, Montreal). Four equal sized 1 ml aliquots of the platelet suspension, containing 109 platelets/ml were incubated with 0.5 PCi 14C-DGLA for five minutes. At the beginning of the incubation ethanol in concentrations of 0, 33, 100 and 300 mg/lOO ml (7.2, 21.7 and 65.2 x 10-3 M) was added to the suspensions The reaction was stopped after five minutes by addition of l/10 volume The suspension was then extracted three times with of 10% formic acid. The three ethyl acetate fractions obtained from each ethyl acetate. The extract was alcohol concentration were pooled and dried under vacuum. then taken up with 5 ml chloroform/methanol (2/l, v/v). Recovery of radioactive material in the extract was checked by taking 50 ~1 of the chloroform/methanol and counting by liquid scintillation. Recovery was in the range 80-95% in most experiments. The chloroform/methanol extract was then reduced in volume to 1 ml under Thin layer chromatography was carried out on dry prepurif ied nitrogen. 500 pg precoated, prescored silica gel G Uniplates (Analtech). Plates were activated by heating to llO°C for 1 hour immediately prior to use. methanol:acetic acid:water (90:8:1: The solvent system was chloroform: Fla, and thromboxane Bl) were run 0.8). Reference compounds (PGs El, at the same time and visualized by phosphomolybdic acid spray followed by brief heating. The bands on the plates corresponding to the reference PGEl, PGFla and TxBl were scraped off and eluted with 20 ml acetone. Each elution was then evaporated to dryness and counted by 1 iquid scintillation (Beckman 100 LS counter). Using the same batch of iments were carried out compounds.
platelets at the same time exactly similar with 14C-AA and PGEZ, PGF2a, and TxB2 as
120
experreference
TxB2
30 r 0
f m E
2520 -
iz g 158 Q
lo-
1 s
5OL
0
33
100
300
0
33
loo
300
0
33
100
Fig. 1. The effects of ethanol on arachidonic acid metabolism, Each column shows the mean and SEM for six platelet incubations. Results are expressed as the number of counts appearing in the bands corresponding to authentic PGEZ, PGF2 alpha and TxB2 standards.
121
300
TXBl
0
33
100
300
0
33
100
xxx
T
300
0
33
loo
300
ETHANOL CONCENTRATION (mg/lOOml)
Fig. 2. The effects of ethanol on dihomogammalinolenic Each column shows the mean and SEM for six metabol i sm. Statistical analysis was by paired t test: incubations. values for differences from incubations without ethanol Results are expressed as xx - p
122
acid platelet p are: the authentic
RESULTS The effects of the concentrations of ethanol used on the conversion of AA There was no ef feet on the formation to its products are shown in Fig. 1. of either PGE2 or PGF2a. TxB2, formed in much higher amounts than the evidence of a non-significant inhibtwo primary PGs, showed some slight ition of its formation at the two higher ethanol concentrations. Figure 2 shows the results with the DGLA incubations. Under basal conditions approximately equal amounts of PGEl and TxBl were produced, with only a slightly smaller amount of PGFla. Ethanol caused a dose dependent and statistically significant rise in the formation of both PGEl and TxBl and a dose dependent but not significant rise in PGFlcr formation. The absolute counts in each band are shown in figs. 1 and 2. Figure 3 is derived from these and shows the percentage changes in the sums of PGE, PGF and TxB formed from either AA or DGLA. The dramatic effect of ethanol on DGLA metabolism is clearly contrasted with the lack of any effect on AA metabolism. DISCUSSION Differences in the products formed by human platelets from DGLA and AA, similar to those reported here, have been previously described (2,3). In general with AA as substrate, TxB2 is the dominant product, whereas with DGLA the products are more evenly distributed. These differences in the pattern of products are usually be1 ieved to be due to differences in metabolism at stages beyond the cycle-oxygenase enzyme. The observations reported in the present paper, while confirming the previous reveal in addition a new phenomenon. work, While ethanol has no clear effect on the formation of any of the three products of AA metabolism analysed, it substantially enhances the formation of the three DGLA products. The similarity of the effect on ail three DGLA products suggests that the main action of ethanol is at the cycle-oxygenase level or earlier. Small differences between the ethanol effects on PGEl, TxBl and PGFlc( probably represent experimental variation but small effects beyond the cycle-oxygenase stage cannot be ruled out completely. There are a number of possible explanations for these results. The first is that ethanol did not affect PG synthesis directly but enhanced conversion of DGLA to AA via the desaturase enzyme, leading to formation of 2 series products from DGLA incubation. This can be ruled out for two reasons. First the extra material formed from DGLA under the influence of alcohol should have reflected the pattern of AA metabolism with far more TxB being formed. In fact at the highest concentration of alcohol used, the extra material formed was almost equally divided between TxBl and PGEl. Second the PGEl and PGE2 spots resulting from DGLA and AA incubations respectively were run using authentic PGEl and PGE2 standards on silver nitrate plates which separate these two PGs. The product from DGLA was indeed PGEl and that from AA was PGEE, ruling out a major effect of ethanol on the desaturase as an explanation of these results.
