Leukotrienes increase levels of prostanoids in cerebrospinal fluid in piglets

PROSTAGLANDINS

LEUKOTRIENES INCREASE LEVELS OF PROSTANOIDS IN CEREBROSPINAL FLUID IN PIGLETS

David W. Buslja and Charles W. Leffler Department of Physiology and Biophysics University of Tennessee Memphis, Tennessee 38163 ABSTRACT We investigated effects of exogenous leukotrienes (C~, D~, or E4) on levels of prostanolds in cerebrospinal fluid in newborn pigs (1-5 days). A "closed" cranial window was placed over the parietal cortex. Pial arterial diameter was measured with a microscope and electronic micrometer system. Levels in cerebrospinal fluid (CSF) of 6-keto-Prostaglandin F1a (6-keto-PGFi~), Thromboxane B 2 (TXB2) , and Prostaglandin E 2 (PGE2) were measured by radioimmunoassay. Topical application of leukotrlenes C4, D~, or E 4 (5,000 ng/ml) similarly constricted pial arteries by 15 ± 2% (n = 14) (mean ± SEM). In addition, leukotrlenes increased levels of 6-keto-PGFi~ from 806 ± 136 to 1,612 ± 304 pg/ml (n = 13), TXB 2 from 161 ± 31 to 392 ± 81 pg/ml (n = i0), and PGE 2 from 2,271 ± 342 to 4,636 ± 740 pg/ml (n = 13). Each type of leukotrlene had similar effects on prostanoid synthesis. In other experiments (n = 5), we found that 2.0 ng/ml PGE 2 in CSF dilated plal arteries by 24 ± 8% and that 1.0 ng/ml PGI 2 dilated pial arteries by 15 ± 6%. These results indicate that leukotrienes are able to increase levels of prostanoids in cerebral cortex. INTRODUCTION Leukotrlenes are a group of biologically active compounds which are synthesized by various tissues in response to injury (1,2). Leukotrienes have potent effects on airway smooth muscle tone, blood flow, and vascular permeability (1,2). These effects of leukotrlenes may or may not be modulated by secondary release of prostanoids

(3-13). Leukotrlenes are synthesized in brain during several pathological conditions (14,15). We recently have shown that leukotrlenes, when applied to the surface of the parietal cortex, are potent constrictors of cerebral arteries in newborn pigs (16). Whether prostanoids modulate this response is unknown. The purpose of this study was to determine whether leukotrienes increase levels of prostanoids in piglet cerebral cortex. METHODS Twenty-two piglets (1-5 days, 1.0-1.4 kg) of either sex were used. They were anesthetized initially with ketamlne hydrochloride (33 mg/kg, im) and acepromazlne (3.3 mg/kg, im), and anesthesia was maintained with ~-chloralose (30-50 mg/kg initially, followed with 5 mg/kg.hr, iv). The piglets were intubated and ventilated with room air. Catheters were inserted into the right femoral artery to record blood pressure and to sample for blood gases and pH, and into the

