Summary: Indomethacin is known to attenuate quite markedly the increase in CBF during hypercapnia. Hy percapnia is, in all likelihood, mediated by the acid ...
Journal of Cerebral Blood Flow and Metabolism 13:724-727 © 1993 The International Society of Cerebral Blood Flow and Metabolism Published by Raven Press, Ltd., New York
Short Communication
Indomethacin Abolishes Cerebral Blood Flow Increase in Response to Acetazolamide-Induced Extracellular Acidosis: A Mechanism for Its Effect on Hypercapnia? Qiong Wang, *Olaf B. Paulson, and Niels A. Lassen Department of Clinical Physiology/Nuclear Medicine, Bispebjerg Hospital, and *Department of Neurology, Rigshospital, Copenhagen, Denmark
Summary: Indomethacin is known to attenuate quite markedly the increase in CBF during hypercapnia. Hy percapnia is, in all likelihood, mediated by the acid shift at the level of the smooth muscle cells of the cerebral arterioles. We therefore investigated the effect of indo methacin on the CBF increase caused by acetazolamide (Az) , a drug that induces brain extracellular acidosis, which triggers its effect on CBF. We compared the results to the inhibitory effect of indomethacin on the CBF in crease during hypercapnia. Indomethacin but not di clofenac, another potent cyclooxygenase inhibitor, was
found to block almost completely the CBF increase caused by Az-induced extracellular acidosis or by CO2, but it did not influence the CBF increase produced by sodium nitroprusside or papaverine. The results suggest that indomethacin exerts its action on CO2 reactivity by a nonprostaglandin-mediated mechanism that directly in terferes with the regulation of cerebrovascular tone me diated by extracellular pH. Key Words: Acetazolamide Cerebral blood flow-Diclofenac-Extracellular pH Hypercapnia-Indomethacin.
The effect of indomethacin, an inhibitor of cyclo oxygenase, on hypercapnic cerebral vasodilation has been intensively investigated over the last de cade. The substance almost completely blocks the increase in CBF caused by hypercapnia in several species including rat (Sakabe and Siesj6, 1979; Dahlgren et al., 1981; Quintana et al., 1988). How ever, experimental evidence argues against a pros taglandin-mediated mechanism since several other inhibitors of cyclic oxygenase, including diclofenac, do not duplicate the effects of indomethacin (Pick ard et al., 1977; Amano and Meyer, 1981; Eriksson et al., 1983; Quintana et al., 1988), and since hyper capnia is not associated with altered prostaglandin production (Ellis et al., 1982; Eriksson et al., 1983; McCalden et al., 1984). The mechanism of the strik-
ing effect of indomethacin on hypercapnic CBF re mains undetermined. Hypercapnic cerebral vasodilation is, in all like lihood, mediated by the acid shift around and/or inside the smooth muscle cells of the cerebral arte rioles (Gotoh et al., 1961; Lassen, 1968). We found it of interest, therefore, to study whether indometh acin would inhibit the cerebral vasodilation caused by acetazolamide (Az), because this drug induces a local acidosis in the brain that triggers the increase in CBF just as CO2 does (Ehrenreich et al., 1961; Severinghaus and Cotev, 1968; Heuser et al., 1975; Bickler et al., 1988; Lassen, 1990). Also, it is pos sible to separate the extracellular and intracellular acidosis caused by Az by controlling respiration: When animals are hyperventilated, Az causes a clear-cut acidosis only in the extracellular compart ment (Severinghaus and Cotev, 1968; Bickler et al., 1988). By comparing the effects of indomethacin on the CBF response to CO2 and to Az, we wanted to explore the mechanism of the pH-sensitive vasomo tor reactions of the brain. The effect of another po-
Received September 14, 1992; final revision received February 5, 1993; accepted February 24, 1993. Address correspondence and reprint requests to Dr. N. A. Lassen at Department of Clinical Physiology/Nuclear Medicine, Bispebjerg Hospital, DK-2400, Copenhagen NV, Denmark.
