- Email: [email protected]
Regional Differences of Callosal Connections in the Granular Zones of the Primary Somatosensory Cortex in Rats
Brain Research Bulletin, Vol. 43, No. 3, pp. 341–347, 1997 Copyright q 1997 Elsevier Science Inc. Printed in the USA. All rights reserved 0361-9230/97 $17.00 / .00
PII S0361-9230(97)00018-X
Regional Differences of Callosal Connections in the Granular Zones of the Primary Somatosensory Cortex in Rats T. HAYAMA1 AND H. OGAWA Department of Physiology, Kumamoto University School of Medicine, Honjo 2-2-1, Kumamoto 860, Japan [Received 20 July 1996; Revised 16 December 1996; Accepted 23 December 1996] ABSTRACT: The primary somatosensory cortex (SI) in rats is cytoarchitectonically divided into three zones: the granular, peri-, and dysgranular zones. To examine callosal connections in the granular zone that bears representation of the body somatosensory map, the distribution of lectin-conjugated horseradish peroxidase labels was explored in the right SI after single or multiple injections into the granular zone of the left SI. After injections in the upper and lower jaw regions, many labeled cell bodies and dense terminal labeling were found in the regions homotopical to the injection sites. Both kinds of labels were densely seen in layers II–III and V, less densely in layer VI. Densely labeled terminals were also observed in layer I. In layer IV, many terminals and a few cell bodies were labeled in the septa, while labeling inside the barrels was sparse or absent. After injections in other regions, i.e., those representing the facial whiskers, fore- and hindlimbs, or trunk in the granular zone, labeled callosal cell bodies and terminals were sparse or absent, except in the septa of the posteromedial barrel subfield representing the facial whiskers. The results clearly show that the density of callosal connections in the granular zone differs in different subfields, and that at least the jaw regions in the granular zones of both hemispheres are directly interconnected, in contrast to the previous assumption that only the dysgranular zone mediates information transmission between the granular zones of both sides. Q 1997 Elsevier Science Inc.
or their absence in all subfields of the SI body map have been described only in rats. The rat SI is cytoarchitectonically divided into three zones: the granular (GZ), peri- (PGZ), and dys- (DGZ) granular zones [7]. The GZ is characterized by barrels in a well-developed layer IV, which receive inputs from the principal thalamic relay [8,19]; the other two zones have a less-developed layer IV and receive thalamic inputs from the posterior thalamic nucleus [23]. It has been repeatedly reported that the callosal cell bodies and terminals are confined to the DGZ, and were found to be sparse in the GZ in which the somatotopical body map was depicted [1,30,34]. However, some callosal interconnections were found in the septa between the barrels representing the facial vibrissae [30]. Bilateral projections of the inferior dental nerve [14] and facial vibrissae [31] to the subfields of the GZ in rat SI have also been reported. Furthermore, the ipsilateral projections from the vibrissae were shown to be mediated through callosal fibers [31], suggesting the existence of callosal connections in the GZ. On this basis, we reexamined the callosal connections of the GZ in the rat SI with tract tracing, and disclosed differences in the density of the callosal connections in the different SI subfields. MATERIALS AND METHODS
KEY WORDS: Barrel, Septum, WGA-HRP, Cytochrome oxidase histochemistry.
Animals Animals were handled in strict adherence to the guidelines for animal treatment issued by our institution and by the Physiological Society of Japan. Eighteen Sprague–Dawley albino rats (200–350 g), male or female, were anesthetized with amobarbital sodium (80 mg/kg) and mounted on a stereotaxic instrument. Part of the parieto-temporal bone was removed with a dental drill to expose the SI on the left side.
INTRODUCTION In many species of mammals, the primary somatosensory cortex (SI) has a precise somatotopical map where, in general, contralateral body surface and bilateral face are represented [29,36]. The corpus callosum connects the SI cortices of both hemispheres. The relationship of callosal connections to subfields in the SI body map has been studied in many mammalian species, such as monkeys [4,22], rabbits [25], rats [1,30,34], mice [37], squirrels [13], cats [3,11,21,27], and raccoons [11,15]. In general, the midline or axial portions of the body have been reported to have callosal connections; for other body portions, some species differences have been reported. Sparse callosal connections 1
Tracer Injection Wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP; ca 4%; Toyobo) was dissolved in a 0.1 M KCl and 0.05 M Tris-buffer (pH 8.0) solution. A glass micropipette, with a tip diameter of 10–20 mm, containing the tracer, was positioned in appropriate SI subregions with the aid of the SI somatosensory
To whom requests for reprints should be addressed.
