Systemic Nitric Oxide Synthase Inhibition Does Not Affect Brain Oxygenation during Cortical Spreading Depression in Rats: A Noninvasive Near-Infrared Spectroscopy and Laser-Doppler Flowmetry Study

Abstract

Cortical spreading depression (CSD) has been implicated in the migraine aura and in stroke. This study demonstrates near-infrared spectroscopy (NIRS) for the first time as capable of noninvasive on-line detection of CSD in the pentobarbital-anesthetized rat. CSD was accompanied by a brief and rapid increase of regional CBF (by laser-Doppler flowmetry) to 200–400% baseline. NIRS demonstrates that this hyperperfusion is associated with concentration increases of oxyhemoglobin, while deoxyhemoglobin decreases. Simultaneously, oxygen partial pressure, measured on the brain surface with a solid-state Polarographic probe, was shown to be raised by at least 14 mm Hg during CSD. Oxygen-dependent phosphorescence life-time quenching measurements confirmed this finding. NIRS data on cytochrome aa₃, however, showed a CSD-related shift toward a more reduced state, despite raised blood oxygenation. This may suggest either limited O₂ transport from the blood to mitochondria or decreased oxygen utilization during CSD as supposed by theories about compartmentalization of energy metabolism favoring glycolytic rather than aerobic energy supply during CSD. However, the data on cytochrome aa₃, warrant caution and are discussed critically. Nitric oxide synthase inhibition by systemic application of N′-nitro-l-arginine had no significant effect on the perfusion response or the tissue Po₂ during CSD. During most CSD episodes, a brief decrease in MABP by 4–8 mm Hg was noted that might be caused by functional decortication during CSD.

Keywords

Blood pressure Cerebral blood flow Cerebral ischemia Coupling Hemoglobin Oxygen Migraine

Cortical spreading depression (CSD) is a wave of initial neuronal excitation followed by sustained depolarization of neuronal and glial cells that spreads at 2–5 mm/min over the cerebral cortex. It is characterized by temporary loss of specific cell function, i.e., flattening of EEG. CSD was first described 50 years ago (Leão, 1944) and has not ceased to intrigue researchers ever since. Due to the associated massive ionic changes, it is known to challenge the energy metabolism of cortical parenchymal cells (Mies and Paschen, 1984; Gault et al., 1994) without causing any damage to the affected brain tissue (Nedergaard and Hansen, 1988). An acute increase of regional CBF (rCBF) follows the ionic changes with little delay (Lauritzen, 1987; Duckrow, 1993), suggesting that it matches increased demand for substrate and oxygen. In the hemodynamically compromised penumbra of focal ischemia, though, CSD-like periinfarct depolarizations have been shown to enhance ischemic injury (Iijima et al., 1992; Mies et al., 1993; Back et al., 1994).

Apart from its clinical implication in stroke, CSD has become more and more attractive in the pathophysiological scenario of migraine, mainly by virtue of accumulating evidence for neuronal mechanisms found in the pursuit of Moskowitz's theory of neurogenic inflammation in migraine (Moskowitz and MacFarlane, 1993; Lauritzen, 1994). Nevertheless, CSD has defied numerous attempts to detect it in the human cerebral cortex in vivo (e.g., McLachlan and Girvin, 1994), and the possibilities of invasive investigation are limited. Near-infrared spectroscopy (NIRS) as a spectrophotometric technique, up to now, has been used clinically for monitoring of cerebral and cardiovascular oxygen sufficiency (Jöbsis, 1977; Edwards et al., 1991). It operates with infrared light near the visible end of the spectrum and penetrates even dense tissue such as bone easily. It is absorbed mainly by three chromophores: oxyhemoglobin [HbO₂], deoxyhemoglobin [Hb], and the mitochondrial cytochrome aa₃ in its oxidized state [CytO]. A laser is pulsed successively in different wavelengths and the light is transmitted to the skull by fiber optics. Photons that have passed the tissue and leave the skull again, carrying the spectrophotometric information, are collected by another fiber bundle. The signal is amplified in a photomultiplier tube, and optical densities of the tissue for the different wavelengths are calculated continuously. From changes in these optical densities, concentration changes of the chromophores can be derived (Cope and Delpy, 1988; Cope et al., 1988; Wray et al., 1988).

The mediators of the blood flow response to CSD in intact brain remain uncertain. As a potent vasodilator, nitric oxide (NO) plays a role in the coupling of CBF and metabolism during neuronal activation in the rat (Dirnagl et al., 1993). Colonna et al. (1994) demonstrated the relevance of NO to CSD-related arteriolar dilatation in the rabbit. Goadsby and colleagues (1992) were able to abolish CSD-associated hyperemia by systemic inhibition of nitric oxide synthase (NOS) in the cat. For the rat, the Moskowitz group found that systemic NOS inhibition did not significantly influence rCBF during CSD (Zhang et al., 1994).

