633 Alterations in Local Cerebral Glucose Utilization during Hemorrhagic Hypotension in the Rat Helen E. Savaki, Helen Macpherson, and James McCulloch From the Wellcome Surgical Institute, University of Glasgow, Glasgow, United Kingdom SUMMARY. The alterations in local cerebral glucose utilization in 58 anatomically discrete regions which occur during a period of hemorrhagic hypotension have been investigated in conscious rats, using the quantitative autoradiographic 14 C-deoxyglucose technique. Hemorrhagic hypotension (mean arterial pressure reduced by approximately 50 mm Hg) effected significant increases in glucose utilization in eight areas of the central nervous system, namely, the nucleus of the tractus solitarius (glucose utilization increased by 38%), the dorsal motor nucleus of the vagus (by 36%), locus coeruleus (by 38%), lateral habenular nucleus (by 40%), periventricular nucleus of the hypothalamus (by 41%), paraventricular nucleus of the hypothalamus (by 97%), supraoptic nucleus (by 86%), and the interstitial nucleus of the stria terminalis (by 84%). In five of these eight areas (nucleus of the tractus solitarius, dorsal motor nucleus of the vagus, paraventricular and supraoptic nuclei, and the interstitial nucleus of the stria terminalis), a significant relationship could be demonstrated between the level of glucose utilization and mean arterial blood pressure. In the majority of the CNS regions examined (neocortex, hippocampus, thalamus, extrapyramidal and motor areas), hemorrhagic hypotension was without significant effect upon local cerebral glucose utilization. The results provide direct evidence of the functional involvement of specific brain areas of conscious rats (thus obviating complicating anesthetic influences) in the response of the CNS to hemorrhagic hypotension. (Circ Res 50: 633-644, 1982) THE neuroanatomical and functional organization of the mechanisms by which systemic blood pressure is regulated has been characterized in detail (Palkovits and Zaborszky, 1977). Within the medulla, the in- volvement of the medial nucleus of the solitary tract (NTS), the dorsal motor nucleus of the vagus (DM X ), the nucleus ambiguus, the paramedian reticular nu- cleus, and the cuneate nucleus, among others, have been demonstrated neuroanatomically or, more usu- ally, electrophysiologically (Palkovits and Zaborszky, 1977). Neuronal activity within the two magnocellular nuclei in the hypothalamus, the supraoptic (SO) and paraventricular (PAV) nuclei, which mediate the in- creased release of antidiuretic hormone during arterial hypotension (Clark and Silva, 1967), has been shown to be influenced by stimulation of both the carotid sinus and aortic depressor nerves (Calaresu and Cir- iello, 1979). In addition to the hypothalamus, a large number of supramedullary regions [including the lo- cus coeruleus, the cerebral cortex, and the interstitial nucleus of the stria terminalis (NIST) inter alia] have been reported to exert a modulatory influence upon cardiovascular control (see Korner, 1971; Palkovits and Zaborszky, 1977; Hilton and Spyer, 1980, for reviews). However, many previous investigations, particularly the elegant electrophysiological studies, do contain a number of well-recognized limitations, most notably distortions of normal functional activity resulting from the use of general anesthetics. Anes- thesia elevates the threshold for baroreceptor activa- tion, thus depressing synaptic transmission of baro- receptor inputs to the medulla (Miura and Reis, 1969). Supramedullary regions, which have been implicated in cardiovascular regulation, are much more suscep- tible to the action of anesthetics than are those in the medulla (Price, 1960; Peiss and Manning, 1964; Kor- ner et al., 1968). The autoradiographic 2-deoxyglucose technique (Sokoloff et al., 1977) has provided neuroscientists with a potent tool with which to investigate functional events within the central nervous system (CNS) of conscious animals (thus avoiding complicating anes- thetic influences). The conceptual basis for this novel investigative aproach is derived from two premises. First, the energy requirements of cerebral tissue are derived almost exclusively from the aerobic catabo- lism of glucose (Sokoloff et al., 1977). Second, func- tional activity within any region of the CNS is inti- mately and directly related to energy consumption within that region (Kennedy et al., 1975; Sokoloff, 1981). The 2-deoxyglucose technique already has been employed with success to provide new insight into CNS processes in a wide range of physiological and pharmacological manipulations (Sokoloff, 1981; McCulloch, 1982). In the present study, we have examined the alterations in local cerebral glucose utilization which occur in response to graded hem- orrhagic hypotension in conscious rats. As the auto- radiographic 2-deoxyglucose technique provides the simultaneous determination of glucose utilization in all neuroanatomically defined regions and nuclei within the CNS, the present study provides a com- by guest on July 5, 2018 http://circres.ahajournals.org/ Downloaded from
Transcript
633
Alterations in Local Cerebral Glucose Utilization duringHemorrhagic Hypotension in the Rat
Helen E. Savaki, Helen Macpherson, and James McCullochFrom the Wellcome Surgical Institute, University of Glasgow, Glasgow, United Kingdom
SUMMARY. The alterations in local cerebral glucose utilization in 58 anatomically discrete regionswhich occur during a period of hemorrhagic hypotension have been investigated in conscious rats,using the quantitative autoradiographic 14C-deoxyglucose technique. Hemorrhagic hypotension(mean arterial pressure reduced by approximately 50 mm Hg) effected significant increases inglucose utilization in eight areas of the central nervous system, namely, the nucleus of the tractussolitarius (glucose utilization increased by 38%), the dorsal motor nucleus of the vagus (by 36%),locus coeruleus (by 38%), lateral habenular nucleus (by 40%), periventricular nucleus of thehypothalamus (by 41%), paraventricular nucleus of the hypothalamus (by 97%), supraoptic nucleus(by 86%), and the interstitial nucleus of the stria terminalis (by 84%). In five of these eight areas(nucleus of the tractus solitarius, dorsal motor nucleus of the vagus, paraventricular and supraopticnuclei, and the interstitial nucleus of the stria terminalis), a significant relationship could bedemonstrated between the level of glucose utilization and mean arterial blood pressure. In themajority of the CNS regions examined (neocortex, hippocampus, thalamus, extrapyramidal andmotor areas), hemorrhagic hypotension was without significant effect upon local cerebral glucoseutilization. The results provide direct evidence of the functional involvement of specific brain areasof conscious rats (thus obviating complicating anesthetic influences) in the response of the CNS tohemorrhagic hypotension. (Circ Res 50: 633-644, 1982)