341 / 2a3c 2367 Mp 341 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
342
HAYAMA AND OGAWA
map of Chapin and Lin [6]. The pipette was inserted through a tiny incision in the dura, roughly perpendicular to the cortical surface, to a cortical depth of 0.5–0.7 mm. The tracer solution was injected by a pressure-injection system [32] in a total of 49 sites in the 18 rats. The volume of injected solution was 10–20 nl (45 sites) or 40–200 nl (4 sites). Five rats received a single injection, and the remaining 13 rats received multiple injections at two to six sites in SI. Histology After a survival period of 24–28 h, the animals were deeply anesthetized by an overdose of amobarbital sodium and perfused through the heart with ca. 50 ml of a cold phosphate buffer (0.1 M, pH 7.3) followed by 200–300 ml of a cold phosphate-buffered 10% formalin solution. After removal from the skull, the brains were kept for 4–12 h in a cold phosphate buffer containing 20% sucrose without postfixation. The left cerebral hemispheres receiving tracer injections were removed, flattened between two slide glasses, and soaked in the sucrose solution for an additional 4–24 h. The flattened tissue was cut tangentially (40–50 mm) on a freezing microtome. The right cerebral hemispheres were processed in the same way for 12 rats, and cut coronally (50 mm) for the remaining 6 rats. For both tangential and coronal preparations, alternate brain sections were treated for peroxidase activity following Mesulam’s protocol [28] or for cytochrome oxidase (CO) activity [35]. Coronal sections processed for HRP histochemistry were subsequently counterstained with thionin. Brain sections were examined after a conventional dehydration procedure. Data Analysis Patch-like distribution patterns of CO activity, distribution of the reaction products of HRP histochemistry, and distribution of many brain vessels were projected and traced on paper. HRP distribution was further examined with a light microscope using bright-field or polarized optics. The locus of the injection centers was superimposed onto CO activity in a flattened cortex, using the loci of blood vessels as reference landmarks. The diameter of the injection sites tended to be larger with more superficial brain sections. To compare the range of different injections, the diameters of the injection sites were measured at the level of layer IV on a brain section prepared for HRP activity. Because of the different shrinkage rate between brain tissues prepared for HRP reaction and those prepared for CO activity, the measured values were normalized to the size on the preparations for CO activity, using rates of distance between blood vessels in the neighboring brain preparations. Individual injections were pooled into ‘‘large’’ ( ¢1.5 mm), ‘‘medium’’ ( ¢1.0 mm, and õ1.5 mm), and ‘‘small’’ ( õ1.0 mm) groups based on the normalized diameter. Unavoidably, the injection sites included areas in the septa, often the PGZs surrounding the GZs, and sometimes the DGZs intercalating the GZs and regions outside the SI. Each injection site was classified with a body region in the SI map on which a larger area of injection sites was located. After tracer injections in the left SI, cell bodies and terminals were labeled in the homotopical regions in the right hemisphere. Label density in the right SI was classified into three groups: dense (clearly visible with the naked eye), moderate (clearly recognizable with a microscope), and sparse (quite sparse at the microscopic examination). Transported HRP was recognized in other portions in the ipsi- and contralateral cortices, i.e., motor regions and regions lateral to the SI, which probably included the secondary somatosensory area, the parietal ventral somato-
sensory area, and the parietal rhinal area, as defined by Fabri and Burton [12]. However, we did not describe them in this article. RESULTS Somatosensory Body Map in SI The body map of the rat SI has been repeatedly identified in the tangential plane of layer IV, the granular layer in the cortex [6,12,24]. A typical tangential distribution of CO activity in layer IV of a flattened cortex is shown in Fig. 1, labeled after Chapin and Lin [6]. Patches of high CO activity, corresponding to the barrels in each subfield, were evident in the whisker and upper and lower jaw regions, and less evident in the forelimb and hindlimb regions. The trunk region displayed a lower CO activity and its boundaries were, therefore, less evident. Difference in the Density of Callosal Connections Among the Subfields Table 1 summarizes the experimental cases. All of the cases were divided into two groups: (1) those receiving small to moderate tracer injections (cases 1–13 in Table 1), and (2) those receiving large ones (cases 14–18). In the 13 animals of group 1, 36 injection sites covered all subfields of the GZ in SI. As shown in Table 1, injections into the lower jaw (LJ) and upper jaw (UJ) regions produced dense labeling in the homotopical portions in the GZ of the opposite hemisphere. However, after injections into the other regions, dense labeling was not observed, clearly showing the regional difference of callosal connections in the GZ in SI. Two rats received injections in the LJ region (cases 1 and 2). In case 1, the animal received a medium-sized injection in the central portion and a small one in the caudomedial portion. In the contralateral SI, we found dense labeling in the central portion of the LJ region
FIG. 1. SI somatosensory map revealed by the distribution of cytochrome oxidase activity in a tangentially sectioned right hemisphere. Terms indicating regions in the map are derived from Chapin and Lin [6]. FBP, front buccal pad; FL, forelimb; HL, hindlimb; LJ, lower jaw; N, nose; RV, rostral small vibrissae; T, trunk; W, facial whisker. The term upper jaw in the present study includes FBP, N, and RV. Arrows r and l point to rostral and lateral directions. Scale bar, 1 mm.