DECEMBER 1986 VOL. 32 NO. 6

803

PROSTAGLANDINS

right femoral vein for injection of drugs and fluids. Body temperature was maintained at 37°-38"C with a water circulating heating pad. Following retraction of the scalp, an opening approximately 2 cm in diameter was made in the skull over the parietal cortex with bone rongeurs. The dura was cut and reflected over the cut bone edge. Special care was taken to avoid touching the brain. A cranial window, similar to that described by Levasseur et al. (17) was placed into the opening and cemented into place with dental acrylic. After the dental acrylic hardened, the space under the window was filled with artificial cerebrospinal fluid (CSF) through needles incorporated into the sides of the window. The composition of CSF was 220 mg KCI, 132 mg MgCI2, 221 mg CaCI2, 7,710 mg NaCI, 402 mg urea, 665 mg dextrose, 2,066 mg NaHCO 3 per liter, pH = 7.33; ~CO^ = 46 ~ dg; PO = 43 mm Hg. Pial arteries were observed with a W~ild trinocular microscope, and diameter measured with a television camera mounted on the microscope, a video monitor, and a video micrometer (model VPAI000, FOR-A-CORPORATION) (16). Experimental Design The CSF under the window was flushed out with fresh CSF 4 times at 5-minute intervals. After the fourth flush, cerebral arterial diameter was measured and, at the end of five minutes, a 300 ~i sample of CSF was collected from one of the outflow needles for prostanoid analysis. Then, leukotriene C 4 (LTC 4) (n = 5), LTD~ (n = 6), or LTE 4 (n = 3) was placed under the window at concentrations of I, I0, I00, 1,000, and 5,000 ng/ml CSF (synthetic leukotrienes were gifts from J. Rokach, Merck-Frosst, Canada, Inc.). Pial artery diameter was measured at peak response for each dose (usually 2-4 minutes) and, after 5 minutes at 5,000 ng/ml, a 300 ~i sample of CSF was taken for prostanoid analysis. The response for each dose largely was sustained for the 5-minute period. We did not examine the length of this response. In 5 other animals, diameter of pial arteries was measured during control conditions and following infusion under the window of 0.2 and 2.0 ng/ml PGE 2 or 0.5 and 1 ng/ml prostaglandin 12 (PGI 2) (Cayman). Each dose was left under the window for 5 mlnutes. Values for peak responses were recorded (usually 1-2 minutes). Dilation at each dose usually was sustained for PGE 2 but faded for PGI 2. For PGE_, the vehicle was artificial CSF, and for PGI~, the vehicle was TK~S buffer at pH I0. The responses to the vehlcles were compared to those of vehicle plus prostanoids. Neither vehicle changed pial artery diameter detectably. Indomethacin tryhydrate (a gift from Merck; I0 mg/kg, iv) was given 20 minutes prior to this protocol to eliminate endogenous production of prostanoids. Effects of indomethacin on pial arteries are sustained for at least 90 minutes, which is longer than the time needed to complete these experiments (unpublished observations). In pilot experiments in 3 piglets, we examined feasibility of examining pial arterial responses to leukotrienes before and after administration of vehicle or indomethacin. Before vehicle, pial arterial diameter was 120 ± 27 ~m during the control condition, and

804

DECEMBER 1986 VOL. 32 NO. 6

PROSTAGLANDINS

Table I.

Diameter (pm) Arterial Blood Pressure (mm Hg)

Effects of Leukotrienes (C4, D4, or E4) on Pial Arterial Diameter

Control

I ng/ml

I0 ng/ml

202 ±23

195 i23

189 a ±22

62 ±3

61 ±4

61 ±3

I00 ng/ml 184 ab i22 61 ±3

1,000 ng/ml

5,000 ng/ml

177 abc ±20

171 abcd ±21

61 ±3

62 ±3

Values are means ± SEM for 14 vessels from 14 animals. Arterial blood pH = 7.34 + 0.02, PCO 2 = 32 + I mm Hg, and P02 = 80 ± 3 mm Hg. bP < 0.05, Compared Ccompared ~Compared

compared to control. to i ng/ml. to I0 ng/ml. to I00 ng/ml.

5000

l--t 4000

/i

Control

/ / i

After L e u k o t r i e n e s

/~

///// /i/// ////i

m

E

I"/I

"/J Ir/.,

/ / / / f

///// I/ill ///// /////

3000

Q, T

2000

///// // // // // // //i// ///// IIIII

r/Z/l, I/Ill /////

1000

IIIII

////# I/I// II/// Ilili

I//// /I///

6 - K e t o - PGF l(X N=13 Fig. I.

TXB2 N=IO

PGE 2

N----13

Prostanoid synthesis during infusion of cerebrospinal fluid containing no leukotrienes or containing leukotrienes (C4, D4, or E~). Values are means ± SEM. *P < 0.05, compared to control.