123
240 220 a
&
200
-
s 0 0
180 -
ik a@
160 -
z
100 80 I
I
I
0
33
100
ETHANOL
CONCENTRATION
I
300 (mg/lOOml)
Fig. 3. Summary of the effects of ethanol on arachidonic acid and dihomogammalinolenic acid metabolism. For each substrate the sum of the counts appearing in the PGE, PGF and TxB bands in the absence of ethanol is expressed as 100%. The sums of the counts at the various ethanol concentrations are then shown as percentages of this control value. Absolute counts are shown in figs 1 and 2.
124
There are tinguished 1.
at least on the
three basis
DGLA and AA are ethanol enhances
other possible explanations of currently available data.
metabolised the transport
by
2.
DGLA and AA are metabolised by ethanol alters the effectiveness without changing it for AA.
3.
There
are
different
which cannot These are:
the same cycle-oxygenase of DGLA but not of AA to the same of this
cycle-oxygenases
cycle-oxygenase system with
involved
in
be dis-
system but the active
system regard to
DGLA and
site.
but DGLA
AA metabol
i sm.
There have been several previous reports of effects of ethanol possibly Co11 ier et al (4) reported that etharelated to changes in PG synthesis. nol enhanced the tone of rat stomach strips in a dose related manner. The half maximal effect was produced by a concentration of ethanol of The effect could be blocked by aspirin. about 3.3 g/loo ml (7.2 x 10-l M). that the inhibition of gastric secretion by Karppanen et al (5) reported 1.8 M) in anesthetized rats could be 8 g/l00 ml, 10% ethanol (about reduced by pre-treatment of the animals with either indomethacin or mefeBoth observations are consistent with enhanced formation of namic acid. PGs under the influence of ethanol. However, since PGs of both series have been reported to cause contractions of rat stomach strips and to either 1 or 2 series PGs could be involved. inhibit gastric secretion, Bul 1 Two reports have investigated formation of 2 series PGs from AA. seminal vesicles increased PG production by about 50% when incubated Guinea pig with 2.5-5.0 g/l00 ml ethanol (in the 10” M range) (6). platelets failed to show changes in conversion of AA to TxB2 and PGs either when the platelets were incubated in vitro with ethanol up to 135 were taken from animals chronically mg/lOO ml (2.9 x 10 -4 M) or when they Much higher concentrations of ethanol (threshold treated with ethanol (7). 0.27 M) inhibited formation of TxB2 in human platelets (71, an effect consistent with the small non-significant inhibition of TxB2 formation observed in the present study. Whatever the mechanism, the results leave 1 i ttle doubt that clinically relevant concentrations of alcohol significantly enhance conversion of DGLA to PGs and related products without much affecting AA conversion. This observation is likely to prove of physiological and clinical significance. In human platelets the products of DGLA metabolism tend to inhibit, whereas those of AA metabolism tend to enhance aggregation (3). Human platelets produce PGEl and even under basal conditions more PGEl than PGE2 is formed (2,8). Attempts have been made with some success to enhance platelet PGEl formation and inhibit aggregation by dietary means, using DGLA or its immediate precursor, garnnalinolenic acid (9,lO). A whole new field is opened up by the possibility that dietary manipulation could be reinforced by agents which selectively enhance PGEl formation from DGLA, without increasing metabolism of AA. PGEl not only inhibits platelet aggregation. It is a potent dilator of coronary and other blood vessels. It is able to reduce infarct size following coronary occlusion in conscious dogs (11)) to inhibit platelet aggregation and improve clinical
