A wealth of recent research has extensively explored the phenomenology of cerebral blood flow (CBF) changes during the wake-sleep cycle both in the spatial (regional) and in the temporal (circadian) domains (Franzini, 2000; cf. also Maquet, 1997). The underlying physiological regulatory mechanisms are, however, understood only in part: few studies have addressed the issues of chemical systemic regulation (Parisi et al., 1988; Santiago et al., 1984), autoregulation (Grant et al., 1998) and microcirculation (Zoccoli et al. 1996). No data exist on other basic mechanisms of CBF control during sleep.

The present paper deals with a central issue in CBF regulation during the wake-sleep cycle: the role of O₂ in coupling metabolism and blood flow in the brain (for a review, cf., Lenzi et al., 1999).

REM Sleep as a Condition of Brain Activation

REM sleep is generally accompanied by an increase in CBF, oxygen and glucose uptake (for reviews cf., Franzini, 1992; Greenberg, 1980; Lenzi et al., 1999; Madsen and Vorstrup, 1991; Maquet, 1995). The increases in CBF and glucose uptake are well correlated with each other and in most cases are substantially higher than the increase in oxygen uptake (Chao et al., 1989; Clapp et al., 1980). Moreover, the arteriovenous difference in oxygen content is reduced and the venous oxygen content is higher in REM than in NREM sleep (Madsen et al., 1991; Meyer and Toyoda, 1971; Santiago et al., 1984). As far as the percentage of oxygenated hemoglobin is concerned, increments (Hoshi et al., 1994; Onoe et al., 1991), decrements (Munger et al., 1998) and wide fluctuations (Shiotsuka et al., 1998) have been reported in REM sleep.

Finally, an increase in tissue PO₂ was found by polarographic techniques (Garcia-Austt et al., 1968) in the transition from NREM to REM sleep. The circulatory and metabolic changes reported for the NREM to REM sleep transition fit quite well with those observed in other instances of neural activation. In neural activation, an increase in glucose uptake is generally paralleled by an increase in BF (for a review see Lenzi et al., 1999). O₂ consumption increases to a lesser extent than glucose uptake and BF (sensory stimulation [Fox et al., 1988]; mental work [Madsen et al., 1995]; seizures [Brodersen et al., 1973; Katsura et al., 1994]). In other instances, different results are reported: O₂ uptake increases to the same extent as glucose uptake (activation of the alpha-chloralose-anesthetized rat sensorimotor cortex [Hyder et al., 1996, 1997]), or does not change (vibrotactile stimulation in man, [Fujita et al., 1999; Ohta et al., 1999]). Blood oxygenation increases (Hoshi and Tamura, 1993; Kato et al., 1993; Krueger et al., 1996; Kwong et al., 1992; Turner et al., 1993) as well as lactate production, as indicated by the increased magnitude of the lactate arteriovenous difference (Madsen et al., 1998) and tissue lactate concentration (Fellows et al., 1993; Madsen et al., 1999; Prichard et al., 1991; Sappey-Marinier et al., 1992; cf., Villringer and Dirnagl, 1995). Polarographic measurements indicate an increase in tissue PO₂(cf., Lenzi et al., 1999). Worth recalling is the increase in neuron firing rate both in REMS (Steriade and Hobson, 1976) and during cortical activation (cf., Roland, 1993).

Based on these striking similarities, the transition from NREM to REM sleep may be consistently viewed as an instance of neural activation extending to the whole brain, as indicated by the substantial increase in BF and metabolism observed in all brain regions (Franzini, 1992). Thus, in order to better model the circulatory changes and related regulatory aspects, the transition from NREM to REM sleep will be hereafter considered an instance of neural activation and jointly analyzed with other instances of neural activation. This assumption makes it possible to draw on a larger amount of data available.

The decreased arteriovenous difference in O₂ content, the increased proportion of oxygenated hemoglobin and the increased PO₂ values found by polarographic measurements were taken as an index that O₂ availability at the level of oxidative enzymes was high, thus excluding that O₂ utilization is limited by diffusion and that O₂ is a vasodilator signal in REM sleep. On the contrary, the present study will show that in microregions at mid-distance from capillaries O₂ diffusion may be a limiting factor of O₂ consumption. Consequently O₂ may be a mediator coupling metabolism and blood flow during REM sleep as well as in other instances of neural activation.

Theoretical Considerations

The above data raise some main questions: a) Why, in most cases, is the increase in O₂ uptake lower than that of glucose? Why, in other cases, is it not?, and b) What is the role of O₂ in regulating CBF in REM sleep, as well as in other instances of neural activation?

The current view is that O₂ utilization is normally set by tissue metabolic activity and only exceptionally may become diffusion limited when PO₂ along the capillary falls below a critical level (cf., the model of Schumacker and Samsel, 1989). However, oxygen uptake of brain tissue has been evaluated 35-42 ml/min/kg (Clarke and Sokoloff, 1993; Ohta et al., 1999), which is much larger than the value of 5 ml/min/kg considered in this model.

Lactate production occurs normally in the brain. In concomitance with neural activation, astrocytes take in K+ and glutamate, the energy required being produced by anaerobic glycolysis and lactate release; lactate in its turn is thought to be aerobically metabolized by neurons (Magistretti and Pellerin, 1996; Pellerin and Magistretti, 1994; Poitry-Yamate et al., 1995). However, in this way glucose eventually would be metabolized oxidatively with no explanation for glucose taken up in excess with respect to O₂, or for the increase in the magnitude of lactate arteriovenous difference.