Considering the study data on typical blood flow changes in CSD, we tested in our study (a) NIRS as a promising tool to detect CSD noninvasively by means of its hemodynamic sequelae. If we are able to monitor CSD by NIRS in the rat model, we can venture to search for it noninvasively in humans, too. In addition, we expected NIRS (b) to yield valuable noninvasive on-line information about the oxygenation of cortical tissue during CSD. Furthermore, we tested the hypothesis (c) as to whether inhibition of NOS by systemic application of N′-nitro-l-arginine (l-NNA) was able to alter CSD itself or the response to it.

METHODS AND MATERIALS

Preparation

We used male Wistar rats of 280–320 g in weight. The animals were anesthetized with pentobarbital (100 mg/kg body wt i.p.). Anesthesia was maintained by intermittent application of further intraperitoneal barbiturate doses (5 mg), aiming at absence of a reaction to a tail pinch. After tracheotomy, they were ventilated (Effenberger small-animal ventilator, Pfaffin-gen, Germany), and end-expiratory CO₂ was monitored (Heyer Artema MM204, Sweden) throughout the experiment. Body temperature was kept at 37.5 ± 0.5°C by an electrical heating pad. A femoral artery and vein were cannulated for permanent monitoring of MABP (RFT Biomonitor, Zwönitz, Germany), intermittent sampling of arterial blood gases (AVL Compact I, Graz, Austria), and continuous infusion of saline (1 ml/hr). The animals were fixed in a stereotaxic frame, and two symmetrical open cranial windows were installed in the frontal bones. The dura was carefully removed to expose the frontal cortex of both hemispheres for elicitation of CSD, and artificial CSF [for composition, see Levasseur et al. (1975)] was superfused. As illustrated in Fig. 1, 5 and 10 mm occipitally from this triggering site, burr holes were drilled with a saline-cooled drill. At both trephinations, the wicks of calomel electrodes were placed for registration of direct current (DC) potential deflection. DC potential was measured differentially between the two electrodes by a differential voltage meter (FD223; WPI, Sarasota, FL, U.S.A.), so that the velocity of the CSD could be calculated from the distance of the two peaks of the resulting biphasic curve. For measurements of brain surface oxygen partial pressure (Po₂) in the burr hole at 10-mm distance from the triggering site, a Polarographic solid-state oxygen probe (Licox-Po₂ monitoring system; GMS, Mielkendorf, Germany) was slit through the dura and placed on the cortical surface by a micromanipulator. In the same area, the transmitting and the receiving optodes (fiber bundle diameter 3.4 mm) of the four-wavelength near-infrared spectrometer (NIRO 500; Hamamatsu Photonics, Japan) were positioned over the intact skull. To sample the cortex of the right hemisphere, they were arranged at an angle of ˜70° and ˜6 mm apart from each other perpendicular to the direction of propagation of CSD (van der Zee et al., 1990). rCBF was measured by laser-Doppler flowmetry (LDF) (Vasamedics BPM², St. Paul, MN, U.S.A.) through the thinned bone (Dirnagl et al., 1989; Lindauer et al., 1993). Since the laser signals of both LDF and NIRS interfered with each other if employed in the same site, we chose to measure rCBF symmetrically in the opposite hemisphere, supposing that CSDs behave in an analogous fashion in both hemispheres when triggered at the same time.

Fig. 1.

Schematic view of the experimental setup. The skull is intact apart from an open cranial window over the frontal poles and burr holes for calomel electrodes and a Polarographic solid-state oxygen probe; regional cerebral blood flow (rCBF) is measured by laser-Doppler flowmetry (LDF) through the thinned bone; the sending and collecting near-infrared spectroscopy (NIRS) optodes are placed over the preserved skull and measure changes of the regional blood oxygenation. Cortical spreading depression (CSD) is triggered in the frontal region of both hemispheres by simultaneous superfusion of high potassium aCSF, the propagating deflection wave of the cortical direct current potential (DC) is followed by calomel electrodes. NIRS, LDF, and Polarographic Po₂ probe are placed 10 mm from triggering site.

Protocol

CSDs were triggered by topical application of 150 mM KCl solution for ˜30 s, after which it was replaced again by artificial CSF. The interval between these episodes was 45 min.

In the experimental group (14 animals), l-NNA (10 mg/kg body wt i.v.) was injected slowly after the third CSD had passed, and after a further 45 min, another CSD was triggered as described. We compared the CSD before with the CSD 45 min after l-NNA application.

In the control group (nine animals), NOS was not inhibited. A control with d-NNA was not done since other studies have shown that only the l-enantiomer is capable of NOS inhibition (e.g., Dirnagl et al., 1994). Of subsequently triggered CSDs in intervals of 45 min, the third and fourth served as time matches to the CSD pre- and post-l-NNA in the experimental group.