THE neuroanatomical and functional organization ofthe mechanisms by which systemic blood pressure isregulated has been characterized in detail (Palkovitsand Zaborszky, 1977). Within the medulla, the in-volvement of the medial nucleus of the solitary tract(NTS), the dorsal motor nucleus of the vagus (DMX),the nucleus ambiguus, the paramedian reticular nu-cleus, and the cuneate nucleus, among others, havebeen demonstrated neuroanatomically or, more usu-ally, electrophysiologically (Palkovits and Zaborszky,1977). Neuronal activity within the two magnocellularnuclei in the hypothalamus, the supraoptic (SO) andparaventricular (PAV) nuclei, which mediate the in-creased release of antidiuretic hormone during arterialhypotension (Clark and Silva, 1967), has been shownto be influenced by stimulation of both the carotidsinus and aortic depressor nerves (Calaresu and Cir-iello, 1979). In addition to the hypothalamus, a largenumber of supramedullary regions [including the lo-cus coeruleus, the cerebral cortex, and the interstitialnucleus of the stria terminalis (NIST) inter alia] havebeen reported to exert a modulatory influence uponcardiovascular control (see Korner, 1971; Palkovitsand Zaborszky, 1977; Hilton and Spyer, 1980, forreviews). However, many previous investigations,particularly the elegant electrophysiological studies,do contain a number of well-recognized limitations,most notably distortions of normal functional activityresulting from the use of general anesthetics. Anes-thesia elevates the threshold for baroreceptor activa-tion, thus depressing synaptic transmission of baro-
receptor inputs to the medulla (Miura and Reis, 1969).Supramedullary regions, which have been implicatedin cardiovascular regulation, are much more suscep-tible to the action of anesthetics than are those in themedulla (Price, 1960; Peiss and Manning, 1964; Kor-ner et al., 1968).
The autoradiographic 2-deoxyglucose technique(Sokoloff et al., 1977) has provided neuroscientistswith a potent tool with which to investigate functionalevents within the central nervous system (CNS) ofconscious animals (thus avoiding complicating anes-thetic influences). The conceptual basis for this novelinvestigative aproach is derived from two premises.First, the energy requirements of cerebral tissue arederived almost exclusively from the aerobic catabo-lism of glucose (Sokoloff et al., 1977). Second, func-tional activity within any region of the CNS is inti-mately and directly related to energy consumptionwithin that region (Kennedy et al., 1975; Sokoloff,1981). The 2-deoxyglucose technique already hasbeen employed with success to provide new insightinto CNS processes in a wide range of physiologicaland pharmacological manipulations (Sokoloff, 1981;McCulloch, 1982). In the present study, we haveexamined the alterations in local cerebral glucoseutilization which occur in response to graded hem-orrhagic hypotension in conscious rats. As the auto-radiographic 2-deoxyglucose technique provides thesimultaneous determination of glucose utilization inall neuroanatomically defined regions and nucleiwithin the CNS, the present study provides a com-
by guest on July 5, 2018http://circres.ahajournals.org/
prehensive analysis of the integrated functional alter-ations which occur in the brain in response to ahypotensive stimulus.
MethodsPreparation of the Animals
The experiments were performed on 15 male Sprague-Dawley rats weighing between 350 and 400 g. Prior to theexperiment, the animals were maintained under a controlledlighting and thermal environment, and were allowed freeaccess to food and water. On the day of the experiment,catheters were inserted into one femoral vein (to permit theadministration of the HC-deoxyglucose) and both femoralarteries (to allow the intermittent sampling of arterial bloodand the continuous monitoring of arterial blood pressure)during a brief period of light halothane anesthesia (1%halothane for approximately 30 minutes). The incision siteswere closed and infiltrated with local anesthetic. A loose-fitting abdominal-pelvic plaster cast was applied to theanimals for restraint, and the cast then taped to a supportinglead brick. The rats were allowed to recover from the effectsof anesthesia for at least 2 hours before any further manip-ulations were performed.
Hypotension was induced 50 minutes prior to the mea-surement of glucose utilization by controlled removal ofarterial blood, and was maintained at the same levelthroughout the measurement period. In four of the 15animals, insulin (0.2-0.4 unit) was administered intrave-nously 30 minutes prior to the measurement of glucoseutilization to maintain arterial plasma glucose within thelimits necessary for rigorous quantification of rates of glu-cose utilization (Savaki et al., 1980). Only those 15 animalsin which constant levels of arterial blood pressure andconstant arterial plasma glucose concentrations were main-tained throughout the measurement period, within the pre-determined limits (Savaki et al., 1980), were included in thisstudy. An additional six animals subjected to hemorrhagichypotension, which failed to meet these rigorous criteria,were excluded from the study, and no calculation of glucoseutilization was made from autoradiograms from these ani-mals. The pattern of altered glucose utilization in theseanimals, which is readily discernible from visual inspectionof the autoradiograms, was indistinguishable from that inanimals with similar levels of arterial blood pressure whichfulfilled the criteria necessary for quantification of the ratesof glucose utilization. Body temperature, hematocrit, bloodpressure and arterial pH, carbon dioxide and oxygen tensionwere monitored throughout the experimental procedure.
