| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Laboratoire de Neurophysiologie, Faculté de Médecine, Université Laval, Quebec G1K 7P4, Canada; and 2Laboratory of Neurobiology, Department of Biosciences, University of Helsinki, FIN-00014 Helsinki, Finland
Submitted 3 February 2004; accepted in final form 24 March 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Niedermeyer and Lopes da Silva 1999). Several studies have also demonstrated the involvement of glial cells in generating slow local field potentials during spreading depression (Somjen 1973; Somjen and Trachtenberg 1979), seizures (Amzica and Steriade 2000; Caspers et al. 1987; Heinemann and Walz 1998; Somjen 1973), and sleep (Amzica and Neckelmann 1999). Besides these concepts, pioneering studies in the early 1950s–1970s have proposed that slow potential shifts are generated at the interface between cerebrospinal fluid (CSF) and blood as a function of the partial pressure of CO2 (PCO2) (Held et al. 1964; Hornbein and Sorensen 1972; Kjällquist and Siesjö 1968; Sorensen and Severinghaus 1970; Tschirgi and Taylor 1958). Such potentials were suggested to originate across the blood–brain barrier (BBB) (Revest et al. 1993; Woody et al. 1970).
In spite of the work cited above, CO2-dependent DC potential shifts (DC shifts) are often attributed to dipoles along the apical dendrites of cortical neurons (Speckmann and Elger 1999). Moreover, as recent technical developments enable routine recording of full-band EEG, including DC potentials in humans (Vanhatalo et al. 2002, 2003), it becomes essential to elucidate relative contributions of neuronal and glial versus nonneuronal (blood, CSF) mechanisms to the genesis of slow EEG shifts. A recent study (Voipio et al. 2003) has proposed an electrophysiological model, focusing on the role of the BBB that accounts for various mechanistic aspects in the generation of DC shifts resulting from hyperventilation (HV) and hypoventilation (hv) manipulations.
In this study we examined a wide spectrum of elements that might be involved in the genesis of CO2-induced DC shifts. We start by testing an easily repeatable paradigm (HV/hv maneuvers) that produces reliable DC responses. If these slow shifts were generated across the BBB, they should be blocked by a disruption of the BBB. On the other hand, if respiratory DC shifts were generated within cortical neuronal and/or glial circuits, it could be anticipated that neuronal and/or glial membrane potentials would undergo corresponding variations. Furthermore, a dendritic generator of CO2-related DC shifts should elicit a cortical laminar distribution similar to the one resulting from synchronous neuronal activity such as, for instance, evoked potentials (Freeman and Nicholson 1975; Mitzdorf and Singer 1979) or sleep K-complexes (Amzica and Steriade 1998). The results from the present study, including both scalp and invasive recordings, support the view that CO2-dependent DC shifts are attributable to nonneuronal generators. Results from this article were previously presented in an abstract form (Amzica et al. 2002).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Acute experiments were carried out on 43 adult cats of both sexes, deeply anesthetized with ketamine and xylazine (10–15 mg/kg and 2–3 mg/kg, respectively, intramuscularly). The anesthesia was continuously monitored by the EEG, heart rate, and end-tidal CO2 concentration. Additional doses of anesthetic were given at the slightest tendency toward an activated EEG pattern or accelerated pulse (>110 beats/min). General surgical procedures included: cephalic vein canulation for systemic liquid delivery, lidocaine infiltration of all pressure points or incision lines, muscle paralysis with gallamine triethiodide, and tracheal cannulation. Cats were artificially ventilated throughout all experiments with room air. Skin temperature in the near proximity of the electrodes was also recorded by means of a copper-constantan thermocouple. Body temperature was maintained between 37 and 39°C with a heating pad. Glucose [5% solution, 10 ml, intravenous (iv)] was given every 3–4 h during experiments. Some cats (n = 10) also underwent cannulation of one of the external carotid arteries for direct injection of drugs into the brain vasculature.
In 18 cats, craniotomy was used to expose the cerebral cortex (suprasylvian gyrus) and to allow the insertion of both intracellular and extracellular glass micropipettes. In this preparation, the stability of intracellular recordings was enhanced by cisternal drainage, hip suspension, pneumothorax, and filling of the hole in the calvarium with a 4% agar solution. The latter constituted a firm pressure foot on the cortex preventing upward lifting of the brain. In some experiments (n = 6) the variations of the intracranial pressure were measured by inserting a needle, connected to a vertical graded CSF column, in the cisterna magna. Care was taken to avoid any liquid spillage around the insertion point to maintain pressurization of the subarachnoid compartment.
