Locus coeruleus input affects glucose metabolism in activated rat barrel cortex

Brain Research Bulletin,

Vol.

19, pp. 495-499. B Pergamon Journals

0361-9230/87$3.00 + .OO

Ltd., 1987.Printed in the U.S.A.

Locus Coeruleus Input Affects Glucose Metabolism in Activated Rat Barrel Cortex REBECCA

L. CRAIK,

PETER

J. HAND

AND

BARRY

E. LEVIN’

Department of Physical Therapy, Beaver College, Glenside, PA 19038 of Animal Biology, School of Veterinary Medicine and Department of Neurology School of Medicine, University of Pennsylvania, Philadelphia, PA 19104 Neurology Service, Veterans Administration Medical Center, E. Orange, NJ 07019 and Department of Neurosciences, New Jersey Medical School, Newark, NJ 07103

Department

Received

22 June

1987

CRAIK, R. L., P. J. HAND AND B. E. LEVIN. Locus coeruleus input affects glucose metabolism in activated rat barrel BRAIN RES BULL 19(4) 495-499, 1987.-Significant depletion (>SO%) of neocortical norepinephrine (NE), 2 weeks after unilateral 6-hydroxydopamine lesions of the locus coeruleus, led to a small (8%), ipsilateral decrease in total C3

cortex.

vibrissa column 14C-2-deoxyglucose uptake, but a larger (2432%) increase in the areal extent of this uptake into the metabolic representation of both the C3 column and barrel of the rat somatosensory (SmI) cortex during stimulation of the contralateral C3 facial vibrissae. This suggests a predominantly inhibitory role for NE in modulating SmI oxidative metabolism during physiologic stimulation. Norepinephrine

2-Deoxyglucose

Somatosensory cortex (SmI)

THE locus coeruleus (LC) contains noradrenergic neurons which project widely throughout the central nervous system [2] but the specific role of the LC in cerebral function remains unclear. Norepinephrine (NE) liberated from LC axons appears to modulate neuronal activity in various neocortical areas [12,18] and may affect blood-brain barrier permeability [7,19]. No convincing effect of LC activity upon neocortical blood flow [lo] has been demonstrated and there is disagreement about whether the LC has a physiologic effect upon neocortical metabolism [ 1, 4, 13,201. The barrel subfield of the rodent somatosensory cortex (SmI) receives extensive input from the LC [ 161. The noradrenergic axons from the LC lie within the centers of the individual barrels where they overlap with thalamo-cortical afferents carrying signals from individual, contralateral facial vibrissae via trigeminal afferents [ 1,231. The cytoarchitectural barrels lie primarily in layer IV of the SmI cortex [25] and stimulation of an individual facial vibrissa leads to increased glucose utilization in the corresponding layer IV barrel and an associated spindle-shaped column extending from cortical laminae I-Via [3-51. We have used this precise and readily identifiable sensory system to examine the effect of depletion of cortical NE upon the metabolic response to a physiologically relevant stimulus in the awake rat.

Barrel

METHOD

Animals and Surgical Procedures Eleven male Sprague-Dawley rats (200-300 g) were injected with pargyline (50 mg/kg, IP) 15-30 min prior to surgery and then anesthetized with Chloropent (Fort Dodge Labs). Stereotaxic injections with 6-hydroxydopamine HBr (6-OHDA; 6 pg free base in 3 ~1 0.9% NaCl containing 1 mg/ml ascorbic acid, pH 5.5) were made into either the left or right LC (1.1 mm posterior, 1.3 mm lateral to the intra-aural line and 7.5 mm down from cranial surface, jaw bar down 2.5 mm) [6]. Six additional rats served as uninjected controls. Estimation of Local Cerebral Glucose Utilization (LCGU) Two to 3 weeks following LC lesions, the rats were anesthetized with halothane and all facial vibrissae were trimmed to their bases except for the C3 vibrissae bilaterally. After a 3-4 hr recovery period, bilateral C3 vibrissal stimulation was begun with a mechanical device at 4-5 Hz [5] 3 min prior to, and continued for 45 min following an intravenous pulse of 30 &i of 14C-2-deoxy-d-glucose (2-dg; New England Nuclear; 49.2 mCi/mmol) in 0.5 ml of 0.9% NaCl. A baseline arterial sample (0.2 ml) was drawn for serum glu-

‘Requests for reprints should be addressed to Barry E. Levin, M.D., Neurology Service (127), VA Medical Center, E. Orange, NJ 07019.

