regional changes in brain 14c-2-deoxyglucose uptake after feeding-inducing intrahypothalamic...
TRANSCRIPT
Brain Research Bullerin, Vol. 24, pp. 249-255. 0 Pergamon Press plc, 1990. Printed in the U.S.A. 0361-9230/W $3.00 + .OO
Regional Changes in Brain 14C-2-Deoxyglucose Uptake After
Feeding-Inducing Intrahypothalamic Norepinephrine Injections
KX& N. NOBRECA AND DONALD V. COSCINA
Section of Biopsychology, Clarke Institute of Psychiatry 250 College Street, Twontcl, Ontario, M5T II?8 Canada
Received 24 July 1989
NGBREGA, J. N. AND D. V. COSCINA. Regional changes in bruin “C-2-deoxyglucose uptake afier feeding-inducing intrahypothalamic norepinephtine injections. BRAIN RES BULL 24(2) 249-255, 1990. -Although norepineptine (NE} injections into the paraventricular h~~~~~s (PVN) have been extensively documented to induce feeding in satiated rats, there have been few systematic attempts to elucidate the neural circuitry subserving this response. In this study quantitative “C-2-deoxygIucose Cz4C-2DG) autoradiography was used to map regional brain changes induced by PVN NE injections. Male Wistar rats, bearing PVN cannulae and previously shown to be positive responders for NE-induced feeding, were given 125 @Xkg 14C-2DG IV immediately following a PVN injection of either 40 nmol NE or vehicle, then hilled 45 min later. 14C-2DG uptake was examined in 97 brain structures using computerized densitometry. PVN NE injections resulted in small, localized changes in brain “C-2DG uptake. Forebrain structures affected included the somatosensory parietal cortex (+ 15%). the CA3 hippocampal field (- 8%). and the reticular thalamic nucleus (+14%}. Midbrain changes involved the anterior pretectal area (+8%> and the central gray area (- 11%). At the hindbrain level, the lateral reticular nucleus showed the most pronounced changes of all brain regions examined ( - 24%0), followed by the nucbus of the solitary tract (- 16%) and the laterodorsal tegmental nucleus (+16%). No changes were seen in the median eminence or in other hypothalamic areas. This pattem of results largely agrees with recent proposals for the circuitry of a PVN-hindbrain system subserving NE-induced as well as hypothalamic lesion-induced feeding effects. In addition, however, they suggest the possibility that altered activity in some forebrain structures may also be involved in the NE response.
i4C-2DG autoradiography Paraventricular hypothalamus Norepinephrine Feeding behaviour Rat
INJECTIONS of norepinephrhe (NE) in the medial hypothalamus are known to induce feeding in satiated rats (1 O), the area of the paraventricular nucleus (PVN) seeming to be a particularly sensi- tive region in this respect (9). Although this phenomenon has been extensively documented, there is a paucity of information on the neural circuitry involved. The PVN is a heterogeneous nucleus, whose various cell groups receive a diverse pattern of neuronal inputs from brainstem and forebrain structures (3, 4, 18, 29, 38) and likewise send efferent projections to a number of limbic forebrain areas (21), autonomic brainstem areas and spinal cord (1, 15, 21, 28, 39). In addition to having been implicated in feeding behaviour, the PVN has well established neuroendocrine and autono~c functions (8, 14, 23, 24, 31, 32). fn this context, one area of interest has been the extent to which neural circuits involved in PVN-driven feeding overlap with autonomic/neuro- endocrine circuits (13, 26, 37).
Previous research aimed at elucidating efferent pathways in- volved in PVN-induced feeding responses have made use of knife-cut lesions placed at various levels along the neuraxis (2, 16, 4U). These studies have produced Ernest evidence for the exclusion of certain forebrain connections and for the likely
involvement of a number of brainstem structures in the PVN- driven noradrenergic feeding circuit. Valuable as such information is, it is nonetheless constrained by two potential limitations of knife-cuts as a circuit mapping technique. One is the need to make a priori decisions about the placement of a necessarily finite number of cuts. The second and potentially more serious ligation relates to the possibility that certain cuts may affect feeding by interfering with parallel ascending or descending sensorimotor systems instead of, or in addition to, circuits involved in the feeding response per se (16).
In the present study we sought to contribute to the mapping of the functional anatomy of the PVN NE feeding circuitry by using a different approach, namely “C-2-deoxyglucose (‘4C-2J3G) au- toradiography, in order to map alterations in regional brain activity associated with intrahypothaIamic NE injections. Use of this technique was expected to provide three potential benefits. First, it should make it possible to examine changes in a large number of brain areas simultaneously, thus obviating the need to make a priori decisions as to relevant brain areas. Second, it should provide indications as to the phys~ologic~ direction of specific regional changes (increased vs. decreased metabolic activity),
249
thereby contributing a degree of functional specificity to the mapping of the neuronal circuitry involved in this response. And finally, the overall pattern of regional changes observed might contribute to elucidating the extent to which the NE feeding response represents a “pure” behavioural response vs. a second- ary consequence of neuroendocrine/autonomic effects.