Measurement of Local Cerebral Glucose UtilizationA detailed description of the theory and operational
procedures for determining the rate of cerebral glucoseutilization with 14C-deoxyglucose has been published pre-viously (Sokoloff et al., 1977), and the present experimentswere conducted in a manner similar to that described. Themeasurement of glucose utilization was initiated by admin-istration of an intravenous pulse of 125 jnCi/kg 2-deoxy-D-[l-'4C]glucose. Fourteen timed arterial blood samples werethen drawn during the succeeding 45 minutes. The bloodsamples were immediately centrifuged and the plasma wasassayed for concentration of I4C (by liquid scintillationcounting) and of glucose (by automated enzymatic assayemploying glucose oxidase). Approximately 45 minutesafter the pulse of 14C-DG, the animal was decapitated andthe brain removed and frozen in isopentane chilled with
dry ice (—45°C). The brain then was coated with embeddingmedium and stored at —50°C prior to sectioning. About 300coronal sections, consisting of three serial 20-fim sectionstaken every 200 /im throughout the CNS, with additionalsections being retained in the medulla and hypothalamus,were prepared in a cryostat (—22°C). These sections wereexposed with medical x-ray film (Kodak SB-5) in light-tightx-ray cassettes for approximately 6 days. Adjacent sectionswere also stained with cresyl ciolet for precise histologicalidentification of nuclei of particular interest, by referenceto the atlases of Konig and Klippel (1963) and Zeman andInnes (1963). Local tissue concentrations of 14C were deter-mined by quantitative densitometric analysis with a com-puter-based densitometer (Quantimet, Cambridge Instru-ments) with a variable frame size (e.g., area in which theaverage glucose utilization was measured), by reference to10 calibrated 14C-methyl methacrylate standards (rangingfrom 44 to 1475 nCi/g) which were exposed together withthe brain sections. For each brain region in every animal, 12determinations of absorbance (bilateral measurement in sixsections) were made, using a fixed predetermined framesize. The smallest frame size used, for brain structures suchas the locus coeruleus, was 0.02 mm2, whereas the largestone used, for structures like the caudate nucleus, was 0.25mm2. The rate of glucose utilization in each region of theCNS was calculated from (1) the concentration of 14C in thisregion, (2) the concentration of 14C and glucose in arterialplasma samples during the experimental period, and (3) theappropriate constants for the rat, by means of the opera-tional equation derived by Sokoloff et al. (1977).
Statistical AnalysisData are presented as means ± SEM. Statistical differences
in measured variables in the two groups of rats (i.e., animalswith mean arterial pressures, MAP, in the range 50-75 mmHg) were analyzed using a one-tailed t-test, and the Bon-ferroni correction for a level was employed to maintain theoverall a-level at 0.05 (i.e., for 13 degrees of freedom, thecritical value of r was 3.94). The coefficient of correlation(r) for the relationship between local glucose utilization andthe level of MAP in each animal (y = a + b loge x where x= MAP and y = local glucose utilization) was calculated bya least squares approach, and the Bonferroni inequality wasemployed to maintain overall significance at 0.05 level (i.e.,for 13 degrees of freedom, the critical value of r was 0.737).
Results
GeneralHemorrhagic hypotension effected alterations in a
number of parameters other than local cerebral glu-cose utilization. Hypotension was associated withmild behavioral depression with animals displayingreduced spontaneous movement. Even at the mostsevere level of hypotension (arterial blood pressuresof 50 mm Hg), consciousness was preserved and allrats displayed normal responses to auditory and tac-tile stimuli. Hypotension was also associated withhyperventilation, and a significant relationship (r =0.901) could be demonstrated between arterial bloodpressure (x mm Hg) and arterial carbon dioxide ten-sion (y mm Hg) (y = -86.2 + 25.3 Iogex). Withhemorrhage, the large vessel hematocrit tended todecline, although the correlation with blood pressure
by guest on July 5, 2018http://circres.ahajournals.org/
Savaki et a/./Cerebral Glucose Use during Hypotension 635
was only modest (r = 0.536). As a consequence of theuse of insulin, the correlation between blood pressureand arterial plasma glucose at the measurement ofcerebral glucose utilization was poor (r = 0.411),although hemorrhage always resulted in an initialhyperglycemia. Rectal temperature was not affectedby hemorrhagic hypotension.
Local Cerebral Glucose Utilization
Glucose utilization was determined in 58 anatomi-cally discrete regions of the central nervous systemduring graded hemorrhagic hypotension. In the ma-jority of the regions examined (50 of the 58 regionsinvestigated), there was no significant difference (P> 0.05) between the rates of glucose utilization in
TABLE 1Glucose Utilization in Caudal Regions of the CNS during
Hemorrhagic Hypotension
Cerebral glucoseutili2ation
(jumol/100 g per min)
Group I(n =9)
MAP: 80-130mm Hg
Group II(n =6)
MAP: 50-75mm Hg
MedullaNucleus tractus solitarius 51 ± 2Dorsal motor nucleus of va- 50 ± 3
Group I MAP: 104 ± 7 mm Hg; Group II MAP: 59 ± 4 mm Hg.* P < 0.05.
animals with arterial pressures in the range 80-130mm Hg and those in the range 50-75 mm Hg (Tables1 and 2). Of these areas, only in the medial forebrainbundle did the alterations in glucose utilization evenapproach statistical significance (t = 3.63); in contrast,in the 49 remaining regions, the calculated r-valueswere less than 2.5. In none of these areas could asignificant relationship (P > 0.05) be demonstrated
by guest on July 5, 2018http://circres.ahajournals.org/
between the rate of glucose utilization and the levelof arterial blood pressure (for examples, see Figures1-5).