DECEMBER 1986 VOL. 32 NO. 6

805

PROSTAGLANDINS

109 ± 26 ~m (-II ± 4%) and 106 ± 26 ~m (-15 ± 4%) following infusion of 1,000 and 5,000 ng/ml LTE4, respectively (5 vessels from 3 piglets). After 20-30 minutes walt with plain CSF under the window, we repeated this protocol. Control diameter was ii0 ± 26 Bm and diameter was 116 ± 32 ~m during 1,000 ng/ml LTE~ and Iii ± 27 ~m during 5,000 ng/ml LTE 4. We find that pial arterial responses to another constrictor stimulus, namely noreplnephrine, is present after repeated administration (18). Prostanold Analysis We determined concentrations in CSF of 6-keto-PGFla , TXB2, and PGE 2 using radioimmunoassay (19). Antibodies to 6-keto-PGF1a, TXB2, and PGE 2 were produced in rabbits immunized with prostanoids coupled to thyroglobulin using the mixed anhydride method. Crossreactlvities of the prostanold antibodies with other, known, biologically relevant eicosanoids tested were all below 1%. Values for these three prostanolds were below detectable levels when we used 5,000 ng/ml leukotrienes as a sample. The assays were performed in gelatin-Tris buffer using the appropriate trltiated elcosanold. After 24 hours of incubation at 4°C, the free fraction was separated from the fraction bound to antibody by precipitating the rabbit antibodies with anti-rabbit y-globulln and 60% saturated ammonium sulfate. Data were handled by computer, with determlnation of secondorder regression of free tracer over tracer bound to antibody against unlabeled eicosanold by the method of least squares. All unknowns were assayed at three dilutions with parallelism between the unknown dilution curve and the standard curve required before the result was used. Sample dilutions used in the present study allowed analysis of eicosanoid concentrations between i00 and 50,000 pg/ml. Statistical Analysis All values are reported as mean ± SEM. Vessel diameter and arterial blood pressure during leukotriene applications were compared using an analysis of variance for repeated measures, and, if significant, pairwise comparisons were made using the Student-Newman-Kuels test. Vessel diameter during application of 2 doses of PGI 2 or PGE 2 were compared to control values using one-tailed paired t-tests with the Bonferronl correction for a-level. Prostanoid levels in CSF were compared between control samples and after topical application of 5,000 ng/ml leukotrienes using two-tailed paired t-tests. An a-level of p < 0.05 was used in all statistical tests. RESULTS Topical application of leukotrienes to the cortical surface constricted pial arteries by 6 ± 2% at I0 ng/ml, 9 + 2% at I00 ng/ml, 12 + 2% at 1,000 ng/ml, and 16 + 2% at 5,000 ng/ml (Table i). At the highest dose, magnitude of effects were 19 _+ 3% (n = 5) for LTC~, 12 + 1% (n = 6) for LTD4, and 17 _+ 5% (n = 3) for LTE~. Arterial blood pressure did not change under these conditions. Topical application of leukotrienes resulted in an increase in CSF of levels of 6-keto-PGFla by 119 + 34%, TXB 2 by 267 + 158%, and PGE 2 by 117 + 20% (Figure i). Responses to individual leukotrienes