METHOD
In designing the experiment, two important meth~oIog~ca1 choices were made. The first was to keep the number of screening NE injections to a minimum. This was intended to minimize possible secondary tissue reactions associated with repeated injec- tions, such as gliosis. This was suggested in a pilot experiment where we observed regions of increased *4C-2DG uptake around the injection site in rats that had undergone multiple injections. In our general experience, one or two trials are sufficient to verify the occurrence of significant feeding in response to PVN NE injec- tions, although multiple injections are desirable when determina- tion of the exact magnitude of feeding scores is important. The second decision was not to allow the animals to feed during the actual 14C-2DG-NE trial. This was intended to minimize possible confounding effects on local cerebra1 glucose utilization (LCGU) associated with the sensorimotor aspects of feeding, as well as potential meal-associated alterations in blood glucose levels, which could interfere with lJC-2DG uptake.
Subjects
Twenty male Wistar rats (Charles River. Montreal) weighing 300-325 grams at the time of cannula impiantation were used. Rats were housed in single hanging wire-mesh cages in an otherwise empty room. Purina Chow pellets and water were available at all times. Temperature was kept at 21 ~fi 1°C. Room lights were on between 0800 and 2000 hr.
Surge? and NE Testing
One week after arrival in the laboratory, all rats were implanted with 26 ga stainless steel guide cannulae (Plastic Products, Roanoke, VA) under sodium pentobarbital anesthesia. The fol- lowing stereotaxic coordinates were used to place the cannula tip 1 mm above the PVN: 7.2 mm anterior to the interaural line, 0.3 mm lateral to the midsagittal sinus and 7.0 mm ventral to the surface of the skull. with the head flat between lambda and bregma. in the days following surgery rats were handled daily. Starting one week after surgery rats were screened for NE-induced feeding. Each rat received one NE and one vehicle injection 3-4 days apart, in counterbalanced order. Forty nmoles of NE (d- 1-norepinephrine bitartrate, Regis, Chemical Co., Morton Grove, IL; dose calculated as free base) were injected in a 0.5 ~1 volume
of deionized water vehicle over a 30-set period. The 33 ga injecting cannula protruded 1 .O mm beyond the tip of the guide cannula. Control injections consisted of a similar volume of deionized water. Food consumption adjusted for spillage was measured one hour after injections.
Under light methoxyfluor~e anesthesia rats that showed pas-
itive feeding responses were impfanted with femoral artery and vein catheters, which were threaded under the skin and exited at the back of the neck. Rats were returned to home cages with no plaster casts or other means of restraint. %Z-2DG procedures were performed 6 to 24 hr later using the complete protocol for quantitative dete~ination of local cerebral glucose utilization (35). The procedure began with collection of a 140 ~1 blood sample from the arterial line for determinations of plasma glucose and background radioactivity counts. Forty nmol NE or vehicle
iVEH) was then injected into the hypothala~iic cannula over ;L 30-see period. This was inlmediately followed by a pulse injection of ‘%Z-2DG (New England Nuclear, 56 mCi/mmol; 100-125 @i/kg body weight) through the venous line. No food or water were present in the test cage. Arterial blood samples were collected at the prescribed intervals to determine the time course of “C-2DG specific activity and glucose in plasma (35). Each blood sample drawn was replaced with an equal volume of isotonic saline. Five rats received in~ahypo~alamic NE or VEH injections followed by IV ‘JC-2DG but did not undergo the blood sampling procedures. This was done to assess the potential confounding effect of blood withdrawing procedures on regional lJC-2DG uptake after NE. Forty-five minutes after the ‘“C-2DG injection rats were killed with sodium pentobarbital injection through the venous line. Brains were removed. frozen in Freon 12 and stored at - 70°C until cryostat sectioning. Twenty micron section\ were taken at 0.3-mm intervals throughout the brain. collected onto gelatinized slides, placed on a slide warmer and then apposed to SB-5 film for 7 days in the presence of calibrated “C methyl- metacrylate standards.
After X-ray films were developed, brain sections were fixed in formalin vapour for 24 hr and then stained with cresyl violet. Brain regional densitometric analyses were performed with the MCID imaging system (Imaging Research, St. Catherines, Ontario). The system allowed overlaying of cresyl violet and autoradiographic images of the same brain section, Once both images were aligned, a brain region delineated on the cresyl violet image provided boundaries for densitometric readings in the corresponding region of the autoradiographic plane. This overlaying procedure greatly contributed to anatomical definition and sampling accuracy. Nine- ty-seven brain areas were examined, as defined in the atlas of Paxinos and Watson (20). Densitometric readings were performed with 8 bit resolution (256 gray levels per image point), converted to label accumulation values (nCi/g tissue), usually through a 3rd degree polynomial function, and then to glucose utilization values (nmol/g tissue/min). The Savaki et (11. (25) modification of the Sokoloff operational equation was used, since it allows for
possible plasma glucose fluctuations during the lJC-2DG incor- poration time.