In sharp contrast to the pattern of response in therest of the CNS, hemorrhagic hypotension was asso-ciated with significant increases in glucose utilizationin eight regions (Tables 1 and 2). In five of these eightareas (nucleus of the tractus solitarius, dorsal motornucleus of the vagus, paraventricular and supraopticnuclei, and interstitial nucleus of the stria terminalis),a significant correlation could be demonstrated be-tween the rates of glucose utilization and mean arterialpressures (Figs. 2-5). The differential pattern of re-sponse in each of the eight areas, which displayedincreased rates of glucose utilization during hypoten-sion, could be discerned readily from visual inspec-tion of the autoradiograms (Figs. 6-10).
Discussion
The present study provides a comprehensive de-scription of the distribution and magnitudes of thelocal alterations in function-related glucose utilizationwhich constitute the response of the CNS to hemor-rhagic hypotension. In the overwhelming majority ofbrain areas examined, the rate of glucose utilizationwas not significantly altered during the period ofhypotension. Areas that displayed this pattern ofresponse included all primary auditory and visualregions, all extrapyramidal and motor areas, mostmyelinated fiber tracts, and most anatomical compo-nents of the limbic system, and the failure of hemor-rhagic hypotension to alter glucose utilization in themajority of these areas would be generally consistent
with the well-defined functional roles, quite distinctfrom cardiovascular regulation, which these areassubserve. In a few small nuclei, located mainly in themedulla, pons, and diencephalon, increased rates ofglucose utilization were observed which were relatedto the degree of hypotension.
MedullaThe medial portion of the nucleus of the solitary
tract (NTS) is not only the first synapse of the affer-ents in the 9th and 10th cranial nerves from severaldifferent peripheral baroreceptor sites, but also theprojecting area of a number of higher centers of thebrain which have been implicated in the control ofarterial blood pressure. In addition, the morphologicaland electrophysiological evidence points to the pres-ence of a large number of interneurones within theNTS, which further emphasizes the complexity of theneuronal network within this nucleus (Palkovits andZaborszky, 1977). The NTS plays an inhibitory rolein the central regulation of blood pressure; in the rat,electrical stimulation within the region of the NTSelicits hypotension and its ablation results in hyper-tension (Scherrer, 1967; Doba and Reis, 1973; Chal-mers, 1975). Thus, the observation of a marked alter-ation in functional activity (as reflected in an alteredrate of glucose utilization) in the NTS during a periodof hemorrhagic hypotension is consistent with therole this nucleus plays in cardiovascular regulation.As a reduction in blood pressure leads to deactivationof the baroreceptors and an inhibitory input to theNTS via the afferents from the baroreceptors, it mighthave been anticipated, from a simplistic viewpoint,
HIPPOCAMPUS
Hi
CEREBELLAR HEMISPHERES
40 60 80 100 120 40 60 80 100 120 140
PARIETAL CORTEX
- .£
PYRIFORM CORTEX
60 80 100 CO
BLOOD PRESSURE (mm Kg)
60 80 WO 120 140
BLOOD PRESSURE ( m m H g )
FIGURE 1. Relationship between glucose utilization and mean arterial blood pressure in the hippocampus (stratum lacunosum moleculare (r= 0.390), cerebellar hemisphere (r = 0.326), posterior parietal cortex (r = 0.577), and pyriform cortex (r = 0.54). Data from individual rats arerepresented as a single point. No significant relationship could be demonstrated in any of these four regions.
by guest on July 5, 2018http://circres.ahajournals.org/
Savaki et a/./Cerebral Glucose Use during Hypotension 637
s P i
1stlit
120
XX)
80
60
BO
160
140
LATERAL THALAMICNUCLEUS
60
4 0
20
INTERNAL CAPSULE
I 120
100
80
60
100
80
60
40
40 60 80 100 120 140 4 0
140
120
100
80
60
4 0
20
60 80 100 120 14(
INTERSTITIAL NUCLEUS OFTHE STRIA TERMINALIS
40 60 80 100 120 140
LATERAL GENICULATE140
120
100
80
40 60 80 100 120 140
MEDIAL GENICULATE
40 60 80 100 120
BLOOD PRESSURE (mm Hg)
140 40 120 14060 80 100
BLOOD PRESSURE (mmHg)
FIGURE 2. Relationship between glucose utilization and mean arterial blood pressure in six subcortical brain areas. Data from individual ratsare represented as a single point. A significant correlation can be demonstrated between these two variables in the interstitial nucleus of thestria terminalis (a - 326, b - -59, r = 0.815, P < 0.01), but not in the lateral habenular nucleus (r = 0.730), lateral thalamic nucleus (r =0.562), lateral geniculate (r = 0.344), medial geniculate (r = 0.044), or internal capsule (r = 0.024).
