- Email: [email protected]
Prostanoids and cell injury
Prostanoids and Cell Injury JOHN T. FLYNN, PhD Cells respond to injury with a variety of mechanisms, one of which is the synthesis and release of various vasoactive factors. These factors can influence the intra- and extracellular environments of the cell and thus affect the overall rate of cellular survival. Products of arachidonic acid metabolism represent one such system. Studies have demonstrated the synthesis and release of various prostanoids during global injuries such as circulatory shock, burn injury, myocardial infarction, and severe trauma. We have carried out studies in an isolated, perfused rabbit liver preparation to investigate the cellular stimuli and cellular mechanisms by which the archidonic acid cascade is stimulated under these conditions. Both hypoxia and metabolic poisoning with dinitrophenol are stimuli for enhanced pmstanoid production in vitro. This eicosanoid production is associated with evidence of severe cellular injury. In contrast, when endogenous prostanoid production is stimulated in a noninjured liver with phospholipase A*, the production is transient and is not associated with cellular injury. Another stimulus, bacterial endotoxin, has no effect on prostanoid biosynthesis in vifro even though endotoxin is a very potent stimulus in vim Activated complement has been shown to be a potent stimulus for arachidonic acid metabolism both in viva and in vitro. It thus becomes apparent that arachidonic acid metabolism can be influenced by both receptor and nonreceptor related mechanisms in the injured cell. In addition, the eicosanoids released by the injured cell are apparently able to participate in the homeostatic response of the cell to the injury. There also appear to be specific situations in which inappmpriate eicosanoid production is related to the pathophysiofogic consequences of the injury process. Additional studies are required to characterize these relationships.
The involvement of blood-borne, or vasoactive, substances in circulatory shock has been under investigation for well over 100 years. Periodically, there is a flurry of interest in a newly discovered, or rediscovered, group of compounds. These bursts of interest have centered on such substances as the catechol-
From the Department of Thomas
Jefferson
of Physiology, University,
Jefferson
Medical
College
Philadelphia.
Supported in part by Grant No. 28023 from the National tutes of Health, Institute of General Medical Sciences.
Insti-
Presented at the First International Conference on the Basic Mechanisms and Clinical Management of Shock, Merrillville, Indiana, September 10-l 1, 1982, and accepted for publication at that time. Address reprint requests to Dr. Flynn: Jefferson Medical College, Department of Physiology, 1020 Locust Street, Philadelphia, PA 19107. Key Words: 20
Cells, prostaglandins,
shock.
amines, bradykinin, the steroids, the renin-angiotensin system, and so on, including the latest: the opiates. The interest in the involvement of prostaglandin-like materials in circulatory shock began in the early 1970s and is still very much in progress because of the discovery of new subgroups of prostanoids. The hypothesis with which we are dealing is shown in Figure 1. It states that products of the arachidonic acid cascade are elaborated by the cell during periods of injury. We suggest that these newly synthetized products can participate in the homeostatic response of that cell to the injury. Conversely, we also believe that under specific conditions, prostaglandin-like materials may play a role in the pathophysiology of the shock state. The first test of this hypothesis is to determine whether the arachidonic acid cascade is activated in viro during periods of circulatory shock. Table 1 summarizes several studies of that phenomenon. Independently of the type of insult or the species, prostanoid concentrations in plasma or lymph always increase during the shock state. We have also shown that in hemorrhagic, endotoxic,’ and splanchnic artery occlusion (SAG) shock,? this increase in the plasma prostanoid concentration is due to an increased rate of de nova synthesis, not to a decreased rate of metabolism. So, there is no doubt that the arachidonic acid cascade is being stimulated during circulatory shock and other forms of in V~VOglobal injury. The next two points to be addressed are the how and why of the system. Our laboratory has studied these questions using the classic techniques of stimulation and ablation. Our overall approach has been to use the same stimulus for injury in an intact animal model and then to compare those in ~ivo results with results obtained in vitro with a perfused rabbit liver model. Figure 2 shows a schematic of this perfusion model. The primary advantage of this system for studying prostaglandin production is that it is of a nonrecirculating design. The perfusate leaving the liver is either sampled or discarded; it does not re-enter the reservoir. By knowing the prostanoid concentration in the effluent; the flow rate of perfusate, which is constant, and the liver weight, we can very precisely calculate the rate of prostaglandin synthesis per gram of wet tissue weight. We have used this model to assess changes in endogenous prostaglandin synthesis during hypoxia, metabolic poisoning, and so on. An important aspect of circulatory shock is tissue hypoxia or anoxia. Figure 3 shows changes in the rate of hepatic production of PGF,, when the perfusate is
FLYNN W PROSTANOIDS
MEMBRANE PHOSPHOLIPIDS
I,-
Phospholipase AZ, Ca++
ARACHIDONICACID FIGURE 1. Hypothesis for a role of prostaglandinlike substances in circulatory shock and cellular injury. Products of the amchidonic acid cascade are elaborated by the cell during periods of injury. These products can participate in the homeostatic response of that cell to the injury or may be involved in the pathophysiologic consequences of the injury.
5 HPETE
12 HPETE
15 HPETE
Leukotrienes PG12
equilibrated with 95% N,-5% CO2 instead of oxygen, beginning at time zero. Over two and a half hours, there is a modest 25% increase in the rate of PGF,, production. If 0.1 p,g/g/min of arachidonic acid was infused into the liver concurrent with hypoxia, the rate of PGF,, production was increased by fourfold. This increased rate of arachidonic acid metabolism had a salutary effect on the liver. While hypoxic livers became very edematous, the arachidonic-acid-treated livers increased in weight by only 5~%.~The same profile was seen with the rate of release of lactic dehydrogenase by the livers. There was little release from the livers in the sham group, significant release of lactate dehydrogenase (LDH) during hypoxia, and an inhibition of LDH release by the arachidonic acid treatment. The results of this study suggest that first, hypoxia does weakly stimulate prostanoid production in
FIGURE 2. Rabbit liver perfusion apparatus. The system is of a flow-through design that permits the calculation of rates of prostanoid synthesis in picograms released per minute per gram wet weight. P = pressure sensor; T = temperature sensor: TE = temperature exchanger.