In three separate animals that received no l-NNA, we employed the phosphorescence life-time quenching method for Po₂ measurement (Oxyspot; Medical Systems Corp. U.S.A., GMS Mielkendorf, Germany) to confirm our Polarographic oxygen measurements. This optical method uses oxygen-dependent phosphorescence kinetics of intravenously given porphyrins [in our case meso-tetra-(4-sulfonatophenyl)-porphine] activated by intermittent flashes of light (Vanderkooi et al., 1987). Its main advantage in our setup was that no probe had to be in actual mechanical contact with the cortex, thus avoiding pressure artifacts.

Data analysis

Cortical DC potential, [HbO₂], [Hb], [CytO], rCBF, Po₂, and MABP were measured continuously. Data were acquired every 2.5 s by a PC running ASYST data acquisition software (Macmillan); Po₂ data were acquired separately every 5 s with the manufacturer's software. Total hemoglobin [HbT] was calculated from raw data by addition of [HbO₂] and [Hb]. Data of [HbO₂], [Hb], [HbT], [CytO], and Po₂ were compared regarding their maximum deflection from baseline. NIRS data are expressed in arbitrary units that will be discussed later. rCBF was determined as relative change compared with the baseline, defined as 100%. The velocity of CSD propagation was calculated from the time that passed between the registration of DC potential shift with either calomel electrode. We compared the CSD before with the CSD after NOS inhibition within the experimental group. We also compared the CSD before and after l-NNA of the experimental group with the corresponding CSD (nos. 3 and 4) of the control group. A paired two-tailed Student's t test was performed to assess differences between data within one group and an unpaired test comparing the data of different groups; p < 0.05 was considered significant. All values are presented as means ± SD.

RESULTS

Physiology

Physiological parameters were kept in normal ranges throughout the experiment and are documented for the instances of measured CSD in Table 1. MABP increased in the experimental group significantly after systemic l-NNA application. During CSD we noticed a brief dip of MABP by ˜5–8 mm Hg (Table 2) that lasted ˜1–3 min.

Table 1.

Physiology data for experimental (n = 14) and control groups (n = 9)

	MABP (mm Hg)	pH	PCo₂(mm Hg)	Po₂(mm Hg)
Exp. group (n = 12)
CSD 3 without l-NNA	113 ± 14^a	7.39 ± 0.02	33 ± 5	116 ± 14
CSD 4 with l-NNA	134 ± 15^{a, b}	7.39 ± 0.04	30 ± 6	122 ± 28
Control group (n = 6)
Control CSD 3	119 ± 20	7.41 ± 0.03	31 ± 3	102 ± 20
Control CSD 4	116 ± 14^b	7.40 ± 0.02	30 ± 3	99 ± 8

Values are means ± SD. Po₂, respiratory oxygen partial pressure. No significant dif-ferences occurred in cortical spreading depression (CSD) of the control group that matched the time course of the experimental group.

Significant increase (p < 0.01) in MABP after N′-nitro-l-arginine (l-NNA) application.

Significant difference (p < 0.01) between MABP in CSD 4 (control group) and CSD 4 (experimental group with l-NNA).

Table 2.

MABP dip occurring during last cortica spreading depression (CSD) before and first after NO synthase (NOS) inhibition by N′-nitro-l-arginine (l-NNA) in experimental group

	MABP baseline(mm Hg)	MABP dip(mm Hg)	Difference(mm Hg)
CSD without l-NNA (n = 11)	108.1 ± 14.9^a	100.0 ± 15.6^a	8.1 ± 2.6^c
CSD with l-NNA (n = 11)	128.6 ± 17.1^b	124.6 ± 15.2^b	4.1 ± 4.6^c

MABP dropped during CSD significantly (^ap < 0.01, ^bp < 0.05) compared with baseline. This brief dip was significantly (^cp < 0.05) lower under NOS inhibition with l-NNA (see also Fig. 3).

Velocity of DC potential deflection, rCBF, Po₂, and NIRS data are listed in Table 3. A typical recording of the DC potential shift is shown on the bottom of Fig. 2. The biphasic curve results from differential recording of the same wave of CSD passing the first calomel electrode at 5-mm distance from the triggering point ˜90 s after the start (negative deflection) and the second one 5 mm more occipital after another 90 s (positive deflection). CSD velocity was not changed by l-NNA.

Table 3.