that hemorrhagic hypotension would result in a de-creased rate of glucose utilization in the NTS, and notthe increase which was observed. However, the rela-tive energetic demands (as reflected in glucose utili-zation in vivo) of neuronal excitation and inhibitionare not well defined at the present time, although theavailable evidence does point to Na+/K+ pump activ-ity within the nerve terminals as being the majorenergy consuming process which is reflected in al-tered glucose use (Schwartz et al., 1979; Mata et al.,1980). It is not possible to ascribe the alterations inglucose use in the NTS during hypotension to aparticular energy-consuming process, at least at the
present level of resolution, or to which structuralelements (for example, the terminals of the afferentsfrom the baroreceptors or the nerve terminals of theinterneurones). Modifications of the 2-deoxyglucosetechnique which may resolve the latter question, atleast qualitatively, are becoming available (Des Ro-siers and Descarries, 1978; Sejnowski et al., 1980).The results do provide a clear demonstration of thefunction-related increase in glucose utilization in theNTS during hemorrhagic hypotension. Moreover, asadministration of the chemoreceptor stimulant, dox-apram (Dopram; Robins) did not increase glucose usein the NTS (or in any of the other nuclei which
by guest on July 5, 2018http://circres.ahajournals.org/
FIGURE 3. Relationship between glucose utilization and mean arterial pressure in six nuclei of the medulla. Data from individual rats arerepresented as a single point. A significant correlation can be demonstrated between these two variables in the nucleus of the solitary tract(a = 796, b = —31, r = 0.820) and the dorsal motor nucleus of the vagus (a = 179, b = -28, r = 0.763), but not in the inferior olive (r = 0.339),the paramedian reticular nucleus (r = 0.295), nucleus ambiguus (r = 0.286), or cuneate nucleus (r = 0.014).
displayed significant increases in the present study)(unpublished observations), it seems unlikely that thepattern of glucose utilization during hypotension re-flects the attendant chemoreceptor stimulation.
Our attempts to identify secondary synaptic sitesof the baroreceptor reflex arc in the medulla, on thebasis of locally altered glucose utilization, enjoyedonly limited success. Increased glucose utilization wasobserved in the dorsal motor nucleus of the vagus(DMX), which receives a neuronal input from the NTS(Norgren, 1978) and is itself the nucleus of origin ofthe vagal cardio inhibitory fibers. However, in thenucleus ambiguus, to which the NTS also projects(Morest, 1967; Palkovits and Zaborszky, 1977; Nor-gren, 1978), no consistent alteration in glucose utili-zation during hypotension was found, althoughclearly demonstrable increases in glucose utilization
were observed in some hypotensive animals (see Fig.1, as an example). There was no evidence in thepresent studies that glucose utilization in two otherkey medullary nuclei (the paramedian reticular nu-cleus, which receives carotid sinus nerve afferents(Miura and Kitamura, 1979) and the external cuneatenucleus, which receives an input from the aortic de-pressor nerve (Ciriello and Calaresu, 1978)), was al-tered during sustained hemorrhagic hypotension.
Hypothalamus
The importance of several hypothalamic nuclei inthe response of the CNS to hypotension has beenwell established, and the observation of increasedglucose utilization in the four specific hypothalamicareas during hemorrhagic hypotension is consistent
by guest on July 5, 2018http://circres.ahajournals.org/
Savaki et a/./Cerebral Glucose Use during Hypotension
, „ LOCUS COERULEUS
40
639
40
RAPHE NUCLEUS
40 60 80 100 0 0 140
PONTINE RETICULAR FORMATION
80 SO
40 60 80 CO 120 140
SUBSTANTIA NIGRAZONA COMPACT*
'33-o
FIGURE 4. Relationship between glucose utilization and mean arterial blood pressure in four regions of the brain stem. Data from individualrats are represented as a single point. No significant correlation can be demonstrated between these two variables in either the locus coeruleus(r = 0.696), the pontine reticular formation (r — 0.506), substantia nigra, pars compacta (r = 0.123) or raphe nucleus (r = 0.118).
with the available neuroanatomical and functionalevidence. In contrast, glucose utilization in other hy-pothalamic nuclei, such as the suprachiasmatic nu-cleus [which is involved in the regulation of circadianrhythms rather than circulatory control (see Schwartzet al., 1980)], was not influenced by the level of meanarterial pressure.
In the present study, the proportionately greatestincreases in glucose utilization during hypotensionwere observed in the supraoptic and paraventricularnuclei. Cell firing rates in these two nuclei are influ-enced by the stimulation of carotid sinus and aorticdepressor nerves (Calaresu and Ciriello, 1979), andthe involvement of these two nuclei in the increasedrelease of antidiuretic hormone during arterial hypo-tension has long been recognized (Clark and Silva,1967). Moreover, the periventricular nucleus alsoprojects to the posterior pituitary (Sherlock et al.,1975), and this nucleus displayed increased glucoseuse during the period of hypotension.
Locus Coeruleus
The involvement of the locus coeruleus in cardio-vascular regulation has long been suggested (Chai andWang, 1962; Ward and Gunn, 1976), and the presentstudies provide definitive evidence for the occurrenceof increased glucose utilization (reflecting, presum-ably, increased functional demand) in this key regionduring hemorrhagic hypotension. In view of the in-tegrated nature of the response of the CNS to inducedhypotension, it is of interest to note the known neu-roanatomical interconnections that exist between the
nuclei in which increased glucose utilization was ob-served in the present studies. The locus coeruleusreceives afferent fibers from the paraventricular nu-cleus (Swanson, 1977), the interstitial nucleus of thestria terminalis (Swanson, 1976), and the catechola-mine-containing cell bodies in the region of the vagalnuclear complex (Lindvall and Bjorklund, 1978). Fi-bers arising from the locus coeruleus innervate thelateral habenular nucleus, periventricular and para-ventricular nuclei (Kobayashi et al., 1974; Lindvalland Bjorklund, 1978), the NTS and dorsal motornucleus of the vagus (Loizou, 1969; McBride andSuttin, 1976).
