Time-Dependent Inhibition of Oxotremorine-Induced Cerebral Hyperemia by Nω-Nitro-l-Arginine in Cats
Background and Purpose Oxotremorine (OXO) is a cholinergic agonist that increases cerebral blood flow (CBF) when administered intravenously. We tested the hypothesis that OXO causes a dose-related increase in CBF in cats via a muscarinic mechanism that involves stimulation of nitric oxide synthase.
Methods Halothane-anesthetized male cats were studied under controlled ventilation. In three groups we measured cerebral blood flow (CBF; microspheres) during 30 minutes of intravenous OXO infusion at doses of 0.5 (n=3), 5 (n=6), or 50 μg · kg−1 · min−1 (n=6). The role of muscarinic receptor activation in the CBF response to OXO (50 μg · kg−1 · min−1) was assessed by determining the effect of atropine sulfate (2 mg · kg−1, n=6) pretreatment in a separate group of cats. The role of nitric oxide synthase was assessed by determining the CBF response to OXO (50 μg · kg−1 · min−1) either 30 (n=6) or 60 minutes (n=5) after administration of 50 mg/kg Nω-nitro-l-arginine (LNA).
Results CBF to forebrain (pre-OXO, 144±12 mL · min−1 ·100 g−1) was unchanged with OXO 0.5 or 5 μg · kg−1 · min−1 but increased at 10 (209±26 mL · min−1 · 100 g−1) and 30 minutes (243±35 mL · min−1 · 100 g−1) of OXO infusion at 50 μg · kg−1 · min−1 (P<.05). Atropine sulfate prevented OXO-induced hyperemia at 10 minutes of infusion but not at 30 minutes of infusion (135±12% of pre-OXO). LNA decreased baseline CBF by approximately 50%. Treatment with LNA 30 minutes before OXO did not affect the extent of OXO-induced hyperemia (CBF, 142±15% of pre-OXO at 10 minutes and 153±18% of pre-OXO at 30 minutes of OXO infusion). Treatment with LNA 60 minutes before OXO ablated OXO-induced hyperemia.
Conclusions In halothane-anesthetized cats, OXO (50 μg · kg−1 · min−1) increases forebrain CBF by a muscarinic mechanism that involves stimulation of nitric oxide synthase. The ability of nitric oxide synthase inhibitors to block agonist-induced nitric oxide–mediated vasodilation (response to OXO) is time dependent and may not be predicted by ability of the inhibitor to significantly decrease basal CBF.
Oxotremorine is a cholinergic agonist that has been shown to increase CBF in rats without alteration in CMRO2.1 In that study, the mechanism for OXO-induced hyperemia appeared to involve production of NO or an NO-containing compound because it was ablated by L-NAME, an inhibitor of NOS. This finding is consistent with the mechanism of cerebral vasodilation by other muscarinic agonists.1 2 3
Because anesthetics modulate the cerebrovascular response to fastigial nucleus stimulation4 5 and expression of cholinergic drug–induced effects on brain nuclei6 and on cerebral blood vessels,7 we recently tested the effect of pentobarbital and isoflurane on the cerebral vascular response to OXO.8 We found that OXO-induced hyperemia did not occur during pentobarbital anesthesia but was robust during isoflurane anesthesia. Although OXO-induced hyperemia was not related to an increase in CMRO2, it was only partially inhibited by L-NAME and only in a limited number of brain regions. Therefore, this previous study8 in dogs led us to conclude that OXO-induced hyperemia is only partially attributable to a mechanism involving NO. Although we demonstrated substantial reduction in whole-brain NOS activity after L-NAME in the previous study,8 we were unable to measure what effect L-NAME had specifically on the NOS pool responsible for OXO-induced hyperemia. In the present study, we considered the possibility that inhibition of the NOS pool responsible for OXO-induced hyperemia may have a different time dependency than is revealed by measurement of whole-brain NOS activity.
In halothane-anesthetized cats, we recently demonstrated9 that OXO-induced hyperemia is markedly attenuated during ischemic reperfusion. We speculated that during reperfusion an alteration in NO production by endothelium was the mechanism for altered OXO-induced hyperemia. However, the role of NOS in OXO-induced hyperemia in halothane-anesthetized cats was never tested. Therefore, the goal of the present study was to determine the role of NOS in the mechanism of OXO-induced hyperemia in halothane-anesthetized cats.
In this study in halothane-anesthetized cats, we tested the hypotheses that (1) intravenous OXO administration produces dose-related cerebral hyperemia; (2) OXO-induced hyperemia occurs via a mechanism involving stimulation of muscarinic receptors; (3) OXO-induced hyperemia occurs via a mechanism that involves stimulation of NOS; and (4) efficacy of an NOS inhibitor, LNA, for inhibition of OXO-induced hyperemia is time dependent.