PGE2
the liver in vitro, and second, that the stimulation of the arachidonic acid cascade allowed the liver cells to better tolerate the injury. Not being satisfied with a 25% stimulation, we made the assumption that the perfused rabbit liver was quite capable of anaerobic metabolism and that we had not really induced a shock-like injury. We produced a more severe energy deficit by adding 100 p.mol/l 2,4 dinitrophenol (DNP) to the reservoir (Fig. 4).4 This agent uncouples oxidative phosphorylation from most, if not all, energy sources. There is a significant increase in the rate of PGF,, production to over 1000 pg/g/min with DNP treatment. A similar change in the rate of prostacyclin release to over 3000 pg@nin was observed with DNP (Fig. 5). The severity of the model is dramatized by an almost linear increase in the rate of loss of acid phosphatase activity from the liver cells.
RESERVOIR
WATER
BATH
25 b
0
25
50
75
125
0
m
hypoxia
sham
L
I 90
Time Cminl
60
I
1 150
I 120
I
FIGURE 5 (right). Prostacyclin production following 100 pmolil dinitrophenol (DNP). Values are the mean ? SEM for seven livers. Asterisks signify P < .05 compared with the control. pre-DNP value. Prostacyclin was measured by its stable metabolite. 6-keto PGF,,. (From Flynn JT. Adv Shock Res 1981:5:149-162.)
FIGURE 3 tabore). Changes in the rate of PGFZa synthesis by the perfused rabbit liver on a percent of control basis. The zero value represents rates of PGF, release of 33 -C 4. 3 1 L 5, and 30 2 8 nanograms of PGF?, released per minute by the perfused livers of the sham, hypoxia, and hypoxia plus arachidonic acid groups, respectively. Six livers were studied in each group. (From Flynn JT. Prostaglandins 1979:17:39-52.) FIGURE 4 (abore, right). Rate of PGFzu release in sham and dinitrophenol (DNP) treated livers. The DNP was added to the perfusate at a concentration of 100 pmolil at time zero. Asterisks signify P < .05 compared with values in the sham group. Six livers were studied in each group. Values are the mean ? SEM. (From Flynn JT. Adv Shock Res 1981;5: 149-162.
30
I
. hypoxia 6 AA
l
l
3000
control (7)
0
500
1000
1500
2000
I 60
TIME
I
IN MINUTES
60
(1)
TIME IN MINUTES
30
DNP
o sham l
30
:ontrol
’
’
’_
’
120
I
AFTER
90
(‘I)
DNP
120
AFTER DNP
90
*
180
*
180
30 90 120 150
10 uM lidocaine
FIGURE 8 (right). Effect of 10 umoV1 dinitrophenol and 10 urnoh I lidocaine on the rate of acid phosphatase released by the perfused liver. Asterisks denote significant differences between the values in each group and those of the sham plus vehicle group at each time point. Stars denote P < .05between the noted groups and the lidocaine alone group. The effect of dinitrophenol plus lidocaine was to exacerbate the degree of injury. (From Flynn JT. Adv Shock Res 1983:10:149-159.)
FIGURE 7 (above, right). Thromboxane production in perfused livers receiving vehicle, 10 pmol/l dinitrophenol, or 10 umol/l dinitrophenol plus 10 umol/l lidocaine. The range of thromboxane release was quite narrow compared with prostacyclin release (Fig. 6). No significant differences were observed. (From Flynn JT. Adv Shock Res 1983;10:149-159.)
160
10 PM DNP 8 lo @I lidocaine
TlME IN MINUTES
60
.-.--
---0
FIGURE 6 (above). Rate of prostacyclin release in isolated, perfused rabbit livers treated with vehicles, 10 urnoh dinitrophenol, or 10 umol/l dinitrophenol plus 10 urnoh lidocaine. All values are the mean * SEM for six observations in each group. Asterisks signify P < .05when the value in the noted group is compared with the corresponding value in the sham plus vehicle group. Note that lidocaine alone moderately but significantly elevated the rate of prostacyclin production in noninjured livers but significantly attenuated the prostacyclin response in injured livers. (From Flynn JT. Adv Shock Res 1983;10:149-159.)
0
sham B vehicle
w......w 10 IJM DNP
B
0
100
200
300
400
500
600
700
800
0
30
30
90
120
1
90
120
TIME IN MINUTES
60
150
150
I
180
180
L
10 ~JMlidocaine
.-.-a
L
lo PM DNP 8 10 @l lidocaine
r-e
TIME IN MINUTES
60
1
10 j~t.ADNP
w-
1
sham b vehicle
-
AMERICAN
TABLE 1.
JOURNAL
OF EMERGENCY
Prostanoid
Synthesis
in Vivo
Hemorrhagic
PGA,, PGE,, PGF,, PGE PGE PGE
Splanchnic artery occlusion Endotoxic
PGF,, 15keto
Anaphylactic
Burn
Numbers
designate
H Volume
during
Prostanoid
Type of Shock
l
MEDICINE
PGF,,
PGE,, PGF,, PGE, PGF PGF,,, PGI,, TxB, PGE, PGF PGE, PGF PG-like materials PGI, TxB, PG-like materials TxB,, PGD, PG-like materials PGE, PGE PGI, references
2, Number
Periods
of Circulatory
Species
Shock
Change in Concentration Increase
Dog Dog Dog Qog Dog Dog
Increase Increase Increase
Dog Dog
Increase
Sheep Baboon Calf Dog Rabbit Rat Cat Guinea Guinea Rat Dog Sheep
pig pig
Increase Increase Increase Increase Increase Increase Increase Increase Increase Increase Increase
Reference’ Flynn et al” Johnson & Selkurt” Collier et a/l3 Jakschik et all4 Flynn et a/l5 Flynn et all6 Herman & VaneI Anderson et a/l8 Demling et aI8 Fletcher et a/l9 Anderson et a/20
Kessler et aI*’ Bult et a122 Cook et a123 Korbut et a124 Anhut et a125 Crutchley et a12’j Okabe et a127 Barac et a128 Harms et a12g
at the end of this article.