The locus coeruleus appears to be involved func-tionally in the response of the nervous system tonoxious or stressful stimuli (Korf et al., 1973; Saavedraet al., 1979), and the stress, associated with the in-duced hemorrhage, may have contributed in part tothe increased glucose utilization we observed in thisnucleus. It should be emphasized that hemorrhagichypotension was not associated with widespread al-terations in glucose utilization in the terminal distri-bution of the locus coeruleus (e.g., neocortex, hippo-campus, thalamus, etc.) (Lindvall and Bjorklund,1978), nor were alterations observed in the pontinereticular formation during hypotension.
Interstitial Nucleus of the Stria Terminalis
The local injection of noradrenaline into the inter-stitial nucleus elicits a reduction in mean arterialpressure (Struyker-Boudier et al., 1974), and this,together with the neuroanatomical connections of the
by guest on July 5, 2018http://circres.ahajournals.org/
FIGURE 5. Relationship between glucose utilization and mean arterial blood pressure in six hypothalamic areas. Data from individual rats arerepresented as a single point. A significant correlation can be demonstrated between these two variables in the paraventricular nucleus (a= 389, b = —72, r = 0.849) and supraoptic nucleus (a = 278, b = —50, r = 0.807), but not periventricular nucleus (r = 0.701) or the medialforebrain bundle (r = 0.684).
nucleus [it has projections to the posterior pituitary,paraventricular, and periventricular nuclei (Swanson,1976; Kelly and Swanson, 1980)], provides the basisfor the tentative involvement of this nucleus in car-diovascular regulation (Palkovits and Zaborszky,1977). The increased glucose utilization that occurs inthe interstitial nucleus during hemorrhagic hypoten-sion provides further corroborative evidence of itsinvolvement in the response of the CNS to cardiovas-cular manipulation. The major afferent connectionsof the interstitial nucleus of the stria terminalis are
derived from the amygdala (Swanson, 1976). How-ever, in spite of the putative involvement of theamygdala in the response of the CNS to stressfulstimuli and in cardiovascular regulation (Palkovitsand Zaborszky, 1977), no significant alteration inglucose use was observed in this region during hem-orrhagic hypotension.
Epithalamus
The medial portion of the lateral habenular nucleusof the epithalamus displayed increased glucose utili-
by guest on July 5, 2018http://circres.ahajournals.org/
Savaki et ai./Cerebral Glucose Use during Hypotension 641
NTS
FIGURE 6. Representative autoradiograms from the medulla. Therelative rates of glucose utilization are directly related to relativeabsorbance. Upper portion: autoradiogram from an animal with amean arterial blood pressure of 125 mm Hg. Absorbance in theNTS is similar to that in adjacent areas. Lower portion: autoradi-ograms from an animal with a mean arterial pressure of 65 mmHg. Absorbance in the NTS is considerably greater than in adjacent
ring within the noradrenergic systems (cf increasedglucose utilization in locus coeruleus, but not in thenigral or raphe complexes).
Cerebral CortexThe deactivating effect on cortical activity of baro-
receptor activation and of the flexibility provided bycortical and diencephalic systems to overall cardio-vascular regulation, have been the subject of muchspeculation (Korner, 1971; Hilton and Spyer, 1980).Electrical stimulation of the cortex can elicit altera-tions in arterial blood pressure, although this appearsto be intimately related to muscular contraction ratherthan cardiovascular control (Hilton and Spyer, 1980).In the present study, in conscious animals, normalrates of glucose utilization were preserved in each ofthe 10 functionally and anatomically discrete corticalregions investigated. Even within the cortical bound-
zation in response to hemorrhagic hypotension, andthis observation provides the first definitive evidenceof the involvement of this nucleus in response tohypotensive stress. Although the lateral habenulacontains among the highest concentrations of vaso-pressin found outside of the hypothalamic-neurohy-pophysial system, the nerve fibers containing vaso-pressin appear to be derived from the suprachiasmaticnucleus of the hypothalamus and not the periventric-ular and paraventricular nuclei (Buijs, 1978). Thelateral habenula appears to play a major role in theprocessing of information from limbic and striatalforebrain areas to key nuclei in the brain stem (Her-kenham and Nauta, 1977, 1979). By virtue of itsunique connections with the raphe nucleus and sub-stantia nigra-ventral tegmental area complex (Herken-ham and Nauta, 1979), the lateral habenula would beideally placed to coordinate activity during hypoten-sion within the ascending and descending seroto-ninergic and dopaminergic systems with those occur-
FICURE 7. Representative autoradiograms at the level of the locuscoeruleus (LC). The relative rates of glucose utilization are directlyrelated to relative absorbance. Upper portion: autoradiogram froman animal with a mean arterial blood pressure of 125 mm Hg.Absorbance in the locus coeruleus is similar to that of adjacentregions of the pons. Lower portion: autoradiograms from an animalwith a mean arterial blood pressure of 65 mm Hg. A markedincrease in the relative absorbance of the locus coeruleus can beobserved.
by guest on July 5, 2018http://circres.ahajournals.org/
MBFFIGURE 8. Representative autoradiograms at the level of the poste-rior hypothalamus. Upper portion: autoradiogram from an animalwith a mean arterial pressure of 125 mm Hg. Absorbance ishomogeneous throughout the hypothalamus at this level. Lowerportion: autoradiogram from an animal with a mean arterial pres-sure of 65 mm Hg. In this animal, absorbance in the region of themedial forebrain bundle (MBF) is increased relative to adjacentareas of the hypothalamus.
ary zone (the region whose blood supply is derivedfrom both the anterior and middle cerebral arteriesand which is particularly sensitive to structural dam-age in severe hypotension), the rate of glucose usewas unaltered during moderate hypotension in thepresent study. Whether our failure to observe anyalteration in cortical glucose utilization indicative ofcortical deactivation is related to our use of consciousanimals and the avoidance of complicating anestheticinfluences, or of some other mechanism, remainsunclear.