At the end of the experiments, the animals received a lethal iv dose of sodium pentobarbital (50 mg/kg). All experimental procedures were performed according to National Institutes of Health guiding principles and were also approved by the committee for animal care of Laval University.
Electrodes and recording
Eeg recordings.
Scalp recordings were performed using DC amplifiers with long-term stability better than 1 µV/h and bandwidth DC-300 Hz (A-M Systems, Carlsborg, WA). We used sintered Ag/AgCl electrodes (E220N-LP, In Vivo Metric, Healdsburg, CA) with 12-mm2 active area placed in a holder above the skin, forming a closed space that was filled with a chloride-containing electrode gel (Berner, Helsinki, Finland). The skin beneath the electrodes was incised to short-circuit skin generated potentials. These measures ensured stable DC recordings (Vanhatalo et al. 2002).
DC scalp electrodes were placed as shown in Fig. 1: on the median line [in frontal (F), central (C), and occipital (O) locations] and symmetrically with respect to the midline [2 temporal electrodes (TL, TR) and 2 mastoid electrodes (ML, MR)].
|
).
Intracellular recordings.
Intracellular recordings from neurons and glia in the cerebral cortex (suprasylvian areas 5 and 7) were obtained with glass micropipettes (tip diameter <0.5 µm) filled with potassium acetate (3 M, in situ impedance 35–50 M). In the case of neuronal impalements, only stable recordings with resting membrane potentials more negative than –60 mV, overshooting action potentials, and input resistances >20 M were accepted for analysis. Intraglial recordings were kept only if displaying a stable membrane potential more negative than –70 mV, without current compensation during the whole recording. No action potentials were fired at impalement, exit, or by imposing intracellular depolarizing currents that would bring the membrane potential close to –55 mV. Only glial cells recorded within the gray matter (between cortical surface and a depth of 1.5 mm) were considered for this study. Because changes in the ventilation rate produced significant variations in the volume of the brain (see RESULTS), additional measures had to be taken to preserve the stability of intracellular impalements. For instance, because hypoventilation was associated with expansion of the brain, the agar pressure foot was applied on the cortex with the intracellular pipette in place before the onset of the hypoventilation. In contrast, hyperventilation induced brain shrinkage and thus fewer intracellular recording could be kept during such maneuvers. The intracellular signals were passed through a high-impedance amplifier with an active bridge circuitry (bandwidth DC to 9 kHz). All signals were digitized (20-kHz sampling rate) and stored for off-line analysis.
Reference electrodes. For all types of recordings, a grounded reference electrode (Ag/AgCl) was placed either on the nasium or in the temporal muscles, and similar results were obtained. Nasium reference was used both during skin recordings and several intracerebral recordings, whereas muscle reference was used mainly during intracellular recordings.
Induction of DC shifts
DC shifts were induced by varying the artificial ventilation rate, which finally acts on differential pH values across the BBB attributed to the differences in the pH buffering systems and their volumes (Voipio 1998), and is reflected by the end-tidal CO2 concentration. The abolition of respiration regulatory mechanisms in our artificially ventilated animals ensured that the pH variation was controlled only by the respiration rate. The latter was adjusted as a function of the body weight between 20 and 28 cycles/min to maintain an end-tidal CO2 between 3 and 3.7%. Hyperventilation was elicited by accelerating the ventilation rate 
40 cycles/min, whereas hypoventilation was achieved with rates as low as 10 cycles/min. DC shifts were also induced with thiopental administered through one of the carotids at various concentrations (2, 4, 8, 16, 32, and 64 mg/kg dissolved in 1 ml saline) to study the dose dependency of the response. A disruption of the BBB was induced with sodium dehydrocholate (Spigelman et al. 1983) injected as a 3-ml solution (17.5%) or with mannitol (0.5 g/kg).
Analysis
In this study, most of the time constants of DC shifts were over the range of tens of seconds. Hence, we used occasionally down-sampled EEG signals at 0.3 Hz from which any baseline drifts unassociated with the experimental procedure were removed over the whole experiment. In such time series, each new EEG sample represented the average value of the voltage calculated over 3.3 s (between the preceding and current sample).
