- Email: [email protected]
A comparison of GAD- and GABA-immunoreactive neurons in the first somatosensory area (SI) of the rat cortex
Brain Research, 474 (1988) 192- 196 Elsevier
192 BRE 23212
A compa
son of GADsomatoun
G y
A-immunoreective neurons in (Sl)of tl- rat
first
R. Spreafico l, S. De Biasi 2, C. Frassoni 1, G. Battaglia ~ 1Department of Neurophysiology, Istituto Neurologico 'C. Besta', 20133 Milan (Italy) and 2Department of General Physiology and Biochemistry, Section of Histology and HumanAnatomy, University of Milan, Milan (Italy) (Accepted 9 August 1988)
Key words: Immunocytochemistry; 7-Aminobutyric acid; Glutamic acid decarboxylase; Somatosensory cortex; Rat
Neurons immunoreactive for anti-glutamic acid decarboxylase (anti-GAD) and anti-y-aminobutyric acid (anti-GABA) were compared in adjacent sections from the rat somatosensory cortex (SI). GAD- and GABA-positive neurons in animals either treated or not treated with colchicine were found to occur at a ratio of 1:2. Measurement of areas of GAD- and GABA-immunoreactive neurons confirmed the presence of an 'exuberant' GABA-positive neuronal population not visualized by the GAD antiserum.
Gamma-aminobutyric acid ( G A B A ) has been identified in several mammalian species as the major inhibitory transmitter utilized by cortical neurons. So far, visualization of G A B A e r g i c neurons has been mostly obtained using antisera against glutamic acid decarboxylase ( G A D ) , the G A B A biosynthetic enzyme6,7,9,1 ]. Polyclonal antibodies directed against G A B A itself have been recently developed is and, in many areas of the nervous system, appear to immunostain the same neurons in which immunoreactivity for G A D has been found 3'12'15'18. More recent studies in which the two antisera were tested in the same brain region, have, on the other hand, suggested the presence of a higher number of neurons immunoreactive for the a n t i - G A B A serum than for the anti-GAD one 1'5't6. Since no statistical evaluation was performed, the aim of the present investigation was therefore to verify the discrepancy in the immunostaining of the two antisera in serial sections of the rat somatosensory cortex (SI). Nine rats were anesthetized with chloral hydrate (4% 1 ml/100 g b. wt.) and perfused through the ascending aorta first with saline and then with a solution of 4% paraformaldehyde and 0.1% glutaralde-
hyde in phosphate buffer (PB 0.1 M pH 7.4) followed by 4% paraformaldehyde in the same buffer. Four rats received 0.5-1 /A of colchicine solution (100 ~g//~l) 4 8 - 7 2 h prior to perfusion. Colchicine was injected either in the thalamus (1 animal), or in the sensorimotor cortex (1 animal) or in the cisterna magna (2 animals). After perfusion, brains were removed from the skull and postfixed for 2 h in 4% paraformaldehyde and then rinsed overnight either in PB (8 animals) or in PB containing 30% sucrose (1 animal) at 4 °C. Serial coronal sections cut on a vibratome at 40 ~m (8 animals) or on a cryostat at 12 ktm (1 animal) were processed for immunocytochemistry using either the peroxidase-antiperoxidase (PAP) method 17 or the avidin-biotin procedure (Vector Laboratories). Alternate adjacent sections were incubated overnight at 4 °C in a n t i - G A D serum (kindly provided by Dr. Schmechel: for characterization see ref. 11) diluted 1:1000. and in a n t i - G A B A serum (lmmunonuclear Corp., Stillwater. MN. U . S , A . ) d i l u t e d 1:2500. Control sections were processed with preimmune serum, Some unreacted sections were stained with thionin to identify the first somatosensory area (SI). G A D - and GABA-immunoreactive (IR) neurons in SI were
Correspondence: R. Spreafico, Department of Neurophysiology, Istituto Neurologico 'C. Besta', Via Celoria 11,20133 Milan, Italy. 0006-8993/88/$03.50 k'~ 1988Elsevier Science Publishers B.V. (Biomedical Division)
Fig. 1. A: photomicrograph of a GAD-positive neuron in SI. Note the presence of a large number of labeled terminals some of which surround unlabeled neurons (asterisks) (Scale bar: 30/zm). B: photomicrograph of a G A B A immunoreactive neuron. With this antibody the dendritic arborization and the proximal portion of the axon (arrow) are also evident (Scale bar: 30~m). C and D: low power photomicrographs of two serial sections showing G A D (C) and G A B A (D) immunoreactivity in the rat SI. Labeled terminals and fibers 'en passage' are more evident in the tissue processed with G A D antibody as compared to that immunostained with the antiG A B A serum. In D, GABA-positive neurons are more evident and more evenly distributed through the cortical layers than in C (Scale bar: 200/~m).