Velocity of cortical direct current (DC) potential deflection, increase of regional CBF (rCBF), concentration changes of oxyhemoglobin [HbO₂], deoxyhemoglobin [Hb], total hemoglobin [HbT], and oxidized cytochrome aa₃ [CytO], as well as rise in oxygen partial pressure of cortex surface (Po₂) measured during cortical spreading depression (CSD) before and after NO synthase (NOS) inhibition by systemic application of N′-nitro-l-arginine (l-NNA; 10 mg/kg) in experimental (n = 14) and control (n = 9) groups

Parameter during CSD	n	CSD 3 (without l-NNA)	CSD 4 (with l-NNA)	n	Control CSD 3	Control CSD 4
Velocity of CSD (mm/min)	9	3.64 ± 0.53	3.48 ± 0.43	8	3.21 ± 0.36	3.14 ± 0.38
Basal rCBF (%)	14	100 ± 23^a	85* ± 17	9	100 ± 27	100 ± 39
Max. rCBF (%)	14	321 ± 150	403 ± 190	9	279 ± 102	279 ± 61
Δ [HbO₂] (AU)	14	25.71 ± 6.22	26.79 ± 4.41	9	23.61 ± 6.67	21.78 ± 7.31
Δ [Hb] (AU)	14	-16.61 ± 6.11	-17.16 ± 6.65	9	-16.50 ±7.40	-18.06 ± 6.18
Δ [HbT] (AU)	14	18.55 ± 8.25	18.70 ± 8.66	9	16.38 ± 8.05	12.00 ± 5.24
Δ [CytO] (AU)	14	-5.16 ± 1.24	-4.95 ± 1.00	8	-3.93 ± 0.83	-4.06 ± 0.47
Δ Po₂ (mm Hg)	7	22.0 ± 26.5	18.5 ± 20.3	4	14.33 ± 10.12	15.67 ± 9.81

Hemoglobin and cytochrome aa₃ changes were measured noninvasively by near-infrared spectroscopy, rCBF by laser-Doppler flowmetry, and Po₂ by a Polarographic solid-state probe. A significant (^ap < 0.01) decrease of basal rCBF occurred after systemic NOS inhibition with l-NNA. For basal rCBF, percentage refers to CSD without l-NNA.

Fig. 2.

Typical changes during one single episode of cortical spreading depression (CSD) (CSD was triggered at 30 s). The pattern was unchanged after systemic N′-nitro-l-arginine application. DC, cortical slow potential; HbO₂, oxyhemoglobin; Hb, deoxyhemoglobin; CytO, cytochrome aa₃; LDF, laser-Doppler flowmetry. Please note that while DC, near-infrared spectroscopy (NIRS), and Po₂ are measured in the same site, the laser-Doppler probe for regional CBF (rCBF) was placed symmetrically in the opposite hemisphere. This accounts for a slightly different time lock between the NIRS data (HbO₂, Hb, CytO) and LDF.

NIRS was clearly able to monitor changes of hemoglobin and cytochrome aa₃ signals during CSD, the patterns of which are also shown in Fig. 2. All NIRS signals were above noise range. Oxyhemoglobin increased, indicating arterialization. Deoxyhemoglobin decreased, not fully compensating for the [HbO₂] rise, though. Accordingly, a total hemoglobin increase suggests hyperemia during CSD. During passage of CSD, cytochrome aa₃ showed a distinct shift toward a reduced state (i.e., a concentration decrease of its oxidized fraction [CytO]). NOS inhibition by systemic l-NNA did not alter the response of [HbO₂], [Hb], [HbT], or [CytO] to CSD in the experimental group. NIRS signals were the first to change and took between 6 and 8 min to return to baseline. Although showing a clear overall concentration rise or fall, they sometimes displayed a “dichrotic” curve pattern like the one of [HbO₂], [HbT], and [CytO] in Fig. 2.

The predominant change in brain surface oxygen partial pressure was a rise by ˜15 min Hg, lasting for ≥2 min. NOS inhibition with l-NNA made no difference in this. Some recordings were not unidirectional, but showed an initial dip or a number of undulations that gave rise to suspicion of an artifact, possibly by movement of the brain toward the 300-μm probe as it may be caused by hyperemia or tissue swelling during CSD. The measurements with phosphorescence life-time quenching, however, confirmed the pattern of a prevailing Po₂ increase during CSD, as illustrated in Fig. 3.

Fig. 3.

Brain surface Po₂ by phosphorescence life-time quenching during a single cortical spreading depression. (The recording was restricted to a smaller time window than in Fig. 2.)

The baseline of rCBF dropped significantly as a consequence of l-NNA application in the experimental group, whereas it did not change in the control group. This, alongside with the increase of MABP, indicates inhibition of the endothelial NOS and the resulting increase in vascular resistance (Wei et al., 1993). In the experimental and control group, the response to CSD was a two- to fourfold increase of rCBF, returning to baseline after ˜100 s. With the NOS inhibition in the experimental group, we observed a trend toward a higher blood flow response compared with the control group, which did not reach significance, though.