The undersampled EEG signals served for setting time marks at the onset/offset of DC shifts. A time mark was taken from the point where the baseline deviated for more than 1 SD of the background fluctuation.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Scalp DC shifts induced by respiratory maneuvers
To obtain a reliable pattern of DC shifts (Somjen 1973; Tschirgi and Taylor 1958; Woody et al. 1970), we hyperventilated (HV) and hypoventilated (hv) the paralyzed cat. These respiratory maneuvers led to variations of the expired CO2 corresponding to hypocapnia and hypercapnia, respectively. In all tested cats (n = 25), the EEG recorded from the scalp displayed positive DC shifts after hypocapnia (Fig. 1A) and negative shifts after hypercapnia (Fig. 1B). Usually, the respiratory frequency was modified for 3 min, after which the initial rate was regained. Each animal underwent 
3 cycles of such maneuvers. In accordance with a previous study (Voipio et al. 2003), the responses were strikingly similar in the same animal. The background electrical activity of the brain was dominated by the pattern imposed by ketamine–xylazine anesthesia, consisting of slow waves at about 1 Hz (see insets in Fig. 4) (Amzica and Steriade 1995; Steriade et al. 1996). This pattern was occasionally and transiently affected during the respiratory manipulations: HV increased slow activity (mostly 1–4 Hz), whereas hv reduced these slow components and enhanced fast rhythms (
15–40 Hz).
|
). Occasionally, an overshoot of the DC could follow after the cessation of the respiratory maneuver (Fig. 1B). This could be ascribed to the rapid return of CO2 levels at their control values after a relatively long period of hypoventilation, during which new homeostatic dynamics could have been established.
Disruption of the blood–brain barrier prevents CO2-dependent DC shifts
Following the proposal of Woody and colleagues that CO2-dependent DC shifts of brain tissue may represent BBB potentials (Woody et al. 1970), and in agreement with a recently proposed model (Voipio et al. 2003), we tested whether the disruption of the BBB would generate DC shifts and/or would affect ventilation-induced DC shifts. The break of the BBB was achieved with intracarotid injections of sodium dehydrocholate (DHC; Spigelman et al. 1983) and produced in all tested cases (n = 5) a persistent positive DC shift (Fig. 2A). The injection of an identical volume of saline (SAL in Fig. 2A) did not elicit notable DC shifts.
|
Anesthesia-induced DC shifts
Unilateral injections of a short-acting barbiturate (sodium thiopental) into one of the carotids (n = 5 cats) elicited in all cases positive shifts with higher amplitude on the side of the injection and in the central leads (Fig. 3A). These DC shifts were independent on the occasional variations in expired CO2 concentration. Besides the DC shifts, large doses of thiopental (>24 mg/kg) also produced an activity change in the conventional EEG by replacing the slow oscillatory pattern of ketamine and xylazine anesthesia (Fig. 3A, detail in panel 1) with burst suppression (panel 2 in Fig. 3A), indicative of an essential compromise in neuronal activity. The burst suppression outlasted the DC shift, suggesting that the 2 phenomena reflect different mechanisms (see DISCUSSION). The amplitude of the DC shifts was dependent on the dose of the injected thiopental and had an additive effect if delivered at short time intervals (<5 min; Fig. 3B). The amount of thiopental was varied, whereas the solution volume was always kept the same (1 ml). Small amounts of thiopental (<8 mg/kg) induced relatively low-amplitude DC shifts, whereas higher amounts (>16 mg/kg) were accompanied by higher-amplitude DC shifts. Moreover, the second injection of 32 mg/kg of thiopental triggered a larger DC shift than the first injection with the same amount.
|
This series of results raises the issue of whether the observed phenomenon was dependent on the agent used (thiopental) or on the depth of anesthesia. Thiopental is known to produce vasoconstriction, thus reducing cerebral blood flow (see Sakabe and Nakakimura 1994). To distinguish between blood flow–related and other putative mechanisms, we repeated the same protocol in 5 animals where thiopental was replaced with a volatile agent (isoflurane) (Fig. 4B). This had the advantage of allowing faster reversible manipulations and producing cerebral vasodilation (see Sakabe and Nakakimura 1994). Increasing doses of isoflurane were accompanied by sustained positive DC shifts (see control level before hv in each trace of Fig. 4B). Successive hypoventilations induced initially negative DC shifts with progressively decreasing amplitudes. Then, at a concentration between 2 and 3%, at which conventional EEG started to show suppressed patterns, a positive component was seen in the hv-induced response. Further increases in the anesthetic dose were associated with exclusive positive DC shifts that survived the removal of the isoflurane anesthesia and return to control patterns of conventional EEG (last panel at right in Fig. 4B).