To determine the role played by these newly synthetized prostaglandins in this form of injury, we used lidocaine, a drug reported by Fletcher er al5 to have several effects on the prostaglandin system. When we added 10 pmol/l lidocaine to a reduced DNP dosage of 10 p,mol/l, we observed that DNP alone stimulated prostacyclin production (Fig. 6), as we had observed in the earlier studye In addition, the lidocaine significantly attenuated the DNP-evoked prostacyclin production. If we examine the concurrent production of thromboxane-like materials, we see a different situation (Fig. 7). In comparison with prostacyclin production, DNP does not appear to significantly stimulate thromboxane production. Lidocaine has little effect on the rate of thromboxane synthesis. So, we have divided the system to some degree; that is, we have inhibited prostacyclin and maintained a low rate of thromboxane release during a period of cellular injury. Under these conditions there seems to be a greater degree of injury. In Figure 8, the fine dotted line shows the effect of DNP alone on the rate of release of acid phosphatase, while the heavier broken line shows the release in the presence of DNP plus lidocaine. There is a significantly greater loss of the enzyme into the perfusate in this group. Lidocaine alone does not induce the injury. Similar results were seen with lactic dehydrogenase release. With this enzyme, significantly greater rates of release in the DNP plus lidoCaine group were seen compared with lidocaine alone. No direct effects of lidocaine were apparent. 24
1
From these results and those of the previous study, it appears that endogenously synthetized prostaglandin-like materials can influence the integrity of the cell during periods of cellular injury, with prostacyclinlike materials playing a beneficial homeostatic role; the role of thromboxane-like materials remains to be assessed. We were also interested in the effect of endogenously synthetized prostanoids on the noninjured cell. To study this, we infused a group of livers with phospholipase A,, an enzyme that cleaves arachidonic acid from membrane phospholipids.’ The rate of thromboxane B, production is shown in Figure 9 for both control and PLA, infusions at a rate of 100 units per minute. There is a significant increase in the rate of thromboxane release with PLA,, which returns toward baseline values even during the period of the PLA, infusion. In contrast, Figure 10 shows the rate of prostacyclin release in the same livers. The PLA, infusion results in a long-lasting prostacyclin release, which extends past the end of the infusion period. It is important that the PLA, livers showed no signs whatever of being injured. There was no weight gain, no enzyme release, and no hemodynamic change. So, newly synthetized prostanoids do not appear to significantly affect normal healthy cells, while we have previously seen that they do have various effects on injured cells. Having characterized the prostanoid response of the isolated perfused liver to several different stimuli, we now sought to correlate in viva responses with the in vitro liver data. In a study with Demling,* we studied
FLYNN n PROSTANOIDS
FIGURE 9. Rate of thromboxane Bz release from the isolated perfused rabbit liver by phospholipase A, infusion. Values are expressed as a percent of the control value for each group. All data are presented as the mean 2 SEM for seven livers in the sham group and nine livers in the phospholipase A, infusion group. Horizontal box on the abscissa represents the duration of the phospholipase AZ infusion at a rate of 100 U/min in 0.1 mlimin. Asterisks signify statistically significant differences (p < .05) in the rate of release of TxB, between groups at each iime point. (From Flynn et al. Can J Physiol Pharmacol 1981;59:1268-1273.
r l
.
I *
.
5
10
20 TIME
25
QROUP
PHOSPHOLIPASE
30
35
40
OROUP
45
50
IN MINUTES
liver perfusate without any effect on either prostanoid production or any index of cellular injury.9 We also quantitated the effect of endotoxin in vitro on Cl4 labeled arachidonic acid metabolism by liver slices.1° We could find no differences in either the pattern or the magnitude of arachidonic acid metabolism between control tissue and tissue treated with endotoxin in vitro. It became apparent that the endotoxin-mediated stimulation of prostanoid production in vivo was dependent upon another process or another tissue type. To further determine what that process might be, we did a study with Gee, in which we used a sheep model
both plasma and lung lymph concentrations of PGF,, and PGI, in conscious sheep receiving gram-negative Escherichia coli endotoxin. These animals demonstrated increased concentrations in plasma, and especially lung lymph, of PGF,, and PGI, during an initial phase response to endotoxin characterized by pulmonary hypertension and decreased cardiac output. These values returned toward baseline values during the post-pulmonary hypertensive phase, with prostacyclin in lung lymph remaining elevated. This was quite a marked response in the intact sheep to 2 pg/ kg endotoxin. In a similar in vitro study, we added up to an equivalent of 10 mg/kg of E coli endotoxin to the
FIGURE 10. Rate of 6keto PGF,, release from the perfused rabbit livers by phospholipase A, infusion, expressed on a percent of control period value for each group. Values are expressed as the mean 2 SEM. Seven livers were studied in the sham group and nine in the phospholipase AZ infusion group. Asterisks signify P < .05 compared with the sham group at each time period. Duration of phospholipase AZ infusion is denoted by the horizontal box on the abscissa. (From Flynn er al. Can J Physiol Pharmacol 1981;59:12681273.)
15
SHAM
180
.
SHAM
.
PHOSPHOLIPASE
GROUP QROUP
180
1
trol con-,
5
10
I
I
I
I
I
I
1
I
15
20
25
30
35
40
45
50
TIME
IN MINUTES
25
AMERICAN
JOURNAL
OF EMERGENCY
MEDICINE
m
Volume 2, Number 1
CONTROL B VEHICLE
.
A ZYMOSAN ACTIVATED PLASMA FIGURE 11. Rate of thromboxane release by the isolated rabbit liver following the infusion of zymosan-activated plasma. The complement-activated plasma was infused during the period noted by the hatched bar at a rate of 1.0 mUmin into a perfusate flow rate of 120 mUmin. Asterisks signify P < .OS for values between groups at each time point. (Flynn JT, Hellerman P: unpublished observation. 1
0
15
30
45
60
75
90
105
120
TIME IN MINUTES
and characterized the in viva prostanoid response to an infusion of complement activated plasma (unpublished observation). Almost immediately after the infusion was begun, there was a significant rise in pulmonary arterial pressure, an increase in the pulmonary lymph flow rate, and a stimulation in endogenous thromboxane and prostacyclin release. This type of hemodynamic and prostanoid response after activated complement infusion is quite similar to the response following endotoxin administration. In both of these in vivo models, we have abolished the early hemodynamic phase with cycle-oxygenase or thromboxane synthetase inhibitors. We then again used the in vitro liver perfusion and tested the effect of complement-activated plasma (Fig. 11) (unpublished observation). There was a stimulation of thromboxane B, production, which demonstrated the same temporal characteristics we observed with both complement administration in viva and the phospholipase A, infusion in vitro. That is, the enzyme is regulated and is turned off by product inhibition, cofactor depletion, or perhaps free radical feedback.
TABLE 2.