RESULTS
Feeding Response
Hypothalamic injections of 40 nmol NE induced significant feeding responses in I7 of the 20 rats implanted (mean 2 s.e.m.: 2.86~0.98 for NE vs. 1.13rtO.46 for VEH). The three rats that did not eat in response to NE were excluded from the experiment. Examination of cannula placements on autoradiograms or cresyI violet sections confirmed that the PVN was affected in all cases. Cannula tips were usually situated in the most anterior aspect of the nucleus.
Local Cerebral Glucose Utilization
Whole brain means were similar for NE and VEH rats (NE meanrts.e.m.=98.7+4.2; VEH mean=101.32%3.1 ~rno~l~ g tissue/min). All rats had blood glucose levels well within the normal range (7.9 to 8.7 mmol/l) . Rectal temperature measured IO min after the NE and %-2DG injections were also within normal bounds (37.2 to 37S”C). This suggested that the intrahypotha- lamic injections did not cause deviations in physiological param- eters that might confound the interpretation of 14C-2DG uptake data (17).
As stated above, LCGU data were not obtained for all rats. In addition to the rats that had been intentionally spared the blood
14C-2DG UPTAKE AFI’ER PVN NE INJECTIONS 251
TABLE 1
CHANGES IN ‘%2DG UF’I-AKE INDUCED BY INTRAHYF’OTHALAMIC NE INJECTIONS
(I) Within-Group Analysis
% Change Injectemoninjected side
VEH NE
(II) Between-Group Analysis
Bilateral Normalized 14C-2DG Uptake
% Change
VEH NE NENEH
Hindbrain
Ambiguus n.
Cerebellum
granular layer
molecular layer
interpositus n.
lateral n.
medial n.
Cuneate n.
Dorsal tegment. n.
Gigantocellular n.
Hypoglossal n.
Inferior olive
Lat. dorsal teg. n.
Lat. reticular n.
Medial longit. fast.
Pontine n.
Pontine reticular n.
Raphe obscurus
Solitary tract n.
Spinal trigeminal n
Superior olive
Trigeminal motor n.
Vestibular n.
Midbrain
Anterior pretectal a.
Central gray a.
Inferior colliculus
Lateral lemniscus n.
Raphe dorsalis
Raphe medialis
Red n.
Subst. nigra compacta
Subst. nigra reticulata
Superior colliculus
Ventral tegmental a.
4.5 k 2.5
5.1 + 2.4
8.1 k 10.0
5.3 * 3.1
5.5 k 2.6
-6.6 k 3.2
6.6 f 3.2
0.5 + 0.7
-0.6 2 2.9
0.7 2 1.7
3.2 k 2.5
-2.5 + 3.5
4.5 -+ 7.9
3.2 2 2.2
0.3 + 2.8
4.6 + 4.2
0.3 2 2.5
6.2 k 7.8
3.4 5 2.0
-1.1 f 2.3
1.3 k 1.3
3.7 * 2.1
4.3 k 4.4
1.6 ‘- 2.8
0.4 2 1.7
4.1 2 3.6
-0.4 k 0.8
0.7 ‘- 1.9
Accumbens n.
Amygdala
basolateral n.
lateral n.
medial n.
post. cortical n.
Bed n. stria termin.
Caudate-putamen
dorsolateral
dorsomedial
Claustrum
2.7 2 1.7
-1.4 k 2.7
-1.6 2 5.8
3.3 k 5.1
0.2 k 0.3
-0.1 k 2.0
1.0 f 3.4
6.8 * 5.3
2.0 k 2.4
4.1 * 3.7
2.9 + 3.1 93 * 5 (7)
0.5 2 1.9
-2.0 k 2.9
0.9 2 3.4
6.6 2 4.7
-6.2 + 5.2
2.7 ” 9.5
-1.1 k 1.1
-1.7 * 2.0
3.9 t 3.5
5.4 * 2.1*
-3.5 k 7.2
13.6 k 7.0
-0.2 + 1.5
3.8 2 3.3
2.2 2 3.2
-3.1 k 2.7
-1.6 k 5.9
-0.9 k 2.6
88 k 5 (6)
82 2 7 (9)
118-c 5(7)
123 ” 10 (7)
110 + 1 (6)
94 * 6 (5)
109 f 3 (7)
85 k 3 (6)
90 k 3 (6)
93 k 6 (6)
98 + 1 (6)
86 r 7 (4)
66 2 4 (7)
85 2 4 (7)
89 k 3 (7)
69 2 5 (6)
89 k 4 (7)
86 + 4 (7)
136 ? 14 (7)
81 + 2 (4)
136 k 7 (7)
0.4 + 1.8
-2.5 f 1.4
-3.2 + 1.8
0.0 k 5.1
-1.4 + 1.6
-3.4 ? 3.1
0.3 t 2.1
-0.7 + 1.3
-3.6 + 2.2
106 ? 3 (7)
90 k 3 (6)
182 + 11 (7)
138 + 6 (6)