Concluding CommentsIn the present investigations, the quantitative au-
toradiographic 2-deoxyglucose technique has beenemployed successfully to map the function-related
alterations in local cerebral glucose utilization whichoccur during a period of sustained hemorrhagic hy-potension. Although the distribution of the highlyfocal alterations in metabolic activity are consistent,in general, with the functional organization of themechanisms by which the CNS responds to reductionin arterial blood pressure, it should be emphasizedthat hemorrhagic hypotension constitutes an ex-tremely complex stimulus for the CNS. The followingfactors—stress associated with hemorrhage, the ele-vation in the levels of circulating catecholamines andother humoral agents (each of which may, per se,have effects upon local cerebral metabolic activity),possible local direct effects of the reduced arterialpressure, etc.—all may have contributed to the patternof altered glucose utilization observed during sus-tained hemorrhagic hypotension. The present studiesdo provide a detailed description of the integrated
FIGURE 9. Representative autoradiograms at the level of the anteriorhypothalamus. Upper portion: autoradiogram from an animal witha mean arterial pressure of 125 mm Hg. Absorbance is homoge-neous throughout the hypothalamus at this level. Lower portion:autoradiogram from an animal with a mean arterial pressure of 65mm Hg. Marked elevations in the relative absorbance in PAV andSO nuclei can be readily discerned.
by guest on July 5, 2018http://circres.ahajournals.org/
Savaki ef a/./Cerebral Glucose Use during Hypotension 643
NIST
FIGURE 10. Representative autoradiograms at the level of the septalregion. Upper portion: autoradiogram from an animal with a meanarterial blood pressure of 125 mm Hg. Absorbance throughout theseptum is relatively homogeneous. Lower portion: autoradiogramfrom an animal with a mean arterial pressure of 65 mm Hg. Amarked elevation in the relative absorbance in the NIST can bereadily discerned. NIST: interstitial nucleus of the stria terminalis.
response of the CNS to a complex hypotensive stim-uli, and may provide the basis from which one mayconduct detailed investigations, with the 2-deoxyglu-cose technique in conscious rats (thus avoiding thecomplicating influences of anesthetic agents), of theresponses of the CNS to more subtle cardiovascularmanipulations (such as sino-aortic deafferentation,alone and in conjunction with hypotension, in spon-taneously and experimentally induced models of hy-pertension, etc.).
This work was supported by the Wellcome Trust and theMigraine Trust.
We are grateful to the technical and secretarial staff of theWellcome Surgical Institute for their assistance with these investi-gations. Dr. Ian Ford provided valuable statistical advice.
Dr. Savaki held a Wellcome Trust Postdoctoral Fellowship.Received June 15, 1981; accepted for publication January 26,
1982.
ReferencesBuijs RM (1978) Intra- and extrahypothalamic vasopressin and
oxytocin pathways in the rat. Cell Tissue Res 192: 423-435Calaresu FR, Ciriello J (1979) Electrophysioiogy of the hypothala-
mus in relation to central regulation of the cardiovascular system.In Nervous System and Hypertension, edited by P Meyer, HSchmitt. New York, Wiley Flammarian, pp 129-136
Chai CY, Wang SC (1962) Localisation of central cardiovascularcontrol mechanisms in lower brain stem of the cat. Am J Physiol202: 25-30
Chalmers JP (1975) Brain amines and models of experimentalhypertension. Circ Res 36: 469-480
Ciriello J, Calaresu FR (1978) Vagal bradycardia elicited by stimu-lation of the external cuneate nucleus in the cat. Am J Physiol235: R286-R293
Clark BJ, Silva MRE (1967) An afferent pathway for the selectiverelease of vasopressin in response to carotid occlusion andhaemorrhage in the cat. J Physiol (Lond) 191: 529-542
Des Rosiers MH, Descarries L (1978) Adaption de la methode audesoxyglucose a l'echelle cellulaire. CR Acad Sci (Paris) 287:153-156
Doba N, Reis DJ (1973) Acute fulminating neurogenic hypertensionproduced by brainstem lesions in the rat. Circ Res 32: 584-593
Herkenham M, Nauta WJH (1977) Afferent connections of thehabenular nuclei in the rat. A horseradish peroxidase study witha note on the fibre of passage problem. J Comp Neurol 173:123-145
Herkenham M, Nauta WJH (1979) Efferent connections of thehabenular nuclei in the rat. J Comp Neurol 187: 19-48
Hilton SM, Spyer KM (1980) Central nervous regulation of vascularresistance. Annu Rev Physiol 42: 399-411
Kelly ], Swanson LW (1980) Additional forebrain regions projectingto the posterior pituitary: Preoptic region, bed nucleus of thestria terminalis and zona incerta. Brain Res 197: 1-9
Kennedy C, Des Rosiers MH, Jehle JW, Reivich M, Sharpe F,Sokoloff L (1975) Mapping of functional neural pathways byautoradiographic survey of local metabolic rate with [14C]-deox-yglucose. Science 187: 850-853
Kobayashi RM, Palkovits M, Kopin 1R, Jacobowitz DM (1974)Biochemical mapping of noradrenergic nerves arising from therat locus coeruleus. Brain Res 77: 269-279
Konig JFR, Klippel RA (1963) The Rat Brain: A Stereotaxic Atlas.Baltimore, Williams & Wilkins
Korf J, Aghajanian GK, Roth RH (1973) Increased turnover ofnorepinephrine in the rat cerebral cortex during stress: Role ofthe locus coeruleus. Neuropharmacology 12: 933-938