806

DECEMBER 1986 VOL. 32 NO. 6

PROSTAGLANDINS

were very similar. For example, PGE 2 values for control versus leukotriene applications were: 1,947 ± 813 and 4,318 ± 1,689 pg/ml, respectively, for LTC~ (n = 4); 2,295 ± 519 and 4,311 ± 1,001 pg/ml, respectively, for LTD~ (n = 6); 2,655 £ 487 and 5,710 ± 1,669 pg/ml, respectively, for LTE 4 (n = 3). Pial arterial diameter was 182 ± 26 ~m before indomethacin and 158 ± 28 ~m after administration of indomethacin (n = 4; p <0.05, via one-way paired t-test). Topical application of PGE 2 (n = 5) dilated pial arteries from 150 ± 23 ~m to 170 ± 26 Bm (14 ± 5%) at 0.2 ng/ml and to 183 ± 26 ~m (24 ± 8%) at 2.0 ng/ml. Topical PGI 2 (n = 5) dilated pial arteries from 159 ± 23 ~m to 169 ± 24 ~m (7 ± i%) at 0.5 ng/ml and to 186 ± 33 ~m (15 ± 6%) at 1.0 ng/ml. DISCUSSION The major findings of the present study in anesthetized, newborn pigs are: I) leukotrienes constrict cerebral arteries, 2) leukotrienes increase levels of prostanoids in CSF and 3) CSF levels of prostanoids induced by leukotrienes are within the vasoactive range. Thus, it seems likely that the total response elicited by topical application of leukotrienes represents a combination of constrictor and dilator actions. We have found previously in piglets that leukotrienes induce significant constriction of pial arteries (16). Our results concerning vascular effects of leukotrienes are in general agreement with those reported in adult rats, mice, and hamsters (20,21,22). However, it has been reported that leukotrienes do not constrict adult human or rabbit cerebral arteries (23,24). The reason for differences in findings is unclear, but may be due to different methods of administration of leukotrienes (intra-arterial versus topical application), different doses, and species variation. Another factor could be the types or amounts of dilator prostaglandins released in response to leukotrienes. Previous studies have found that leukotrienes induce prostanoid synthesis in some but not in other tissues. For example, indomethacin has been found to augment (4) or attenuate (3) constriction of airway smooth muscle by leukotrienes. In addition, leukotriene infusion into the pulmonary circulation results in increased synthesis of prostanoids (5). In peritoneal macrophages (6), smooth muscle cells (7), and endothelial cells (8), leukotrienes induce augmented synthesis of prostanoids. On the other hand, prostanoid synthesis does not appear to be affected by leukotrienes in the renal (12), mesenteric (12), and coronary (9,10,11) circulations. In the present study, we have shown that topical application of leukotrienes increases levels of prostanoids in CSF, and that levels of PGE 2 and PGI 2 produced are within the vasodilator range for pial arteries. Pial arteries are important resistance vessels in the cerebral circulation (25). In a previous study, we found that exogenous norepinephrine constricted pial arteries and increased levels of CSF prostanoids by similar amounts in piglets (18). ~ether synthesis of prostanoids is a direct effect of leukotrienes on cells or whether it represents secondary effects due to cerebrovascular

DECEMBER 1986 VOL. 32 NO. 6

807

PROSTAGLANDINS

constriction is unclear. It seems likely that the total cerebrovascular response involves a constrictor component due to leukotrienes and a dilator component due to PGE 2 and PGI 2. ~e only directly assessed vasoactive capabilities of PGE 2 and PGI 2 because levels of these prostanoids were much higher after leukotriene application than TKB 2. It is possible that thromboxane A 2 contributes in a minor way to the total pial arterial response. Another approach to test our hypothesis concerning prostanoid modulation of leukotriene vascular responses ~ u l d be to determine magnitude of constriction by leukotrienes before and after inhibition of cyclooxygenase by indomethacin. In another study, we found that indomethacin potentiated constriction of pial arteries to exogenous norepinephrine (18). However, we found in preliminary experiments that this approach was not feasible (see Methods). Although pial arteries constricted during application of leukotrienes before generation of the first dose-response curve, the subsequent response 20-30 minutes after administration of vehicle was inhibited severely. Reasons for this attenuation of response to leukotrienes are unclear. ~e did not feel comfortable comparing one group of animals pretreated with indomethacin with another group pretreated with vehicle because of the expected large intra-animal variability. ~e believe that the relatively high levels of prostanoids in subarachnoid CSF during control conditions, compared to levels reported in ventricular and cisternal CSF, and quickly-frozen whole brain, reflect the unique topography of the cortical surface. Prostanoid concentrations in whole brain would be expected to be low since prostanoids are not stored and do not accumulate in cells. Cerebral blood vessels, perivascular nerves, neurons, and glial cells, are active producers of prostanoids (26,27). Since the cortical surface is irregular, it has a large surface area. The CS~ layer over this area is thin, and in intimate contact with blood vessels and nerves. Therefore, it is not surprising that prostanoids produced by brain accumulate in significant amounts in cortical subarachnoid CSF. In our preparation, the pial arteries were responsive to constrictor and dilator stimuli, indicating that the cortical surface was not damaged or inflamed. There was no blood in the CSF under the window. Advantages of this method are: I) we are studying intact, undamaged brain; 2) multiple samples of CSF can be removed for prostanoid analysis; and 3) responses of cerebral arteries and CSF prostanoids levels can be correlated. Leukotrienes are synthesized in brain during pathological conditions such as concussive brain injury, subarachnoid hemorrhage, and cerebral ischemia (14,15). Conditions such as asphyxia and/or cerebral ischemia may occur at or soon after birth, and intraventricular hemorrhage is particularly prevalent in small birth-weight infants. Synthesis of leukotrienes in response to cerebral stress may result in inappropriate constriction of cerebral arteries and may possibly lead to further cerebral ischemia. Synthesis of vasodilator prostanoids in response to the presence of leukotrienes may limit this vasoconstriction. On the other hand, the presence of prostanoids has been shown to be necessary for permeability changes induced