Effect of Cobra Venom
Factor
(CVF) on the Prostanoid
WBC (103/mm3) Control 5 min 10 min 15 min 30 min 90 min 120 min 15 min
26
CVF CVF CVF post post post post
CVF CVF CVF endotoxin
This does not seem to be a permanent feedback inhibition, in that given 10 to 15 minutes, an identical response can be evoked. A similar profile was seen for prostacyclin synthesis on a much smaller scale. Table 2 presents preliminary information on one rabbit and is an attempt to demonstrate the prostanoidcomplement relationship in vivo (unpublished observation). In this rabbit, we measured the leukocyte count, plasma PGI, and TxB, and the CH,, complement titer. We infused cobra venom factor at 25 units/ kg over a IS-minute period. This factor activates the endogenous complement system, as confirmed by the drop in the CH,, value and the leukopenia. Concurrently there was a transient increase in plasma prostacyclin and a more prolonged increase in thromboxane. After the cobra venom infusion was completed, the leukocyte count rebounded, the prostanoids varied somewhat, and the complement titer remained depressed. The interesting finding was that when a massive dose of endotoxin was given to this animal, there was no apparent leukopenia, no hemodynamic change, and no sign of the massive release
4.6 2.8 3.3 3.2 3.2 4.6 5.9 5.4
Response
to Endotoxin
PGI,
TXB,
(w/ml)
(w/ml)
0.134 0.437 0.164 0.100 0.203 0.100 0.380 0.620
1.045 1.136 1.362 2.122 1 S96 1.969 0.470 0.433
%o 606 30 71 36 79 42 125 116
FLYNN w PROSTANOIDS
of thromboxane that we have seen in earlier models. We did notice a modest increase in prostacyclin. It must also be noted that this rabbit was in stable condition for 45 minutes after a dose of 5 mg/kg endotoxin but died within 50 minutes because of circulatory collapse. In summary, a substantial amount of information suggests that prostaglandin-like materials are synthetized and released during periods of cellular injury. These materials, when released, can play a role in both the homeostatic response to, and/or the pathophysiologic consequences of, the injury. Future investigations should be directed toward defining the role of these substances in specific forms of injury and in elucidating the mechanisms of interaction between the prostanoids and other humoral systems during the shock state.
REFERENCES 1. Flynn JT, Lefer AM. Prostaglandin metabolism during circulatory shock. Biochim Biophys Acta 1977;497:775-784. 2. Flynn JT, Bridenbaugh GA, Lefer AM. Clearance of prostaglandin F,, during circulatory shock. Life Sci 1975; 17:1699-1706. 3. Flynn JT. Influence of the arachidonic cascade on the in vitro hepatic response to hypoxia. Prostaglandins 1979;17:39-52. 4. Flynn JT. Effect of 2,4-dinitrophenol on the rate of hepatic prostaglandin production. Adv Shock Res 1981;5: 149-162. 5. Fletcher JR, Ramwell PW. Lidocaine or indomethacin improves survival in baboon endotoxic shock. J Surg Res 1978;24:154-160. 6. Flynn JT. Effect of lidocaine on hepatic prostanoid production in vitro following 2,4-dinitrophenol administration. Adv Shock Res 1983;10:149-159. 7. Flynn JT, Henry JM, Perkowski SZ. Phospholipase A, stimulated release of prostaglandins from the isolated, perfused rabbit liver. Can J Physiol Pharmacol 1981; 59:1268-1273. 8. Demling RH, Smith M, Gunther R, et al. Pulmonary injury and prostaglandin production during endotoxemia in conscious sheep. Am J Physiol 1981;240:H348-H353. 9. Flynn JT. Lack of a direct stimulatory effect of endotoxin on hepatic prostanoid synthesis. Ramwell, Samuelsson, Paoletti, (eds). Abstracts, 4th International Prostaglandin Conference. Washington: Raven Press, 1979:36. 10. Flynn JT. Hepatic arachidonic acid metabolism following in vitro perfusion and endotoxin administration. Prostaglandins Leukotrienes Med 1982;9:363-371. 11. Flynn JT, Appert HW, Howard JM. Arterial prostaglandin A,, E,, and F,, concentrations during hemorrhagic shock in the dog. Circ Shock 1974;2:155-163. 12. Johnson PA, Selkurt EE. Effect of hemorrhagic shock on renal release of prostaglandin E. Am J Physiol 1976; 230:831-838.
13. Collier JG, Herman AG, Vane JR. Appearance of prostaglandins in the renal venous blood of dogs in response to acute systemic hypotension produced by bleeding or endotoxin. J Physiol (Lond) 1972;230:17P-18P. 14. Jakschik BA, Marshall GR, Kourik JL, et al. Profile of circulating vasoactive substances in hemorrhagic shock and their pharmacologic manipulation. J Clin Invest 1974; 54~842-852. 15. Flynn JT, Bridenbaugh GA, Lefer AM. Release of prostaglandin F,, during splanchnic artery occulsion shock. Am J Physiol 1976;230:684-690. 16. Flynn JT, Bridenbaugh GA, Lefer AM. Plasma prostaglandin F,, and 15-keto-F,, concentrations during splanchnic artery occlusion shock. Proc Sot Exp Biol Med 1976; 151:193-197. 17. Herman AG, Vane JR. Release of renal prostaglandins during endotoxin induced hypotension. Eur J Pharmacol 1976;39:79-90. 18. Anderson FL, Jubiz W, Tsagaris TJ, et al. Endotoxin induced prostaglandin E and F release in dogs. Am J Physiol 1975;228:410-414. 19. Fletcher JR, Ramwell PW, Herman CM. Prostaglandins and the hemodynamic course of endotoxin shock. J Surg Res 1976;20:589-594. 20. Anderson FL, Tsagaris TJ, Jubiz W, et al. Prostaglandin F and E levels during endotoxin induced pulmonary hypertension in calves. Am J Physiol 1975;228:1479-1482. 21. Kessler E, Hughes RC, Bennett presence of prostaglandin-like dogs with endotoxin shock. 85-94.