In order to account for the excess glucose utilization that is mostly observed during activation, it has been suggested that glucose oxidation may be near maximal capacity at rest (oxidative capacity limitation) (Van den Berg, 1986; cf., Fox et al., 1988). The hypothesis of "aerobic" glycolysis, i.e., a stimulation of nonoxidative glucose utilization despite adequate oxygen supply, was also taken into account by Ueki et al. (1988) (for a discussion cf., Villringer and Dirnagl, 1995). Krueger et al. (1996), based on transient effects observed on activation and deactivation in magnetic resonance (MR) images in man, initially proposed that a progressive up-regulation of oxidative phosphorylation occurs during activation: as a consequence, the uncoupling of oxygen delivery and oxidative metabolism observed in the transition between functional states would disappear at the steady state, thus eliminating the oxidative capacity limitation. However, alternative interpretations of MR images were later proposed (Buxton and Frank, 1997; Buxton et al., 1998) based on a model taking into account the biomechanical properties of the brain vascular system, in particular the conflicting effects of dynamic changes in both blood oxygenation and blood volume; calculations based on the model show pronounced transients in the deoxyhemoglobin content and the blood oxygenation level dependent (BOLD) signal measured with functional MR imaging, even in the presence of tight coupling of cerebral blood flow and oxygen metabolism. Thus, the hypothesis of a progressive up-regulation of oxidative phosphorylation was no longer required to explain the transient effects in MR images. Of course, the same oxidative capacity limitation hypothesis could also be invoked to explain the excess glucose uptake in REM sleep: little reserve would be available in oxidative machinery for a further increase in glucose oxidation and extra energy requirements would be partly satisfied by non-oxidative metabolism and lactate production.

Data, however, exist that cannot be explained on the basis of the hypothesis of oxidative capacity limitation:

1) In different experimental conditions, scattered PO₂ values have been found in the brain, with measures ranging from arterial PO₂ to zero (Padnick et al., 1999; Shinozuka et al., 1989), values close to zero augmenting with decreasing arterial PO₂ (Kozniewska et al., 1987; Shinozuka et al., 1989), decreasing with increasing arterial PO₂ and almost disappearing at arterial PO₂ ~300 torr (Shinozuka et al., 1989) (Figure 1).

2) Tissue concentration of H+, related to tissue lactate concentration, as well as cerebral blood flow, changed accordingly: [H+] and BF decreased with increasing arterial PO₂ in the range 100 to 300 mmHg. Data consistent with those of Shinozuka et al. (1989), were also found by Fennema et al. (1989): an increase in O₂ inspired fraction from 0.3 to 1.0 determined an early increase in brain tissue PO₂, followed by a slow return to baseline value due to a slow decrease in local BF.

3) Cytochrome-c oxidase (Cyt-Ox) oxidation was found to increase in the brain under different conditions (neural activation, [Heekeren et al., 1999]); 7% CO₂ in inspired air at near normoxia (126 mmHg PaO₂, [Quaresima et al., 1998]), but not 15% CO₂ in inspired air at hyperoxia (353 mmHg PaO2, [Hoshi et al., 1997]), while it decreased when arterial blood pressure dropped below the autoregulation limit (Cooper et al., 1998). According to measurements in vivo, Cyt-Ox is not fully oxidated in vivo (82% oxidation, [Cooper et al., 1998]). To the extent Cyt-Ox oxidation may increase with CBF increasing beyond its normal value, this strongly suggests that Cyt-Ox oxidation is not maximal in normal conditions ("maximal" does not necessarily mean 100% oxidation, other factors being implicated) and may increase by increasing tissue oxygenation. Even if different explanations are proposed (cf., Quaresima et al., 1998), these data are compatible with the presence of hypoxic microregions in the brain.

In order to explain these data, additional hypotheses are to be considered. The hypothesis here proposed is that a limit may exist in the capacity of O₂ to pass from erythrocytes to mitochondria (oxygen diffusion limitation), due to all the steps involved in this passage, i.e., O₂ unbinding from hemoglobin and O₂ diffusion within the capillary and in the surrounding tissue. Due to O₂diffusion limitation, brain microregions lying at mid-distance between capillaries may become hypoxic. The presence of such hypoxic microregions would be affected by different factors: a) with increasing metabolic rate, O₂ consumption in pericapillary microregions increases and the radial PO₂ fall around the capillary becomes steeper, so that, in spite of CBF increases augmenting capillary PO₂, lower PO₂ values may occur at mid-distance between capillaries. This concept is classically described in respiratory physiology (see Figure 2). As a consequence, in microregions far from capillaries the increase in metabolic rate is accompanied by a decrease in PO₂, so that partial recourse to non-oxidative glucose metabolism occurs; and b) changes in arterial PO₂ affect blood-tissue PO₂ gradient and O₂ diffusion; in spite of CBF changes observed with changing arterial PO₂, hypoxic microregions would enlarge in arterial hypoxia and shrink in arterial hyperoxia. It should be noted that capillary network architecture varies, depending on a variety of factors, including tissue type (gray or white matter), but also random processes intrinsic to capillary network development. As a consequence, hypoxic microregions do not necessarily form around all capillaries, but only in such capillary sets where diffusion distances are high with respect to local metabolic rates.

In conclusion, the above data show that O₂ diffusion in brain tissue is a major problem. The presence of a significant number of spots with a PO₂ close to zero indicates that in some microregions O₂ supply may be at its limit and partial recourse to anaerobic glucose metabolism, with lactate production, occurs. Oxidative capacity limitation and O₂ diffusion limitation may coexist in the brain, the former showing up on activation in microregions close to capillaries, where O₂ availability is abundant, the latter in microregions far from capillaries, where the increase in cell metabolic rate is concomitant with a decrease in PO₂. These hypoxic microregions might be the site of origin of vasodilatory messages, acting through different mechanisms and providing error signals helping keep blood flow adequate to metabolic needs, both in short (vasomotor responses) and long-term (plastic adjustments of capillary density and cell enzyme activity, see below) regulations. Based on the number and/or size of hypoxic microregions, local blood flow would be set at such a level as to provide appropriate O₂ supply to most of the tissue, leaving however a very small percentage of tissue hypoxic; this would guarantee against exceedingly high, and thus uneconomic, local BF levels. Obviously, the feedback regulation based on O₂ diffusion in the tissue may coexist with other regulations, including not only feedback, but also anticipatory regulations, characterized by arteriolar vasodilatation directly controlled by neuronal activation.