Materials and Methods
This study was approved by the animal care and use committee at The Johns Hopkins Medical Institutions. Adult male cats (3.7±0.1 kg) were anesthetized with halothane (4%), orally intubated, and mechanically ventilated with supplemental oxygen to maintain arterial blood gases within normal limits. A single dose of pancuronium bromide (0.2 mg/kg IV) was administered to facilitate surgical preparation. Halothane concentration was maintained at 1.5% to 2% during surgical preparation and was increased for signs of cardiovascular stimulation during surgery but was not changed after obtaining baseline measurements.
Catheters were placed in both femoral veins and advanced into the inferior vena cava for administration of fluids and drugs. Another catheter was inserted into the descending thoracic aorta through a femoral artery for measurement of MABP, blood gases, hemoglobin concentration, and glucose levels. A catheter placed in the contralateral femoral artery was used for withdrawal of blood into heparinized syringes to minimize the increase in MABP expected during LNA administration and for the reference sample withdrawal during microsphere injection. A left atrial catheter was placed through a left thoracotomy for injection of radiolabeled microspheres. A ligature was loosely placed around the descending aorta (level of the diaphragm, below the tip of the catheters used for monitoring of MABP) for control of perfusion pressure as needed during the protocol. The cat was then placed in the prone position with the external auditory meatus approximately 3 cm above the level of the heart.
A catheter was placed in the sagittal sinus for withdrawal of cerebral venous blood for measurement of oxygen content for subsequent calculation of CMRO2. ICP was measured with a catheter placed in the left lateral cerebral ventricle through a burr hole. Brain temperature was measured with a thermistor placed through a burr hole into the superficial frontal cortex. Rectal and brain temperatures were maintained at 38±1°C with a heating pad and heating lamp throughout each experimental protocol. MABP and ICP were measured continuously. CPP was calculated as the difference between MABP and ICP. Arterial and ICP transducers (Statham P23 Db) were referenced to the right atrium and recorded on a Gould-Brush recorder. CVR was calculated by dividing CPP by blood flow to cerebrum.
Blood samples were analyzed for pH, Paco2, and Pao2 at the time of each microsphere injection with a Radiometer ABL3 analyzer. Oxygen saturation and hemoglobin were measured spectrophotometrically with a Radiometer Hemoximeter OSM3 so that arterial and cerebral venous O2 content could be calculated. Blood glucose was measured at the time of each measurement of CBF (YSI Glucose Analyzer). CMRO2 was calculated by multiplying the difference of arterial to sagittal sinus O2 content by blood flow to cerebrum.
CBF was measured using the reference-sample radiolabeled-microsphere technique,10 as previously used in our laboratory.11 12 13 Briefly, microspheres 16±0.5 μm in diameter (Dupont-NEN) were injected into the left atrium, and a reference sample was withdrawn with a Harvard pump. Before injection, the microspheres were agitated in an ultrasonic mixer for at least 20 minutes and vigorously shaken with a vortex mixer. For each flow measurement, approximately 1×106 spheres were injected into the left atrial catheter and flushed with 5 mL normal saline over a period of 20 seconds. Five isotopes (153Gd, 114mIn, 113Sn, 103Ru, 95Nb, or 46Sc) were injected in random sequence. The number of microspheres injected in each cat was chosen to allow at least 400 spheres to be delivered to the smallest tissue sample. The reference sample was withdrawn at 1.94 mL · min−1 beginning just before microsphere injection and continued for 90 seconds after completion of the infusion of the flush solution. The microsphere injection did not alter arterial blood pressure. At the end of each experiment, the cat was killed with KCl, and the brain was removed for processing. The brain was placed into 10% buffered formalin for 2 to 10 days and was then sectioned to determine blood flow to hindbrain (brain stem and cerebellum) and cerebrum (all brain except brain stem and cerebellum). After sectioning, each tissue sample was weighed and placed into 15-mL scintillation vials for analysis in an autogamma scintillation spectrometer (Packard Minaxi Auto-Gamma 5000 series) with a 3-in through-hole NaI crystal. The energy windows were set at 68 to 170 keV for 153Gd, 174 to 230 keV for 114mIn, 360 to 440 keV for 113Sn, 450 to 560 keV for 103Ru, 690 to 820 keV for 95Nb, and 830 to 1200 keV for 46Sc. The overlap of activity from high-energy isotopes into low-energy windows was corrected by differential spectroscopy to obtain corrected counts. Blood flow to each brain region was calculated from the formula Qt=Ct×Qr/Cr, where Qt is the flow to the tissue, Ct is the corrected counts per minute in the tissue sample, Qr is the reference flow rate, and Cr is the corrected counts per minute in the reference sample. Blood flow is expressed in milliliters per minute per 100 g tissue by normalizing for tissue weight.