Intracranial recordings associated with DC shifts
From the data presented so far it is clear that the production of DC shifts was consistently accompanied by changes of blood circulation parameters (e.g., cerebral blood flow, pH), known to have important regulatory functions in the brain. The ICP was among the factors that were affected by respiratory manipulations. The variation of the ICP that took place under the present experimental conditions was measured with a catheter filled with artificial CSF introduced in the subdural space at the level of the cisterna magna. Care was taken not to compromise the ICP by exposing only a very small surface of the atlanto-occipital dura and by avoiding any liquid spillage around the needle. Hyperventilation induced a relatively low decrease in ICP of 2.25% (n = 4 cats), whereas hv was associated with a more pronounced increase in ICP of 42.5% (n = 4).
Clear macroscopic movement of the brain was observed through the craniotomy window during respiratory manipulations. The displacement of the brain was read on the micropositioner carrying the pipette by continuously tracking the electrical contact between the recording electrode and the cortical surface. Hyperventilation was consistently associated (n > 100) with a drop of the brain surface within a range of 44 to 62 µm (on average 53 ± 3 µm; n = 25 selected episodes) for an average CO2 concentration decrease of 0.85%. In all cases hypoventilation induced brain expansion. On average, the uplifting of the cortex was 2.3 ± 0.3 mm (range 1.6 to 2.7 mm, n = 25) for an average increase of the CO2 concentration of 1.3%.
Intracortical recordings of DC shifts
Because of the gross movements of brain tissue seen during HV/hv as described above, stable intracortical recordings of DC shifts could be recorded under the following conditions only. First, recordings were performed after drainage of the CSF. Second, a firm pressure foot of agar was placed over the cortex and around the electrodes. The electrodes were made of a bundle of 4 glass pipettes with one tip on the surface of the cortex and 3 tips at various depths of the cortex and underlying white matter (Fig. 5A). The polarity of DC shifts was the same as recorded on the scalp and within lateral ventricles (Fig. 5B) and no polarity reversal could be observed (n = 12 animals). At the recording site the agar pressure foot prevented brain expansion induced by hv. Because brain shrinkage was still possible with HV, only small variations of the expired CO2 concentration (0.5%) were imposed in such experiments to reduce the magnitude of the brain displacement. Under such conditions, HV-induced DC shifts were positive at all recorded depths. The absence of polarity reversal at different depths directly argues against a neuronal generation of ventilation-induced responses.
|
To further examine the possible role of pyramidal apical-dendritic generators in relation to PCO2 responses, we performed intraneuronal recordings (n = 27). However, as described below, we did not observe membrane potential changes that might possibly have accounted for the DC shifts. Because of the small amplitude of intraparenchymal ventilation-induced DC shifts, which is within the range of normal baseline activity of neurons and glia, their contribution to the recorded intracellular potential with a reference outside the brain was negligible.
The epoch presented in Fig. 6 contains an intraneuronal recording in a cat that was displaying spontaneous recurrent spike-wave seizures (around one seizure/min). This kind of recording is important because it is well known that epileptic discharges are characterized by high-amplitude membrane potential variations that are reflected as ample deflections in the EEG. Figure 6A contains undersampled time series of the signals to assess global trends in their evolution. (The undersampling, however, does not conceal the true amplitude of the DC shift recorded respectively during seizures and hv manipulation.) The undersampling of the intraneuronal membrane potential occurred after clipping of the action potentials. The panel with the neuronal firing rate displays the instantaneous discharge frequency of action potentials and suggests the limits of successive seizures (6 seizures during the first half of the recording). The panel with the neuronal membrane potential below contains voltage values sampled at the respiration strokes. They are shown to disclose the maximal depolarizing amplitude reached during seizures. The period B underlined with gray is expanded below to show the complete evolution and recording quality of the neuronal and EEG recording during one of these seizures. Further details in the inset at the right show that epileptic discharges are made of paroxysmal depolarizing shifts starting at a more depolarized membrane potential than during the control slow oscillation, and reaching plateaus higher than the action potential threshold, thus inactivating spike generation. The superimposition of membrane potential histograms in Fig. 6D1 depicts a bimodal distribution for both conditions (seizure-free slow oscillation and epileptic discharges). On average, the neuron was depolarized from –69.3 ± 12.9 mV during the former condition to –61.5 ± 18.1 mV during the latter.
|
Comparison of membrane potential distributions during periods containing seizure-free slow oscillations with equivalent periods of hv (Fig. 6D2) further emphasizes the lack of systematic effect of hv on the neuronal polarization. In the case depicted here, the slow oscillation was only slightly more ample during hv than during control conditions, with no significant change in the average activity (–69.3 ± 12.9 mV vs. –69.5 ± 14.9 mV). At the level of the entire neuronal database (n = 27), no significant statistical difference (t-test, P < 0.05) could be detected for the average membrane potential before and during the respiratory maneuver. The average hyperpolarization was –1.3 ± 0.2 mV (range –2.3 to +0.4 mV; both depolarizations and hyperpolarizations were recorded in individual neurons).