495

496

CRAIK, HAND AND LEVIN

FIG. 1. Autoradiogram showing the uptake of ‘“C-2-deoxyglucose into the C3 vibrissa barrels and adjacent columns in the SmI of a rat receiving bilateral C3 vibrissal stimulation at 4-5 Hz for 45 min following administ~tion of W-2-deoxyglucose. The boundaries of the C3 metabolic barrel in Iayer IV and the adjacent metabolic column are less clearly defined and the area1 extent LCGU is greater on the side ipsilateral to an LC lesion yielding 66% cortical NE depletion (left side; open arrow) than contralateral, intact side (right side; closed arrow).

case (automated glucose oxidase method; Beckman) and ‘F-Z-dg determination (liquid scintillation counting). Six additional 0.2 ml arterial samples for glucose and IF activity were drawn at prescribed intervals over the 4.5 min following 2-dg injection [22]. Animals were killed with an overdose of sodium pentobarbital, the brains quickly removed, and 20 mg samples of frontal cortex removed from each side for determination of NE content by high performance liquid chromato~~hy 1151. Brains were then rapidly frozen in liquid Freon XII. Subsequent 20 ym coronal sections of these brains, along with 14C standards, were apposed to Kodak SB-5 x-ray film for 7-9 days and developed by standard techniques. Resulting autoradiograms were read on 2 different occasions by 2 different observers who had no knowledge of the origin of the samples, using computer-assisted micr~ensitomet~ (Drexel University Image Analysis System). A pseudo-color image of a given autoradiogram was generated by cotorcoding the range of gray values within thresholds determined from computer-assisted histograms of the original image. Increased LCGU in the metabolically activated C3 barrel (cortical lamina II&-IV) and column (lamina I-Via) were then identified in this image by selecting the color co~es~nding to the highest density readings as compared to colors representing lower densities in the surrounding cortex (approximately 15% difference). The area1 extent of the metabolic uptake in these structures was computed by outlining the

cortex lateral of the in the

margins of the barrel and column on the computer-generated image and computing the area within these margins based on the number of pixels (precaIibrated in mm?. LCGU was calculated by computing the integrated density of the autoradiographic image within these boundaries using the method of Sokoloff et al. [22]. Readings were taken in all of the sections containing the metabolically activated C3 barrels and columns from both sides of the SmI cortex. Since approximately 1-3 sections through the area of the activated barrel were lost for any given brain, areas rather than volumes were used. This method of assessing barrel and column margins gave a 1% inter- and intra-observer variability. Statistics

Mean values of the 5 largest areas from each C3 barrel and column were taken as representative of the maximum extent of those areas on each side of the brain. These data were compared between sides in individual animals by paired t-test and between lesion and control animals by t-test for unpaired samples. Correlation between resultant area1 size or LCGU and cortical NE concentrations was made by Pearson’s correlation. RESULTS

LC lesions produced frontal cortex NE depletions ranging from 0.2 to 7% of levels in both the contralateral cortex,

NOREPINEPHRINE

AND BARREL CORTEX

497

r=O.f38

0

20

40

NE

80

60

(% of

100

Control)

FIG. 2. Plot of the norepinephrine (NE) concentration (as a percent of the intact control side) in the frontal cortex ipsilateral to a locus coeruleus (LC) lesion with 6hydroxydopamine made 2 weeks prior to testing versus the areal size of the metabolic representation of the C3 column in the SmI cortex. Data points are for individual animals with unilateral LC lesions (n= 11) where “column area” is the ratio of the areas of the C3 column on the side of the LC lesion (“L”) to that on the control (“C”) side (X 100). The solid line is the regression of the C3 column area upon the NE concentration as a percent of control where the correlation coefftcient (“r”)=0.88. The horizontal, dotted line represents a Lesion/Control ratio of 1.0; the data point to the right of this line represents the mean value for control rats (n=6) &SEM (vertical and horizontal bars).