194 TABLE
I
GA D- and GA BA-positive neurons counted in SI and L G N o f colchicine-treated (n = 4) and untreated (n = 4) rats T h e a n i m a l u s e d for c r y o s t a t sections is not i n c l u d e d in this T a b l e .
Rat
Colchicine
686 786 187 5187 1886 2486 1386 6187
Cortex
-
Thai. Cortex Cist. m a g n a Cist. m a g n a
L GN
GAD
GABA
GAD/GABA
GAD
GA BA
GA D/GA BA
1828 355 136 748 1342 852 1095 -
3908 1155 232 1577 3038 1804 2259 -
0.46 0.30 0,58 0.47 0.44 0.47 0.48 -
-
170 152 162 142 308
0.56 0.94 1.12 0.94 0.90
cell bodies of different sizes of shapes, often revealing the proximal and also distal portions of their dendritic arborization (Fig. 1B). With respect to the pattern of G A D immunolabeling, neurons staining by anti-GABA serum showed a more even distribution throughout all cortical layers (Fig. 1D) and G A B A IR puncta were less numerous. None of the cells labeled by either anti-GAD or anti-GABA serum possessed the morphological characteristics of cortical pyramidal cells. No staining was observed in control sections. In sections from animals not treated with colchicine, counting of labeled cells in SI showed that GAD-IR neurons were less numerous than the G A B A - I R ones, the G A D / G A B A ratio being about 0.5 (Table I). The study of laminar distribution of IR neurons showed that the 'exuberant' population of GABA-positive cells is not segregated but distributed through all cortical layers. This datum was fur-
counted on vibratome sections and plotted by means of an x - y plotter connected to a Leitz microscope at a magnification of 400 x. Serial cryostat sections were used to count and measure the areas of immunoreactive neurons using a MOP-Videoplan computer connected to a Zeiss microscope. Only neurons with a visible nucleolus in the plane of the section were considered. The anti-GAD serum labeled the cytoplasm of small- to medium-sized neurons, either round or fusiform in shape (Fig. 1A). Only the very proximal portion of some dendrites was occasionally labeled. Large numbers of puncta, interpreted as terminals or cross-sectioned fibers en passage, were intensely immunoreactive (Fig. 1A). The laminar distribution of G A D - I R neurons and puncta (Fig. 1C) was similar to that previously described in the cortex of different mammalian species 6'7"9. The anti-GABA serum intensely stained the cytoplasm and nuclei of neuronal
30, N.
GAD
96 144 183 134 371
GABA
3N|."