125
status in humans with pre-infarction angina and acute myocardial infarction (12), to reduce vasospasm and improve ulceration in Raynaud’s disease (13) and dramatically to relieve pain and promote ulcer healing in patients with advanced 1 imb arteriosclerosis (14). There are obvious 1 imits to the usefulness of exogenous PGEl but the possibility of selectively enhanced endogenous PGEl production is promising. There may indeed al ready be evidence of its value in ischaemic heart disease. A recent epidemiological study of ischaemic heart disease in countries with a basically Western life style showed strong negative correlations with both alcohol and polyunsaturated fat consumption (15). If alcohol is in part exerting its effect by converting DGLA to PGEl, the combination of alcohol and polyunsaturates would seem particularly desirable. Other possible consequences of selective enhancement of PGEl formation relate to control of 2 series PG synthesis and to immunological function. Feinstein et a.1 (16) have shown in platelets that PGEl is able to inhibit mobilisation of free arachidonate and formation of its products. It is unknown at present whether this is a general mechanism but if it is, selective enhancement of PGEl formation may be of value in the control of situations where overproduction of 2 series PG is important. Such a mechanism could perhaps explain the inhibition of arachidonate mobilisation and 2 series PG formation by an ethanol-containing vehicle following UV irradiation of the skin (17). There is a good deal of evidence that PGEl may have a specific role in regulation of thymus and T lymphocyte function and this may be worth exploring (18,19). The control of infections by alcohol consumption may prove to have a scientific basis! Further investigation is required before the importance, if any, of enhanced PG 1 series formation in the actions of ethanol can be elucidated. The fact that the thresholds of the effects of ethanol on PG production Platelets from and on human behaviour are so close is of interest. a disease in which there is behaviour similar to that patients with mania, seen in mild alcoholic intoxication, have an enhanced abi 1 ity to convert DGLA to PGEl in the presence of ADP (2). Other evidence that PGEl may be important in brain function relates to its effects on nerve conduction (21) and to the evidence that low levels may be involved in schizophrenia (22) * In conclusion we have formation of 1 series physiology and clinical
demonstrated PGs. This medicine.
that ethanol may selectively observation may prove important
enhance in both
ACKNOWLEDGMENTS We thank the Fisher Family this study.
Muscular Dystrophy Association Foundation and Mrs. Sydney
REFERENCES
126
of Duder
Canada, Dr. for financial
David Roy, the support for
1.
J, Wakabayashi Hamberg M, Svensson structure of two prostaglandin Proc Nat Acad aggregation.
2.
Different Lagarde M, Gharib A, Dechavanne M. donic acid and dihomogammalinolenic acids prostaglandin synthetase. Biochimie 59:
3.
Falardeau P, Hamberg M, Samuelsson eicosatrienoic acid in human Acta 441: 193-200, 1976.
4.
Collier
HOJ, glandin affecting
McDonald-Gibson biosynthesis gut function.
5.
Karppanen H, Puurunen Eur in the rat.
6.
Collier
HOJ, glandin 1: 702,
Smith blood
by
CP. The platelets.
8,11,14Biophys
Stimulation of prostain vitro of some drugs 58: 193-9, 1976.
indomethacin 35: 221-3,
WJ, Saeed capsaicin,
and platelet
utilization of arachiby human platelet 935-7, 1977.