Abbreviation used: Az, acetazolamide.
724
INDOMETHACIN IN pHe-SENSITIVE MECHANISM IN BRAIN tent cyclooxygenase inhibitor, diclofenac, on Az and CO2-induced CBF increases was also studied. MATERIALS AND METHODS The experiments were performed in male Wistar rats, weighing 270-350 g, that were anesthetized with 0. 6% halothane170% N20, paralyzed by suxamethonium (40 mg/kg), and artificially ventilated. The surgical prepara tion was performed as described previously (Wang et al. , 1992). CBP was measured by the intracarotid 133Xe in jection method as adapted for small rodents (Hertz et al. , 1977). With this method, after an intracarotid injection of a 10- to 60-I.Ll bolus of 133Xe (10 mCiiml; Amersham), the clearance of 133Xe from the brain was determined by an external detector with a heavily collimated Nal(Th) crys tal placed over the head ipsilateral to the injection site. The output was recorded by an analog ratemeter, and CBP was determined from the 10- to 15-s initially steepest slope of the semilogarithmic clearance curve. Blood gases were analyzed in small samples of arterial blood using a Radiometer gas analyzer. Systemic blood pressure was followed using a conventional capacitance manometer. With a heating pad and infusion of blood from a donor animal when needed, body temperature (37"C) and blood volume were maintained. The effects of indomethacin (Dumex, Denmark) and diclofenac (Sigma) on the CBP increases induced by CO2 and Az (Diamox; Lederle Parenterals, Carolina, Puerto Rico) were studied on 66 rats. CBP was first measured at normocapnia 10 min after bolus intravenous injection of 10 mg/kg indomethacin or of the same dose of diclofenac. Then, hypercapnia was induced by adding 5% CO2 to the inspired gas or administering 50 mg/kg Az s.c. CBP was again determined 15 min after CO2 or 30 min after Az. With Az, two series of studies were carried out. In the first series, ventilation was not changed after Az admin istration. In the second, ventilation was increased with the aim of maintaining brain Pe02 at control levels by a procedure described in detail by Bickler et al. (1988). Briefly, 2% CO2 was added to the inspired gas during the control period, increasing ventilation to maintain normo capnia. The added CO2 was removed just after Az admin istration because a profound impairment of pulmonary CO2 exchange caused by Az made it difficult to maintain
725
brain surface Pe02 by increasing ventilation alone (Bick ler et ai. , 1988). As indomethacin was found to inhibit markedly the CO2- and Az-induced increase in CBP, its effects on the pH-independent CBP increases produced by 0. 05 mg/kg sodium nitroprusside (Sigma) or 1 mg/kg papaverine (Aygehus, Denmark) were also studied. Both compounds were dissolved in saline and administered by intracarotid infusion for 1 min (0.15 mUmin). CBP was determined 1 min later.
Statistics The Mann-Whitney U test was used for comparing val ues under control conditions and during interventions. All values are presented as means ± SD. A value of p 140 mm Hg. The physiolog ical variables recorded during the various interven tions are given in Table 1. In the group treated by Az with constant ventilation and the group with in halation of CO2, Paco2 increased while pH de creased. As illustrated in Fig. 1, CBF rose in the animals after administration of Az alone, both when venti lation was increased and when it was kept constant. With constant ventilation, the increase in CBF was from a control level of 84 ± 9 to 225 ± 39 mll00 g-I min-I, an effect of the same magnitude as with in halation of 5% CO2 giving a Paco2 of 68 mm Hg. After indomethacin, normocapnic control CBF was reduced to 56 ± 8 ml 100 g-I min-I; the Az CBF dropped from 225 ± 39 to 77 ± 8 mll00 g-I min-I and the CO2 CBF from 240 ± 29 to 85 ± 13 ml 100 g-I min-I. When Az was combined with induced hyperventilation, Az still increased CBF to 157 ±