+ neu¢ons
C o u n t e d neurons 136 Mean Area 123+39
1
20-
2o-t
lO
lO-
-I- n e u r o n s
Counted Mean
neurons
Area
232
166,7 +46.5
- - 1
1OO
200
300
pm 2
4~)O
100
200
300
Fig. 2. D i s t r i b u t i o n o f a r e a s o f G A D - a n d G A B A - p o s i t i v e n e u r o n s m e a s u r e d on a d j a c e n t 12/~m c r y o s t a t sections.
pm 2
400
195 ther confirmed by computer analysis of the cellular density in different layers as a ratio between the number of IR neurons and the cross-sectional areas of different layers in which the neurons were counted. In one animal in which labeled neurons were also counted in the adjacent lateral geniculate nucleus (LGN), a G A D / G A B A ratio of 0.56 was found. Similar counts were performed on sections from colchicine-treated rats, to determine whether the numerical difference between GAD- and G A B A - I R neurons might be due to a rapid transport of the enzyme to the terminals. Colchicine treatment, regardless of the injection site, increased the number of G A D - I R neurons in LGN (the G A D / G A B A ratio became about 1.0), but not in S1 (the G A D / G A B A ratio remained about 0.5) (Table I). Measurements of the perikaryal size of labeled neurons performed on adjacent cryostat sections showed that the mean area of G A D - I R neurons was 123 _ 39/xm 2 (n = 136), whereas that of G A B A - I R ones was 166.7 _+ 46.5/am 2 (n = 239). The distribution of the areas (Fig. 2) showed the existence of a population of large G A B A - I R neurons, with areas ranging between 200 and 250 ~m 2, not present among the G A D - I R neurons. The present study therefore demonstrates the existence of an 'exuberant' population of neurons stained by the anti-GABA serum which is not detected by the anti-GAD serum. This discrepancy is unlikely to be merely related to technical factors. In fact, although so far no quantitative evaluations have previously been performed, other authors using different anti-GAD and anti-GABA sera also reported the visualization of higher numbers of G A B A - I R than G A D - I R neurons in different areas of the nervous system 1,5,10,16. This also suggests that this exuberant population of GABA-positive neurons is not confined to a specific layer, but that it is evenly distributed through all cortical layers. Nevertheless, it should be pointed out that this result was obtained from thick sections. A more accurate evaluation of the incidence of GABA- and/or G A D - I R neurons among the cellular population within the different layers is now in progress in our laboratory using thin sections. This methodology should be more precise avoiding the prob-
lem of limited penetration of the two antibodies and the possible overestimation of unlabeled neurons. Preliminary results, obtained in our laboratory using another polyclonai anti-GABA serum (Seralab) confirm the data reported here and therefore rule out the possibility that the observed exuberant G A B A - I R population might be due to an unspecific staining of the anti-GABA serum used in this study. In previous experiments (unpublished data), a comparison of G A B A - I R neurons counted in SI of rats perfused with two different fixatives (2.5% glutaraldehyde and 0.5% paraformaldehyde or 4% paraformaldehyde and 0.1% glutaraldehyde) did not show any significative difference in the number of G A B A - I R cells that might be related to the fixation procedure employed in this study. It seems unlikely that the polyclonal anti-GAD serum used in this study recognizes only one of the two known allosteric forms of GAD 2 (GAD1 and GAD2), since GAD2 has never been detected in the nervous system. Biochemical investigations have recently shown the existence of a G A B A biosynthetic route through succinic semialdehyde and polyamine, which does not involve GAD 4, that might explain the lower number of G A D - I R cells detected in SI. This metabolic pathway is, however, mainly present in the embryo and seems to be quantitatively insignificant in the adult compared to the more common pathway through G A D s'13'14. The exuberant G A B A - I R neurons detected in the cortex might simply be represented by neurons that are capable of uptake of G A B A from the local environment, but do not synthesize and use it as neurotransmitter. If true, however, this should be a characteristic limited to some cortical cells, since no exuberant G A B A - I R population was observed in LGN after colchicine treatment. This hypothesis has to be verified using specific inhibitors of G A B A uptake.
This work was partially supported by the P. Zorzi Association for Neurosciences, the Ministry of Public Education Grant C-179007H and the CNR Grant 88.00685.04.