DISCUSSION

NIRS changes were the first to be noticed and the last to return to baseline, which indicates that the NIRS optodes have the largest sampling volume of all probes employed. In the face of this spatial resolution, the collected NIRS information refers simultaneously to those cortical areas that are affected by CSD as well as to those that are either not yet or no longer in the wave front. This allows only for interpretation of gross changes and makes drawing conclusions from subtle patterns, such as the occasionally occurring “dichrotic” curves, difficult. Employment of optodes with a smaller optical diameter should improve the spatial resolution.

The measured increase in oxyhemoglobin concentration corresponds well to the hyperperfusion phase, as it is known to be due to arterial (Colonna et al., 1994) and arteriolar (Shibata et al., 1992; Wahl et al., 1994) dilatation. This arterialization appears to exceed oxygen extraction because deoxyhemoglobin is cleared from the tissue, resulting in lowered [Hb] concentration. An increase in brain surface Po₂ in CSD was shown with two different methods and may reflect the increased arterialized perfusion in the cortex. This finding supports data of Back et al. (1994). However, the oxygen measurements of this study are hampered by the fact that both used methods restricted to superficial structures (see later). No difference between NOS inhibition and normal CSD was found. However, with the given standard deviations, a type II error is not unlikely; that is, the difference in the Po₂ response of NOS-inhibited and noninhibited CSD may have been too low to be detected.

Cytochrome aa₃ shows a reduction in spite of the raised blood oxygenation. This is in discordance with previous findings of other workers on cytochrome aa₃ spectrophotometry in the visible light range (Jöbsis et al., 1977; Sylvia and Rosenthal, 1979). However, NADH reflectance studies of the mitochondrial oxidation state during CSD produced evidence for mitochondrial reduction during CSD in anesthetized gerbils (Mayevsky et al., 1980) and cats (Dora et al., 1984). Due to the temporal resolution of both Polarographic and phosphorescence-quenching Po₂ measurements, an initial brief dip in Po₂ may have been missed. Furthermore, both methods sample the cortical surface and are unlikely to reflect the tissue oxygen partial pressure in deeper layers that are sampled by NIRS. Other workers who implanted micro-electrodes to deeper structures reported biphasic responses of tissue Po₂ (Tsacopoulus and Lehmenkühler, 1977; Dietrich et al., 1994) or found an increase with superficial oxygen sensors while probes in deeper layers recorded a monotonic decrease of oxygen tension (LaManna et al., 1989), so that the cytochrome aa₃ reduction in our experiments may well have been accompanied by an actual drop of tissue Po₂. However, there remains still the puzzling coexistence of raised blood oxygenation ([HbO₂]↑ [Hb]↓) and apparent decreased mitochondrial oxygenation ([CytO]↓) as both are recorded with the same method, namely, NIRS. The discordant findings of different groups in different animals and models seem to indicate that the primary variables may be signal penetration depth and depth of anesthesia.

The shift of cytochrome aa₃ toward a more reduced state during increased blood oxygenation could be interpreted as increased oxygen demand of mitochondria during CSD that is unmet despite excessive blood oxygenation, possibly by a diffusion barrier (Hotez et al., 1977; Kreisman et al., 1981) or as physiologically decreased oxygen utilization by mitochondria during CSD. While the first speculation would stress a pathological aspect of CSD, the second supports theories of compartmentalization of energy metabolism that suppose a glycolytic rather than aerobic energy supply of ionic pumps during CSD (Fox et al., 1988; Andersen and Marmarou, 1992). The assumption of a temporary drop in tissue Po₂ that we mentioned would put more weight on the first speculation. Although excessive blood oxygenation during CSD does suggest that oxygen is not the coupling agent between metabolism and blood flow, this seems obvious only under the presumption that no bottleneck exists for oxygen transfer from hemoglobin to cytochrome aa₃.

Though much work has been done on the validation of multiple wavelength NIRS measurements of cytochrome aa₃ (Kariman and Burkhart, 1985; Wray et al., 1988; Ferrari et al., 1990), the question must be raised of the degree to which the [CytO] data are based on changes other than in absorption. Because of the low concentration of cytochrome aa₃ compared to hemoglobin (Cooper et al., 1994), this signal is prone to distortions by hemoglobin spillover, by cytochrome c contamination, or by alterations of the optical properties of the tissue as they may be caused by cell swelling during CSD (Lipton, 1973; Bureš et al., 1974b; Hansen and Olsen, 1980; Jing et al., 1994).

The Hamamatsu NIRO 500 expresses data as micromolar changes, provided a differential path length factor is given. For the human head, this factor has been estimated (van der Zee et al., 1992), but in our setup it is unknown. Simultaneous optical path length measurements, however, would provide the opportunity to calibrate arbitrary units to micromolar. Nevertheless, arbitrary units are suitable for comparisons of the kind we undertook.