Glucose is not Diffusion Limited

D-glucose crosses cell membranes by facilitative diffusion in a manner that is stereospecific, saturable, nonconcentrative, and not requiring energy or sodium. Even if the brain glycolytic rate increases with glucose availability (Harada et al., 1993), the brain normally extracts 10% of glucose from arterial blood (Edvinsson et al., 1993; Villringer and Dirnagl, 1995), so that glucose consumption can increase in the absence of any rise in CBF by increasing glucose extraction only; this has been shown in somatosensory stimulation in the presence of scopolamine, which abolishes the CBF response but not the increase in glucose uptake (Ogawa et al., 1994); the same was also observed in seizures induced by topical application of penicillin on the cortical surface in rats: the increase in glucose utilization was not accompanied by a similar increase in blood flow (Mies et al., 1981). This indicates that glucose has no major effects on CBF regulation, albeit some minor effects have been described: hyperglycemia lowers resting CBF (Duckrow, 1995; Knudsen et al., 1992), while "during hypoglycemia, cerebral blood flow does not increase significantly until peripheral glucose levels are very low (2.0 mmol/l), that is well below the blood glucose threshold for impairment of cognitive functions (3.0 mmol/l)" (Thomas et al., 1997). On the other hand, the rate-limiting step for glucose catabolism in the brain is normally phosphorylation (Edvinsson, et al., 1993) and glucose transport is not rate-limiting under ordinary circumstances (Harik et al., 1994; Hein et al., 1975). With increasing metabolic activation, unidirectional glucose flow increased with glucose consumption, always exceeding it by ~60% (Cremer et al., 1983), thus confirming that even in the case of strong metabolic activation no diffusion limit is reached for glucose. Only in seizures, when the glucose metabolic rate increased more than four-fold, did glucose consumption exceed blood-tissue glucose transport, thus showing that glucose may become diffusion limited in extreme situations; however, endogenous brain glucose accounted for most of the increase in glucose utilization, and glycolysis, at least in an early period, was not limited by the blood-tissue diffusion rate (Fujikawa et al., 1989; Miller et al., 1982). For this reason, diffusion limitation for glucose is less compelling than for O₂, given the lack of endogenous stores of O₂.

The capacity of glucose to diffuse within brain tissue is much higher than that of O₂ (cf., Lenzi et al., 1999, p. 172), so that substrate requirements should be much easier to satisfy for glucose than for O₂. In fact, glucose uptake is in excess with respect to O₂ uptake during conditions of functional activation such as REM sleep. It is noteworthy that, since anaerobic ATP production is less efficient than the oxidative production (two ATP molecules instead of 33 per glucose molecule), glucose consumption vs. ATP production sharply increases as anaerobic glycolysis starts up, thus increasing glucose/ O₂ ratio, as occurs in REM sleep (Chao et al., 1989; Clapp et al., 1980).

Is Oxygen Diffusion Limited?

The hypothesis that a "substantial diffusion limitation of oxygen transfer" could also occur in brain tissue was initially advanced by Gjedde et al. (1991), and Kuwabara et al. (1992). According to this hypothesis, if O₂ consumption is limited by O₂ diffusion, only a decrease in diffusion distances, i.e., capillary recruitment, could increase O₂ consumption. During sensory stimulation, however, these authors reported indirect evidence of an increase in the density of perfused capillaries, but no corresponding increase in O₂ consumption. Thus, there was no evidence of O₂ diffusion limitation. Now it is well established that about 95% capillaries are open for plasma perfusion in all conditions (Goebel et al., 1990). This also occurs in a state of low metabolic rate like NREM sleep, and there is no capillary recruitment in REM sleep in rats (Zoccoli et al., 1996) even in the presence of CBF increments up to 70-100% (Zoccoli et al., 1994). Data on erythrocyte flow indicate that as much as 91% of capillaries are erythrocyte perfused at rest in the brain (Villringer et al., 1994). The lack of capillary recruitment, however, does not rule out O₂ diffusion limitation.

In fact, O₂ diffusion capacity in the brain is much lower than that of glucose (for discussion see Lenzi et al., 1999), so that relevant PO₂ gradients can be measured in the tissue (see above) and venous PO2 is by no means representative of average tissue PO2. Moreover, O₂ unbinding from hemoglobin significantly hampers blood-tissue O₂ transfer and increases O2 diffusion resistance: the O₂ release time was evaluated as ~200-300 ms (Honig et al., 1984), which is the same magnitude as the red blood cell transit time in brain capillaries (100-300 ms) (Hudetz, 1997); as a consequence, O₂ unbinding is reduced as blood flow velocity increases, as occurs during neural activation in the absence of capillary recruitment (REM sleep) (Zoccoli et al., 1996); the resulting mean intracapillary resistance fraction to O₂ diffusion was estimated as ~0.37 in muscle (Tuhin and Popel, 1996) and ~0.50 in the lung (West, 1990). All this makes the assumption of O₂ diffusion limitation appropriate.

On the other hand, the brain normally extracts up to 50% of oxygen from arterial blood (Edvinsson et al., 1993; Villringer and Dirnagl, 1995) and an increase in O₂ consumption can conceivably occur only if BF increases. This makes O₂ a good candidate for blood flow-metabolism coupling in the different conditions of the wake sleep cycle.

DISCUSSION

The wide dispersion of PO₂ values in brain tissue and the presence of PO₂ values close to zero, both depending on arterial PO₂ (Padnick et al., 1999; Shinozuka et al., 1989) clearly show that O₂ diffusion in brain tissue is a relevant problem. Tissue H+ concentration, presumably related to lactate concentration, as well as cerebral blood flow, change accordingly to arterial PO₂. All this can be explained by the present model: arterial hyperoxia reduces the size and/or number of hypoxic microregions, thus reducing their vasodilatatory effect and then BF, until tissue PO₂ distribution returns to baseline. The way arterial hyperoxia affects glycolysis is conceivably linked to O₂ diffusion in the tissue.

The increase in average tissue PO₂ reported during activation is compatible with the presence of hypoxic microregions, larger than at rest, resorting to partial anaerobic glucose metabolism. In fact, during activation, vasodilatation and consequent blood flow and blood volume changes may occur over a larger spatial domain than the actual site of neuronal activity (Malonek and Grinvald, 1996; Menon et al., 1995; Turner and Grinvald, 1994), thus producing an increase in average tissue PO₂.