EN, et al. Evidence for the material in the plasma of J Lab Clin Med 1973;81:
22. Bult H, Beetens J, Vercruysse P, et al. Blood levels of 6keto-PGF,,, the stable metabolite of prostacyclin during endotoxin induced hypotension. Arch Int Pharmacodyn Ther 1978;236:285-286. 23. Cook JA. Wise WC, Halushka PV. Elevated levels in the rat during endotoxic shock. 1980;65:227-230.
thromboxane J Clin Invest
24. Korbut R, Ocetkiewicz A, Gryglewski RJ. The influence of hydrocortisone and indomethacin on the release of prostaglandin-like substances during circulatory shock in cats which was induced by an intravenous administration of rabbit blood. Pharmacol Res Commun 1978;10:371-385. 25. Anhut H, Peskar BA, Bernauer W. Release of 15-keto-13,14 dihydrothromboxane B, and prostaglandin D, during anaphylaxis as measured by radioimmunoassay. Arch Pharmacol 1978;305:247-252. 26. Crutchley DJ, Piper PJ, Seale JP. The nature of prostaglandin-like substances released from guinea pig lungs in anaphylaxis. Eur J Pharmacol 1977,44:319-323. 27. Okabe E, Oyama M, Tanaka T, et al. Prostaglandin E, and cyclic nucleotides during anaphylactic shock in rats. Jpn J Pharmacol 1980;30:773-781. 28. Barac G, Van Caneghem P, Deby C. Prostaglandins in burninduced acute oligo-anuria and oedema. Arch Int Physiol Biochim 1975;83:612-613. 29. Harms BA, Bodai BI, Smith M, et al. Prostaglandin and altered microvascular integrity after burn Surg Res 1981;31:274-280.
release injury. J
27
The involvement of blood-borne, or vasoactive, substances in circulatory shock has been under investigation for well over 100 years. Periodically, there is a flurry of interest in a newly discovered, or rediscovered, group of compounds. These bursts of interest have centered on such substances as the catechol-
From the Department of Thomas
Jefferson
of Physiology, University,
Jefferson
Medical
College
Philadelphia.
Supported in part by Grant No. 28023 from the National tutes of Health, Institute of General Medical Sciences.
Insti-
Presented at the First International Conference on the Basic Mechanisms and Clinical Management of Shock, Merrillville, Indiana, September 10-l 1, 1982, and accepted for publication at that time. Address reprint requests to Dr. Flynn: Jefferson Medical College, Department of Physiology, 1020 Locust Street, Philadelphia, PA 19107. Key Words: 20
Cells, prostaglandins,
shock.
amines, bradykinin, the steroids, the renin-angiotensin system, and so on, including the latest: the opiates. The interest in the involvement of prostaglandin-like materials in circulatory shock began in the early 1970s and is still very much in progress because of the discovery of new subgroups of prostanoids. The hypothesis with which we are dealing is shown in Figure 1. It states that products of the arachidonic acid cascade are elaborated by the cell during periods of injury. We suggest that these newly synthetized products can participate in the homeostatic response of that cell to the injury. Conversely, we also believe that under specific conditions, prostaglandin-like materials may play a role in the pathophysiology of the shock state. The first test of this hypothesis is to determine whether the arachidonic acid cascade is activated in viro during periods of circulatory shock. Table 1 summarizes several studies of that phenomenon. Independently of the type of insult or the species, prostanoid concentrations in plasma or lymph always increase during the shock state. We have also shown that in hemorrhagic, endotoxic,’ and splanchnic artery occlusion (SAG) shock,? this increase in the plasma prostanoid concentration is due to an increased rate of de nova synthesis, not to a decreased rate of metabolism. So, there is no doubt that the arachidonic acid cascade is being stimulated during circulatory shock and other forms of in V~VOglobal injury. The next two points to be addressed are the how and why of the system. Our laboratory has studied these questions using the classic techniques of stimulation and ablation. Our overall approach has been to use the same stimulus for injury in an intact animal model and then to compare those in ~ivo results with results obtained in vitro with a perfused rabbit liver model. Figure 2 shows a schematic of this perfusion model. The primary advantage of this system for studying prostaglandin production is that it is of a nonrecirculating design. The perfusate leaving the liver is either sampled or discarded; it does not re-enter the reservoir. By knowing the prostanoid concentration in the effluent; the flow rate of perfusate, which is constant, and the liver weight, we can very precisely calculate the rate of prostaglandin synthesis per gram of wet tissue weight. We have used this model to assess changes in endogenous prostaglandin synthesis during hypoxia, metabolic poisoning, and so on. An important aspect of circulatory shock is tissue hypoxia or anoxia. Figure 3 shows changes in the rate of hepatic production of PGF,, when the perfusate is
FLYNN W PROSTANOIDS
MEMBRANE PHOSPHOLIPIDS
I,-
Phospholipase AZ, Ca++
ARACHIDONICACID FIGURE 1. Hypothesis for a role of prostaglandinlike substances in circulatory shock and cellular injury. Products of the amchidonic acid cascade are elaborated by the cell during periods of injury. These products can participate in the homeostatic response of that cell to the injury or may be involved in the pathophysiologic consequences of the injury.
5 HPETE
12 HPETE
15 HPETE
Leukotrienes PG12
equilibrated with 95% N,-5% CO2 instead of oxygen, beginning at time zero. Over two and a half hours, there is a modest 25% increase in the rate of PGF,, production. If 0.1 p,g/g/min of arachidonic acid was infused into the liver concurrent with hypoxia, the rate of PGF,, production was increased by fourfold. This increased rate of arachidonic acid metabolism had a salutary effect on the liver. While hypoxic livers became very edematous, the arachidonic-acid-treated livers increased in weight by only 5~%.~The same profile was seen with the rate of release of lactic dehydrogenase by the livers. There was little release from the livers in the sham group, significant release of lactate dehydrogenase (LDH) during hypoxia, and an inhibition of LDH release by the arachidonic acid treatment. The results of this study suggest that first, hypoxia does weakly stimulate prostanoid production in
FIGURE 2. Rabbit liver perfusion apparatus. The system is of a flow-through design that permits the calculation of rates of prostanoid synthesis in picograms released per minute per gram wet weight. P = pressure sensor; T = temperature sensor: TE = temperature exchanger.
PGE2
the liver in vitro, and second, that the stimulation of the arachidonic acid cascade allowed the liver cells to better tolerate the injury. Not being satisfied with a 25% stimulation, we made the assumption that the perfused rabbit liver was quite capable of anaerobic metabolism and that we had not really induced a shock-like injury. We produced a more severe energy deficit by adding 100 p.mol/l 2,4 dinitrophenol (DNP) to the reservoir (Fig. 4).4 This agent uncouples oxidative phosphorylation from most, if not all, energy sources. There is a significant increase in the rate of PGF,, production to over 1000 pg/g/min with DNP treatment. A similar change in the rate of prostacyclin release to over 3000 pg@nin was observed with DNP (Fig. 5). The severity of the model is dramatized by an almost linear increase in the rate of loss of acid phosphatase activity from the liver cells.
RESERVOIR
WATER
BATH
25 b
0
25
50
75
125
0
m
hypoxia
sham
L
I 90
Time Cminl
60
I
1 150
I 120
I
FIGURE 5 (right). Prostacyclin production following 100 pmolil dinitrophenol (DNP). Values are the mean ? SEM for seven livers. Asterisks signify P < .05 compared with the control. pre-DNP value. Prostacyclin was measured by its stable metabolite. 6-keto PGF,,. (From Flynn JT. Adv Shock Res 1981:5:149-162.)