100 + 2 (6)
108 f 3 (7)
97 5 3 (7)
82 t 5 (6)
82 t 2 (7)
104 + 2 (7)
82 k 4 (7)
-0.8 2 2.3 104 2 6 (7)
-10.5 2 3.1t
-21.2 2 10.2
1.4 k 2.4
0.7 k 5.4
-0.0 + 3.1
95 + 3 (5)
90 + 7 (5)
78 ” 4 (7)
87 k 1 (5)
80 2 2 (7)
-2.0 + 9.5
-0.7 * 7.0
-3.0 2 9.2
-11.7 2 3.2*
106 t 4 (6)
97 + 2 (6)
106 k 2 (6)
1132 4(5)
91 + 3 (8) -2
88 f 3 (10) 0
80 k 3 (8) -2
111 Zk 5 (10) -6
114 f 1 (6) -7
106 f 3 (5) -4
99 + 7 (4) 5
120 + 5 (10) 10
82 2 5 (5) -4
94 2 9 (6) 4
90 k 2 (6) -3
114 k 4 (4)? 16t 65 + 3 (5)* - 24*
60 c 4 (4) -9
81 k 3 (10) -5
86 5 2 (10) -3
75 ” 4 (8) 9
75 k 3 (lo)? - 16t
88 * 2 (10) 2
145 ‘- 6 (10) 7
84 k 3 (4) 4
136 k 5 (10) 0
115 k 2 (1o)t 8? 80 + 3 (lo)* -11*
182 + 6 (10) 0
126 + 3 (10) -9
97 + 3 (10) -3
107 k 2 (10) -1
99 + 3 (10) 2
77 + 7 (9) -6
83 2 3 (10) 1
103 2 2 (10) -1
75 2 5 (9) -8
102 2 3 (10) -2
104 k 4 (8) 9 96 k 15 (6) 7 73 f 2 (7) -6 83 k 5 (4) -5 77 k 2 (8) -4
115 + 5 (5) 8 109 + 4 (5) 12 109 + 5 (5) 3 116 k 8 (5) 3
252
TABLE I
CONYiUEI)
(I) Within-Group Analysis
% Change lnjected/Noninjected side
VEH
Cerebral cortex
cingulate anterior
cingulate posterior
entorhinal
frontal
infralimbic
occipital
pyrifornl
SS parietal
temporal
Diagonal band horiz.
Globus pallidus
Habenula lateral
Habenula medial
Hippocampus
CA1
CA2
CA3
dentate gyrus
pyramidal layer
subiculum
Hypothalamus
anterior a.
dorsomedial n.
lateral a.
posterior n.
ventromedial n.
lnterpeduncular n.
Mamillary n.
Median eminence
Olfactory bulb (glom.)
Olfactory tubercle
Preoptic a. lateral
Preoptic a. medial
Septum lateral n.
Septum medial n.
Subthalamic n.
Thalamus
anteroventral n
centromedial n.
laterodorsal n.
lateral geniculate n.
lateral posterior n.
medial geniculate n.
posterior n.
reticular n.
ventrolateral n.
ventromedial n.
ventroposterior n.
Zona incerta
-0.7 t I.1
-2.0 t 1.7
-2.3 k 3.6
-3.5 + 2.3
-3.2 I 2.3
4.8 t 4.2
- 1.7 x 2.3
-1.2r I.1
2.3 2 3.4
2.6 -t 2.3
0.7 k 3.0
2.5 + 3.0
-5.x i- 4.2
I.0 f 3.9
-5.1 2 2.7
-1.7 2 3.6
-4.7 t- 2.6
-3.5 JT 3.6
-2.5 ? 4.4
-1.1 2 I.8
-2.4 i 1.8
0.3 + 2.3
~ I.4 i- 3.4
1.6-c 1.9
2.8 t- 5.6
0.2 2 3.1
~4.2 2 2.8
I.22 6.1
7.8 5 2.6*
-3.5 r 2.0
5.3 2 2.7
I.2 k 2.1
1.3 + 2.3
-0.5 i- 3.8
_ 2.3 k 3.2
5.6 i- 2.7
4.2 t 2.8
-0.6 I 1.3
-0.7 f 1.7
- 3.1 k 2.6
NE VEH
-3.8 L 1.4”: l18? 4(7)
- 1.4 5 1.x II7 i 5 (7)
-9.2 i 2.1: 89 k 2 (7) -3.0 k 2.9 105 + 5 (7) -5.3 + 2.3’ 108 + 2 (4) -2.1 i 2.4 104 t 3 (7) -1.52 1.2 120 t 4 (7) -2.8 + 3.6 116 ? 3 (6) -3.5 k 2.2 l30+- 3(7)
-0.4 k 1.x I08 t 3 (7)
-5.0 + 3.4 81 i 3 (7) -5.6 t 2.4* 121 2 3 (7)
PI.9 i 1.x 109 2 6 (6)
-3.4 lr 2.x
PO.9 +- 3.7
-2.9 2 2.7
~8.0 i 5.9
PO.8 f 2.4
- 1.2 t 3.9
XI lr s (7)
86 i 2 (7)
86 t 2 (7)
81 2 5 (6) 100 2 3 (7)
96 t 2 (7)
-0.1 2 1.1
0.0 i 1.9
-0.9 k 2.3
-0.3 + I.8
-0.3 t 2.4
0.9 i 4.1
-1.92 1.0
~2.6 ? 2.3
3.0 -+ 3.3
0.2 t 2.5
-5.9 k 5.9
86k 5(7)
96 2 8 (7)
83 f 3 (7)
88 r 4 (7) 81 ? 5 (7)
122 i 3 (5)
126 i- 4 (5)
76 +- 6 (4)
99 t I I (4)
II3 ? 3 (3)
88 f 3 (8)
82 I 3 (7)
75z 3(7,
85 i 4 (6) 107 2 2 (4)
~5.3 i 3.6
I.9 r 2.5
-0.4 t 1.9
-2.5 k 2.8
4.5 -t 4.3
-4.6 +- 2.3
-13.3 -+ 9.4
0.2 k 1.8
0.5 -t_ 3.7
~ 1.7 t 3.8
-3.2 i 3.2
I29 2 6 (7)
93 ? I (4)
112 f 3 (7) 98 2 I (7)
107 -t 4 (7) 121 t 3 (7)
107 F 6 (7) 94 k 2 (5)