808

DECEMBER 1986 VOL. 32 NO. 6

PROSTAGLANDINS

by leukotrienes cerebral stress circulation.

to occur (28,29). Permeability changes following could have detrimental effects on the cerebral

ACKNO~'LEDGMENTS ~e thank M.L. Gray, J. Giddens, J. Broughton, and M. Jackson for excellent technical assistance. This work was supported by grants from the National Institutes of Health and the Tennessee Affiliate of the American Heart Association. Dr. Leffler is an Established Investigator of the American Heart Association, with funds contributed in part by the Tennessee Affiliate. REFERENCES I) Samuelsson, B., P. Borgeat, S. Hammarstr~m, and R.C. Murphy. Leukotrienes: a new group of biologically active compounds. Adv. Prost. Thromb. Res. 6:1. 1980. 2) Stjernschantz, J. The le-ukotrienes. Med. Biol. 62:215. 1984. 3) Duncan, P.G. and J.S. Douglas. Influences of gender and maturation on responses of guinea-pig airway tissues to LTD 4. Eur. J. Pharmacol. 112:423. 1985. 4) Leitch, A.G., E.J. Corey, K.F. Austen, and J.M. Drazen. Indomethacin potentiates the pulmonary response to aerosol leukotriene C~ in the guinea pig. Ann. Rev. Respir. Dis. 128:639. 1983. 5) Sirois, P., P. Borgeat, A. Jeanson, $. Roy, and G. Girard. The action of leukotrienes B 4 (LTB 4) in the lung. Prost. Med. 5:429. 1980. 6) Feuerstein, N., M. Foegh, and P.W. Ramwell. Leukotrienes C~ and D 4 induce prostaglandin and thromboxane release from rat peritoneal macrophages. Br. J. Pharmacol. 72:389. 1981. 7) Clark, M.A., M. Cook, S. Mong, and S.T. Croake. The binding of leukotriene C4 and leukotriene D~ to membranes of a smooth muscle cell line (BC3H I) and evidence that leukotriene induced contraction in these cells is mediated by thromboxane, protein and RNA syntheses. Eur. J. Pharmacol. 116:207. 1985. 8) Clark, M.A., D. Littlejohn, $. Mong, and S.T. Croake. Effect of leukotrienes, bradykinin and calcium ionophore (A 23187) on bovine endothelial cells: release of prostacyclin. Prostaglandins 31:157. 1986. 9) Panzenbeck, M.J. and G. Kaley. Leukotriene D 4 reduces coronary blood flow in the anesthetized dog. Prostaglandins 25:661. 1983. I0) Boyd, L.M., D. Ezra, G. Feuerstein, and R.E. Goldstein. Effects of FPL-55712 or indomethacin on leukotriene-induced coronary constriction in the intact pig heart. Eur. J. Pharmacol. 89:307. 1983. II) Wo----odman, O.L. and G.J. Dusting. Coronary vasoconstriction induced by leukotrienes in the anaesthetized dog. Eur. J. Pharmacol. 86:125. 1983. 12) Feigen, L.P. Differential effects of leukotrienes % , D 4 and E~ in the canine renal and mesenteric vascular beds. J. Pharmacol. Exp. Therap. 225:602. 1983.