FIGURE 3 tabore). Changes in the rate of PGFZa synthesis by the perfused rabbit liver on a percent of control basis. The zero value represents rates of PGF, release of 33 -C 4. 3 1 L 5, and 30 2 8 nanograms of PGF?, released per minute by the perfused livers of the sham, hypoxia, and hypoxia plus arachidonic acid groups, respectively. Six livers were studied in each group. (From Flynn JT. Prostaglandins 1979:17:39-52.) FIGURE 4 (abore, right). Rate of PGFzu release in sham and dinitrophenol (DNP) treated livers. The DNP was added to the perfusate at a concentration of 100 pmolil at time zero. Asterisks signify P < .05 compared with values in the sham group. Six livers were studied in each group. Values are the mean ? SEM. (From Flynn JT. Adv Shock Res 1981;5: 149-162.
30
I
. hypoxia 6 AA
l
l
3000
control (7)
0
500
1000
1500
2000
I 60
TIME
I
IN MINUTES
60
(1)
TIME IN MINUTES
30
DNP
o sham l
30
:ontrol
’
’
’_
’
120
I
AFTER
90
(‘I)
DNP
120
AFTER DNP
90
*
180
*
180
30 90 120 150
10 uM lidocaine
FIGURE 8 (right). Effect of 10 umoV1 dinitrophenol and 10 urnoh I lidocaine on the rate of acid phosphatase released by the perfused liver. Asterisks denote significant differences between the values in each group and those of the sham plus vehicle group at each time point. Stars denote P < .05between the noted groups and the lidocaine alone group. The effect of dinitrophenol plus lidocaine was to exacerbate the degree of injury. (From Flynn JT. Adv Shock Res 1983:10:149-159.)
FIGURE 7 (above, right). Thromboxane production in perfused livers receiving vehicle, 10 pmol/l dinitrophenol, or 10 umol/l dinitrophenol plus 10 umol/l lidocaine. The range of thromboxane release was quite narrow compared with prostacyclin release (Fig. 6). No significant differences were observed. (From Flynn JT. Adv Shock Res 1983;10:149-159.)
160
10 PM DNP 8 lo @I lidocaine
TlME IN MINUTES
60
.-.--
---0
FIGURE 6 (above). Rate of prostacyclin release in isolated, perfused rabbit livers treated with vehicles, 10 urnoh dinitrophenol, or 10 umol/l dinitrophenol plus 10 urnoh lidocaine. All values are the mean * SEM for six observations in each group. Asterisks signify P < .05when the value in the noted group is compared with the corresponding value in the sham plus vehicle group. Note that lidocaine alone moderately but significantly elevated the rate of prostacyclin production in noninjured livers but significantly attenuated the prostacyclin response in injured livers. (From Flynn JT. Adv Shock Res 1983;10:149-159.)
0
sham B vehicle
w......w 10 IJM DNP
B
0
100
200
300
400
500
600
700
800
0
30
30
90
120
1
90
120
TIME IN MINUTES
60
150
150
I
180
180
L
10 ~JMlidocaine
.-.-a
L
lo PM DNP 8 10 @l lidocaine
r-e
TIME IN MINUTES
60
1
10 j~t.ADNP
w-
1
sham b vehicle
-
AMERICAN
TABLE 1.
JOURNAL
OF EMERGENCY
Prostanoid
Synthesis
in Vivo
Hemorrhagic
PGA,, PGE,, PGF,, PGE PGE PGE
Splanchnic artery occlusion Endotoxic
PGF,, 15keto
Anaphylactic
Burn
Numbers
designate
H Volume
during
Prostanoid
Type of Shock
l
MEDICINE
PGF,,
PGE,, PGF,, PGE, PGF PGF,,, PGI,, TxB, PGE, PGF PGE, PGF PG-like materials PGI, TxB, PG-like materials TxB,, PGD, PG-like materials PGE, PGE PGI, references
2, Number
Periods
of Circulatory
Species
Shock
Change in Concentration Increase
Dog Dog Dog Qog Dog Dog
Increase Increase Increase
Dog Dog
Increase
Sheep Baboon Calf Dog Rabbit Rat Cat Guinea Guinea Rat Dog Sheep
pig pig
Increase Increase Increase Increase Increase Increase Increase Increase Increase Increase Increase
Reference’ Flynn et al” Johnson & Selkurt” Collier et a/l3 Jakschik et all4 Flynn et a/l5 Flynn et all6 Herman & VaneI Anderson et a/l8 Demling et aI8 Fletcher et a/l9 Anderson et a/20
Kessler et aI*’ Bult et a122 Cook et a123 Korbut et a124 Anhut et a125 Crutchley et a12’j Okabe et a127 Barac et a128 Harms et a12g
at the end of this article.