/ 2a3c 2367 Mp 342 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
CALLOSAL CONNECTIONS OF THE RAT SI
343 TABLE 1
SUMMARY OF ALL EXPERIMENTAL CASES Injection Sites in Left SI
Callosal Label in Right SI
Case No.
s
Summary of all 18 experimental cases. Circles show the number, site, and diameter of tracer injections in the left SI. Injections were classified by diameter into small ( õ1.0 mm), medium ( ¢1.0 mm and õ1.5 mm), and large ( ¢1.5 mm), denoted by different-sized circles. Star size corresponds to the density of callosal labels in the GZ homotopical to the injection sites; large and small filled stars denote dense labeling visible with the naked eye and moderate labeling recognizable with a microscope. Small unfilled stars denote sparse labeling or none at all. LJ, lower jaw; UJ, upper jaw; W, facial whisker; T, trunk; FL, forelimb; HL, hindlimb. ‘‘Others’’ means the DGZs or SII. The injections, areas of which did not include any portions of the GZ, were classified with ‘‘Others.’’ Letters ‘‘t’’ and ‘‘c’’ in the column ‘‘s’’ indicate that the right hemisphere was sectioned tangentially or coronally. The amount of tracer in each injection was ca. 200 nl for one of the two sites of case 17 and for case 18; it was ca. 50 nl for cases 11 and 14, and ca. 15 nl for other injection sites. Survival time of the animal was 36–48 h for cases 16, 17, and 18, and 24–29 h for other cases.
but no labeling in the caudomedial portion. In case 2, after an injection into the central portion of the LJ region, moderate density of labeling was found in the homotopical portion of the contralateral SI. Four rats received injections in the UJ region (cases 2, 3, 4, and 5); the eight injection sites covered whole UJ region except for a medial portion of the nose region. In the contralateral SI, we saw dense labeling in the region homotopical to the injection sites. The five animals in group 2 received a total of 13 injections. The difference in the density of callosal connections in the granular subfields was similar to that found in group 1, except for the UJ region. In cases 15 and 16, three injection sites were located on a medial portion of the nose region, into which no injections were made in the animals of group 1. Sparse labeling, or none at all, was observed in the homotopical portion of the contralateral SI. In case 18, the injection site covered the whole area of the LJ region. In the contralateral hemisphere, dense labeling was found
in the whole LJ area except for the caudomedial portion where the labeling was less dense. Labeling Patterns in Each Subfield LJ and UJ subfields. Callosal labeling was found in the tangential preparations for the LJ (cases 1, 2, and 18) and UJ regions (cases 2 and 4), and in the coronal preparations for the LJ (case 14) and UJ regions (cases 3, 5, and 17). In all four tangential preparations, a similar labeling pattern was observed in both regions except for the LJ region of case 2. Labeled cell bodies and terminals were found in both supra- and infragranular layers, and were denser in the former layers. In layer IV, the granular layer, honeycomb-like label patterns were seen; these labels were located in the septa because the pattern was complementary to the patch-like patterns of CO activity in a neighboring tangential section. In the septa many terminals and a few cell bodies were
/ 2a3c 2367 Mp 343 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
344
HAYAMA AND OGAWA
labeled, whereas labels inside the barrels were quite sparse. In the LJ region of case 2 we saw labels only in the supragranular layer. Figure 2 shows an example of the tangential distribution of labels in the UJ region of the right hemisphere after multiple tracer injections into the homotopical region (case 4). In the coronal preparations (cases 3, 5, 14, and 17), we observed labeled axons traveling through the corpus callosum and the external capsule, which entered the cortex at the site homotopical to the injection site in the opposite hemisphere. In the cortex, labeled cell bodies and terminals were seen in