Glial cells are known to display very slow variations of their membrane potential during spreading depression or epileptic seizures. However, reliable intraglial recordings (n = 45) during HV/hv failed to show significant variations of the membrane potential related to CO2-dependent DC shifts (Fig. 7). Both maneuvers induced a global depolarization of the glial membrane. The average membrane potential of glial cells was –84.6 ± 11.9 mV before and –84.3 ± 12.8 mV during hv-induced responses (n = 45 maneuvers), indicating a very small change in the membrane potential, with a mean of 0.26 mV and a range from –4.8 to 4.8 mV. HV-induced responses were associated with an average change of 2.7 mV (range –9.2 to 2 mV; n = 31 maneuvers), with a mean of –85.1 ± 11.7 mV before and –82.4 ± 11.0 mV during the HV. In both cases (HV and hv), membrane potential values before and after the respiratory maneuvers were not statistically different (paired 2-tailed Student's t-tests). Moreover, at the individual level, both hyperpolarization and depolarization of the membrane potential were encountered in both hv and HV cases (Fig. 7, A2 and B2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Speckmann and Elger 1999). Our conclusion is based on the following lines of evidence: 1)Epileptic discharges are paroxysmal events during which neuronal depolarizations attain their peak values. Notably, however, DC components associated with epileptic discharges in neurons are of much lesser amplitude than DC shifts elicited by HV/hv.
2)The ventilation-induced DC shifts occur in the absence of any parallel change in the neuronal membrane potential (Fig. 6).
3)Burst suppression is a state with scarce neuronal responsiveness (Steriade et al. 1994). Hence, it is noteworthy that respiration-related DC shifts were also present during neuronal quiescence such as anesthesia-induced burst suppression (Fig. 4).
4)Although glial cells are able to sustain very slow potentials, especially during seizures (Amzica and Neckelmann 1999; Caspers et al. 1984; Heinemann and Walz 1998; Somjen 1973), our data show no variation of the glial membrane potential in relation to respiratory-induced DC shifts, suggesting that the latter are not the reflection of glial activities. Similarly, a previous study (Carpenter et al. 1974) found inconsistent changes of the glial membrane potential as a function of CO2 levels. The small variations recorded in glial cells during both hv and HV were not correlated with DC shifts and were not consistently of the same polarity.
5)No extracellular, spatial polarity reversal of the DC shift responses was seen across or within the neocortex in recordings spanning from the cortical surface to the white matter and to deep brain structures (Fig. 5). This finding provides further support for the view that the DC shifts are not caused by cortical current dipoles generated in response to neuronal and/or glial activity.
The present data, summarized in the 5 arguments above exclude the possibility that CO2-dependent DC shifts recorded in the EEG are caused by neuronal or glial generators. Thus it is important to pay attention to the existence of steady potential gradients across blood vessels and the BBB that have been described by several teams (Held et al. 1964; Hornbein and Sorensen 1972; Revest et al. 1993; Sorensen et al. 1978; Tschirgi and Taylor 1958; Voipio et al. 2003). Their dependency on variations of pH was repeatedly emphasized (Hornbein and Pavlin 1975; Sorensen and Severinghaus 1970; Voipio et al. 2003; Woody et al. 1970).
Our main experimental paradigm relied on controlling the respiration rate of the paralyzed animal, and all our data are in agreement with the hypothesis that the BBB is the site of generation of the CO2-dependent DC shifts. In particular, we show that BBB disruption produces a prominent positive voltage shift, and prevents subsequent DC shifts of either polarity as seen in HV/hv-maneuvers (Fig. 2). In humans, CSF potential is positive with respect to blood (Sorensen et al. 1978) but the polarity of the DC shift after BBB disruption is unknown. From our experiments in cats, a positive DC shift consecutive to BBB disruption (which assumes canceling of the BBB polarization) suggests that the resting polarity of the CSF–blood potential is negative. Thus the polarization of the BBB appears to be opposite between humans and cats. This corroborates also with the different polarities of ventilation-induced DC shifts reported in previous studies: for instance, hv was associated with mostly positive responses in humans (Voipio et al. 2003) and with negative responses in cats (Woody et al. 1970; also present data). Voipio and colleagues (2003) proposed that this apparent incongruity may also be attributable to the locations of the reference and recording electrodes among studies and species.