TABLE

1

EFFECT OF NEOCORTICAL NE DEPLETION ON SmI C3 VIBRISSA COLUMN AND BARREL LOCAL CEREBRAL GLUCOSE UTILIZATION (LCGU)

Column Area (mm*)

Barrel

LCGU (I*.mol/lOOg/min)

Intact Controls Left Right

1.141 t 0.034 1.109 f 0.030

109 110

NE Depletion Intact LC Lesion

1.109 * 0.074 1.374 2 o.o51*t

103 *5 93.9 * 4.2*t

k.5 24

Area (mm”)

LCGU (PmoUlOO g/min)

0.373 * 0.020 0.375 f 0.025

114. * 9 113 ? 6

0.351 ? 0.021 0.438 2 0.24*t

115 2 6 108 ? 6

Values are mean f SEM for metabolic areas and absolute rates of LCGU for intact controls (n=6) and rats with unilateral 6-hydroxydopamine LC lesions producing >50% depletion of ipsilateral cortical NE (n=8). *=p
ipsilateral to an intact LC and to intact control (209+ 17 r&g). In the intact SmI cortex, the metabolic

rats

column activated by contralateral C3 vibrissal stimulation was discrete, spindle shaped and extended from lamina I to Via (Fig. 1). Eight of the animls had >50% NE depletion [46.2+ 14.0 rig/g (meanaSEM); range 0.4-94.1 rig/g]] as compared to the contralateral cortex with intact LC input. As compared to those in the intact cortex, C3 vibrissa columns ipsilateral to such effective LC lesions were larger

and had less distinct margins (Fig. 1). Overall, their mean areas were 24% larger than metabolically activated C3 columns in contralateral SmI cortex with intact LC innervation and 22% larger than those in brains of intact rats (Table 1). Overall, there was a highly significant, negative correlation between the areal extent of LCGU in the metabolic representation of the C3 column ipsilateral to a LC lesion and cortical NE content given as a percent of the contralateral side without LC lesions (Fig. 2; r=0.88, p=O.OOl). Specifically, in

498

CRAIK, HAND AND LEVIN

those rats with >50% cortical NE depletion, the ratio of the area1 extent of LCGU in the C3 column ipsilateral to the LC lesion to that on the contralateral side ranged from 1.05 to 1.42 and there was a significant, negative correlation (r=0.70; pO. 1). Nor was there an effect of NE depletion on overall C3 barrel LCGU.

DISCUSSION

The major effect of NE released from LC axon terminus in the barrel subfield of the rat SmI cortex appears to be a suppression of the radial spread of metabolic activation resulting from contralateral vibrissal stimulation. In the absence of NE, enhanced metabolic activation occurred well outside the normal boundaries of the C3 barrel and column. An interesting feature of these findings was the fact that column but not barrel areas correlated with the degree of NE depletion on the side of a LC lesion. Thus there was a graded response to NE depletion within the column and an “all or none” response seen in the barrel. Since the primary input from ventrobasal thalamus is to the barrel neurons in cortical layer IV [21], these differing responses may be a reflection of such structural and physiological differences. This may also explain why there was decreased LCGU associated with the enlarged column but not barrel. There are also differences in the distribution of noradrenergic receptors in the barrel and remaining column which may explain the contrasting effects of NE depletion. Barrels [24] (as well as cerebral microvessels [9]) contain predominantly &adrenoreceptors. By contrast, P,-receptors predominate in the remainder of the SmI cortical laminae [24] and other neocortical areas [ 171. Although it is not clear what subtype of receptor is responsible, noradrenergic activation of P-adrenoreceptors is known to inhibit neuronal firing rates in some cortical regions [12,18] and LC stimulation decreases LCGU specificaily in parietal cortex [ 11. Our studies have not addressed the issue of the mechanism by which NE exerts its effect on barrel metabolism. NE may directly modulate cellular metabolism in neurons and glia of the activated cortex f&13] or it may modulate barrel metabolism by a primary effect on