196 1 Agardh, E., Bruun, A., Ehinger, B., Ekstrom, P., Van Veen, T. and Wu, J.Y., Gamma-aminobutyric acid and glutamic acid decarboxylase immunoreactive neurons in the retina of different vertebrates, J. Comp. Neurol., 258 (1987) 622-630. 2 Bowman, B.C. and Rand, M.J., Textbook of Pharmacology, Blackwell, London, 1985. 3 De Biasi, S., Frassoni, C. and Spreafico, R., GABA immunoreactivity in the thalamic reticular nucleus of the rat. A light and electron microscopical study, Brain Research, 399 (1986) 143-147. 4 Erdo, S., Peripheral GABAergic mechanisms, Trends Pharmacol. Sci., 6 (1985) 205-208. 5 Fitzpatrick, D., Lund, J.S., Schmechel, D.E. and Towles, A.C., Distribution of GABAergic neurons and axon terminals in the macaque striate cortex, J. Comp. Neurol., 264 (1987) 73-91. 6 Houser, C.R., Hendry, S.H.C., Jones, E.G. and Vaugh, J.E., Morphological diversity of immunocytochemically identified GABA neurons in the monkey sensory-motor cortex, J. Neurocytol., 12 (1983) 617-638. 7 Keller, A. and White, E.L., Distribution of glutamic acid decarboxylase immunoreactive structures in the barrel region of mouse somatosensory cortex, Neurosci. Lett., 66 (1986) 245-250. 8 Lauder, J.M., Hail, V.K.M., Menderson, P., Verdooru, T. and Towle, A.C., Prenatal ontogeny of the GABAergic system in the rat brain: an immunocytochemical study, Neuroscience, 19 (1986) 465-493. 9 Lin, C.S., Lu, S.M. and Schmechel, D.E., Glutamic acid decarboxylase and somatostatin immunoreactivities in rat visual cortex, J. Comp. Neurol., 244 (1986) 369-383. 10 Newton, B.W. and Maley, B.E., A comparison of GABAand GAD-like immunoreactivity within the area postrema
of the rat and cat, J. Comp. Neurol., 255 (1987) 208-216. 11 Oertel, W.H., Schmechel, D.E., Tappaz, H.L. and Kopin, I.J., Production of specific antiserum to rat brain glutamic acid decarboxylase by injection of an antigen-antibody complex, Neuroscience, 6 (1981) 2689-2700. 12 Ottersen, O.P. and Storm-Mathisen, J., Glutamate and GABA containing neurons in the mouse and rat brain, as demonstrated with a new immunocytochemical technique, J. Comp. Neurol., 229 (1984) 374-392. 13 Seiler, A. and Sarhan, S., Metabolic routes of GABA formation in chick embryo brain, Neurochem. Int., 5 (1983) 625-633. 14 Shank, R.P. and Campbell, G., Metabolic precursors of glutamate and GABA. In L. Hertz, E. Krammer, E.G. McGeer and A. Schousboe (Eds.), Glutamine, Glutamate and GABA in the Central Nervous System, Alan R. Liss, New York, 1983, pp. 355-369. 15 Somogyi, P., Hodgson, A.J., Chubb, I.W., Penke, B. and Erdei, A., Antisera to gamma aminobutyric acid. II. Immunocytochemical application to the central nervous system, J. Histochem. Cytochem., 33 (1985) 240-248. 16 Somogyi, P., Hodgson, A.J., Smith, A.D., Nunzi, M.G., Gorio, A. and Wu, J.Y., Different populations of GABAergic neurons in the visual cortex and hippocampus of cat contain somatostatin- or cholecystokinin-immunoreactive material, J. Neurosci., 4 (1984) 2590-2603. 17 Sternberger, L.A., lmmunocytochemistry, 2nd edn., Wiley, New York, 1979. 18 Storm-Mathisen, J., Leknes, A.K., Bore, A.T., Vaaland, J.L., Edminson, P., Haugh, F.M.S. and Ottersen, O.P., First visualization of glutamate and GABA in neurons by immunocytochemistry, Nature (Lond.), 301 (1983) 517-520.
