4 Guillery, R. W. (1966) J. Comp. NeuroL 46, 80-129 5 Lieberman, A. R. (1973) Brain Res. 59, 35-39 6 0 h a r a , P. T., Lieberman, A. R., Hunt, S. P. and Wu, J. Y. (1983) Neuroscience 8, 189-211 7 Gabbott, P. L. A., Somogyi, M., Stewart, M. G. and Hamori, J. (1986) Exp. Brain Res. 61,311-322 8 Montero, V. M. and Singer, W. (1984) Exp. Brain Res. 56, 115-125 9 Bowery, N. G. (1989) Trends Pharmacol. Sci. 10, 401-407 10 Bowery, N. G., Bittiger, H. and Olpe, H. R. (1990) GABAB Receptors in. Mammalian Function, J. Wiley & Sons 11 Bowery, N. G., Hudson, A. L. and Price, G. W. (1987) Neuroscience 20, 365-383 12 Singer, W., POppel, E. and Creutzfeldt, O. (1972) Exp. Brain Res. 14, 210-226 13 Sillito, A. M. and Kemp, J. A. (1983) Brain Res. 277, 63-77 14 Andersen, P., Eccles, J. C. and Sears, T. A. (1964) J. PhysioL 174, 370-399 15 Purpura, D. P. and Cohen, B. (1962) J. NeurophysioL 25, 621-635 16 Roy, J. P., Clerq, M., Steriade, M. and Desch~nes, M. (1984) J. NeurophysioL 51, 1220-1235 17 Yamamoto, C. (1974) Exp. Brain Res. 19, 271-281 18 Kelly, J. S., Godfraind, J-M. and Maruyama, S. (1979) Brain Res. 168, 388-392 19 Hirsch, J. and Burnod, Y. (1987) Neuroscience 23,457-468 20 Crunelli, V., Haby, M., Jassik-Gerschenfeld, D., Leresche, N. and Pirchio, M. (1988) J. PhysioL 399, 153-176 21 Soltesz, I., Lightowler, S., Leresche, N. and Crunelli, V. (1989) Neuroscience 33, 23-33 22 Crunelli, V., Kelly, J. S., Leresche, N. and Pirchio, M. (1987) J. PhysioL 384, 587-602 23 Sillito, A. M., Murphy, P. C. and Salt, T. E. (1990) Neuroscience 34, 273-280 24 Soltesz, I., Haby, M., Leresche, N. and Crunelli, V. (1988) Brain Res. 448, 351-354 25 Kerr, D. I. B., Ong, J., Prager, R. H., Gynther, B. D. and Curtis, D. R. (1987) Brain Res. 405, 150-152

26 Hu, B., Steriade, M. and Desch~nes, M. (1989) Neuroscience 31, 13-24 27 Bloomfield, S. A. and Sherman, S. M. (1988)J. NeurophysioL 60, 1924-1945 28 Hablitz, J. J. and Thalmann, R. H. (1987) J. Neurophysiol. 58, 160-171 29 Newberry, N. R. and Nicoll, R. A. (1985) J. Physiol. 360, 161-185 30 Hasuo, H. and Gallagher, J. P. (1988) Neurosci. Lett. 86, 77-81 31 Connors, B. W., Malenka, R. C. and Silva, L. R. (1988) J. Physiol. 406, 443-468 32 Bormann, J. (1988) Trends Neurosci. 11, 112-116 33 Deisz, R. A. and Prince, D. A. (1989) J. Physiol. 412,513-542 34 G~hwiler, B. H. and Brown, D. A. (1985) Proc. Nat/Acad. Sci. USA 82, 1558-I 562 35 Steriade, M., Deschanes, M., Domich, L. and Mulle, C. (1985) J. Neurophysiol. 54, 1473-1497 36 Thomson, A. M. (1990) Neurosci. Left. Suppl. 38, $89 37 Dolphin, A. C. and Scott, R. H. (1986) Br. J. Pharm. 88, 213-220 38 Baumfalk, U. and Albus, K. (1988) Brain Res. 463,398-402 39 Crunelli, V., Lightowler, S. and Pollard, C. E. (1989) J. Physiol. 413,543-561 40 Coulter, D. A., Huguenard, J. R. and Prince, D. A. (1989) J. Physiol. 414, 587-604 41 Hernandez-Cruz, A, and Pape, H-C. (1989) J. Neurophysiol. 61, 1270-1283 42 Suzuki, S. and Rogawski, M. A. (1989) Proc. NatlAcad. Sci. USA 86, 7228-7232 43 McCormick, D. A. (1989) Trends Neurosci. 12, 215-221 44 Clarke, K. A. (1983) Neuropharmacology 22, 1231-1235 45 Gloor, P. and Fariello, R. G. (1988) Trends Neurosci. 11, 63-68 46 Leresche, N., Jassik-Gerschenfeld, D., Haby, M., Soltesz, I. and Crunelli, V. (1990) Neurosci. Lett. 113, 72-77 47 Thomson, A. M. (1988) Neuroscience 25, 491-502 48 Soltesz, I. etal. (1990) Eur. J. Neurosci. 2,414-429