almost all layers. They were dense in layers II–III and in both the superficial and deep portions of layer V, and less dense in layer VI. Labeled terminals were also densely distributed in layer I. In layer IV, a label pattern reflecting the barrel-septum organization was clearly found in caudal portions of the UJ region, i.e., the regions of rostral small vibrissae and lateral portions of the nose region. In the septa, many terminals and a few cell bodies were labeled; inside the barrels, labels were sparse or absent. In the LJ and rostral UJ regions, the label pattern of the barrel-septum organization was not clear. The barrel-septum structure was also less evident in CO preparations, probably because the axes of the barrels were not parallel to the coronal plane in rostral portions of the cortex. Figure 3 shows laminar distributions of callosal cell bodies and terminals in two experiments in which single and multiple injections were made into the LJ (case 14) and UJ (case 5) regions, respectively. Other subfields. Moderate labeling of callosal cell bodies and terminals was found in the whisker region in the tangential preparations (cases 2, 6, 15, and 16). Labeling patterns reflected the barrel-septum organization in layer IV. Sparsely and moderately labeled cell bodies and terminals were confined to the septa in honeycomb-like pattern with denser labels between barrel rows than between barrel numbers. The labels were continuous with those in the supra- and infragranular layers, where the honeycomb-like label patterns were less evident or absent. In the trunk, forelimb, and hindlimb regions, labels were quite sparse or completely lacking. In two cases, labels slightly invaded the supragranular space of the GZ from the surrounding PGZ or DGZ in the hindlimb (case 16) and trunk (case 9) regions. DISCUSSION Regional Differences of Callosal Connections in the Granular SI Previous researchers [1,30,34], tracing terminal degeneration or transport of HRP and amino acids, concluded that the GZs in the rat SI do not have callosal inputs across all layers, while the DGZs intercalating or surrounding the GZs [1,30,34], and the septa between the barrels [30], both have dense callosal connections. Ivy et al. [18] reported a wide distribution of callosal cell bodies in layer Va in the rat GZ. However, the authors [18] did not specifically mention where those neurons were distributed in the SI somatosensory map. The present study, using neuronal tracing with WGA-HRP, disclosed regional differences in the density of the callosal connections in the GZ of SI: the connections were dense in the UJ and LJ regions, and were sparse in the other regions, as summarized in Fig. 4. In the jaw regions, the connections were homotopic and reciprocal. The callosal cell bodies and terminal fields were distributed in almost all of the cortical layers, but mainly in layers II–III and V, and moderately in layer VI and in the septa of layer IV. In layer I, dense callosal terminals, but not callosal cell bodies, were found. Discrepancies between the previous and the present results may be due to differences in the employed neuronal tracing methods: Wise and Jones [34] and Akers and Killackey [1] concluded
FIG. 2. Tangential distribution of anterograde and retrograde labels in the UJ region of the primary somatosensory cortex in the right hemisphere after multiple injections of WGA-HRP into the homotopical region of the left hemisphere ( case 4 ) . ( A ) Tangential distribution of barrels in the rostrolateral portion of the SI revealed by cytochrome oxidase histochemistry at the level of layer IV. ( B ) Label distribution in the supragranular layer, 50 mm superficial to the section shown in A. ( C ) Label distribution at the level of layer IV, 50 mm deeper than in ( A ) . ( D ) Label distribution in the infragranular layer, 200 mm deeper than in ( A ) . Photographs B, C, and D were taken with polarized optics. For abbreviations, see legend to Fig. 1. Arrowheads in A to D show corresponding blood vessels. Scales, 1 mm. The scale in B applies to C and D.