The first protocol was conducted to establish a dose-response relationship for OXO on CBF in cats. We evaluated three doses of OXO: 0.5 (OXO-0.5, n=3), 5 (OXO-5, n=5), and 50 (OXO-50, n=5) μg · kg−1 · min−1. In each cat, baseline values were recorded 30 minutes after the surgical preparation was completed. A second baseline measurement was then obtained 30 minutes after 10-mL infusion of saline (1 mL · min−1). After the second baseline measurement, each cat was administered 2 mg · kg−1 atropine methyl bromide (in a total volume of 10 mL saline) by intravenous infusion over 10 minutes to prevent the systemic effects of subsequent OXO-induced muscarinic stimulation. Immediately after completion of the infusion of atropine methyl bromide, all variables were measured and the infusion of OXO commenced. All variables were measured again at 10 and 30 minutes of OXO infusion. At the end of this protocol, each cat was killed with an intravenous injection of KCl, and the brain was removed for subsequent CBF determination.
The second protocol was conducted to determine the mechanism of cerebral vasodilation from OXO infusion in halothane-anesthetized cats. We evaluated the effect of atropine sulfate (AS group, n=6) and LNA administered either 30 (LNA-30, n=6) or 60 minutes (LNA-60, n=5) before administration of OXO. In each cat, baseline values were recorded 30 minutes after completion of the surgical preparation. A second measurement of all variables was then obtained 30 minutes after 10-mL infusion of saline (AS group) or 50 mg · kg−1 LNA in a total volume of 40 mL saline (LNA-30). In a separate cohort of cats, the second measurement was obtained 60 minutes after LNA (LNA-60). The dose and timing of LNA administration were chosen on the basis of previous work from our laboratory demonstrating rapid (<30 minutes) and substantial (>70%) inhibition of brain NOS in cats.14 Each cat was then administered 2 mg · kg−1 atropine methyl bromide (LNA group) or 2 mg · kg−1 atropine sulfate (AS group) in a total volume of 10 mL saline by intravenous infusion over 10 minutes. After completion of the infusion of atropine sulfate or atropine methyl bromide, all variables were measured and the infusion of OXO commenced. All variables were measured again at 10 and 30 minutes of OXO infusion (50 μg · kg−1 · min−1). In two cats in the LNA-60 group, at the end of the experiment the OXO infusion was stopped, and the CBF response to hypoxia (Fio2, 0.08 to 0.10) was determined. This was done to determine whether any loss in CBF response to OXO was specific or simply due to a generalized loss of cerebral vascular reactivity. At the end of this protocol, each cat was killed with an intravenous injection of KCl, and the brain was removed for subsequent CBF determination.
Data are presented as mean±SEM. Two-way ANOVA for between (group) and within (time) subjects design was used to compare CBF and physiological parameters. Significance was assumed when P<.05. If a significant group-time interaction was demonstrated, one-way ANOVA was performed for between-group effects at individual time points and for within-group effects (repeated measures). Post hoc comparisons were made with the Student-Newman-Keuls test.
Table 1⇓ shows blood gas data and CPP for the dose-response protocol. All values remained within a normal physiological range throughout the protocols, and there were no differences among groups.
Fig 1⇓ shows forebrain blood flow, CVR, and CMRO2 for the dose-response protocol. OXO-0.5 produced no effect on CBF or CVR. OXO-5 resulted in no change in CBF or CMRO2 but showed a late reduction in CVR. OXO-50 resulted in a 65±16 mL · min−1 · 100 g−1 increase (144±10% of pre-OXO) in CBF at 10 minutes and a 100±24 mL · min−1 · 100 g−1 increase (167±12% of pre-OXO) at 30 minutes. CVR was decreased by 0.21±0.07 mm Hg · mL−1 · min−1 · 100 g−1 (71±8% of pre-OXO) at 10 minutes and by 0.28±0.05 mm Hg · mL−1 · min−1 · 100 g−1 (61±4% of pre-OXO) at 30 minutes. Despite differences among groups in CBF, there was no difference among groups in CMRO2. Blood flow to hindbrain was similar among groups during baseline conditions (mL · min−1 · 100 g−1: OXO-0.5, 90±5; OXO-5, 88±12; and OXO-50, 118±12) and was unchanged during OXO infusion.