Although, because of its time course after BBB disruption (Fig. 2), the DC shift could imply an underlying spreading depression, this possibility can be ruled out by the following arguments: 1) the DC shift starts synchronously in virtually all EEG leads, which would not be the case of spreading depression, known to propagate at a very low speed (a few mm/min according to Somjen 2001), and 2) although of lesser amplitude, the EEG still contains signals that could be assigned to neuronal and/or glial phasic activities. Further insights concerning the neuronal reactivity could be gained from studying the neuronal response to the BBB breakdown, although this was beyond the scope of the present study.
DC shifts were also elicited by 2 anesthetic agents such as thiopental and isoflurane in a dose-dependent manner (Figs. 3 and 4). None of them has any known impact on the BBB permeability. Although thiopental can induce arterial vasoconstriction (Tsuji and Chiba 1987), isoflurane induces vasodilation (Sakabe and Nakakimura 1994). The ensuing decreases and respective increases in cerebral blood flow can thus be safely ruled out as causes for the positive DC shift. In contrast, both are well-known agents that reduce the cerebral metabolic rate and O2 consumption (Hall and Murdoch 1990; Hoffman et al. 1998). The DC shifts induced by ventilation maneuvers were affected by the depth of anesthesia. At low doses of anesthetics, the DC shift response was gradually reduced and, with larger concentrations, a component of opposite polarity appeared in parallel with a likely decrease of the cerebral metabolic rate during burst-suppression patterns. Similarly, hypoxic and ischemic insults led to reversed responses (Woody et al. 1970).
Because ventilation DC shifts result from the polarization of the BBB, it might be speculated that, while keeping the intactness of the barrier, a dramatic impairment of the metabolic function could lead to the reversal of the voltage generator across the BBB because of a modification of ionic homeostasis and/or isoelectric line, as observed during ischemia (Gido et al. 1994; Siesjö 1992a,1992b). Studies on other tight epithelia have shown that a change in transepithelial potential difference may reflect an altered apical and/or basolateral membrane potential of epithelial cells (Mustonen 1998). Similarly, the polarization of the BBB results from a series connection of 2 membrane potentials, and the cellular mechanisms at the BBB level of the DC shifts cannot be worked out from the present data.
Moreover, it is clear that the CO2-related DC shifts reported here are not the only source of DC potentials in the brain. As previously demonstrated, slow events can also be generated within networks of neurons and/or glia during various states such as epileptic discharges (Amzica and Steriade 2000; Caspers et al. 1987; Heinemann and Walz 1998; Somjen 1973) or spreading depression (Somjen 1973; Somjen and Trachtenberg 1979). Nevertheless, pH changes occur both during seizures (Duffy et al. 1975; Wang and Sonnenschein 1955) and spreading depression (see Somjen and Tombaugh 1998), thus making it difficult to evaluate the respective contributions of neurons/glia versus pH to the recorded DC potentials.
In conclusion, variations in brain pH/CO2 consecutive to ventilation maneuvers generate persistent DC shifts that not only can be recorded within brain tissue but are also clearly seen in the extracranial EEG. We provide here, for the first time, conclusive evidence that electrical activity of neurons and glial cells is not the source of these DC shifts. Our data are in full agreement with a key role for BBB in the generation of the pH/CO2-dependent DC shifts.
In light of the present data, it is clear that much work is needed to elucidate the relative contributions and interactions of nonneuronal and neuronal/glial generators to DC shifts seen in both invasive and noninvasive recordings of brain activity. The tight link between neuronal activity and hemodynamic responses (Gjedde 2002) calls for caution in the mechanistic interpretation of slow EEG signals related to a number of situations, including cognitive tasks and preparatory states as well as pathophysiological brain activity (Vanhatalo et al. 2003; see Voipio et al. 2003).
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: F. Amzica, Laboratoire de Neurophysiologie, Faculté de Médecine, Université Laval, Quebec G1K 7P4, Canada (E-mail: florin.amzica{at}phs.ulaval.ca).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Amzica F, Nita DA, Vanhatalo S, Voipio J, and Kaila K. Origin of DC potential shifts in the EEG: an in vivo study in cat. Soc Neurosci Abstr 793.3, 2002.
Amzica F and Steriade M. Short- and long-range neuronal synchronization of the slow (<1 Hz) cortical oscillation. J Neurophysiol 73: 20–38, 1995.