that the GZ had an acallosal nature using a degeneration method, which is less sensitive than WGA-HRP tracing. This may also account for their previous negative finding of callosal connections in the septa [1,34], which were instead detected in the pres-
/ 2a3c 2367 Mp 344 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
CALLOSAL CONNECTIONS OF THE RAT SI
345
FIG. 3. Laminar distribution of callosal cell bodies of origin and terminals. The upper figure in A shows the injection sites of two experiments (cases 5 and 14); the lower figure sketches the rostrocaudal levels of the coronal sections, shown in B and C, of the opposite hemisphere. (B) Coronal label distributions in the ‘‘b’’ level of the lower figure in A after tracer injections in the UJ jaw region shown by ‘‘b’’ in the upper figure in A. (C) Coronal label distributions in the ‘‘c’’ level of the lower figure in A after tracer injections in the LJ region shown by ‘‘c’’ in the upper figure in A. Consecutive sections rostral to C had denser labels in the supragranular layer.
ent study. Olavarria et al. [30] also concluded that the GZ had a generally acallosal nature by observing the tangential distribution of callosal labels in flattened cortex preparations with CO and HRP histochemistry, a method similar to ours. However, close inspection of their Fig. 1 reveals that the callosal connections are denser in the UJ and LJ regions than in other subfields of the GZ. However, these authors [30] stressed the generally acallosal nature of the GZ and did not specifically mention regional differences among its subfields.
Pathways Linking the SI of Both Hemispheres Akers and Killackey [1] proposed a scheme of somatosensory information flow between the rat SI of both hemispheres based on findings that the DGZ was connected with callosal fibers while the GZ was not [1,34]. Sensory messages conveyed from the specific thalamic nuclei to layer IV of the GZ in SI [8,19] should have been, therefore, transmitted indirectly through the DGZs to the contralateral SI. Chapin et al. [7], examining the connections of cytoarchitectonic SI subfields, disclosed the information flow from the GZ to the DGZ through the PGZ. The present finding of dense callosal connections in the jaw regions of the GZ indicates that such a scheme [6,33] cannot be applied to all the subfields of the GZ, and that another pathway of interhemispheric interaction, at least for the jaw regions, complements the indirect pathway through the DGZ. Somatosensory information, transmitted from the thalamus to the barrel portions of layer IV [8,19], can thus be conveyed to the supra-, and infragranular layers and to the septum area [2,10], and can then be directly transferred to the GZ of the homotopical SI in the opposite hemisphere. The Midline Rule for Callosal Connections
FIG. 4. Regional difference in the density of reciprocal callosal connections in the granular zone of the primary somatosensory cortex. Shaded regions had denser connections than other regions. For abbreviations, see legend to Fig. 1.
Studies on the visual system have revealed that the corpus callosum connects border regions between 17 and 18 areas of two cerebral hemispheres, which represent the visual field lying close to the vertical meridian [5,9,11,16,26]. These callosal connections have been considered to have the function of fusing images of the right and left visual hemifields, known as the ‘‘midline rule’’ or ‘‘midline fusion theory’’ of callosal organization of the sensory system [17]. In the somatosensory system, the rela-
/ 2a3c 2367 Mp 345 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
346
HAYAMA AND OGAWA
tionship of callosal connections to the subfields of the somatosensory body map in SI has been examined in many animal species such as monkeys [4,22], rabbits [25], rats [1,30,34], mice [37], squirrels [13], cats [3,11,21,27], and raccoons [11,15]. All these species, except rats, were reported to have callosal connections at least in regions representing the midline or axial portions of the face and trunk. The callosal connection of the somatosensory system has also been explained on the basis of the ‘‘midline rule’’ as a neuronal substrate for providing a continuous somatosensory image of the right and left sides [13,17,21,25,27,37]. Because the generally acallosal nature of the GZs in the rat SI has been stressed [1,30,34], rats have hitherto represented the only exception to the midline rule [20]. However, the dense callosal connections in the jaw regions detected in the present study indicate that the midline rule is applicable to rats as well. Callosal connections were found also in the present study to be sparsely distributed in the whole trunk region of the rat SI, inconsistent with the midline rule. However, before drawing definite conclusions, some technical problems should be taken into account. The trunk region is relatively small and its boundaries are not clearly defined due to the much lower CO activity than in other subfields of the GZ. Previous researchers have drawn its boundaries differently, in particular those between the proximal portions of the forelimb and the ventral axis portion of the trunk [6,12,24]. In the present study, we could detect a slight invasion of callosal labels from the surrounding regions into the supragranular layer of the trunk region. Hence, any definite conclusions must be drawn only after experiments in which HRP and CO histochemistry are combined with electrophysiological mapping to disclose the configuration of the trunk region. Callosal Fibers May Mediate Ipsilateral Somatosensory Potential Pidoux and Verley [31] recorded somatosensory field potentials in the rat SI by mechanically stimulating bilateral facial vibrissae, and they concluded, by making unilateral lesions in SI, that the field potentials evoked ipsilaterally were mediated by callosal fibers. We previously reported bilateral projections to the LJ region, a subfield of the GZ of the rat SI from the inferior dental nerve [14], although pathways for the ipsilateral projections remain to be elucidated. The LJ region was shown in the present study to have dense callosal connections, while the posteromedial barrel subfield representing the facial vibrissae had sparse to moderate callosal connections ([30] and the present study). Because such sparse to moderate connections could be sufficient to mediate the field potentials evoked by stimulating the ipsilateral vibrissae, the dense callosal connections in the LJ region could represent a substrate for the pathway of SI field potentials evoked from the ipsilateral inferior dental nerve. ACKNOWLEDGEMENTS
The authors wish to express their thanks to Dr. A. Rosen for corrections of the English text.