This increase in average tissue PO₂ may explain the transient increase in Cyt-Ox oxidation observed during neural activation (Heekeren et al., 1999), as well as when CBF is forced up by hypercapnia at near normoxia (Quaresima et al., 1998), but not by hypercapnia at hyperoxia (Hoshi et al., 1997). Other explanations have been preferred (Cooper et al., 1998; Heekeren et al., 1999), but tissue oxygenation may be one of several factors affecting Cyt-Ox oxidation in the brain. Changes in arterial PO₂ within physiological limits induced slighter (Hampson et al., 1990) or not detectable changes (Cooper et al., 1998; Hoshi et al., 1997; Wickramasinghe et al., 1995) in Cyt-Ox oxidation. This discrepancy may depend on the fact that such changes are within the noise range of NIRS measurement.

From the above data it appears that at rest local CBF regulation can maintain Cyt-Ox oxidation nearly constant at a value lower than 100% (82%, cf., Cooper et al., 1998) even in the presence of wide changes in arterial PO₂ within physiological limits. This suggests that local BF regulation is tuned at maintaining Cyt-Ox oxidation nearly constant with changing PaO₂; only if CBF regulatory mechanisms are overcome Cyt-Ox oxidation may undergo major changes. Interestingly, the increase in CBF induced by hypercapnia augments Cyt-Ox oxidation only at normoxia, but not at hyperoxia: this correlates with the disappearance of hypoxic spots at hyperoxia (Shinozuka et al., 1989) and may indicate that Cyt-Ox oxidation is already maximal by the effect of hyperoxia and cannot augment any further by the effect of CBF increase. To conclude, through the presence of hypoxic microregions, acting as sensors, CBF regulatory system helps optimize BF, guaranteeing a constant level of Cyt-Ox oxidation while avoiding unnecessary BF.

At any rate, the hypothesis that O₂ diffusion limitation exists in the brain, albeit restricted to microregions far from capillaries, is in particular required to explain a) the presence in brain tissue of spots with PO₂ close to zero, and b) the fact that H+ concentration is reduced by increasing arterial PO₂ in the range 100 to 300 mmHg. The fact that cytochrome-c oxidase oxidation augments by the effect of increased CBF at normoxia but not at hyperoxia strongly suggests that Cyt-Ox oxidation is not normally maximal and may increase by the effect of increased tissue oxygenation.

The hypothesis that O₂ diffusion limitation exists in brain tissue is also consistent with the following data:

1) Increased capillary density is observed in brain regions characterized by a high metabolic rate. As an example, the average intercapillary distance is ~25-40 µm for gray and ~80 µm for white matter (Klein et al., 1986; Padnick et al., 1999). Capillary density is positively correlated with Cyt-Ox activity, and reciprocal patterns exist for lactate dehydrogenase and Cyt-Ox activity within laminated gray matter structures (Borowsky and Collins, 1989); Cyt-Ox is a marker for neuronal activity (Wong-Riley, 1989). Within cells, "the size of clusters of mitochondria is limited by the magnitude of the O₂ gradient needed to provide adequate O₂ concentrations for mitochondrial function within the clusters" (Jones, 1986). Microvascular anatomy is closely associated with oxidative capacity during development (Tuor et al., 1994). This can be summarized as follows: apart from genetic factors, capillary density and oxidative enzyme activity develop in parallel with metabolic rate, thus matching substrate supply and utilization, as well as waste products removal, with the average local energy need. It may be assumed that in the different conditions of everyday life if the local energy need exceeds the average level on which the O₂ utilization system was tuned, an O₂ diffusion limit may show up.

2) Brain capillary density increases in hypobaric hypoxia (Harik et al., 1995; LaManna et al., 1992), but is not affected by hyperglycemia (Kikano et al., 1989; Knudsen et al., 1991). Of course, reducing intercapillary distance favors the blood-tissue diffusion of all substrates, but this is of benefit only for those substrates whose diffusion is a limiting factor. Such a substance may not be glucose (see above), but, if any, O₂, in particular during hypobaric hypoxia, when the blood-tissue PO₂ gradient is reduced.

3) Ambient hypoxia affects sleep architecture by reducing Total Sleep Time and in particular the percentage of REM sleep in humans (Anholm et al., 1992; Mizuno et al., 1993), cats (Baker and McGinty, 1979; Huertas and Mc Millin, 1968) and rats (Laszy and Sarkadi, 1990; Pappenheimer, 1977). According to Huertas (Huertas and Mc Millin, 1968), "a mechanism closely related to the metabolism of oxygen in the brain must play an important role in the production of paradoxical sleep." The sleep disturbances may be reduced by enrichment of room air with O₂ (Luks et al., 1998; West, 1995). Moreover, in the chronic pontine cat hyperoxia increases REM sleep by 85% (Arnulf et al., 1998). Local CBF regulation during REM sleep is fully effective and aims at maintaining adequate O₂ supply to the brain: a recent work by Walker et al. (1999), reported a dramatic increase in CBF during REM sleep in response to hypoxic stimulation. However, regulatory mechanisms may fail in extreme conditions and REM sleep alterations ensue.

As far as O₂ utilization during REM sleep is concerned, it may be argued that in normoxic conditions, according to the above model of CBF regulation, O₂ availability is at its limit only in hypoxic microregions, where partial recourse to anaerobic glucose metabolism occurs. In hypoxic conditions, such microregions enlarge and alterations in REM sleep become manifest. Interestingly, ambient hypoxia affects blood-tissue O₂ diffusion not only by reducing the blood-tissue O₂ gradient, but also through the effects of hyperventilation on the hemoglobin dissociation curve: a left shift occurs as a consequence of hypocapnia and alkalosis, O₂ unbinds from hemoglobin at capillary PO₂ values lower than at normoxia and tissue hypoxia occurs, independently of the amount of O₂ supplied by blood flow. Enrichment of ambient air with O₂ not only increases capillary PO₂, but also reduces hyperventilation, blood hypocapnia (and alkalosis) and tissue hypoxia. Of course, habituation is also to be considered.

The present hypothesis of oxygen diffusion limitation also helps to explain why O₂ consumption during neural activation sometimes increases as much as glucose consumption, sometimes to a lesser extent. In the first condition, the resting energy level is low, the increase in O₂ utilization required by activation is reached within the limits of O₂ diffusion capability and local BF regulation, so that only aerobic glucose utilization occurs. In the second condition, the resting energy consumption is high, in spite of local BF regulation, the increase in O₂ utilization required by the activated state would overcome the limits of O₂ diffusion, at least in microregions at mid-distance from capillaries, so that partial recourse to anaerobic glucose utilization occurs.