192 BRE 23212
A compa
son of GADsomatoun
G y
A-immunoreective neurons in (Sl)of tl- rat
first
R. Spreafico l, S. De Biasi 2, C. Frassoni 1, G. Battaglia ~ 1Department of Neurophysiology, Istituto Neurologico 'C. Besta', 20133 Milan (Italy) and 2Department of General Physiology and Biochemistry, Section of Histology and HumanAnatomy, University of Milan, Milan (Italy) (Accepted 9 August 1988)
Key words: Immunocytochemistry; 7-Aminobutyric acid; Glutamic acid decarboxylase; Somatosensory cortex; Rat
Neurons immunoreactive for anti-glutamic acid decarboxylase (anti-GAD) and anti-y-aminobutyric acid (anti-GABA) were compared in adjacent sections from the rat somatosensory cortex (SI). GAD- and GABA-positive neurons in animals either treated or not treated with colchicine were found to occur at a ratio of 1:2. Measurement of areas of GAD- and GABA-immunoreactive neurons confirmed the presence of an 'exuberant' GABA-positive neuronal population not visualized by the GAD antiserum.
Gamma-aminobutyric acid ( G A B A ) has been identified in several mammalian species as the major inhibitory transmitter utilized by cortical neurons. So far, visualization of G A B A e r g i c neurons has been mostly obtained using antisera against glutamic acid decarboxylase ( G A D ) , the G A B A biosynthetic enzyme6,7,9,1 ]. Polyclonal antibodies directed against G A B A itself have been recently developed is and, in many areas of the nervous system, appear to immunostain the same neurons in which immunoreactivity for G A D has been found 3'12'15'18. More recent studies in which the two antisera were tested in the same brain region, have, on the other hand, suggested the presence of a higher number of neurons immunoreactive for the a n t i - G A B A serum than for the anti-GAD one 1'5't6. Since no statistical evaluation was performed, the aim of the present investigation was therefore to verify the discrepancy in the immunostaining of the two antisera in serial sections of the rat somatosensory cortex (SI). Nine rats were anesthetized with chloral hydrate (4% 1 ml/100 g b. wt.) and perfused through the ascending aorta first with saline and then with a solution of 4% paraformaldehyde and 0.1% glutaralde-
hyde in phosphate buffer (PB 0.1 M pH 7.4) followed by 4% paraformaldehyde in the same buffer. Four rats received 0.5-1 /A of colchicine solution (100 ~g//~l) 4 8 - 7 2 h prior to perfusion. Colchicine was injected either in the thalamus (1 animal), or in the sensorimotor cortex (1 animal) or in the cisterna magna (2 animals). After perfusion, brains were removed from the skull and postfixed for 2 h in 4% paraformaldehyde and then rinsed overnight either in PB (8 animals) or in PB containing 30% sucrose (1 animal) at 4 °C. Serial coronal sections cut on a vibratome at 40 ~m (8 animals) or on a cryostat at 12 ktm (1 animal) were processed for immunocytochemistry using either the peroxidase-antiperoxidase (PAP) method 17 or the avidin-biotin procedure (Vector Laboratories). Alternate adjacent sections were incubated overnight at 4 °C in a n t i - G A D serum (kindly provided by Dr. Schmechel: for characterization see ref. 11) diluted 1:1000. and in a n t i - G A B A serum (lmmunonuclear Corp., Stillwater. MN. U . S , A . ) d i l u t e d 1:2500. Control sections were processed with preimmune serum, Some unreacted sections were stained with thionin to identify the first somatosensory area (SI). G A D - and GABA-immunoreactive (IR) neurons in SI were
Correspondence: R. Spreafico, Department of Neurophysiology, Istituto Neurologico 'C. Besta', Via Celoria 11,20133 Milan, Italy. 0006-8993/88/$03.50 k'~ 1988Elsevier Science Publishers B.V. (Biomedical Division)
Fig. 1. A: photomicrograph of a GAD-positive neuron in SI. Note the presence of a large number of labeled terminals some of which surround unlabeled neurons (asterisks) (Scale bar: 30/zm). B: photomicrograph of a G A B A immunoreactive neuron. With this antibody the dendritic arborization and the proximal portion of the axon (arrow) are also evident (Scale bar: 30~m). C and D: low power photomicrographs of two serial sections showing G A D (C) and G A B A (D) immunoreactivity in the rat SI. Labeled terminals and fibers 'en passage' are more evident in the tissue processed with G A D antibody as compared to that immunostained with the antiG A B A serum. In D, GABA-positive neurons are more evident and more evenly distributed through the cortical layers than in C (Scale bar: 200/~m).