Korner PI (1971) Integrative neural cardiovascular control. PhysiolRev 51: 312-367
Korner PI, Ungsford G, Strarr D, Uther JB, Ward W, White SW(1968) The effects of chloralose-urethane anesthesia on the localand autonomic components of the circulatory response to arterialhypoxia. J Physiol (Lond) 199: 283-302
Lindvall O, Bjorklund A (1978) Organization of catecholamineneurons in the rat central nervous system. In Handbook ofPsychopharmacology, vol 9, edited by LL Iversen, SD Iversen,SH Snyder. pp 139-231
Loizou LA (1969) Projections of the nucleus locus coeruleus in thealbino rat. Brain Res 15: 563-566
Mata M, Fink DJ, Gainer H, Smith CB, Davidsen L, Savaki HE,Schwartz WJ, Sokoloff L (1980) Activity-dependent energy me-tabolism in rat posterior pituitary primarily reflects sodium pumpactivity. J Neurochem 34: 213-215
McBride RL, Suttin J (1976) Projections of the locus coeruleus andadjacent pontine tegmentum in the cat. J Comp Neurol 165:265-284
McCuIloch J (1982) Mapping functional alterations in the CNS withuC-deoxygIucose. In Handbook of Psychopharmacology, vol 15,edited by LL Iversen, SD Iversen, SH Snyder. in press
Miura M, Kitamura T (1979) Postsynaptic potential recorded frommedullary neurones following stimulation of carotid sinus nerve.Brain Res 162: 261-272
Miura M, Reis DJ (1969) Termination and secondary projections of
by guest on July 5, 2018http://circres.ahajournals.org/
carotid sinus nerve in the cat brain stem. Am J Physiol 217:142-153
Morest DK (1967) Experimental study of the projections of thenucleus of the tractus solitarius and the area postrema in the cat.J Comp Neurol 130: 277-300
Norgren R (1978) Projections from the nucleus of the solitary tractin the rat. Neuroscience 3: 207-218
Norgren R, Leonard C (1973) Ascending central gustatory path-ways. J Comp Neurol 150: 217-238
Palkovits M, Zaborszky L (1977) Neuroanatomy of central cardio-vascular control. Nucleus tractus solitarii: Afferent and efferentneuronal connections in relation to the baroreceptor reflex arc.Prog Brain Res 47: 9-34
Peiss CN, Manning JW (1964) Effects of sodium pentobarbital onelectrical and reflex activation of the cardiovascular system. CircRes 14: 228-235
Price HL (1960) General anesthesia and circulatory homeostasis.Physiol Rev 40: 187-218
Saavedra JM, Kvetnansky R, Kopin IR (1979) Adrenaline, nor-adrenaline and dopamine levels in specific brain stem areas ofacutely immobilized rats. Brain Res 160: 271-280
Savaki HE, Davidsen L, Smith C, Sokoloff L (1980) Measurementof free glucose turnover in brain. J Neurochem 35: 495-502
Scherrer J (1967) Inhibition of sympathetic discharge by stimulationof the medulla oblongata in the rat. Acta Neuroveg (Wien) 29:56-74
Schwartz W], Smith CB, Davidsen L, Savaki HE, Sokoloff L, MataM, Fink DJ, Gainer H (1979) Metabolic mapping of functionalactivity in the hypothalamo-neurohypophysial system of the rat.Science 205: 723-725
Schwartz WJ, Davidsen LC, Smith CB (1980) In vivo metabolic
activity of a putative circadian oscillator, the rat suprachiasmaticnucleus. J Comp Neurol 189: 157-167
Sejnowski TJ, Reingold SC, Kelley DB, Gelperin A (1980) Locali-zation of 3H-2-deoxyglucose in single molluscan neurones. Na-ture 287: 449-451
Sherlock DA, Field PM, Raisman G (1975) Retrograde transport ofhorseradish peroxidase in the magnocellular neurosecretory sys-tem of the rat. Brain Res 88: 403-414
Sokoloff L (1981) Localization of functional activity in the centralnervous system by measurement of glucose utilization withradioactive deoxyglucose. J CBF Metab 1: 7-36
Sokoloff L, Reivich M, Kennedy C, Des Rosiers MH, Patlak CS,Pettigrew KD, Sakurada O, Shinohara M (1977) The l4C-deox-yglucose method for the measurement of local cerebral glucoseutilization: Theory, procedure and normal values in the consciousand anesthetized albino rat. J Neurochem 28: 897-916
Struyker Boudier HAJ, Smeets GWM, Brouwer GM, van RossumJM (1974) Hypothalamic alpha adrenergic receptors in cardio-vascular regulation. Neuropharmacology 13: 837-846
Swanson LW (1976) An autoradiographic study of the efferentconnections of the preoptic region in the rat. J Comp Neurol 167:227-256
Swanson LW (1977) Immunohistochemical evidence for a neuro-physin-containing autonomic pathway arising in the paraventric-ular nucleus of the hypothalamus. Brain Res 128: 356-363
Ward DG, Gunn CG (1976) Locus coeruleus complex: elicitation ofa pressor response and a brain stem region necessary for itsoccurrence. Brain Res 107: 401-406
Zeeman W, Innes JRM (1963) Craigie's Neuroanatomy of the Rat.New York, Academic Press
by guest on July 5, 2018http://circres.ahajournals.org/
http://circres.ahajournals.org/content/50/5/633World Wide Web at:
The online version of this article, along with updated information and services, is located on the
http://circres.ahajournals.org//subscriptions/
is online at: Circulation Research Information about subscribing to Subscriptions:
http://www.lww.com/reprints Information about reprints can be found online at: Reprints:
document. Permissions and Rights Question and Answer about this process is available in the
located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is
can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:
by guest on July 5, 2018http://circres.ahajournals.org/