Table 2⇓ shows blood gas data and CPP for the atropine sulfate and LNA protocol. Values for pH, Paco2, Pao2, and hemoglobin were similar among groups and over time. CPP was similar among groups under baseline conditions but was increased after LNA administration in both groups receiving this drug despite an attempt to minimize the increase in MABP with controlled hemorrhage.
Fig 2⇓ shows blood flow to cerebrum and hindbrain, and CVR to cerebrum. Atropine sulfate treatment did not change CBF in cerebrum or hindbrain but was associated with an increase in CVR compared with baseline conditions. Atropine sulfate treatment completely prevented OXO-induced hyperemia in cerebrum at 10 minutes of infusion. At 30 minutes of infusion, CBF increased by 29±12 mL · min−1 · 100 g−1 (135±12% of pre-OXO), and CVR decreased by 0.39±0.15 mm Hg · mL−1 · min−1 · 100 g−1 (74±9% of pre-OXO).
Treatment with LNA resulted in greater than 50% reduction in blood flow to both cerebrum and hindbrain, and there was no further reduction in CBF at 60 minutes compared with 30 minutes after LNA administration. In the LNA-30 group, 10 minutes of infusion with OXO increased blood flow to cerebrum by 18 mL · min−1 ·100 g−1 (142±15% of pre-OXO) and decreased CVR by 0.77±0.34 mm Hg · mL−1 · min−1 · 100 g−1 (78±10% of pre-OXO); 30 minutes of infusion with OXO increased blood flow to cerebrum by 24±9 mL · min−1 · 100 g−1 (153±18% of pre-OXO) and decreased CVR by 0.89±0.37 mm Hg · mL−1 · min−1 · 100 g−1 (74±11% of pre-OXO). There was no change in blood flow to hindbrain.
In the LNA-60 group, treatment with OXO resulted in no change in CBF or CVR. However, CBF was more than doubled (cerebrum, 115% and 108% increase from prehypoxic blood flow) in both animals tested for subsequent reactivity to hypoxia (oxygen content, 5.5 and 2.8 mL · dL−1; Pao2, 35 and 29 mm Hg).
Intravenous administration of OXO in halothane-anesthetized cats results in dose-related cerebral vasodilation that is not associated with a change in CMRO2. OXO-induced hyperemia is prevented by muscarinic-receptor blockade. In halothane-anesthetized cats, the muscarinic-receptor effects of OXO are mediated by NOS. OXO-induced cerebral hyperemia in cats is attenuated to a greater extent 60 minutes, compared with 30 minutes, after systemic administration of the NOS inhibitor LNA despite similar marked reduction in unstimulated CBF at both time points.
Sensitivity of the cerebral vasculature in halothane-anesthetized cats appears to be different than that of fentanyl/nitrous oxide–anesthetized rats. In rats, infusion of 1 μg · kg−1 · min−1 OXO resulted in dilation of pial vessels by approximately 30% within 5 minutes of the start of the infusion,15 and 5 and 50 μg · kg−1 · min−1 resulted in substantial and similar cerebral hyperemia.1 In cats, we observed no hyperemia at 0.5 and 5 μg · kg−1 · min−1, although CVR was decreased 30 minutes after 5 μg · kg−1 · min−1 infusion was started. Another difference between the response to OXO in halothane-anesthetized cats and fentanyl/nitrous oxide– anesthetized rats1 is the regional distribution of CBF changes. In cats, changes were isolated to forebrain regions, whereas in rats they occurred both in forebrain and hindbrain regions. The etiology of this difference may be related to differences in control mechanisms for CBF in different species or to differences in how anesthetics affect these control mechanisms. For example, OXO-induced hyperemia is robust in isoflurane-anesthetized dogs but is absent in pentobarbital-anesthetized dogs.8
NOS activity can be assessed by either measuring enzyme activity in brain biopsy samples or by measuring the physiological response to an intervention that is known to work by an NO-mediated mechanism. We previously demonstrated that intravenous administration of 50 mg · kg−1 LNA in cats causes rapid (within 30 minutes) inhibition of NOS14 as assessed in whole-brain biopsy samples that are analyzed ex vivo. Therefore, OXO-induced hyperemia occurred at a time when whole-brain NOS activity was greatly diminished (or absent). We believe that these data underscore a need for testing physiological NOS activity as a positive control before reaching conclusions regarding the role of NOS activation in a physiological or pharmacological paradigm.