B. Metabolism of Biochim platelets.
WJ, Saeed SA. effects drugs: Br J Pharmacol
J. Ethanol, J Pharmacol
McDonald-Gibson biosynthesis 1975.
SN, by
by
T, Samuelsson B. Isolation endoperoxides that cause Sci USA 71: 345-9, 1974.
and 1976.
gastric
secretion
SA. Stimulation of prostaethanol and tyramine. Lancet
7.
Pennington thesis
effect of ethanol Prostaglandins
on thromboxane Med 2: 43-50,
8.
Lagarde M, Dechavanne M, Rigaud platelet prostaglandins: international Prostaglandin Abstract Book, p 65.
9.
Sim
10.
Kernoff
PBA, Will is AL, Stone KJ, Davies botic potential of dihomo-gamma-linoleic Med J 2: 1441-4, 1977.
11.
Jugdutt
B I, Hutchins GM, Bulkley BH, Becker LC. of PGEl and PGE2 on myocardial infarct size occlusion in conscious dogs. international Conference, Washington, May 1979, Abstract
12.
Bierenbaum ML, Oudhof JH. Platelet hyperaggregabil nary disease managed with PGEl. International Conference, Washington, May 1979, Abstract
13.
Data
M, Durand J. Basal PGEl is more elevated Meeting, Washington,
syn1979.
level of human than PGE2. May 1979,
AK, McGraw AP. The activity of y-linolenate and dihomo-ylinolenate methyl esters in vitro and in vivo on blood platelet function in non-human primates and in man. Thrombosis Res lo: 385-7, 1977.
JL, Martin WB, DePree HE, Molony BA, El for Raynaud’s disease. International ence, Washington, May 1979, Abstract
127
JA,
McNicol GP. Antithromacid in man. Brit
Stern
Dissimilar effects after coronary Prostaglandin Book p 57. i ty in acute Prostaglandin Book, p 10.
coro-
PM. Prostaglandin Prostaglandin ConferBook, p 23.
14.
Olsson
AG, Carlson LA. Prostaglandin El in ischemic peripheral artery disease. International Prostaglandin Conference, Washington May 1979, Abstract Book, p 89.
15.
St.Leger AS, Cochrane AL, Moore F. Factors associated mortality in developed countries with particular consumption of wine. Lancet 1: 1017-20, 1979.
16.
Feinstein MB, Becker EL, Fraser C. Thrombin, collagen and A23187 stimulated platelet arachidonate metabolism: differential inhibition by PGEl, local anesthetics and a serine-protease inhibitor. Prostaglandins 14: 1075-94, 1977.
17.
Black
AK, Greaves MW, Hensby CN, Plummer of indomethacin on arachidonic acid F2a levels in human skin 24 h after Br J Clin Pharmacol 6: 261-6, 1978.
18.
Zurier
RB, Sayadoff DM, Torrey treatment in NZB/NZW mice.
19.
Horrobin DF, Manku MS, Oka M, Morgan Ghavur T, Schweitzer M, Karmali ation of T lymphocyte function.
with cardiac reference to
NA, Warin AP. The effects and prostaglandin E2 and UVB and UVC irradiation.
SB, Rothfield Arthritis
NF. Prostaglandi Rheum 20: 723-8,
n E 1977.
RO, Cunnane SC, Ally Al, RA. The nutritional regulMed Hypotheses 5: 969-85,
1979.
20.
Ef feet of ADP on PGE formation in blood Abdul la YH, Hamadah K. platelets from patients with depression, mania and schizophrenia. Br J Psychiat 127: 591-5, 1975.
21.
Horrobi n DF, conduct Prostag
22.
Reconciliation Schizophrenia: Horrobin DF. and opioid concepts and pros tag 1andin, Lancet 529-531, 1979. ( March 10)
Durand on and andins
Prostaglandin El LG, Manku MS. interferes with iocal anesthetic 14: 103-8, 1977.
128
of the
mod i f ies action.
nerve
the dopamine, role of the pineal.