TABLE 1. Physiological parameters after different treatments Group
Paco2 (mm Hg)
Pa02 (mm Hg)
7.27 ± 0.04
60.7 ± 7.6 59. 0 ± 5.0
93 ± 9
7.24 ± 0.04
6 1. 5 ± 6. 1
165 ± 18 170 ± 12 179 ± 8
104 ± 13 1 12 ± 12
7.37 ± 0.02 7.38 ± 0.02 7.39 ± 0.03
4 1. 5 ± 2.2 4 1. 6 ± 1. 7 40.9 ± 3. 7
204 ± 22 199 ± 18 20 1 ± 25
7.23 ± 0.02 7.23 ± 0.03 7. 2 1 ± 0.03
68. 2 ± 2.7 69. 5 ± 4. 4 68. 9 ± 5. 5
164 ± 1 1
MABP (mm Hg)
pH
109 ± 8 1 1 1 ± 15
Constant ventilation Az (n = 6) Indo + Az (n = 7) Dido + Az (n = 7) Hyperventilation Az (n = 7) Indo + Az (n = 6) Dido + Az (n = 6) Hypercapnia CO2 (n = 12) Indo + CO2 ( n = 10) Dido + CO2 ( n = 5) Indo + SNP (n = 5) Indo + papaverine (n = 4)
96 ± 7 1 16 ± 13 122 111 1 18 1 13
15 ± 16 ± 15 ± 12 ±
7.25 ± 0.04
7.39 ± 0.03
40. 3 ± 2. 1
7.40 ± 0.02
39. 3 ± 2. 8
156 169 153 163
11 7 ± 17 ± 13 ±
±
Values are means ± SD; n = no. of rats. Az, acetazolamide; Indo, indomethacin; Dido, didofenac; SNP, sodium nitroprusside.
J Cereb Blood Flow Metab, Vol. 13, No.4, 1993
Q. WANG ET AL.
726 �
I
c:
'E I
'"
200
I
150
-g
100
� .0 l'
50
o 1i
U "
17ot1
250
g
Jo o ;;::
185_
300
87111
OL-____-L���__���L___LU��� Vehicle
o Control
� CO2
Indomethacin
Oielofenae
i/l/li Az + Constant ventilation
[ZJ Az + Hyperventilation
FIG. 1. The mean values of CBF under different conditions.
After indomethacin, basal CBF was reduced to 67%, while much more marked reduction of the response to CO2, acet azolamide (Az), and Az combined with hyperventilation was seen. Oiclofenac had no such effects. The numbers above the bars indicate the percentage increase above the corre sponding control levels. All values are means ± SO. **p < 0.01 versus rats treated by vehicle alone.
17 ml 100 g 1 min 1. In this series where Paco2 and pHa both remained as in the control state (Table 1), indomethacin almost completely abolished the Az induced increase in CBF (Fig. 1). Diclofenac at 10 mg/kg had no effect either on resting or on Az or hypercapnia CBF (Fig. 1). As illustrated in Fig. 2, indomethacin had no ef fect on sodium nitroprusside- or papaverine induced increases in CBF. -
-
DISCUSSION Inhibition of carbonic anhydrase by Az delays the conversion of H2C03 to H+ + HC03 and induces a rapid decrease in the pH of cerebrospinal fluid and brain cells (Severinghaus and Cotev, 1968; Bickler et aI., 1988) as well as an increased CO2 tension in the cerebral cortex (Brzezinski et aI. , 1967; Bickler et aI., 1988) when ventilation is kept constant. This situation mimics hypercapnia (Ehrenreich et aI. , 1961). It is not surprising, therefore, that indometh acin, which so markedly inhibits the CBF reactivity to CO2, should also inhibit the reactivity to Az in animals maintained at constant ventilation. Our re sults indeed show this response and that the inhib itory effect is even quantitatively almost the same as for CO2, However, when animals are hyperventilated ac cording to the method of Bickler et al. (1988) so as to keep constant brain tissue Pco2, then brain intra cellular pH as measured with nuclear magnetic res onance phosphorus spectroscopy remains un changed and only brain extracellular acidosis devel ops after Az. Thus, the method of Bickler et al. (1988), combining Az with increased ventilation, alJ Cereb Blood Flow Metab, Vol. 13, No.4, 1993