Amzica F and Steriade M. Cellular substrates and laminar profile of sleep K-complex. Neuroscience 82: 671–686, 1998.[CrossRef][ISI][Medline]
Amzica F and Steriade M. Neuronal and glial membrane potentials during sleep and paroxysmal oscillations in the neocortex. J Neurosci 20: 6648–6665, 2000.
Carpenter DO, Hubbard JH, Humphrey DR, Thompson HK, and Marshall WH. Carbon dioxide effects on nerve cell function. In: Carbon Dioxide and Metabolic Regulation, edited by Nahas G and Schaefer KE. New York: Springer-Verlag, 1974, p. 49–62.
Caspers H, Speckmann E-J, and Lehmenkühler A. Electrogenesis of slow potentials of the brain. In: Self-Regulation of the Brain and Behavior, edited by Elbert T, Rockstroh B, Lutzenberger W, and Birbaumer N. New York: Springer-Verlag, 1984, p. 27–41.
Caspers H, Speckmann E-J, and Lehmenkühler A. DC potentials of the cerebral cortex: seizure activity and changes in gas pressures. Rev Physiol Biochem Pharmacol 106: 127–178, 1987.[ISI][Medline]
Daly DD and Pedley TA. Current Practice of Clinical Electroencephalography. New York: Raven Press, 1990.
Duffy TE, Howse DC, and Plum F. Cerebral energy metabolism during experimental status epilepticus. J Neurochem 24: 925–934, 1975.[CrossRef][ISI][Medline]
Freeman JA and Nicholson C. Experimental optimization of current source-density technique for anuran cerebellum. J Neurophysiol 38: 369–382, 1975.
Gido G, Kristian T, and Siesjö BK. Induced spreading depressions in energy-compromised neocortical tissue: calcium transients and histopathological correlates. Neurobiol Dis 1: 31–41, 1994.[CrossRef][Medline]
Gjedde A. Coupling of blood flow to neuronal excitability. In: The Neuronal Environment: Brain Homeostasis in Health and Disease, edited by Walz W. Totowa, NJ: Humana Press, 2002, p. 233–257.
Hall R and Murdoch J. Brain protection: physiological and pharmacological considerations. Part II: The pharmacology of brain protection. Can J Anaesth 37: 762–777, 1990.
Heinemann U and Walz W. Contributions of potassium currents and glia to slow potential shifts (SPSs). In: Glial Cells: Their Role in Behaviour, edited by Laming PR, Sykova E, Reichenbach A, Haton GI, and Bauer H. Cambridge, UK: Cambridge Univ. Press, 1998, p. 197–209.
Held D, Fencl V, and Pappenheimer JR. Electric potential of cerebrospinal fluid. J Neurophysiol 27: 942–959, 1964.
Hoffman WE, Charbel FT, Edelman G, and Ausman JI. Thiopental and desflurane treatment for brain protection. Neurosurgery 43: 1050–1053, 1998.[CrossRef][ISI][Medline]
Hornbein TF and Pavlin EG. Distribution of H+ and HCO3– between CSF and blood during respiratory alkalosis in dogs. Am J Physiol 228: 1149–1154, 1975.
Hornbein TF and Sorensen SC. d-c Potential difference between different cerebrospinal fluid sites and blood in dogs. Am J Physiol 223: 415–418, 1972.
Kjällquist Å and Siesjö BK. Regulation of CSF pH—influence of the CSF/plasma potential. Scand J Clin Lab Invest Suppl 102: I:C, 1968.
Mitzdorf U and Singer W. Excitatory synaptic ensemble properties in the visual cortex of the macaque monkey: a current source density analysis of electrically evoked potentials. J Comp Neurol 187: 71–83, 1979.[CrossRef][ISI][Medline]
Mustonen H. Improved intraepithelial two-dimensional cable analysis with application to necturus gastric antral mucosa. Pfluegers Arch 436: 646–652, 1998.[CrossRef][ISI][Medline]
Niedermeyer E and Lopes da Silva F. Electroencephalography: Basic Principles, Clinical Applications and Related Fields. Baltimore, MD: Williams & Wilkins, 1999.
Revest PA, Jones HC, and Abbott NJ. The transendothelial DC potential of rat blood–brain barrier vessels in situ. Adv Exp Med Biol 331: 71–74, 1993.[Medline]
Sakabe T and Nakakimura K. Effects of anesthetic agents and other drugs on cerebral blood flow, metabolism, and intracranial pressure. In: Anesthesia and Neurosurgery, edited by Cottrell JE and Smith DS. St. Louis, MO: Mosby, 1994, p. 129–143.