To determine the role played by these newly synthetized prostaglandins in this form of injury, we used lidocaine, a drug reported by Fletcher er al5 to have several effects on the prostaglandin system. When we added 10 pmol/l lidocaine to a reduced DNP dosage of 10 p,mol/l, we observed that DNP alone stimulated prostacyclin production (Fig. 6), as we had observed in the earlier studye In addition, the lidocaine significantly attenuated the DNP-evoked prostacyclin production. If we examine the concurrent production of thromboxane-like materials, we see a different situation (Fig. 7). In comparison with prostacyclin production, DNP does not appear to significantly stimulate thromboxane production. Lidocaine has little effect on the rate of thromboxane synthesis. So, we have divided the system to some degree; that is, we have inhibited prostacyclin and maintained a low rate of thromboxane release during a period of cellular injury. Under these conditions there seems to be a greater degree of injury. In Figure 8, the fine dotted line shows the effect of DNP alone on the rate of release of acid phosphatase, while the heavier broken line shows the release in the presence of DNP plus lidocaine. There is a significantly greater loss of the enzyme into the perfusate in this group. Lidocaine alone does not induce the injury. Similar results were seen with lactic dehydrogenase release. With this enzyme, significantly greater rates of release in the DNP plus lidoCaine group were seen compared with lidocaine alone. No direct effects of lidocaine were apparent. 24
1
From these results and those of the previous study, it appears that endogenously synthetized prostaglandin-like materials can influence the integrity of the cell during periods of cellular injury, with prostacyclinlike materials playing a beneficial homeostatic role; the role of thromboxane-like materials remains to be assessed. We were also interested in the effect of endogenously synthetized prostanoids on the noninjured cell. To study this, we infused a group of livers with phospholipase A,, an enzyme that cleaves arachidonic acid from membrane phospholipids.’ The rate of thromboxane B, production is shown in Figure 9 for both control and PLA, infusions at a rate of 100 units per minute. There is a significant increase in the rate of thromboxane release with PLA,, which returns toward baseline values even during the period of the PLA, infusion. In contrast, Figure 10 shows the rate of prostacyclin release in the same livers. The PLA, infusion results in a long-lasting prostacyclin release, which extends past the end of the infusion period. It is important that the PLA, livers showed no signs whatever of being injured. There was no weight gain, no enzyme release, and no hemodynamic change. So, newly synthetized prostanoids do not appear to significantly affect normal healthy cells, while we have previously seen that they do have various effects on injured cells. Having characterized the prostanoid response of the isolated perfused liver to several different stimuli, we now sought to correlate in viva responses with the in vitro liver data. In a study with Demling,* we studied
FLYNN n PROSTANOIDS
FIGURE 9. Rate of thromboxane Bz release from the isolated perfused rabbit liver by phospholipase A, infusion. Values are expressed as a percent of the control value for each group. All data are presented as the mean 2 SEM for seven livers in the sham group and nine livers in the phospholipase A, infusion group. Horizontal box on the abscissa represents the duration of the phospholipase AZ infusion at a rate of 100 U/min in 0.1 mlimin. Asterisks signify statistically significant differences (p < .05) in the rate of release of TxB, between groups at each iime point. (From Flynn et al. Can J Physiol Pharmacol 1981;59:1268-1273.
r l
.
I *
.
5
10
20 TIME
25
QROUP
PHOSPHOLIPASE
30
35
40
OROUP
45
50
IN MINUTES
liver perfusate without any effect on either prostanoid production or any index of cellular injury.9 We also quantitated the effect of endotoxin in vitro on Cl4 labeled arachidonic acid metabolism by liver slices.1° We could find no differences in either the pattern or the magnitude of arachidonic acid metabolism between control tissue and tissue treated with endotoxin in vitro. It became apparent that the endotoxin-mediated stimulation of prostanoid production in vivo was dependent upon another process or another tissue type. To further determine what that process might be, we did a study with Gee, in which we used a sheep model
both plasma and lung lymph concentrations of PGF,, and PGI, in conscious sheep receiving gram-negative Escherichia coli endotoxin. These animals demonstrated increased concentrations in plasma, and especially lung lymph, of PGF,, and PGI, during an initial phase response to endotoxin characterized by pulmonary hypertension and decreased cardiac output. These values returned toward baseline values during the post-pulmonary hypertensive phase, with prostacyclin in lung lymph remaining elevated. This was quite a marked response in the intact sheep to 2 pg/ kg endotoxin. In a similar in vitro study, we added up to an equivalent of 10 mg/kg of E coli endotoxin to the
FIGURE 10. Rate of 6keto PGF,, release from the perfused rabbit livers by phospholipase A, infusion, expressed on a percent of control period value for each group. Values are expressed as the mean 2 SEM. Seven livers were studied in the sham group and nine in the phospholipase AZ infusion group. Asterisks signify P < .05 compared with the sham group at each time period. Duration of phospholipase AZ infusion is denoted by the horizontal box on the abscissa. (From Flynn er al. Can J Physiol Pharmacol 1981;59:12681273.)
15
SHAM
180
.
SHAM
.
PHOSPHOLIPASE
GROUP QROUP
180
1
trol con-,
5
10
I
I
I
I
I
I
1
I
15
20
25
30
35
40
45
50
TIME
IN MINUTES
25
AMERICAN
JOURNAL
OF EMERGENCY
MEDICINE
m
Volume 2, Number 1
CONTROL B VEHICLE
.
A ZYMOSAN ACTIVATED PLASMA FIGURE 11. Rate of thromboxane release by the isolated rabbit liver following the infusion of zymosan-activated plasma. The complement-activated plasma was infused during the period noted by the hatched bar at a rate of 1.0 mUmin into a perfusate flow rate of 120 mUmin. Asterisks signify P < .OS for values between groups at each time point. (Flynn JT, Hellerman P: unpublished observation. 1
0
15
30
45
60
75
90
105
120
TIME IN MINUTES
and characterized the in viva prostanoid response to an infusion of complement activated plasma (unpublished observation). Almost immediately after the infusion was begun, there was a significant rise in pulmonary arterial pressure, an increase in the pulmonary lymph flow rate, and a stimulation in endogenous thromboxane and prostacyclin release. This type of hemodynamic and prostanoid response after activated complement infusion is quite similar to the response following endotoxin administration. In both of these in vivo models, we have abolished the early hemodynamic phase with cycle-oxygenase or thromboxane synthetase inhibitors. We then again used the in vitro liver perfusion and tested the effect of complement-activated plasma (Fig. 11) (unpublished observation). There was a stimulation of thromboxane B, production, which demonstrated the same temporal characteristics we observed with both complement administration in viva and the phospholipase A, infusion in vitro. That is, the enzyme is regulated and is turned off by product inhibition, cofactor depletion, or perhaps free radical feedback.
TABLE 2.
Effect of Cobra Venom
Factor
(CVF) on the Prostanoid
WBC (103/mm3) Control 5 min 10 min 15 min 30 min 90 min 120 min 15 min
26
CVF CVF CVF post post post post
CVF CVF CVF endotoxin
This does not seem to be a permanent feedback inhibition, in that given 10 to 15 minutes, an identical response can be evoked. A similar profile was seen for prostacyclin synthesis on a much smaller scale. Table 2 presents preliminary information on one rabbit and is an attempt to demonstrate the prostanoidcomplement relationship in vivo (unpublished observation). In this rabbit, we measured the leukocyte count, plasma PGI, and TxB, and the CH,, complement titer. We infused cobra venom factor at 25 units/ kg over a IS-minute period. This factor activates the endogenous complement system, as confirmed by the drop in the CH,, value and the leukopenia. Concurrently there was a transient increase in plasma prostacyclin and a more prolonged increase in thromboxane. After the cobra venom infusion was completed, the leukocyte count rebounded, the prostanoids varied somewhat, and the complement titer remained depressed. The interesting finding was that when a massive dose of endotoxin was given to this animal, there was no apparent leukopenia, no hemodynamic change, and no sign of the massive release
4.6 2.8 3.3 3.2 3.2 4.6 5.9 5.4
Response
to Endotoxin
PGI,
TXB,
(w/ml)
(w/ml)
0.134 0.437 0.164 0.100 0.203 0.100 0.380 0.620
1.045 1.136 1.362 2.122 1 S96 1.969 0.470 0.433
%o 606 30 71 36 79 42 125 116
FLYNN w PROSTANOIDS
of thromboxane that we have seen in earlier models. We did notice a modest increase in prostacyclin. It must also be noted that this rabbit was in stable condition for 45 minutes after a dose of 5 mg/kg endotoxin but died within 50 minutes because of circulatory collapse. In summary, a substantial amount of information suggests that prostaglandin-like materials are synthetized and released during periods of cellular injury. These materials, when released, can play a role in both the homeostatic response to, and/or the pathophysiologic consequences of, the injury. Future investigations should be directed toward defining the role of these substances in specific forms of injury and in elucidating the mechanisms of interaction between the prostanoids and other humoral systems during the shock state.