Acknowledgements We are grateful to ProfessorA. 44. Sillito, Drs R. W. Horton and D. JassikGerschenfe/dfor their critical comments. We are indebted to our colleaguesDrs M. Haby, S. Lightowler, 44. Pirchio, C E. Pollard and I. 5oltesz who, at different stages, wereinvolved in the work reported m thispaperand to the We//come Trust and the CNRSfor constant financial support.

Dopaminergicinnervationof the cerebralcortex:unexpeded differencesbetweenrodentsandprimates B. Berger, P. Gaspar and C. Verney Until recently, views on the organization and role of the mesotelencephalic dopaminergic (DA) systems were mostly based on studies of rodents, and it was assumed that homology existed across mammalian species. However, recent studies of both human and non-human primates indicate that this might not be so. The mesocortical DA system m primates, which is directly involved in the pathophysiology of severe illnesses such as Parkinson's disease and psychoses, shows substantial differences from that of rodents. These differences include much larger, re-organized terminal fields, a different phenotype for the co-localization of neuropeptides and a very early prenatal development. The dopaminergic (DA) projections to the telencephaIon (striatum, limbic system and cerebral cortex) originate from a large band of neurones, the A10, A9, A8 cell complex, located in the ventral mesencephalon1,2. In contrast to the widespread projections of the noradrenergic and serotonergic systems, the cortical DA innervation [first demonstrated in the rodent in the early 1970s (see references in Ref. 1)] has long been thought to be restricted to just a few cortical areas. The area that has attracted most study has been the prefrontal TINS, Vol. 14, No. I, 1991

cortex, because of its role in cognitive functions ~. However, we have provided evidence recently for a more extensive and heterogeneous cortical DA projection in rodent than has been previously described3,4. Furthermore, we have demonstrated a considerable topographical extension and laminar redistribution of cortical DA projections in humans 5, in parallel with a modified peptidergic phenotype. This seems to confirm the evolutionary trend in primates that was suggested by previous observations on monkey6-8. The large interspecific differences that have been demonstrated for the cortical DA system underline the need to re-evaluate just how much the data obtained from rodents, with regard to functional chemoanatomy, retain a predictive value in primates. Other experimental models whose cortical organizations are closer to that of humans are probably needed to further characterize the role of DA in major disorders such as psychoses and Parkinson's disease.

B. Berger, P. Gaspar and C Verneyare at the INSERMU 106, B~timent de P~diatrie, H6pital Sa/p~tri~re, 47 BId de l'H6pital, 75651 ParisCedex 13, France.

Regional and laminar distribution of the cortical DA projections in primates and rodents The cortical DA input in monkeys6,7 and humans "~is expanded compared with that in rodents so that in the