Siesjö BK. Pathophysiology and treatment of focal cerebral ischemia. Part I: Pathophysiology. J Neurosurg 77: 169–184, 1992a.[ISI][Medline]
Siesjö BK. Pathophysiology and treatment of focal cerebral ischemia. Part II: Mechanisms of damage and treatment. J Neurosurg 77: 337–354, 1992b.[ISI][Medline]
Somjen GG. Electrogenesis of sustained potentials. Prog Neurobiol 1: 201–237, 1973.[Medline]
Somjen GG. Mechanisms of spreading depression and hypoxic spreading depression-like depolarization. Physiol Rev 81: 1065–1096, 2001.
Somjen GG and Tombaugh GC. pH modulation of neuronal excitability and central nervous system functions. In: pH and Brain Function, edited by Kaila K and Ransom BR. New York: Wiley–Liss, 1998, p. 373–393.
Somjen GG and Trachtenberg M. Neuroglia as generator of extracellular current. In: Origin of Cerebral Field Potentials, edited by Speckmann EJ and Caspers H. Stuttgart, Germany: Thieme, 1979, p. 21–32.
Sorensen E, Olesen J, Rask-Madsen J, and Rask-Andersen H. The electrical potential difference and impedance between CSF and blood in unanesthetized man. Scand J Clin Lab Invest 38: 203–207, 1978.[ISI][Medline]
Sorensen SC and Severinghaus JW. Effect of cerebral acidosis on the CSF–blood potential difference. Am J Physiol 219: 68–71, 1970.
Speckmann E-J and Elger CE. Introduction to the neurophysiological basis of the EEG and DC potentials. In: Electroencephalography: Basic Principles, Clinical Applications, and Related Fields, edited by Niedermeyer E and Lopes da Silva F. Baltimore, MD: Williams & Wilkins, 1999, p. 15–27.
Spigelman MK, Zappulla RA, Malis LI, Holland JF, Goldsmith SJ, and Goldberg JD. Intracarotid dehydrocholate infusion: a new method for prolonged reversible blood–brain barrier disruption. Neurosurgery 12: 606–612, 1983.[ISI][Medline]
Steriade M, Amzica F, and Contreras D. Cortical and thalamic cellular correlates of electroencephalographic burst-suppression. Electroencephalogr Clin Neurophysiol 90: 1–16, 1994.[ISI][Medline]
Steriade M, Amzica F, and Contreras D. Synchronization of fast (30–40 Hz) spontaneous cortical rhythms during brain activation. J Neurosci 16: 392–417, 1996.
Tschirgi RD and Taylor JL. Slowly changing bioelectric potentials associated with the blood–brain barrier. Am J Physiol 195: 7–22, 1958.
Tsuji T and Chiba S. Mechanism of vascular responsiveness to barbiturates in isolated and perfused canine basilar arteries. Neurosurgery 21: 161–166, 1987.[ISI][Medline]
Vanhatalo S, Tallgren P, Andersson S, Sainio K, Voipio J, and Kaila K. DC-EEG discloses prominent, very slow activity patterns during sleep in preterm infants. Clin Neurophysiol 113: 1822–1825, 2002.[CrossRef][ISI][Medline]
Vanhatalo S, Holmes MD, Tallgren P, Voipio J, Kaila K, and Miller JW. Very slow EEG responses lateralize temporal lobe seizures: an evaluation of non-invasive DC-EEG. Neurology 60: 1062–1063, 2003.
Voipio J. Diffusion and buffering aspects of H+, HCO3–, and CO2 movements in brain tissue. In: pH and Brain Function, edited by Kaila K and Ransom BR. New York: Wiley–Liss, 1998, p. 45–65.
Voipio J, Tallgren P, Heinonen E, Vanhatalo S, and Kaila K. Millivolt-scale DC shifts in the human scalp EEG: evidence for a nonneuronal generator. J Neurophysiol 89: 2208–2214, 2003.
Wang R and Sonnenschein R. pH of cerebral cortex during induced convulsions. J Neurophysiol 18: 130–137, 1955.
Woody CD, Marshall WH, Besson JM, Thompson HK, Aleonard P, and Albe-Fessard D. Brain potential shift with respiratory acidosis in the cat and monkey. Am J Physiol 218: 275–283, 1970.
This article has been cited by other articles:
|
|
 |
|
  H. Yoshida, T. Kushikata, S. Kabara, H. Takase, H. Ishihara, and K. Hirota Flat Electroencephalogram Caused by Carbon Dioxide Pneumoperitoneum Anesth. Analg., December 1, 2007; 105(6): 1749 - 1752. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