194 TABLE
I
GA D- and GA BA-positive neurons counted in SI and L G N o f colchicine-treated (n = 4) and untreated (n = 4) rats T h e a n i m a l u s e d for c r y o s t a t sections is not i n c l u d e d in this T a b l e .
Rat
Colchicine
686 786 187 5187 1886 2486 1386 6187
Cortex
-
Thai. Cortex Cist. m a g n a Cist. m a g n a
L GN
GAD
GABA
GAD/GABA
GAD
GA BA
GA D/GA BA
1828 355 136 748 1342 852 1095 -
3908 1155 232 1577 3038 1804 2259 -
0.46 0.30 0,58 0.47 0.44 0.47 0.48 -
-
170 152 162 142 308
0.56 0.94 1.12 0.94 0.90
cell bodies of different sizes of shapes, often revealing the proximal and also distal portions of their dendritic arborization (Fig. 1B). With respect to the pattern of G A D immunolabeling, neurons staining by anti-GABA serum showed a more even distribution throughout all cortical layers (Fig. 1D) and G A B A IR puncta were less numerous. None of the cells labeled by either anti-GAD or anti-GABA serum possessed the morphological characteristics of cortical pyramidal cells. No staining was observed in control sections. In sections from animals not treated with colchicine, counting of labeled cells in SI showed that GAD-IR neurons were less numerous than the G A B A - I R ones, the G A D / G A B A ratio being about 0.5 (Table I). The study of laminar distribution of IR neurons showed that the 'exuberant' population of GABA-positive cells is not segregated but distributed through all cortical layers. This datum was fur-
counted on vibratome sections and plotted by means of an x - y plotter connected to a Leitz microscope at a magnification of 400 x. Serial cryostat sections were used to count and measure the areas of immunoreactive neurons using a MOP-Videoplan computer connected to a Zeiss microscope. Only neurons with a visible nucleolus in the plane of the section were considered. The anti-GAD serum labeled the cytoplasm of small- to medium-sized neurons, either round or fusiform in shape (Fig. 1A). Only the very proximal portion of some dendrites was occasionally labeled. Large numbers of puncta, interpreted as terminals or cross-sectioned fibers en passage, were intensely immunoreactive (Fig. 1A). The laminar distribution of G A D - I R neurons and puncta (Fig. 1C) was similar to that previously described in the cortex of different mammalian species 6'7"9. The anti-GABA serum intensely stained the cytoplasm and nuclei of neuronal
30, N.
GAD
96 144 183 134 371
GABA
3N|."