lows one selectively to institute extracellular acido sis. As this acidosis triggers the CBF increase to Az (Severinghaus and Cotev, 1968; Heuser et aI., 1975; Bickler et aI. , 1988; Lassen, 1990), the almost com plete blockage by indomethacin of the increase in CBF under this condition suggests that indometha cin blocks the increase in CBF triggered by extra cellular acidosis. This agrees well with the marked inhibitory effect of indomethacin on hypercapnic CBF increase. However, this might also be a mere coincidence: Perhaps indomethacin nonspecifically suppresses the smooth muscle response to all vasodilator stim uli. However, indomethacin has no effect on the CBF increase in response to hypoxia (Sakabe and Siesjo, 1979; McCalden et aI., 1984), seizures (Ing var et aI., 1981), and hypoglycemia (Nilsson et aI., 1981). In the present study, it was also shown not to influence the percentage increase in CBF caused by other two pH-independent vasodilator stimuli-the nitric oxide donor nitroprusside and the calcium en try blocker papaverine (Fig. 2). Thus, our results would appear to suggest that the complete blockage by indomethacin of the increase in CBF elicited by Az and by CO2 is due to its direct interference with the pHe-sensitive mechanism regulating cerebral 2 vascular tone, a mechanism involving Ca + entry into the cells (West et aI., 1992). It is unlikely that indomethacin exerts its marked effect by directly increasing pHe. Rather indomethacin might influ ence calcium channels (keeping them open?) as this could explain the very rapid (within seconds) inhib itory effect of indomethacin on CBF in hypercapnia (Dahlgren et aI. , 1981) and with Az (data not
_
I
350
.�
300
g
250
I
200
I
� 0 -= "C 0 0 1i
1i .0 �
Q) u
200!!> 256!!> BO!!>
150 100 50 0
Vehicle
c:::J Control
� SNP
Indomethacin
ilZii Popaverine
(SNP; n = 3) and papaverine (n = 7) caused a marked rise in CBF by 80 and 200%, respectively, as seen on the graph. With indomethacin, the basal CBF and the CBF responses to SNP and papaverine were reduced proportionally so that almost the same per centage increases in CBF in response to SNP (n = 5) and papaverine (n = 4) were observed as without indomethacin. **p < 0.01 versus saline-treated group. Values are means ± SO. FIG. 2. Infusion of sodium nitroprusside
INDOMETHACIN IN pHe-SENSITIVE MECHANISM IN BRAIN shown), and this might also explain the species dif ference in its action. Because indomethacin blocks the synthesis of prostaglandins, it was reasoned that prostaglandins were perhaps somehow involved in the action of CO2, However, other potent cyclooxygenase inhib itors, including diclofenac, were found to have no effect on the CBF increase caused by CO2 (Quin tana et ai., 1988). In our studies, diclofenac, in a dose suppressing prostaglandin synthesis massively in the rat brain (Abdel-Halim et ai., 1978), had no effect on the CBF increase caused by Az or by CO2, This strengthens the concept of a nonprostaglandin mechanism of action of indomethacin and therefore supports the basic result of this article which is that indomethacin exerts its remarkable effect on CBF by directly interfering with pHe-sensitive mecha nisms. Acknowledgment: This work was supported by the Al fred Benzon Foundation, the Lundbeck Foundation, and the Danish Health Sciences Research Council. The ex cellent technical assistance of Kim Pape Gadegaard and Kent Pedersen is gratefully acknowledged.