An increase in CBF, as normally observed during activation, is of benefit for O₂ tissue availability even in the absence of any increase in diffusion surface, i.e., in the absence of capillary recruitment. In fact, CBF increase is functional in sustaining average capillary PO₂, blood-tissue PO₂ gradient and O₂ diffusion rate, thus improving tissue O₂ availability. This is clearly displayed in the mathematical model of Buxton and Frank (1997): disproportionately large increases in flow are required to produce a small increase in O2 uptake, but the magnitude of the CBF increase depends strongly on the resting O₂ extraction fraction. In fact, whenever an increase in O₂ blood-tissue transfer is required, e.g., in the presence of increased metabolic rate, arterial hypoxia, hemodilution and others, an increase in CBF is observed.

In conditions of low ambient PO₂ and high local metabolic rate, O₂ supply problems may become stringent. This may be the case of REM sleep at high altitudes. It is well known that REM sleep is profoundly affected by O₂. Hypoxic atmosphere reduces, and hyperoxic atmosphere increases, REM sleep time. According to the present hypothesis, the general increase in whole brain energy needs occurring in REM sleep pushes O₂ utilization towards its upper limits, thus enhancing the relevance of O₂ diffusion limitation. It is also known that REM onset requires safe environmental conditions, due to the deep alteration of homeostatic regulations occurring in this sleep stage (Parmeggiani and Morrison, 1990). In this condition, any pathological decrease in arterial PO₂ and/or O₂ delivery to the brain creates a specific risk factor in REM sleep (Parmeggiani, 1991).

REFERENCES

2. Arnulf I, Sastre J-P, Buda C, Jouvet M. Hyperoxia increases paradoxical sleep rhythm in the pontine cat. Brain Res 1998; 807: 160-6.

3. Baker TL, McGinty DJ. Sleep-waking patterns in hypoxic kitten. Dev Psychobiol 1979; 12: 561-75.

4. Borowsky IW, Collins R. Metabolic anatomy of brain: a comparison of regional capillary density, glucose metabolism, and enzyme activities. J Comp Neurol 1989; 288: 401-13.

5. Brodersen P, Paulson OB, Bolwig TG, Rogon ZE, Rafaelsen OJ, Lassen NA. Cerebral hyperemia in electrically induced epileptic seizures. Arch Neurol 1973; 28: 334-8.

6. Buxton RB, Frank LR. A model for the coupling between cerebral blood flow and oxygen metabolism during neural stimulation. J Cereb Blood Flow Metab 1997; 17: 64-72.

7. Buxton RB, Wong EC, Frank LR. Dynamics of blood flow and oxygenation changes during brain activation: the balloon model. Magn Reson Med 1998; 39: 855-64.

8. Chao CR, Hohimer AR, Bissonnette JM. The effect of electrocortical state on cerebral carbohydrate metabolism in fetal sheep. Dev Brain Res 1989; 49: 1-5.

9. Clapp JF, Szeto HH, Abrams R, Larrow R, Mann LI. Physiologic variability and fetal electrocortical activity. Am J Obstet Gynecol 1980; 136: 1045-50.

10. Clarke DD, Sokoloff L. Circulation and energy metabolism of the brain. In: Basic Neurochemistry. Siegel JH, Agranoff BW, Albers RW, Molinoff P, eds. New York: Raven Press, 1993: p. 645-80.

11. Cooper CE, Delpy DT, Nemoto EM. The relationship of oxygen delivery to absolute haemoglobin oxygenation and mitochondrial cytochrome oxidase redox state in the adult brain: a near-infrared spectroscopy study. Biochem J 1998; 332: 627-32.

12. Cremer JE, Cunningham VJ, Seville MP. Relationships between extraction and metabolism of glucose, blood flow, and tissue blood volume in regions of rat brain. J Cereb Blood Flow Metab 1983; 3: 291-302.

13. Duckrow RB. Decreased cerebral blood flow during acute hyperglycemia. Brain Res 1995; 703: 145-50.

14. Edvinsson L, MacKenzie ET, McCulloch J. Cerebral blood flow and metabolism. New York: Raven Press, 1993.

15. Fellows LK, Boutelle MG, Fillenz M. Physiological stimulation increases nonoxidative glucose metabolism in the brain of the freely moving rat. J Neurochem 1993; 60: 1258-63.

16. Fennema M, Wessel JN, Faithfull NS, Erdmann W. Tissue oxygen tension in the cerebral cortex of the rabbit. J Adv Exp Biol 1989; 248: 451-60.

17. Fox PT, Raichle ME, Mintun MA, Dence C. Nonoxidative glucose consumption during focal physiologic neural activity. Science 1988; 241: 462-4.

18. Franzini C. Brain metabolism and blood flow during sleep. J Sleep Res 1992; 1: 3-16.

19. Franzini C. The peripheral circulation in sleep: cerebral and extracerebral regional blood flows. In: The Principles and Practice of Sleep Medicine. Third Edition. Kryger MH, Roth T, Dement WC, eds. Philadelphia: WB Saunders Company, 2000: pp. 193-203.

20. Fujikawa DG, Dwyer BE, Lake RR, Wasterlain CG. Local cerebral glucose utilization during status epilepticus in newborn primates. Am J Physiol 1989; 256: C1160-7.

21. Fujita H, Kuwabara H, Reutens DC, Gjedde A. Oxygen consumption of cerebral cortex fails to increase during continued vibrotactile stimulation. J Cereb Blood Flow Metab 1999; 19: 266-71.

22. Garcia-Austt E, Velluti R, Villar JI. Changes of brain PO2 during paradoxical sleep in cats. Physiol Behav 1968; 3: 477-85.

23. Gjedde A, Ohta S, Kuwabara H, Meyer E. Is oxygen diffusion limiting for blood-brain transfer of oxygen? In: Brain work and mental activity. Lassen NA, Ingvar DH, Raichle ME, Friberg L. Copenaghen: Munskaard, 1991: pp. 177-84.