Ill i 4 (7)
I17 k 5 (7)
Ill -t 6(7)
105 t 2 (7)
(II) Between-Group Analysis Bilateral Normalized ‘%?DG Uptake
NE
:G Change
NE/VEH
II5 -+ 2 (10) -3 I08 2 5 (8) -8 X6 2 2 (8) 3
106 i 4 (7) I 102 I 3 (IO) 5 104 I 3 (9) 0 131 z 3 (9) 9 134 + 3 (lo): 16: I38 f 4 (8) 6 107 t 8 (8) I 77r ‘(10) -5
123 -’ J(l0) 2 97 F 4 (8) -II
73 zv 2 (9) Xl rf- 3 (IO)
79 i- 2 (lo)*
79 i 3 (6)
98 z 2 (IO)
89 i_ 5 (7)
IO
85 z 4 (9)
86 f 2 (9)
87 2 3 (8)
89 z 2 (9)
79 I 2 (IO)
127 f 6 (7)
I25 z -t(lO)
x7 I I6 (5)
III z 5 (6)
I IO -t 9 (4)
79 z 7 (9)
72 z 3 (IO)
771 2(lO)
95 L 3(lO)
109 t 2 (4)
-I IO
5
I
-2
4
1
I3
I2 _. 3
IO
I2
3
I2
2
125 r 5 (5)
108 -+ IO (3)
109 k 2 (5)
102 i 3 (IO)
I08 I 2 (10)
II9 t 2 (10)
I14 -i 2 (8)
107 ri- 3 (5)t 120 r+_ 3 (9)
135 t 5 (IO)
116 z 5 (IO)
108 + 3 (9)
3
I6
-3
4
I
-2
6
14t
8
IS
4
3
Values are means k s.e.m. Ns in parentheses apply to both analyses. Within-group data (I) are o/o change in nCi/g tissue or pmol of glucose/g tissue/min. Symbols indicate results of paired r-comparisons. Between-group data (II) are pmol glucose/100 g tissue/min or nCi/g tissue normalized by the whole brain mean for each animal and multiplied by 100. Symbols indicate results of independent r-tests. *p<o.o5; tpc0.02.
14C-2DG UPTAKE AFTER PVN NE INJECTIONS 253
sampling procedures, four others had partially obstructed arterial cannulae at the time of 14C-2DG procedures. Therefore, data for these rats were expressed as final label concentration in tissue (nCi/g) and not LCGU. In order to ascertain whether results for all animals could be validly included in the same analysis, each rat’s data were normalized using the whole brain mean as the ratio denominator. Within NE and VEH groups no significant differ- ences were found between ratios formed from nCi/g and those formed from pmol/g min data for any of the regions examined. Normalized ratios were therefore used in all subsequent analyses.
The criterion for inclusion of structures in the analysis was clear anatomical definition in at least three subjects in each of the two groups. Two types of analyses were performed. Since injections were unilateral, percent change between injected vs. noninjected homologous brain areas were examined for both NE- and VEH-injected rats. Data were then reanalyzed after combining right and left brain values for bilateral structures for each rat. These aggregate values were normalized by the overall brain mean. The results of both analyses are shown in Table 1. Ten rats were tested in the NE group and 7 in the VEH group.
As indicated in the table, NE injections resulted in ipsilateral decreases in 14C-2DG uptake in 6 forebrain areas and an increase in the inferior olive. The changes were generally small, in the 5% to 10% range, but were consistently observed across animals. Three structures showed large ipsilateral changes in 14C-2DG uptake in some rats but not others, namely, the lateral nucleus of the amygdala, the lateral reticular nucleus, and the reticular nucleus of the thalamus. There was no obvious difference in cannula placement between rats showing changes in these struc- tures and the rats that did not.