A reduction of basal rCBF due to NOS inhibition is a well described phenomenon (Prado et al., 1992; Dirnagl et al., 1993; Pelligrino et al., 1993; Fabricius and Lauritzen, 1994). In this study, it was found in the same range and was regarded as evidence for successful inhibition of NOS. In a quite similar protocol of l-NNA application, Zhang et al. (1994) measured an inhibition of cortical NOS activity by >80%. The degree of hyperperfusion during CSD depends on the basal vascular tone. It was reported to be absent in awake animals (Duckrow, 1991), it is found in animals under volatile narcotics, and it is high in relatively preconstricted vessels as, for instance, with pentobarbital (Lauritzen, 1987; Wei et al., 1993). The last was the case in this study. This notion is supported by the tendency of an even higher hyperperfusion response when the basal vascular tone was further raised by previous application of l-NNA, a finding similar to recent data of Fabricius and Lauritzen (1994). But these authors, too, showed recently that LDF overestimates rCBF compared to the [¹⁴C]iodoantipyrine method at high-flow rates because the relative contribution of plasma and blood cells to overall flow varies at different flow levels (Fabricius and Lauritzen, 1995). The fact that systemic NOS inhibition did not alter blood flow response to CSD suggests that endothelial NOS is not involved in the mediation of this hyperperfusion. This finding supports recent data of Zhang et al. (1994). Evidence from our laboratory, however, indicates that topical superfusion of l-NNA in a closed cranial window produces a marked decrease of blood flow response to CSD (Dreier et al., 1995). Whether this stands for a spatial effect on endothelial versus parenchymal NOS or for a quantitative difference of NOS inhibition remains to be investigated.

In this study, we did not observe the previously described prolonged hypoperfusion (Lauritzen, 1984; Duckrow, 1991) after the first CSD had passed. The reason for this may be pentobarbital narcosis. We did not, with NOS inhibition, encounter an initial brief dip preceding the hyperperfusion phase as reported by Duckrow (1993).

The brief dip in MABP during CSD may reflect functional decortication by CSD that influences the activity of vegetative brain structures (Bureš et al., 1974a). Although this change is not in a range of any significance to blood flow, this reaction recalls vegetative features occurring during human migraine. The lessening of this MABP change with l-NNA may reflect restricted reactivity of the vascular system after systemic NOS inhibition.

Nitric oxide is not only a vasodilator but also a transmitter [for review see Bredt and Snyder (1994)] and may hence theoretically be involved in CSD propagation itself. But so far there has been no evidence for such involvement, nor did this study produce any: Neither susceptibility of the cortical tissue to CSD nor velocity of the DC deflection wave was influenced by l-NNA.

CONCLUSIONS

In this study, it is shown for the first time that NIRS is capable of detecting CSD noninvasively in the rat. We now investigate patients in the aura phase of migraine with the same method.

Acquisition of spectra over the entire NIR range (e.g., 600–1000 nm) instead of in only four wavelengths remain to be desired. This would make path length measurements possible (for instance, with the water absorption peak) (Matcher and Cooper, 1994) and could yield a much improved separation of cytochrome aa₃ from hemoglobin spectra, thus enabling us to continuously obtain reliable information about intracellular and blood oxygenation at the same time.

The use of optodes with a smaller optical diameter could enhance the spatial resolution, thus providing a sharper temporal correlation to DC and oxygen measurements.

Footnotes

Abbreviations used

Acknowledgment:

This study was supported by the Deutsche Forschungsgemeinschaft (DFG Vi93/7-1, Di454/4-2).

References

Andersen

Marmarou

(1992) Functional compartmentalization of energy production in neural tissue. Brain Res 585:190–195.

Back

Kohno

Hossmann

K-A

(1994) Cortical negative DC deflections following middle cerebral artery occlusion and KCl-induced spreading depression: effect on blood flow, tissue oxygenation, and electroencephalogram. J Cereb Blood Flow Metab 14:12–19.

Bredt

Snyder

(1994) Nitric oxide: a physiologic messenger molecule. Annu Rev Biochem 63:175–195.

Bureš

Burešova

Krivanek

(1974a) Spreading depression as a tool for studying the functional organization of the brain. In: The Mechanism and Applications of Leao's Spreading Depression of Electroencephalographic Activity. Prague, Academia, pp 171–216.

Bureš

Burešova

Krivanek

(1974b) The Mechanism and Applications of Leao's Spreading Depression of Electroencephalic Activity. Prague, Academia.

Colonna

Meng

Deal

Busija

(1994) Nitric oxide promotes arteriolar dilation during cortical spreading depression in rabbits. Stroke 25:2463–2470.

Cooper

Matcher

Wyatt

Cope

Brown

Nemoto

Delpy

(1994) Near infrared spectroscopy of the brain: relevance to cytochrome oxidase bioenergetics. Biochem Soc Trans 22:974–980.