24. Goebel U, Theilen H, Kuschinsky W. Congruence of total and perfused capillary network in rat brains. Circ Res 1990; 66: 271-81.

25. Grant DA, Franzini C, Wild J, Walker AM. Cerebral circulation in sleep: vasodilatory response to cerebral hypotension. J Cereb Blood Flow Metab 1998; 18: 639-45.

26. Greenberg JH. Sleep and the cerebral circulation. In: Physiology in sleep. Orem J, Barnes CD, eds. New York: Academic Press, 1980: pp. 57-95.

27. Hampson NB, Camporesi EM, Stolp BW, Moon RE, Shook JE, Griebel JA, Piantadosi CA. Cerebral oxygen availability by NIRS spectroscopy during transient hypoxia in humans. J Appl Physiol 1990; 69(3): 907-13.

28. Harada M, Sawa T, Okuda C, Matsuda T, Tanaka Y. Effects of glucose load on brain extracellular lactate concentration in conscious rats using a microdyalysis technique. Horm Metab Res 1993; 25: 560-3.

29. Harik SI, Behmand RA, LaManna JC. Hypoxia increases glucose transport at blood-brain barrier in rats. J Appl Physiol 1994; 77: 896-901.

30. Harik SI, Lust WD, Jones SC, Lauro KL, Pundik S, LaManna JC. Brain glucose metabolism in hypobaric hypoxia. J Appl Physiol 1995; 79: 136-40.

31. Heekeren HR, Kohl M, Obrig H, Wenzel R, von Pannwitz W, Matcher SJ, Dirnagl U, Cooper CE, Villringer A. Noninvasive assessment of changes in cytochrome-c oxidase oxidation in human subjects during visual stimulation. J Cereb Blood Flow Metab 1999; 19: 592-603.

32. Hein H, Krieglstein J, Stock R. The effects of increased glucose supply and thiopental anesthesia on energy metabolism of the isolated perfused rat brain. Naunyn Schmiedebergs Arch Pharmacol 1975; 289: 399-407.

33. Honig CR, Gayeski TEJ, Federspiel W, Clark A, Clark P. Muscle O2 gradients from hemoglobine to cytochrome: new concepts, new complexity. Adv Exp Med Biol 1984; 169: 23-38.

34. Hoshi Y, Hazeki O, Kakihana Y, Tamura M. Redox behavior of cytochrome oxidase in the rat brain measured by near-infrared spectroscopy. J Appl Physiol 1997; 83(6): 1842-8.

35. Hoshi Y, Mizukami S, Tamura M. Dynamic features of hemodynamic and metabolic changes in the human brain during all night sleep as revealed by near-infrared spectroscopy. Brain Res 1994; 652: 257-62.

36. Hoshi Y, Tamura M. Detection of dynamic changes in cerebral oxygenation coupled to neuronal function during mental work in man. Neurosci Lett 1993; 150: 5-8.

37. Hudetz AG. Blood flow in the capillary network: a review emphasizing observations with intravital microscopy. Microcirculation 1997; 4: 233-52.

38. Huertas J, Mc Millin JK. Paradoxical sleep: effect of low partial pressures of atmospheric oxygen. Science 1968; 159: 745-6.

39. Hyder F, Chaese JR, Behar KL, Mason GF, Siddeek M, Rothman DL, Shulman RG. Increased tricarboxylic acid cycle flux in rat during forepaw stimulation detected by ¹H-[¹³C]NMR spectroscopy. Proc Natl Acad Sci USA 1996; 93: 7612-7.

40. Hyder F, Rothman DL, Mason GF, Rangarajan A, Behar KL, Shulman RG. Oxidative glucose metabolism in rat brain during single forepaw stimulation: a spatially localized 1H13C nuclear magnetic resonance study. J Cereb Blood Flow Metab 1997; 17: 1040-7.

41. Jones DP. Intracellular diffusion gradients of O2 and ATP. Am J Physiol 1986; 250: C663-75.

42. Kato T, Kamei A, Takashima S, Ozaki T. Human visual cortical function during photic stimulation monitoring by means of near-infrared spectroscopy. J Cereb Blood Flow Metab 1993; 13: 516-20.

43. Katsura K, Folbergrova J, Gido G, Siesjo BK. Functional, metabolic, and circulatory changes associated with seizure activity in the postischemic brain. J Neurochem 1994; 62: 1511-5.

44. Kikano GE, La Manna JC, Harik SI. Brain perfusion in acute and chronic hyperglycemia in rats. Stroke 1989; 20: 1027-31.

45. Klein B, Kuschinsky W, Schroeck H, Vetterlein F. Interdependency of local capillary density, blood flow, and metabolism in rat brains. Am J Physiol 1986; 251: H1333-40.

46. Knudsen GM, Goebel U, Paulson OB, Kuschinsky W. Regional density of perfused capillaries and cerebral blood flow in untreated short term and long term streptozotocin diabetes. J Cereb Blood Flow Metab 1991; 11: 361-5.

47. Knudsen GM, Tedeschi E, Jakobsen J. The influence of hematocrit and blood glucose on cerebral blood flow in normal and diabetic rats. Neuroreport 1992; 3: 987-9.

48. Kozniewska E, Weller L, Hoper J, Harrison DK, Kessler M. Cerebrocortical microcirculation in different stages of hypoxic hypoxia. J Cereb Blood Flow Metab 1987; 7: 464-70.

49. Krueger G, Kleinschmidt A, Frahm J. Dynamic MRI sensitized to cerebral blood oxygenation and flow during sustained activation of human visual cortex. Magn Reson Med 1996; 35: 797-800.

50. Kuwabara H, Ohta S, Brust P, Meyer E, Gjedde A. Density of perfused capillaries in living human brain during functional activation. Prog Brain Res 1992; 91: 209-15.

51. Kwong KK, Belliveau JW, Chesler DA, Goldberg IE, Weisskoff RM, Poncelet BP, Kennedy DN, Hoppel BE, Cohen MS, Turner R, Cheng H-M, Brady TJ, Rosen BR. Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc Natl Acad Sci USA 1992; 89: 5675-9.