We have previously demonstrated that L-NAME inhibited OXO-induced hyperemia in subcortex but not cortex of isoflurane-anesthetized dogs.8 In that study,8 CBF response to OXO was tested 20 minutes after L-NAME administration. Thus, the incomplete inhibition of OXO-induced hyperemia may have been due to inadequate time for inhibition of NOS responsible for OXO-induced hyperemia. Alternatively, it is possible that some of OXO-induced hyperemia in isoflurane-anesthetized dogs occurs by a mechanism independent of NO production. For example, OXO is also known to be an agonist at nicotinic receptors in vivo.16
Systemic administration of high-dose OXO is capable of causing an increase in cerebral glucose consumption.17 18 However, when administered at a low dose, OXO causes hyperemia in cerebrum, which was not correlated with increased metabolism in our study. Therefore, we conclude that in our study OXO-induced hyperemia is not a secondary effect resulting from OXO-induced metabolic activation. It is possible that the arteriovenous method of measuring CMRO2 used in this study may have led to underestimation of focal changes in metabolism. However, we believe that if the increase in flow to the entire cerebrum was metabolically mediated, we would have observed an increase in CMRO2, especially with the venous catheter placed in the superior sagittal sinus. This is consistent with the findings of Pelligrino et al,1 who demonstrated OXO-induced cerebral hyperemia without an increase in CMRO2, and of Scremin et al,19 who demonstrated that physostigmine, an acetylcholinesterase inhibitor, increases CBF without an increase in cerebral glucose utilization. However, this is not consistent with the finding that physostigmine increases CBF in parallel with an increase in CMRO2.20 The mechanism for these differences between the cerebral metabolic effects of OXO and physostigmine is not delineated by our study. However, we speculate that at high doses physostigmine and OXO may have effects on brain independent of those produced by elevated acetylcholine at vascular muscarinic receptors.
In summary, our data demonstrate that intravenous OXO causes dose-dependent cerebral vasodilation that is not associated with cerebral metabolic activation in halothane-anesthetized cats. The mechanism for OXO-induced hyperemia is caused by stimulation of muscarinic receptors and activation of NOS. There was persistent cerebral vasodilation to OXO at a time when CBF (and presumably whole-brain NOS activity) is markedly decreased (30 minutes after LNA administration). Therefore, we speculate that assessment of whole-brain NOS activity measured ex vivo does not necessarily predict physiological enzyme activity.
Selected Abbreviations and Acronyms
|CBF||=||cerebral blood flow|
|CMRO2||=||cerebral oxygen consumption|
|CPP||=||cerebral perfusion pressure|
|L-NAME||=||Nω-nitro-l-arginine methyl ester|
|MABP||=||mean arterial blood pressure|
|NOS||=||nitric oxide synthase|
This study was supported by National Institutes of Health NS-20020. The authors express their gratitude to the staff of the Department of Anesthesiology Research Laboratories for their excellent technical assistance and to Candy Berryman for her excellent secretarial assistance.
- Received April 3, 1995.
- Revision received August 11, 1995.
- Accepted August 15, 1995.
- Copyright © 1995 by American Heart Association
Pelligrino DA, Miletich DJ, Albrecht RF. Diminished muscarinic receptor-mediated cerebral blood flow response in the streptozotocin-treated rat. Am J Physiol. 1992;262:E447-E454.
Faraci FM. Role of nitric oxide in regulation of basilar artery tone in vivo. Am J Physiol. 1990;259:H1216-H1221.
Bedran de Castro MTB, Crystal GJ, Downey HF, Bashour FA. Regional blood flow in canine brain during nicotine infusion: pentobarbital versus chloralose anesthesia. Stroke. 1984;15:690-694.
Clavier N, Kirsch JR, Hurn PD, Traystman RJ. Effect of postischemic hypoperfusion on vasodilatory mechanisms in cat. Am J Physiol. 1994;267:H2012-H2018.
Baldwin WA, Kirsch JR, Hurn PD, Toung WSP, Traystman RJ. Hypothermic cerebral reperfusion and recovery from ischemia. Am J Physiol. 1991;261:H774-H781.
Helfaer MA, Kirsch JR, Traystman RJ. Anesthetic modulation of cerebral hemodynamic and evoked responses to transient middle cerebral artery occlusion in cats. Stroke. 1990;21:795-800.
Traystman RJ, Moore LE, Helfaer MA, Davis S, Banasiak K, Williams M, Hurn PD. Nitro-l-arginine analogues: dose- and time-related nitric oxide synthase inhibition in brain. Stroke. 1995;26:864-869.
Dam M, Wamsley JK, Rapoport SI, London ED. Effects of oxotremorine on local glucose utilization in the rat cerebral cortex. J Neurosci. 1982;2:1072-1078.