REFERENCES Abdel-Halim MS. Sjoquist B, Anggard E (1978) Inhibition of prostaglandin synthesis in rat brain. Acta Pharmacol Toxicol 43:266-272 Amano T, Meyer JS (1981) Prostaglandin and human CBF con trol: effect of aspirin and indomethacin. J Cereb Blood Flow Metab I:S403-S404 Bickler PE, Litt L, Banville DL, Severinghaus JW (1988) Effects of acetazolamide on cerebral acid-base balance. J Appl PhysioI65:422-427 Brzezinski J, Kjallquist A, Siesjo BK (1967) Mean carbon diox ide tension in the brain after carbonic anhydrase inhibition. J Physiol 188:13-23 Dahlgren N, Nilsson B, Sakabe T, Siesjo BK (1981) The effect of indomethacin on cerebral blood flow and oxygen consump tion in the rat at normal and increased carbon dioxide ten sions. Acta Physiol Scand III:475-485 Ehrenreich DL, Bums RA, Alman RW, Fazekas JF (1961) In fluence of acetazolamide on cerebral blood flow. Arch Neu raI5:227-232 Ellis EF, Wei EP, Cockrell CS, Traweek DL, Saady 11, Kontos
727
HA (1982) The effect of O2 and CO2 on prostaglandin levels in the cat cerebral cortex. Circ Res 51:652-656 Eriksson S, Hagenfeldt L, Law D, Patrono C, Pinca E, Wennmalm A (1983) Effect of prostaglandin synthesis inhib
itors on basal and carbon dioxide stimulated cerebral blood flow in man. Acta Physiol Scand 117:203-211
Gotoh F, Tazaki Y, Meyer JS (1961) Transport of gases through brain and their extravascular vasomotor action. Exp Neurol
4:48-58 Hertz MM, Hemmingsen R, Bolwig TG (1977) Rapid and repet itive measurements of blood flow and oxygen consumption in the rat brain using intracarotid xenon injection. Acta Physiol Scand 101:501-503 Heuser D, Astrup J, Lassen NA, Betz E (1975) Brain carbonic acid acidosis after acetazolamide. Acta Physiol Scand 93: 385-390 Ingvar M, Nilsson B, Siesjo BK (1981) Local cerebral blood flow in the brain during bicuculline-induced seizures and the modulating influence of inhibition of prostaglandin synthe sis. Acta Physiol Scand 111 :205-212 Lassen NA (1968) Brain extracellular pH: the main factor con trolling cerebral blood flow. Scand J Clin Lab Invest 22:247251 Lassen NA (1990) Is central chemoreceptor sensitive to intracel lular rather than extracellular pH? Clin Physiol 10:311-319 McCalden TA, Nath RG, Thiele K (1984) The role of prostaglan dins in the hypercapnic and hypoxic cerebrovascular dila tions. Life Sci 34: 1801-1807 Nilsson B, Agardh CD, Ingvar M, Siesjo BK (1981) Cerebrovas cular response during and following severe insulin-induced hypoglycemia: CO2-sensitivity, autoregulation, and influ ence of prostaglandin synthesis inhibition. Acta Physiol Scand III:455-463 Pickard DD, Rose JE, Cooky MBD, Blair 1M, Strathdee A (1977) The effect of salicylate on cerebral blood flow in man. Acta
Neural Scand 56 (suppl 64):422-423 Quintana A, Raczka E, Quintana MA (1988) Effect of indometh acin and diclofenac on cerebral blood flow in hypercapnic conscious rats. Eur J Pharmacol 149:385-388 Sakabe T, Siesjo BK (1979) The effect of indomethacin on the blood flow-metabolism couple in the brain under normal, hypercapnic and hypoxic conditions. Acta Physiol Scand 107:283-284 Severinghaus JW, Cotev S (1968) Carbonic acidosis and cerebral vasodilatation after diamox. Scand J Clin Lab Invest 22 (suppl 102):IE
Wang Q, Paulson OB, Lassen NA (1992) Effect of nitric oxide blockage by W-nitro-L-arginine on cerebral blood flow re sponse to changes in carbon dioxide tension. J Cereb Blood Flow Metab 12:947-953 West DA, Leppla DC, Simard JM (1992) Effects of external pH on ionic currents in smooth muscle cells from the basilar artery of the guinea pig. Circ Res 71:201-209
J Cereb Blood Flow Metab, Vol. 13, No.4, 1993