cerebral blood flow, capillary recruitment, the rate of glucose transport across the blood-brain barrier relative to blood flow (extraction ratio) or by general changes in bloodbrain barrier permeability. However, it has been difficult to show a direct effect of NE released from endogenous LC axons upon cerebral blood flow [10,19]. Also, LCGU and cerebral blood flow are tightly coupled in the rat SmI barrels and columns during vibrissal stimulation so that glucose utilization and not cerebral blood flow would be expected to be the limiting factor were any uncoupling to occur [3,4]. It is also unlikely that changes in blood-brain barrier permeability can explain our results. Although cortical NE depletion can affect blood-brain barrier function, this effect has only been described for water permeability [19] or during major disturbances of overall cerebral metabolism such as occur during status epilepticus associated with hypertension [7]. Therefore, under physiologic conditions, NE appears to exert a direct inhibitory effect upon the spread of metabolic activity beyond the normal limits of the lamina IV barrels and their associated columns in laminae I, II, III and Via of the SmI cortex. Although the number of neocortical /I-adrenoreceptors and isoproterenol-stimulated adenylate cyclase activity increase by SO-70% at 2 weeks after LC lesions which deplete ipsilateral cortical NE levels by more than SOLTO [6,14], denervation supersensitivity of adrenoreceptors on cells surrounding the physiologic C3 barrel and column probably was not responsible for our results. The largest metabolic column areas were seen in rats with the most profound NE depletions (98%-9%) where few noradrenergic terminals would remain to release sufficient NE for activation of even supersensitive receptors. Therefore, a deficiency of NE rather than an excess of noradrenergic function secondary to receptor supersensitivity probably accounted for our findings. Several questions about NE depletion and neocortical glucose utilization have not been addressed here: (1) did the changes in LCGU in the C3 vibrissa barrels and columns occur in the neurons or glia of the SmI cortex: (2) what effect, if any, did noradrenergic denervation of subcortical vibrissa-barrel afferent pathways have upon the observed changes in barrel subfield activation to vibrissal stimulation; (3) were the changes in barrel field metabolism due to a global effect of NE, depletion upon cerebral metabolism or does NE released from LC axons specifically affect cortical sensory processing by modulating afferent signals relaying information about the frequency or directionality of vibrissal deflection [ 11,211. What has been shown is that NE can play a significant role in modulating the metabolic activity of a discrete area of neocortex in response to a physiological stimulus.

This research was funded by the Research Service of the Veterans Administration. We thank Antoinette Colitti for manuscript preparation.

NOREPINEPHRINE

AND BARREL

CORTEX

499

REFERENCES 1. Abraham, W. C., R. L. Delanoy, A. J. Dunn and S. F. Zornitzer. Locus coeruleus stimulation decreases deoxyglucose uptake in ipsilateral mouse cerebral cortex. Bruin Res 172: 387392, 1979. 2. Anden, N. E., A. Dahlstrom, K. Fuxe, K. Larsson, L. Olson and U. Ungerstedt. Ascending monoamine neurons to the teleencephalon and diencephalon. Acta Physiol Stand 67: 313-326, 1966. 3. Ginsberg,