Cope

Delpy

(1988) A system for long term measurement of cerebral and tissue oxygenation in newborn infants by near infrared transillumination. Med Biol Eng Comput 32:1457–1467.

Cope

Delpy

Reynolds

EOR

Wray

Wyatt

van der Zee

(1988) Methods of quantitating cerebral near infrared spectroscopy data. Adv Exp Biol Med 33:183–189.

10.

Dietrich

Feng

Leistra

Watson

Rosenthal

(1994) Photothrombotic infarction triggers multiple episodes of cortical spreading depression in distant brain regions. J Cereb Blood Flow Metab 14:20–28.

11.

Dirnagl

Kaplan

Jacewicz

Pulsinelli

(1989) Continuous measurement of cerebral cortical blood flow by laser-Doppler flowmetry in a rat stroke model. J Cereb Blood Flow Metab 9: 589–596.

12.

Dirnagl

Lindauer

Villringer

(1993) Role of nitric oxide in the coupling of cerebral blood flow to neuronal activation in rats. Neurosci Lett 149:43–46.

13.

Dirnagl

Niwa

Lindauer

Villringer

(1994) Coupling of cerebral blood flow to neuronal activation: role of adenosine and nitric oxide. Am J Physiol 267:H296–H301.

14.

Dora

Gyulai

Kovach

(1984) Determinants of brain activation-induced cortical NAD/NADH responses in vivo. Brain Res 299: 61–72.

15.

Dreier

Görner

Körner

Lindauer

Back

Dirnagl

(1995) Nitric oxide modulates the CBF response to spreading depression. J Cereb Blood Flow Metab (in press).

16.

Duckrow

(1991) Regional cerebral blood flow during spreading cortical depression in conscious rats. J Cereb Blood Flow Metab 11:150–154.

17.

Duckrow

(1993) A brief hypoperfusion precedes spreading depression if nitric oxide synthesis is inhibited. Brain Res 618:190–195.

18.

Edwards

Brown

Cope

Wyatt

McCormick

Roth

Delpy

Reynolds

(1991) Quantification of concentration changes in neonatal human cerebral cytochrome oxidase. J Appl Physiol 71:1907–1913.

19.

Fabricius

Lauritzen

(1994) Examination of the role of nitric oxide for the hypercapnic rise of cerebral blood flow in rats. Am J Physiol 266:H1457–H1464.

20.

Ferrari

Hanley

Wilson

Traystman

(1990) Redox changes in cat brain cytochrome-c oxidase after blood-fluoro-carbon exchange. Am J Physiol 258:H1706–H1713.

21.

Fox

Raichle

Mintun

Dence

(1988) Nonoxidative glucose consumption during focal physiologic activity. Science 241: 462–464.

22.

Gault

Lin

LaManna

Lust

(1994) Changes in energy metabolites, cGMP and intracellular pH during cortical spreading depression. Brain Res 641:176–180.

23.

Goadsby

Kaube

Hoskin

(1992) Nitric oxide synthesis couples cerebral blood flow and metabolism. Brain Res 595:167–170.

24.

Hansen

Olsen

(1980) Brain extracellular space during spreading depression and ischemia. Acta Physiol Scand 108:355–365.

25.

Hotez

Dailey

Geelhoed

Gainer

(1977) The role of oxygen diffusivity in biochemical reactions. Experientia 33:1424–1425.

26.

Iijima

Mies

Hossmann

K-A

(1992) Repeated negative DC deflections in rat cortex following middle cerebral artery occlusion are abolished by MK-801: effect on volume of ischemic injury. J Cereb Blood Flow Metab 12:727–733.

27.

Jing

Aitken

Somjen

(1994) Interstitial volume changes during spreading depression (SD) and SD-like hypoxic depolarization in hippocampal tissue slices. J Neurophysiol 71:2548–2551.

28.

Jöbsis

(1977) Noninvasive infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 198:1264–1267.

29.

Jöbsis

Keizer

LaManna

Rosenthal

(1977) Reflectance spectrophotometry of cytochrome aa3 in vivo. J Appl Physiol 43: 858–872.

30.

Kariman

Burkhart

(1985) Non-invasive in vivo spectrophotometric monitoring of brain cytochrome aa3 revisited. Brain Res 360:203–213.

31.

Kreisman

Sick

LaManna

Rosenthal

(1981) Local tissue oxygen tension—cytochrome aa3 relation in rat cerebral cortex in vivo. Brain Res 218:161–174.

32.

LaManna

McCracken

Paul

Prohaska

(1989) Stimulus-activated changes in tissue temperature in the anesthetized rat. Metab Brain Dis 4:225–237.

33.

Lauritzen

(1984) Long-lasting reduction of cortical blood flow of the brain after spreading depression with preserved autoregulation and impaired CO2 response. J Cereb Blood Flow Metab 4:546–554.