52. LaManna JC, Vendel LM, Farrel R. Brain adaptation to chronic hypobaric hypoxia in rats. J Appl Physiol 1992; 72: 2238-43.

53. Laszy J, Sarkadi A. Hypoxia-induced sleep disturbance in rat. Sleep 1990; 13: 205-17.

54. Lenzi P, Zoccoli G, Walker AM, Franzini C. Cerebral blood flow regulation in REM Sleep: a model for flow-metabolism coupling. Arch Ital Biol 1999; 137: 165-79.

55. Luks AM, van Melick H, Batarse RR, Powell FR, Grant I, West JB. Room oxygen enrichment improves sleep and subsequent day-time performance at high altitude. Resp Physiol 1998; 113: 247-58.

56. Madsen PL, Cruz NF, Sokoloff L, Dienel GA. Cerebral Oxygen/Glucose ratio is low during sensory stimulation and rises above normal during recovery: excess glucose consumption during stimulation is not accounted for by lactate efflux from or accumulation in brain tissue. J Cereb Blood Flow Metab 1999; 19: 393-400.

57. Madsen PL, Hasselbach SG, Hageman LP, Skovgaard Olsen K, Bulow J, Holm S, Wildschiodtz G, Paulson OB, Lassen NA. Persistent resetting of the cerebral oxygen/glucose uptake ratio by brain activation: evidence obtained with the Kety-Schmidt technique. J Cereb Blood Flow Metab 1995; 15: 485-91.

58. Madsen PL, Linde R, Hasselbalch SG, Paulson OB, Lassen NA. Activation-induced resetting of cerebral oxygen and glucose uptake in the rat. J Cereb Blood Flow Metab 1998; 18: 742-8.

59. Madsen PL, Schmidt JF, Wildschiodtz G, Friberg L, Holm S, Vorstrup S, Lassen NA. Cerebral O2 metabolism and cerebral blood flow in humans during deep and rapid-eye-movement sleep. J Appl Physiol 1991; 70: 2597-601.

60. Madsen PL, Vorstrup S. Cerebral blood flow and metabolism during sleep. Cerebrovasc Brain Metab Rev 1991; 3: 281-96.

61. Magistretti PJ, Pellerin L. Cellular bases of brain metabolism and their relevance to functional imaging: evidence for a prominent role of astrocytes. Cerebral Cortex1996; 6: 50-61.

62. Malonek D, Grinvald A. Interactions between electrical activity and cortical microcirculation revealed by imaging spectroscopy: implications for functional brain mapping. Science 1996; 272: 551-4.

63. Maquet P. Contribution a l'etude du sommeil human normal et pathologique. Etude en tomographie a emission de positron. Thèse d'agrégé de l'Enseignement Supèrieur. Université de Liège, 1997: 1-38.

64. Maquet P. Sleep function(s) and cerebral metabolism. Behav Brain Res 1995; 69: 75-83.

65. Menon RS, Ogawa S, Hu X, Strupp JP, Anderson P, Ugurbil K. BOLD based functional MRI at 4 Tesla includes a capillary bed contribution: echo-planar imaging correlates with previous optical imaging using intrinsic signals. Magn Reson Med 1995; 33: 453-9.

66. Meyer JS, Toyoda M. Studies of rapid changes in cerebral circulation and metabolism during arousal and rapid eye movement sleep in human subjects with cerebrovascular disease. In: Cerebral circulation and stroke. Zulch KJ, ed. Berlin: Springer, 1971: pp. 159-63.

67. Mies G, Niebuhr I, Hossman KA. Simultaneous measurement of blood flow and glucose metabolism by autoradiographic techniques. Stroke 1981; 12: 581-8.

68. Miller AL, Shamban AT, Corddry DH, Kiney CA. Cerebral metabolic responses to electroconvulsive shock and their modification by hypercapnia. J Neurochem 1982; 38: 916-24.

69. Mizuno K, Asano K, Okudaira N. Sleep and respiration under acute hypobaric hypoxia. Japan J Physiol 1993; 43: 161-75.

70. Munger DM, Bucher HU, Duc G. Sleep state changes associated with cerebral blood volume changes in healthy term newborn infants. Early Hum Dev 1998; 52: 27-42.

71. Ogawa M, Magata Y, Ouchi Y, Fukuyama H, Yamaguchi H, Kimura J, Yonekura Y, Konishi J. Scopolamine abolishes cerebral blood flow response to somatosensory stimulation in anasthetized cats: PET study. Brain Res 1994; 650: 249-52.

72. Ohta S, Reutens DC, Gjedde A. Brief vibrotactile stimulation does not increase cortical oxygen consumption when measured by single inahalation of positron emitting oxygen. J Cereb Blood Flow Metab 1999; 19: 260-5.

73. Onoe H, Watanabe V, Tamura M, Hayaishi O. REM-sleep associated hemoglobin oxygenation in the monkey forebrain studied using near-infrared spectrophotometry. Neurosci Lett 1991; 129: 209-13.

74. Padnick LB, Linsenmeier RA, Goldstick TK. Oxygenation of the cat primary visual cortex. J Appl Physiol 1999; 86: 1490-6.

75. Pappenheimer JR. Sleep and respiration of rats during hypoxia. J Physiol 1977; 266: 191-207.

76. Parisi RA, Neubauer JA, Frank MM, Santiago TV, Edelman NH. Linkage between brain blood flow and respiratory drive during rapid eye movement sleep. J Appl Physiol 1988; 64: 1457-65.

77. Parmeggiani PL. Physiological risks during sleep. In: Sleep and health risk. Peter JH, Penzel T, Podzus T, von Wichert P, eds. Berlin, Heidelberg: Springer-Verlag, 1991: pp. 119-23.

78. Parmeggiani PL, Morrison AR. Alterations in autonomic functions during sleep. In: Central regulation of autonomic functions. Loewy AD, Spyer KM, eds. Oxford: Oxford University Press, 1990: pp. 367-86.

79. Pellerin L, Magistretti PJ. Glutamate uptake into astrocytes stimulates aerobic glycolisis: A mechanism coupling neuronal activity to glucose utilization. Proc Natl Acad Sci 1994; 91: 1065-9.