M. D., W. D. Dietrich and R. Busto. Coupled forebrain increases of local cerebral glucose utilization and blood flow during physiologic stimulation of a somatosensory pathway in the rat: Demonstration by double-lable autoradiography. Neurology 37: 11-19, 1987. 4. Greenberg, J., P. J. Hand, A. Sylvestro and M. Reivich. Localized metabolic-flow couple during functional activation. Acta Nemo1 Stand 60: Suppl72, 12-13, 1979. 5. Hand, P. J. The 2-deoxyglucose method. In: Neuroanatomical Tract-Tracing Methods, edited by L. Heimer and M. J. Robards. New York: Plenum Press, 1981, pp. 511-538. 6. Harik, S. I., R. B. Duckrow, J. C. LaManna, M. Rosenthal, V. K. Sharma and S. P. Banerjee. Cerebral compensation for chronic noradrenergic denervation induced by locus coeruleus lesion: Recovery of receptor binding, isoproterenol-induced adenylate cyclase activity, and oxidative metabolism. J Neurosci 1: 641-649, 1981. 7. Harik, S. I. and T. McGunigal, Jr. The protective influence of the locus ceruleus on the blood-brain barrier. Ann Nemo1 15: 568-574, 1984. 8. Harik, S. I., J. C. LaManna, A. I. Light and M. Rosenthal. Cerebral norepinephrine: influence on cortical oxidative metabolism in situ. Science 206: 6971, 1979. 9. Harik, S. I., V. K. Sharma, J. R. Wetherbee, R. H. Warren and S. P. Banerjee. Adrenergic receptors of cerebral microvessels. Eur J Pharmacol 61: 207-208, 1980. 10. Heistad, D. D. Summary of symposium on cerebral blood flow: effect of nerves and neurotransmitters. J Cerebral Blood Flow Metab 1: 447-450, 1981. 11. Ito, M. Some quantitative aspects of vibrissa-driven neuronal responses in rat neocortex. J Neurophysiol 46: 705-715, 1981. 12. Katayama, Y., Y. Ueno, T. Tsukiyama and T. Tsubokawa. Long lasting suppression of tiring of cortical neurons and decrease in cortical blood flow following train pulse stimulation of the locus coeruleus in the cat. Bruin Res 21: 173-179, 1981. 13. LaManna, J. C., S. I. Harik, A. I. Light and M. Rosenthal. Norepinephrine depletion alters cerebral oxidative metabolism in the ‘active’ state. Brain Res 204: 87-101, 1981.

14. Levin, B. E., W. P. Battisti and M. Murray. Axonal transport of P-receptors during the response to axonal injury and repair in locus coeruleus neurons. Brain Res 359: 215-223, 1985. 15. Levin, B. E. and A. Beigon. Reset-pine and the role of axonal transport in the independent regulation of pre- and postsynaptic P-adrenoreceptors. Brain Res 311: 39-50, 1984. 16. Lidov, H. G. W., F. L. Rice and M. E. Molliver. The organization of the catecholamine innervation of somatosensory cortex: the barrel field of the mouse. Brain Res 153: 577-584, 1978. 17. Minneman, K. P., A. Hedberg and P. B. Molinoff. Comparison of beta-adrenergic subtypes in mammalian tissues. J Pharmacol Exp Ther 211: 502-508, 1979. 18. Olpe, H.-R., A. Glatt, J. Laszlo and A. Schellenberg.

electrophysiological tical, noradrenergic

Some and pharmacological properties of the corprojection of the locus coeruleus in the rat.

Bruin Res 186: 9-19, 1980. 19. Raichle, M. E., B. K. Harman, J. 0. Eichling and L. G. Sharpe.

Central noradrenergic regulation of cerebral blood flow and vascular permeability. Proc Nat1 Acad Sci USA 72: 3726-3730, 1975. 20. Schwartz,

W. J. 6-Hydroxydopamine lesions of rat locus coeruleus alter brain glucose consumption, as measured by the 2-deoxy-d-[14C]glucose tracer technique. Neurosci Lett 7: 141150, 1978. 21. Simons, D. J. Multiwhisker stimulation and its effects on vibrissa units in rat SmI barrel cortex. Brain Res 276: 178-182, 1983. 22. Sokoloff, L., M. Reivich, C. Kennedy,

M. H. DesRosiers, C. S. Patlak, 0. Pettigrew, 0. Sakurada and M. Shinohara. The [‘4C]deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure, and normal values in the conscious and anesthetized albino rat. J Neurochem 23:

897-916, 1977. 23. Van der Loos, H. Barreloids in the mouse somatosensory thalamus. Neurosci Lett 2: 1-6, 1976. 24. Vos, P., D. Kaufman, E. Mansfield, B. B. Wolfe and P. J.

Hand. &adrenergic receptors are specifically associated with the “whisker barrels” in the somatosensory cortex of the rat. Sot Neurosci Abstr 11: 666, 1985. 25. Woolsey, T. A. and H. Van der Loos. The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. Brain Res 17: 205-242, 1970.