REFERENCES 1. Flynn JT, Lefer AM. Prostaglandin metabolism during circulatory shock. Biochim Biophys Acta 1977;497:775-784. 2. Flynn JT, Bridenbaugh GA, Lefer AM. Clearance of prostaglandin F,, during circulatory shock. Life Sci 1975; 17:1699-1706. 3. Flynn JT. Influence of the arachidonic cascade on the in vitro hepatic response to hypoxia. Prostaglandins 1979;17:39-52. 4. Flynn JT. Effect of 2,4-dinitrophenol on the rate of hepatic prostaglandin production. Adv Shock Res 1981;5: 149-162. 5. Fletcher JR, Ramwell PW. Lidocaine or indomethacin improves survival in baboon endotoxic shock. J Surg Res 1978;24:154-160. 6. Flynn JT. Effect of lidocaine on hepatic prostanoid production in vitro following 2,4-dinitrophenol administration. Adv Shock Res 1983;10:149-159. 7. Flynn JT, Henry JM, Perkowski SZ. Phospholipase A, stimulated release of prostaglandins from the isolated, perfused rabbit liver. Can J Physiol Pharmacol 1981; 59:1268-1273. 8. Demling RH, Smith M, Gunther R, et al. Pulmonary injury and prostaglandin production during endotoxemia in conscious sheep. Am J Physiol 1981;240:H348-H353. 9. Flynn JT. Lack of a direct stimulatory effect of endotoxin on hepatic prostanoid synthesis. Ramwell, Samuelsson, Paoletti, (eds). Abstracts, 4th International Prostaglandin Conference. Washington: Raven Press, 1979:36. 10. Flynn JT. Hepatic arachidonic acid metabolism following in vitro perfusion and endotoxin administration. Prostaglandins Leukotrienes Med 1982;9:363-371. 11. Flynn JT, Appert HW, Howard JM. Arterial prostaglandin A,, E,, and F,, concentrations during hemorrhagic shock in the dog. Circ Shock 1974;2:155-163. 12. Johnson PA, Selkurt EE. Effect of hemorrhagic shock on renal release of prostaglandin E. Am J Physiol 1976; 230:831-838.
13. Collier JG, Herman AG, Vane JR. Appearance of prostaglandins in the renal venous blood of dogs in response to acute systemic hypotension produced by bleeding or endotoxin. J Physiol (Lond) 1972;230:17P-18P. 14. Jakschik BA, Marshall GR, Kourik JL, et al. Profile of circulating vasoactive substances in hemorrhagic shock and their pharmacologic manipulation. J Clin Invest 1974; 54~842-852. 15. Flynn JT, Bridenbaugh GA, Lefer AM. Release of prostaglandin F,, during splanchnic artery occulsion shock. Am J Physiol 1976;230:684-690. 16. Flynn JT, Bridenbaugh GA, Lefer AM. Plasma prostaglandin F,, and 15-keto-F,, concentrations during splanchnic artery occlusion shock. Proc Sot Exp Biol Med 1976; 151:193-197. 17. Herman AG, Vane JR. Release of renal prostaglandins during endotoxin induced hypotension. Eur J Pharmacol 1976;39:79-90. 18. Anderson FL, Jubiz W, Tsagaris TJ, et al. Endotoxin induced prostaglandin E and F release in dogs. Am J Physiol 1975;228:410-414. 19. Fletcher JR, Ramwell PW, Herman CM. Prostaglandins and the hemodynamic course of endotoxin shock. J Surg Res 1976;20:589-594. 20. Anderson FL, Tsagaris TJ, Jubiz W, et al. Prostaglandin F and E levels during endotoxin induced pulmonary hypertension in calves. Am J Physiol 1975;228:1479-1482. 21. Kessler E, Hughes RC, Bennett presence of prostaglandin-like dogs with endotoxin shock. 85-94.
EN, et al. Evidence for the material in the plasma of J Lab Clin Med 1973;81:
22. Bult H, Beetens J, Vercruysse P, et al. Blood levels of 6keto-PGF,,, the stable metabolite of prostacyclin during endotoxin induced hypotension. Arch Int Pharmacodyn Ther 1978;236:285-286. 23. Cook JA. Wise WC, Halushka PV. Elevated levels in the rat during endotoxic shock. 1980;65:227-230.
thromboxane J Clin Invest
24. Korbut R, Ocetkiewicz A, Gryglewski RJ. The influence of hydrocortisone and indomethacin on the release of prostaglandin-like substances during circulatory shock in cats which was induced by an intravenous administration of rabbit blood. Pharmacol Res Commun 1978;10:371-385. 25. Anhut H, Peskar BA, Bernauer W. Release of 15-keto-13,14 dihydrothromboxane B, and prostaglandin D, during anaphylaxis as measured by radioimmunoassay. Arch Pharmacol 1978;305:247-252. 26. Crutchley DJ, Piper PJ, Seale JP. The nature of prostaglandin-like substances released from guinea pig lungs in anaphylaxis. Eur J Pharmacol 1977,44:319-323. 27. Okabe E, Oyama M, Tanaka T, et al. Prostaglandin E, and cyclic nucleotides during anaphylactic shock in rats. Jpn J Pharmacol 1980;30:773-781. 28. Barac G, Van Caneghem P, Deby C. Prostaglandins in burninduced acute oligo-anuria and oedema. Arch Int Physiol Biochim 1975;83:612-613. 29. Harms BA, Bodai BI, Smith M, et al. Prostaglandin and altered microvascular integrity after burn Surg Res 1981;31:274-280.
release injury. J
27