DECEMBER 1986 VOL. 32 NO. 6

809

PROSTAGLANDINS

13)

14)

15)

16)

17)

18)

19)

20)

21) 22)

23)

24)

25)

26)

27)

28)

810

Letts, L.G. and P.J. Piper. The actions of leukotrienes C 4 and D 4 on guinea-pig isolated hearts. Br. J. Pharmacol. 76:169. 1982. Kiwak, K.J., M.A. Moskowitz, and L. Levine. Leukotriene production in gerbil brain after ischemic insult, subarachnoid hemorrhage, and concussive injury. J. Neurosurg. 62:865. 1985. Moskowitz, M.A., K.J. Kiwak, K. Hekimian, an-d-L. Levine. Synthesis of compounds with properties of leukotrienes C 4 and D in gerbil brains after ischemia and reperfusion. Science 224:~86. 1984. Busija, D.W., C.W. Leffler, and D.G. Beasley. Effects of leukotrienes C4, D 4 and E 4 on cerebral arteries of newborn pigs. Pediatr. Res. In press. Levasseur, J.E., E.P. Wei, A.J. Raper, H.A. Kontos, and J.L. Patterson, Jr. Detailed description of a cranial window technique for acute and chronic experiments. Stroke 6:308. I~75. Busija, D.W. and C.W. Leffler. Eicosanoid synthesis ellcited by norepinephrine in piglet parietal cortex. Brain Res., in press, 1986. Leffler, C.W. and D.W. Busija. Prostanoids in cortical subarachnoid cerebrospinal fluid and pial arterial diameter in newborn pigs. Circ. Res. 57:689. 1985. Tagari, P., G.H. BuBol~y, V. Aitker, and D.J. Boullin. Leukotriene D 4 and the cerebral vasculature in vivo and in vitro. Prost. Leuk. Med. 11:281. 1983. Rosenblum, W.l. Co-~stricting effect of leukotrienes on cerebral arterioles of mice. Stroke 16:262. 1985. Mayhan, W.G., G. Sahagun, R. Spector, and D.D. Heistad. Effects of leukotriene C4 on the cerebral microcirculature. Am. J. Physiol. 20:H471-H474. 1986. Kamitani, T., M.H. Little, and E.F. Ellis. Effect of leukotrienes, 12-HETE, histamine, bradykinin, and 5-hydroxytryptamine on in vivo rabbit cerebral arteriolar diameter. J. Cereb. Blood Flow Metabol. 5:554. 1985. Von Hoist, H., E. GranstrDm, S. HammarstrDm, B. Samuelsson, and L. Steiner. Effect of leukotriene C4, D4, prostacyclin, and thromboxane A 2 on isolated human cerebral arteries. Acta Neurochim. 62:177. 1982. Busija, D.W. and D.D. Heistad. Factors involved in physiological regulation of cerebral olood flow. Rev. Physiol. Pharmacol. Biochem. 101:161. 1984. Abdel-Halim, M.S., H. van Hoist, B. Meyerson, C. Sachs, and E. F~nggard. Prostaglandin profiles in tissue and blood vessels from human brain. J. Neurochem. 34:1331-333. 1980. Hedqvist, P. Basic mechanisms of prostaglandin action on autonomlc neurotransmission. Ann. Rev. P~armacol. Toxicol. 17:259279. 1977. Bray, M.A., F.M. Cunningham, A.W. Ford-Hutchinson, and M.J.H. Smith. Leukotriene B4: a raediator of vascular permeability. Br. J. Pharmacol. 72:483. 1981.

DECEMBER 1986 VOL. 32 NO. 6

PROSTAGLANDINS

29)

Peck, M.J., P.J. Piper, and T.J. Willi~as. trienes C 4 and D~ on the microvasculature Prostaglandins 21:315. 1981. Editor:

F. Coceani

The effect of leukoof guinea pig skin.

Received: 5-29-86

DECEMBER 1986 VOL. 32 NO. 6

Accepted:

9-29-86

811