When right and left brain readings were combined for each brain structure, significant differences between NE and VEH groups were seen in 10 brain areas. Increased activity was observed in the somatosensory parietal cortex (+ 15%), reticular nucleus of the thalamus (+ 14%). anterior pretectal area (+8%) and laterodorsal tegmental area (+ 16%). Decreased activity was seen in the CA3 hippocampal field (- 8%). central gray area ( - 11%) , the lateral reticular nucleus ( - 24%)) and the nucleus of the solitary tract (- 16%).
DISCUSSION
The 14C-2DG pattern obtained by comparing homologous structures within groups differed somewhat from that obtained when right and left structures were combined. These two types of analyses thus appear to be contributing information of different sorts. While we have chosen to report both for the sake of completeness, there are two main reasons for suggesting that the latter type may be the more meaningful. First, there is evidence that at least some of the descending projections from PVN cross over to the contralateral side before reaching the dorsal medullary complex, since unilateral PVN lesions have been reported to result in presynaptic degeneration both ipsilateral and contralateral to the lesion side (1). Second, the NE injection volume (0.5 ~1) and the lateral coordinates (0.3 mm from midline) used here, while fairly standard in studies of this type, make it difficult to rule out the possibility of contralateral injection spread in at least some animals. For these reasons the unilateral data might be taken as suggestive at best, but should await confirmation from studies using smaller injection volumes at PVN sites farther removed from the midline.
The pattern of i4C-2DG changes observed in the bilateral analyses shows a substantial degree of overlap with data obtained in the knife-cut mapping studies of McCabe er al. (16) and Weiss and Leibowitz (40). In particular, midbrain and brainstem struc- tures showing alterations in 14C-2DG appear to lie on a descending
pathway proposed by these authors to mediate feeding, including the central gray, the lateral reticular nucleus, and the nucleus of the solitary tract. These brainstem changes were in the direction of decreased metabolic activity, except for the observed increase in the laterodorsal tegmental nucleus. The latter contains a significant number of NE cells projecting to the PVN (27). It is conceivable that the observed increase in label uptake may reflect a compen- satory feedback effect from NE stimulation of presynaptic recep- tors in hypothalamic terminal fields.
Among hindbrain structures showing the largest changes in 14C-2DG uptake, the nucleus of the solitary tract (NST) has been well characterized as a major recipient of descending efferents from the PVN (14, 15, 33, 39,41), and may be a pivotal structure in NE-elicited as well as other types of hypothalamic-driven feeding (5). Sawchenko et al. (30) provided the fist evidence that efferents from the dorsal vagal complex are critical to feeding induced by intrahypothalamic NE. More recent work has concen- trated on the connections between the PVN and this dorsal medullary complex. Thus the NST was the most caudal structure in the descending feeding pathway proposed by McCabe et al. (16) and Weiss and Leibowitz (40) as a result of their knife-cut mapping studies. Immunohistochemical studies have identified the descending connections between PVN and the dorsal medullary complex as being primarily oxytocinergic and vasopressinergic (28), and recent work combining behavioural and anatomical tracing techniques have implicated the oxytocinergic projection between PVN and hindbrain as responsible for the hypothalamic hyperphagia-obesity syndrome (6).
Studies employing electrical stimulation of the PVN have led to the suggestion that the PVN exerts a powerful monosynaptic excitatory control over the NST (22,23). In the present study intrahypothalamic NE injections caused a depression of activity in the NST. This finding is consistent with the possibility that alpha adrenergic receptors exert inhibitory effects on a caudally project- ing, possibly oxytocinergic, excitatory system. The possibility that NE in the PVN has a primarily inhibitory role is also consistent with observations that lesions in that nucleus produce feeding effects similar to NE injections (12,34).
It is also of interest to note some of the brain areas which were not significantly affected by PVN NE injections. No significant changes were seen in the median eminence, through which the major endocrine output from the PVN appears to be directed (15,21). Neither were changes seen in any other hypothalamic nuclei. These observations agree with the conclusions from earlier knife-cut studies (2,40). Other structures lying directly on the path of the proposed descending feeding pathway, such as the nucleus ambiguus (16,40), also showed no changes in 14C-2DG uptake. This suggests that the relevant fibers of this feeding circuit may course through, but do not synapse in, this brain area.
In addition to changes in hindbrain structures, the present analysis also identified 14C-2DG uptake changes in forebrain structures after PVN NE, specifically an increase in activity in the somatosensory parietal cortex and the reticular thalamic nucleus and a decrease in the CA3 hippocampal area. These areas have not previously been implicated in PVN NE feeding effects. The potential contribution of these forebrain areas to NE feeding requires further investigation, perhaps by direct manipulation of these structures in conjunction with PVN-NE feeding tests.