© 1991, ElsevierScience PublishersLtd, (UK) 0166- 2236/91/$0200

21

v /'$ 7

$m

tern

o

ent

cin

m~

ent

6

J

cing pf ins

( ~ 4 (Pf) RAT

,--, i ~

3mm

~

5

MACAQUE

111

~

lOmm

A10-A9 Fig. 1. Schematic coronal sections of rat brain (1-3) and macaque brain (4-7) illustrated at different scales. Scale bars are respectively 3 mm and 10 mm. The graded shades represent topographical landmarks of DA innervation and not density differences. Many arguments (see text) support the hypothesis that there are two classes of cortical DA afferents in rat. These are indicated by distinct blue shades in the cortex and the ventral mesencephalon (only DA neurones projecting to the cerebral cortex are indicated in the A9 and A 10 DA cell groups). In the monkey, the expanded DA projections display a general bilaminar pattem (indicated by distinct blue shades) except in the agranular cortices (cing, sin, pm, ml, ins), which are diffusely innervated. Whether these DA projections in primates are provided by a single or by several different classes of DA neurones has not yet been investigated. Abbreviations: cing, cingulate cortex; ent, entorhinal cortex; ins, insula; m 1, primary motor cortex; par, parietal cortex; pf, prefrontal cortex; pm, premotor cortex; rs, retrosplenial cortex; sm, supplementary motor cortex; tem, temporal cortex; v, visual cortex; vl, primary visual cortex. rat, there are extensive cortical zones that receive minor or no DA projections3,4, whereas in primates these zones are innervated (Fig. 1). They include the motor, premotor and supplementary motor areas, densely innervated in primates, and the parietal, temporal and posterior cingulate cortices that are more lightly innervated, that is most of the primary sensorimotor and association areas. Other areas are equally innervated in rodent and primate either densely, as in the prefrontal, anterior cingulate, insular, pifiform, perirhinal and entorhinal cortices, or poorly, as in the visual areas 3,6,7,9. The laminar distribution of DA terminals also differs (Figs 1, 2, Table I). In primates, the molecular layer is a widespread laminar target, layer I being the most densely innervated lamina throughout the cortical mantle and virtually the only target in the primary visual cortex. In rodents, on the other hand, only the anterior cingulate cortex and, to a lesser extent, the entorhinal cortex receive a dense contingent of DA

22

terminals in the superficial layers I-III. In addition, there are regionally specific patterns of DA laminar distribution in the primate cortex. DA projections are widespread to all cortical layers in the agranular cortices, the motor and anterior cingulate cortices for instance. However, they display a characteristic bilaminar pattern (the largest number of terminals being in layers I and V-VI) in the granular somatosensory and association cortices, which have a layer IV (Fig. 2A). Consequently, the DA innervation is most dense in the motor and anterior cingulate cortex, and displays a decreasing density gradient rostralwards and caudalwards. These phylogenetic differences in the distribution of DA afferents are paralleled by a similar expansion and laminar rearrangement of DA receptors, especially those of the D1 subtype 1°-13. Neuronal targets

Several lines of evidence now indicate that DA-containing varicosities form conventional TINS, VoL 14, No. 1, 1991

synapses 9,~4. Immunocytochemical characterization of dopaminoceptive neurones at the light and electron microscope levels indicates a preferential input to pyramidal neurones. DA terminals in the frontal lobe of rat ~4,~5 (Fig. 3) and monkey ~6 form symmetrical synaptic contacts, preferentially with pyramidal cells on dendritic shafts and spines. A cAMP- and DAregulated phosphoprotein named DARPP-32, which is enriched in neurones beating D~ receptors ~7, also preferentially labels pyramidal neurones. In rodents, a large number of DARPP-32-positive neurones is observed in layer VI, and these correspond to the corticothalamic projecting output neurones ~8. In young monkeys, a high density of DARPP-32-1abelled pyramidal cells is observed, not only in layer VI but also in layer V of all cortical areas s (Fig. 4). This

suggests that DA might modulate primarily the vast array of cortico-subcortical output pyramidal neurones located in the deep cortical layers, as well as a small proportion of neurones in layers V-VI that participate in corticocortical projections. Interestingly, D~ receptors, mapped using the selective D~ antagonist raclopride, also accumulate in layer V ~°, and might be synergistically involved with D1 receptors ~9 in the modulatory effect of DA on these output pyramidal neurones. The labelled basal dendrites of the DARPP32-positive neurones and the prominent tufting of their apical dendrites in layer I form a bilaminar pattern that fits in well with the bilaminar distribution of DA terminals 5-7 and Dz receptors 1°-~3, particularly in the granular cortices (association and somatosensory areas). In these areas, DA terminals could thus

A

Area

Area

46

Area

4

5

Area

~ > ~ ~

+ #,'~:+~':::,:,"+" - { I I ",:~ . : " - - -

•

:,:-"-~,?:'.!::.i T . f : :: : - . - - : : . . ~:+,:.-,'.:: .L-:

. . . . .

.:. . .

.

"

.. " , .:' ."-'!..:, t " .

"+

+

!

.:...._,' +.].

,j.

,+

~:. : :

, . " -"

7" , "

.

~..--.+~

~

~

L+ :~: ~.~:~:'.4+.'~-;

.,.. ~.£.

~

:.

'~:

'" " " "

- -

:~,,:'. -

-

"' " ; "),:

";!., ~. ".:'~-'-

L

" -, ~:.. ".: • '"': • ~..:..1;3. ' .,.". ~ , . ; • . : /.. .. V\_..,...

IV

: . . :+ ' ~ . r~::': ~.-..i :-:~.'~-:+..:.~".,'"; -

+y-