34.

Lauritzen

(1987) Regional cerebral blood flow during cortical spreading depression in rat brain: increased reactive hyperperfusion in low-flow states. Acta Neurol Scand 75:1–8.

35.

Lauritzen

(1994) Pathophysiology of the migraine aura. The spreading depression theory. Brain 117:199–210.

36.

Leão

(1994) Spreading depression of activity in the cerebral cortex. J Neurophysiol 7:359–390.

37.

Levasseur

Wei

Raper

Kontos

Patterson

(1975) Detailed description of a cranial window technique for acute and chronic experiments. Stroke 6:308–317.

38.

Lindauer

Villringer

Dirnagl

(1993) Characterization of the CBF response to somatosensory stimulation: the model and influence of anesthetics. Am J Physiol 264:H1223–H1228.

39.

Lipton

(1973) Effects of membrane depolarization on light scattering by cerebral cortical slices. J Physiol 231:365–383.

40.

Matcher

Cooper

(1994) Absolute quantification of deoxyhemoglobin concentration in tissue near infrared spectroscopy. Phys Med Biol 39:1295–1312.

41.

Mayevsky

Lebourdais

Chance

(1980) The interrelation between brain pO2 and NADH oxidation-reduction state in the gerbil. J Neurosci Res 5:173–182.

42.

McLachlan

Girvin

(1994) Spreading depression of Leão in rodent and human cortex. Brain Res 666:133–136.

43.

Mies

Paschen

(1984) Regional changes of blood flow, glucose, and ATP content determined on brain sections during a single passage of spreading depression in rat brain cortex. Exp Neurol 84:249–258.

44.

Mies

Iijima

Hossmann

K-A

(1993) Correlation between peri-infarct DC shifts and ischemic neuronal damage in rat. Neuroreport 4:709–711.

45.

Moskowitz

MacFarlane

(1993) Neurovascular and molecular mechanisms in migraine headaches. Cerebrovasc Brain Metab Rev 5:159–177.

46.

Nedergaard

Hansen

(1988) Spreading depression is not associated with neuronal injury in the normal brain. Brain Res 449:395–398.

47.

Pelligrino

Koenig

Albrecht

(1993) Nitric oxide synthesis and regional cerebral blood flow responses to hypercapnia and hypoxia in the rat. J Cereb Blood Flow Metab 13:80–87.

48.

Prado

Watson

Kuluz

Dietrich

(1992) Endothelium-derived nitric oxide synthase inhibition. Effects on cerebral blood flow, pial artery diameter, and vascular morphology in rats. Stroke 23:1118–1124.

49.

Shibata

Leffler

Busija

(1992) Pial arteriolar constriction following cortical spreading depression is mediated by prostanoids. Brain Res 572:190–197.

50.

Sylvia

Rosenthal

(1979) Effects of age on brain oxidative metabolism in vivo. Brain Res 165:235–248.

51.

Tsacopoulos

Lehmenkühler

(1977) A double-barreled Pt-microelectrode for simultaneous measurement of pO2 and bioelectrical activity in excitable tissues. Experientia 33:1337–1338.

52.

Vanderkooi

Maniara

Green

Wilson

(1987) An optical method for measurement of dioxygen concentration based upon quenching of phosphorescence. J Biol Chem 262:5476–5482.

53.

van der Zee

Arridge

Cope

Delpy

(1990) The effect of optode positioning on optical pathlength in near infrared spectroscopy of brain. Oxygen Transport to Tissue XII J:79–84.

54.

van der Zee

Arridge

Essenpreis

Potter

Edwards

Wyatt

McCormick

Roth

Reynolds

EOR

Delpy

(1992) Experimentally measured optical pathlengths for the adult head, calf and forearm and head of the newborn infant as a function of inter optode spacing. Adv Exp Med Biol 316:143–153.

55.

Wahl

Schilling

Parsons

Kaumann

(1994) Involvement of calcitonin gene-related peptide (CGRP) and nitric oxide (NO) in the pial artery dilatation elicited by cortical spreading depression. Brain Res 637:204–210.

56.

Wei

Weiss

Sinha

Chi

(1993) Effects of nitric oxide synthase inhibition on regional cerebral blood flow and vascular resistance in conscious and isoflurane-anesthetized rats. Anesth Analg 77:880–885.

57.

Wray

Cope

Delpy

Wyatt

Reynolds

EOR

(1988) Characterization of the near infrared absorption spectra of cytochrome aa3 and hemoglobin for the non-invasive monitoring of cerebral oxygenation. Biochem Biophys Acta 933:184–192.

58.

Zhang

Chopp

Maynard

Moskowitz

(1994) Cerebral blood flow changes during cortical spreading depression are not altered by inhibition of nitric oxide synthesis. J Cereb Blood Flow Metab 14:939–943.