It is important to note that while the present analysis was fairly comprehensive it was by no means exhaustive. Excluded were a number of structures not represented in a sufficiently large number of films for reliable assessment, including several additional amygdaloid and hypothalamic nuclei, various subdivisions of the olfactory bulb, and hindbrain structures caudal to the NST, such as the area postrema. Since many of these structures have anatomical connections to the PVN and/or have been implicated in feeding
754 NOBREGA AND COSCIN~
behaviour, they certainly merit future investigation. A final point fied here are of any relevance to the feeding response must bt, of caution refers to constraints posed by the methodological ascertained with other techniques. Aside from providing corrobo- decisions not to perform extensive baseline testing and not to allow the animals to eat during the 14C-2DG trial. While we found no
rative support for the existence of a specific PVN-hindbrain feeding pathway, the present “C-2DG profile offers a functional
obvious differences in cannula placement among animals, it is neuroanatomical reference against which to compare ’ ‘C2DG conceivable that individual variation in the strength of the elicited profiles to be similarly obtained with other PVN-effective conl- feeding response could have contributed to the overall variation in 14C-2DG uptake and perhaps masked additional localized effects,
pounds such as galanin (7). neuropeptide Y (36). and opiates (I 1.19).
In summary, intrahypothalamic NE injections caused changes in 14C-2DG accumulation in a discrete number of brain areas. Most of these changes were located caudally to the PVN. and on the path of a descending feeding system recently proposed on the bases of knife-cut studies. Whether the forebrain changes identi-
ACKNOWLEDGEMENTS
The authors thank J. Snow. L. M. Dixon and E. de Rooy for technical help.
I.
2.
3
4.
5.
6.
7.
8.
9.
IO.
11.
12.
13.
14.
15.
16.
17.
18.
REFERENCES
Akmayev, I. G.; Vikhereva, 0. V.; Konovalova, L. K. The origin of the hypothalamic-vagal descending pathway: an experimental ultra- structural study. Brain Res. 230:342-345; 198 1. Aravich, P. F.; Sclafani, A.; Leibowitz, S. F. Effects of hypothalamic knife cuts on feeding induced by paraventricular norepinephrine injections. Pharmacol. Biochem. Behav. 16:101-l 11; 1982. Cunningham, E. T.; Sawchenko, P. E. Anatomical specificity of noradrenergic inputs to the paraventricular and supraoptic nuclei of the rat hypothalamus. J. Comp. Neurol. 274:60-76; 1988. Guevara-Aguilar, R.; Jimenez-Montufar. L. L.; Garcia-Diaz. D. E.; Wayner, M. J.; Armstrong, D. L. Olfactory and visceral projections to the paraventricular nucleus. Brain Res. Bull. 20:799-801; 1988. Kirchgessner, A. L.; Sclafani, A. PVN-hindbrain pathway involved in the hypothalamic hyperphagia-obesity syndrome. Physiol. Behav. 42:517-528; 1988. Kirchgessner, A. L.; Sclafani. A.; Nilaver. G. Histochemical identi- fication of a PVN-hindbrain feeding pathway. Physiol. Behav. 42: 529-543; 1988. Kykouli, S. E.; Stanley, B. G.; Leibowitz, S. F. Galanin: Stimulation of feeding induced by medial hypothalamic injection of this novel peptide. Eur. .I. Pharmacol. 122:159-160; 1986. Lawrence, D.; Pittman, Q, J. Interaction between descending paraven- tricular neurons and vagal motor neurons. Brain Res. 332: 158-160; 1985. Leibowitz, S. F. Paraventricular nucleus: A primary site mediating adrenergic stimulation of feeding and drinking. Pharmacol. Biochem. Behav. 8:163-175; 1977. Leibowitz, S. F. Neurochemical systems of the hypothalamus. Con- trol of feeding and drinking behavior and water and electrolyte excretion. In: Morgane, P. J.; Panksepp, J., eds. Handbook of the hypothalamus, vol. 3, part A. New York: M. Dekker; 1980:2991138. Leibowitz, S. F.; Hor, L. Endorphinergic and alpha-noradrenergic systems in the paraventricular nucleus: Effects on eating behavior. Peptides 3:421428; 1982. Leibowitz, S. F.; Hammer, N. J.; Chang, K. Hypothalamic paraven- tricular nucleus lesions produce overeating and obesity in the rat. Physiol. Behav. 27:1031-1040; 1981. Leibowitz, S. F.; Roland, C. R.; Hor, L.; Squllari, V. Noradrenergic feeding elicited via the paraventricular nucleus is dependent upon circulating corticosterone. Physiol. Behav. 32:857-864; 1984. Luiten, P. G. M.; ter Horst, G. J.; Steffens, A. B. The hypothalamus, intrinsic connections and outflow pathways to the endocrine system in relation to the control of feeding and metabolism. Prog. Neurobiol. 