80. Poitry-Yamate CL, Poitry S, Tsacopoulos M. Lactate released by Muller glial cells is metabolized by photoreceptors from mammalian retina. J Neurosci 1995; 15: 5179-91.

81. Prichard J, Rothman D, Novotny E, Petroff O, Kuwabara T, Avison M, Hanstock C, Shulman RG. Lactate rise detected by 1H NMR in human visual cortex during physiologic stimulation. Proc Natl Acad Sci USA 1991; 88: 5829-31.

82. Quaresima V, Springett R, Cope M, Wyatt JT, Delpy DT, Ferrari M, Cooper CE. Oxidation and reduction of cytochrome oxidase in the neonatal brain observed by in vivo near-infrared spectroscopy. Biochem Biophys Acta 1998; 1366: 291-300.

83. Roland PE. Brain activation. New York: John Wiley & Sons, 1993.

84. Santiago TV, Guerra E, Neubauer JA, Edelman NH. Correlation between ventilation and brain blood flow during sleep. J Clin Invest 1984; 73: 497-506.

85. Sappey-Marinier D, Calabrese G, Fein G, Hugg JW, Biggins C, Weiner MW. Effect of photic stimulation on human visual cortex lactate and phosphatase using 1H and 31P magnetic resonance spectroscopy. J Cereb Blood Flow Metab 1992; 12: 584-92.

86. Schumacker PT, Samsel RW. Analysis of oxygen delivery and uptake relationships in the Krogh tissue model. J Appl Physiol 1989; 67: 1234-44.

87. Shinozuka T, Nemoto EM, Winter PM. Mechanisms of cerebrovascular O2 sensitivity from hyperoxia to moderate hypoxia in the rat. J Cereb Blood Flow Metab 1989; 9: 187-95.

88. Shiotsuka S, Atsumi Y, Ogata S, Yamamoto R, Igawa M, Takahashi K, Hirasawa H, Koyama K, Maki A, Yamashita Y, Koizumi H, Toru M. Cerebral blood volume in the sleep measured by near-infrared spectroscopy. Psychiatry Clin Neurosci 1998; 52: 172-3.

89. Steriade M, Hobson JA. Neuronal activity during the sleep-waking cycle. Prog Neurobiol 1976; 6: 155-376.

90. Thomas M, Sherwin RS, Murphy J, Kerr D. Importance of cerebral blood flow to the recognition of and physiological responses to hypoglycemia. Diabetes 1997; 46: 829-33.

91. Tuhin KR, Popel SA. Theoretical predictions of end-capillary PO2 in muscles of athletic and nonathletic animals at VO2MAX. Am J Physiol 1996; 1: H721-37.

92. Tuor UI, Kurpita G, Simone C. Correlation of local changes in cerebral blood flow, capillary density and cytochrome oxidase during development. J Comp Neurol 1994; 342: 439-48.

93. Turner R, Grinvald A. Direct visualization of patterns of deoxygenation and reoxygenation in monkey cortical vasculature during functional brain activation. Proc Soc Magn Res 1994; 1: 430.

94. Turner R, Jezzard P, Wen H, Kwong KK, Lebihan D, Zeffiro T, Balaban RS. Functional mapping of the human visual cortex at 4 and 1-5 Telsa using the oxygenation contrast EPI. Magn Res Med 1993; 1: 277-9.

95. Ueki M, Linn F, Hossman KA. Functional activation of cerebral blood flow and metabolism before and after global ischemia of rat brain. J Cereb Blood Flow Metab 1988; 8: 486-94.

96. Van den Berg C. On the relation between energy transformations in the brain and mental activities. In: Energetics and human information processing. Hocky GRJ, Gaillard AWK, Coles MGH, eds. Boston: Nijhoff, 1986: pp. 131-5.

97. Villringer A, Dirnagl U. Coupling of brain activity and cerebral blood flow: basis of functional neuroimaging. Cerebrovasc Brain Metab Rev 1995; 7: 240-76.

98. Villringer A, Them A, Lindauer U, Einhaupl K, Dirnagl U. Capillary perfusion of the rat brain cortex. An in vivo confocal microscopy study. Circ Res 1994; 75: 55-62.

99. Walker AM, Grant DA, Wild J. Cerebral circulation in sleep: responses to repeated and continuous hypoxia. Sleep Research Online 1999; 2(1): 119.

100. West JB. Oxygen enrichment of room air to relieve the hypoxia of high altitude. Respir Physiol 1995; 99: 225-32.

101. West JB. Respiratory Physiology - the essentials. Baltimore: Williams & Wilkins, 1990: pp. 84.

102. Wickramasinghe YA, Rolfe P, Palmer K, Spencer SA. Investigation of neonatal brain cytochrome redox by NIRS. Dev Brain Res 1995; 89(2): 307-8.

103. Wong-Riley MT. Cytochrome oxidase: an endogenous metabolic marker for neuronal activity. Trends Neurosci 1989; 12: 94-104.

104. Zoccoli G, Bach V, Cianci T, Lenzi P, Franzini C. Brain blood flow and extracerebral carotid circulation during sleep in rat. Brain Res 1994; 641: 46-50.

105. Zoccoli G, Lucchi ML, Andreoli E, Bach V, Cianci T, Lenzi P, Franzini C. Brain capillary perfusion during sleep. J Cereb Blood Flow Metab 1996; 16: 1312-8.

FIGURES

Figure 1.

Scattered Values of Brain Tissue PO₂ at Different Arterial PO₂. PO₂ values extend from arterial PO₂ to 0 mmHg. The scatter decreases with increasing arterial PO₂. This indicates that blood tissue O₂ gradient affects tissue PO₂ values, as predicted by the hypothesis of O₂ diffusion limitation. Modified from Shinozuka et al., 1989

Figure 2.

Tissue PO₂ Between Adjacent Capillaries. PO₂ falls to a minimum which deepens with increasing distance and metabolic rate (I, II and III: adequate, critical and inadequate O₂ delivery for aerobic metabolism, respectively), even if the average capillary PO₂ increases by the effect of BF increase. Modified from West, 1990.

TABLES

This article does not contain any tables.

ACKNOWLEDGMENTS

This work was supported by grants from Ministero dell'Universitá e della Ricerca Scientifica and Consiglio Nazionale delle Ricerche.