+ neu¢ons
C o u n t e d neurons 136 Mean Area 123+39
1
20-
2o-t
lO
lO-
-I- n e u r o n s
Counted Mean
neurons
Area
232
166,7 +46.5
- - 1
1OO
200
300
pm 2
4~)O
100
200
300
Fig. 2. D i s t r i b u t i o n o f a r e a s o f G A D - a n d G A B A - p o s i t i v e n e u r o n s m e a s u r e d on a d j a c e n t 12/~m c r y o s t a t sections.
pm 2
400
195 ther confirmed by computer analysis of the cellular density in different layers as a ratio between the number of IR neurons and the cross-sectional areas of different layers in which the neurons were counted. In one animal in which labeled neurons were also counted in the adjacent lateral geniculate nucleus (LGN), a G A D / G A B A ratio of 0.56 was found. Similar counts were performed on sections from colchicine-treated rats, to determine whether the numerical difference between GAD- and G A B A - I R neurons might be due to a rapid transport of the enzyme to the terminals. Colchicine treatment, regardless of the injection site, increased the number of G A D - I R neurons in LGN (the G A D / G A B A ratio became about 1.0), but not in S1 (the G A D / G A B A ratio remained about 0.5) (Table I). Measurements of the perikaryal size of labeled neurons performed on adjacent cryostat sections showed that the mean area of G A D - I R neurons was 123 _ 39/xm 2 (n = 136), whereas that of G A B A - I R ones was 166.7 _+ 46.5/am 2 (n = 239). The distribution of the areas (Fig. 2) showed the existence of a population of large G A B A - I R neurons, with areas ranging between 200 and 250 ~m 2, not present among the G A D - I R neurons. The present study therefore demonstrates the existence of an 'exuberant' population of neurons stained by the anti-GABA serum which is not detected by the anti-GAD serum. This discrepancy is unlikely to be merely related to technical factors. In fact, although so far no quantitative evaluations have previously been performed, other authors using different anti-GAD and anti-GABA sera also reported the visualization of higher numbers of G A B A - I R than G A D - I R neurons in different areas of the nervous system 1,5,10,16. This also suggests that this exuberant population of GABA-positive neurons is not confined to a specific layer, but that it is evenly distributed through all cortical layers. Nevertheless, it should be pointed out that this result was obtained from thick sections. A more accurate evaluation of the incidence of GABA- and/or G A D - I R neurons among the cellular population within the different layers is now in progress in our laboratory using thin sections. This methodology should be more precise avoiding the prob-
lem of limited penetration of the two antibodies and the possible overestimation of unlabeled neurons. Preliminary results, obtained in our laboratory using another polyclonai anti-GABA serum (Seralab) confirm the data reported here and therefore rule out the possibility that the observed exuberant G A B A - I R population might be due to an unspecific staining of the anti-GABA serum used in this study. In previous experiments (unpublished data), a comparison of G A B A - I R neurons counted in SI of rats perfused with two different fixatives (2.5% glutaraldehyde and 0.5% paraformaldehyde or 4% paraformaldehyde and 0.1% glutaraldehyde) did not show any significative difference in the number of G A B A - I R cells that might be related to the fixation procedure employed in this study. It seems unlikely that the polyclonal anti-GAD serum used in this study recognizes only one of the two known allosteric forms of GAD 2 (GAD1 and GAD2), since GAD2 has never been detected in the nervous system. Biochemical investigations have recently shown the existence of a G A B A biosynthetic route through succinic semialdehyde and polyamine, which does not involve GAD 4, that might explain the lower number of G A D - I R cells detected in SI. This metabolic pathway is, however, mainly present in the embryo and seems to be quantitatively insignificant in the adult compared to the more common pathway through G A D s'13'14. The exuberant G A B A - I R neurons detected in the cortex might simply be represented by neurons that are capable of uptake of G A B A from the local environment, but do not synthesize and use it as neurotransmitter. If true, however, this should be a characteristic limited to some cortical cells, since no exuberant G A B A - I R population was observed in LGN after colchicine treatment. This hypothesis has to be verified using specific inhibitors of G A B A uptake.
This work was partially supported by the P. Zorzi Association for Neurosciences, the Ministry of Public Education Grant C-179007H and the CNR Grant 88.00685.04.