28:l-54; 1987. Luiten, P. G. M.; ter Horst, G. J.; Karst, H.; Steffens, A. B. The course of paraventricular hypothalamic efferents to autonomic struc- tures in medulla and spinal cord. Brain Res. 329:374-378; 1985. McCabe, J. T.; DeBellus, M.; Leibowitz, S. F. Clonidine-induced feeding: Analysis of central sites of action and fiber projections mediating this response. Brain Res. 309:85-104; 1984. McCulloch, J. Mapping functional alterations in the CNS with “C-deoxyglucose. In: Iversen, er al., eds. Handbook of psychophar- macology. vol. 5. New York: Plenum; 1985:321410. McKellar, S.; Loewy, A. D. Organization of some brain stem afferents to the paraventricular nucleus of the hypothalamus in the rat.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Brain Res. 217:351-357: 1981. McLean, S.; Hoebel, B. G. Feeding induced by opiates injected into the paraventricular hypothalamus. Peptides 4:287-292; 1983. Paxinos, G.; Watson, C. The rat brain in stereotaxic coordinates. 2nd ed. Sydney: Academic Press; 1986. Pittman, Q. J.; Blume, H. W.; Renaud, L. P. Connections of the hypothalamic paraventricular nucleus with the neurohypophysis. me- dian eminence, amygdala, lateral septum and midbrain periaqueductal gray: An electrophysiological study in the rat. Brain Res. 2 15: 15-28; 1981. Rogers, R. C.; Nelson, D. 0. Neurons of the vagal division of the solitary nucleus activated by the paraventricular nucleus of the hypothalamus. J. Auton. Nerv. Syst. 10:193-197; 1984. Rogers, R. C.; Hermann, G. E. Gastric-vagal solitary neurons excited by paraventricular nucleus microstimulation. J. Auton. Nerv. Syst. 14:351-362; 1985. Sakaguchi, T.; Ohtake, M. Inhibition of gastric motility induced by activation of the hypothalamic paraventricular nucleus. Brain Res. 335:365-367; 1985. Savaki, H. E.; Davidsen, L.: Smith, C.; Sokoloff, L. Measurement of free glucose turnover in brain. J. Neurochem. 35:495-502; 1980. Sawchenko, P. E.; Swanson, L. W. Central noradrenergic pathways for the integration of hypothalamic neuroendocrine and autonomic responses. Science 214:685-687; 1981. Sawchenko, P. E.; Swanson. L. W. The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat. Brain Res. Rev. 4:275-325; 1982. Sawchenko. P. E.; Swanson, L. W. Immunohistochemical identifi- cation of neurons in the paraventricular nucleus of the hypothalamus that project to the medulla or to the spinal cord in the rat. J. Comp. Neural. 205:260-272; 1982. Sawchenko. P. E.; Swanson, L. W. The organization of forebrain afferents to the paraventricular and supraoptic nuclei of the rat. J. Comp. Neurol. 218:121-144; 1983. Sawchenko. P. E.; Gold, R. M.; Leibowitz, S. F. Evidence for vagal involvement in the eating elicited by adrenergic stimulation of the paraventricular nucleus. Brain Res. 225:249-269; 1981. Shiraishi, T. Gastric related properties of rat paraventricular neurons. Brain Res. Bull. 18:315-323; 1987. Shiraishi, T. Hypothalamic control of gastric acid secretion. Brain Res. Bull. 20:791-797; 1988. Silverman, A. J.; Hoffman, D. L.; Zimmerman, E. A. The descend- ing afferent connections of the paraventricular nucleus of the hypo- thalamus (PVN). Brain Res. Bull. 6:47-61; 1981. Sims, J. S.; Lorden, J. F. Effect of paraventricular nucleus lesions on body weight, food intake and insulin levels. Behav. Brain Res. 22:265-281; 1986. Sokoloff, L.; Reivich, M.; Kennedy, C.; DesRosiers, M. H.; Patlak, C. S.: Pettierew. K. D.: Sakurada, 0.; Shinohara. M. The [LJC]deoxygl&ose method for the measurement of local cerebral glucose utilization. Theory, procedure, and normal values in the conscious and anesthetized albino rat. J. Neurochem. 28:897-916; 1977. Stanley, B. G.; Leibowitz, S. F. Neuropeptide Y: stimulation of feeding and drinking by injection into the paraventricular nucleus.
14C-2DG UPTAKE AFI’ER PVN NE INJECTIONS 255
Life Sci. 35:2635-2642; 1984. 37. Swanson, L. W.; Sawchenko, P. E. Hypothalamic integration: orga-
nization of the paraventricular and supraoptic nuclei. Annu. Rev. Neurosci. 6:269-324; 1983.
38. Tribollet, E.; Dreifirss, J. J. Localization of neurones projecting to the hypothalamic paraventricular nucleus area of the rat: A horseradish peroxidase study. Neuroscience 6:1315-1328; 1981.
39. van der Kooy, D.; Koda, L. Y.; McGinty, J. F.; Gerfen, C. R.; Bloom, F. E. The organization of projections from the cortex,
amygdala, and hypothalamus to the nucleus of the solitary tract in rat. J. Comp. Neurol. 224:1-24; 1984.
40. Weiss, G. F.; Leibowitz, S. F. Efferent projections from the paraven- tricular nucleus mediating alpha,-noradrenergic feeding. Brain Res. 347:225-238; 1985.
41. Zerihun, L.; Harris, M. An electrophysiological analysis of caudally- projecting neurones from the hypothalamic paraventricular nucleus in the rat. Brain Res. 261:13-20; 1983.