196 1 Agardh, E., Bruun, A., Ehinger, B., Ekstrom, P., Van Veen, T. and Wu, J.Y., Gamma-aminobutyric acid and glutamic acid decarboxylase immunoreactive neurons in the retina of different vertebrates, J. Comp. Neurol., 258 (1987) 622-630. 2 Bowman, B.C. and Rand, M.J., Textbook of Pharmacology, Blackwell, London, 1985. 3 De Biasi, S., Frassoni, C. and Spreafico, R., GABA immunoreactivity in the thalamic reticular nucleus of the rat. A light and electron microscopical study, Brain Research, 399 (1986) 143-147. 4 Erdo, S., Peripheral GABAergic mechanisms, Trends Pharmacol. Sci., 6 (1985) 205-208. 5 Fitzpatrick, D., Lund, J.S., Schmechel, D.E. and Towles, A.C., Distribution of GABAergic neurons and axon terminals in the macaque striate cortex, J. Comp. Neurol., 264 (1987) 73-91. 6 Houser, C.R., Hendry, S.H.C., Jones, E.G. and Vaugh, J.E., Morphological diversity of immunocytochemically identified GABA neurons in the monkey sensory-motor cortex, J. Neurocytol., 12 (1983) 617-638. 7 Keller, A. and White, E.L., Distribution of glutamic acid decarboxylase immunoreactive structures in the barrel region of mouse somatosensory cortex, Neurosci. Lett., 66 (1986) 245-250. 8 Lauder, J.M., Hail, V.K.M., Menderson, P., Verdooru, T. and Towle, A.C., Prenatal ontogeny of the GABAergic system in the rat brain: an immunocytochemical study, Neuroscience, 19 (1986) 465-493. 9 Lin, C.S., Lu, S.M. and Schmechel, D.E., Glutamic acid decarboxylase and somatostatin immunoreactivities in rat visual cortex, J. Comp. Neurol., 244 (1986) 369-383. 10 Newton, B.W. and Maley, B.E., A comparison of GABAand GAD-like immunoreactivity within the area postrema
of the rat and cat, J. Comp. Neurol., 255 (1987) 208-216. 11 Oertel, W.H., Schmechel, D.E., Tappaz, H.L. and Kopin, I.J., Production of specific antiserum to rat brain glutamic acid decarboxylase by injection of an antigen-antibody complex, Neuroscience, 6 (1981) 2689-2700. 12 Ottersen, O.P. and Storm-Mathisen, J., Glutamate and GABA containing neurons in the mouse and rat brain, as demonstrated with a new immunocytochemical technique, J. Comp. Neurol., 229 (1984) 374-392. 13 Seiler, A. and Sarhan, S., Metabolic routes of GABA formation in chick embryo brain, Neurochem. Int., 5 (1983) 625-633. 14 Shank, R.P. and Campbell, G., Metabolic precursors of glutamate and GABA. In L. Hertz, E. Krammer, E.G. McGeer and A. Schousboe (Eds.), Glutamine, Glutamate and GABA in the Central Nervous System, Alan R. Liss, New York, 1983, pp. 355-369. 15 Somogyi, P., Hodgson, A.J., Chubb, I.W., Penke, B. and Erdei, A., Antisera to gamma aminobutyric acid. II. Immunocytochemical application to the central nervous system, J. Histochem. Cytochem., 33 (1985) 240-248. 16 Somogyi, P., Hodgson, A.J., Smith, A.D., Nunzi, M.G., Gorio, A. and Wu, J.Y., Different populations of GABAergic neurons in the visual cortex and hippocampus of cat contain somatostatin- or cholecystokinin-immunoreactive material, J. Neurosci., 4 (1984) 2590-2603. 17 Sternberger, L.A., lmmunocytochemistry, 2nd edn., Wiley, New York, 1979. 18 Storm-Mathisen, J., Leknes, A.K., Bore, A.T., Vaaland, J.L., Edminson, P., Haugh, F.M.S. and Ottersen, O.P., First visualization of glutamate and GABA in neurons by immunocytochemistry, Nature (Lond.), 301 (1983) 517-520.