REFERENCES 1. Akers, R. M.; Killackey, H. P. Organization of corticocortical connections in the parietal cortex of the rat. J. Comp. Neurol. 181:513– 538; 1978. 2. Armstrong–James, M.; Fox, K.; Das–Gupta, A. Flow of excitation within rat barrel cortex on striking a single vibrissa. J. Neurophysiol. 68:1345–1358; 1992.
3. Barbaresi, P.; Fabri, M.; Conti, F.; Manzoni, T. D-[3H]Aspartate retrograde labelling of callosal and association neurones of somatosensory areas I and II of cats. J. Comp. Neurol. 263:159–178; 1987. 4. Beck, P. D.; Kaas, J. H. Interhemispheric connections in neonatal owl monkeys (Aotus trivirgatus) and Galagos ( Galago crassicaudatus). Brain Res. 651:57–75; 1994. 5. Berlucchi, G.; Gazzaniga, M. S.; Rizzolatti, G. Microelectrode analysis of transfer of visual information by the corpus callosum. Arch. Ital. Biol. 105:583–596; 1967. 6. Chapin, J. K.; Lin, C.-S. The somatic sensory cortex of the rat. In: Kolb, B.; Tees, C. R., eds. The cerebral cortex of the rat. Cambridge, MA: MIT Press; 1990:341–380. 7. Chapin, J. K.; Sadeq, M.; Guise, J. L. U. Corticocortical connections within the primary somatosensory cortex of the rat. J. Comp. Neurol. 263:326–346; 1987. 8. Chmielowska, J.; Carvell, G.; Simons, D. J. Spatial organization of thalamocortical and corticothalamic projection systems in the rat SmI barrel cortex. J. Comp. Neurol. 285:325–339; 1989. 9. Choudhury, B. P.; Whitteridge, D.; Wilson, M. E. The function of the callosal connections of the visual cortex. Q. J. Exp. Physiol. 50:214–219; 1965. 10. Diamond, M. E.; Armstrong–James, M.; Ebner, F. Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus. J. Comp. Neurol. 318:462–476; 1992. 11. Ebner, F. F.; Myers, R. E. Distribution of corpus callosum and anterior commissure in cat and raccoon. J. Comp. Neurol. 124:353– 366; 1965. 12. Fabri, M.; Burton, H. Ipsilateral cortical connections of primary somatic sensory cortex in rats. J. Comp. Neurol. 311:405–423; 1991. 13. Gould, H. J.; Kaas, J. H. The distribution of commissural terminations in somatosensory areas I and II of the grey squirrel. J. Comp. Neurol. 196:489–504; 1981. 14. Hayama, T.; Hashimoto, K.; Ogawa, H. Projection of the inferior dental nerve to the primary somatosensory cortex in rats. Neurosci. Lett. 164:13–16; 1993. 15. Herron, P.; Johnson, J. I. Organization of intracortical and commissural connections in somatosensory cortical areas I and II in the raccoon. J. Comp. Neurol. 257:359–371; 1987. 16. Hubel, D. H.; Wiesel, T. N. Cortical and callosal connections concerned with the vertical meridian of visual fields in the cat. J. Neurophysiol. 30:1561–1573; 1967. 17. Innocenti, G. M. General organization of callosal connections in the cerebral cortex. In: Jones, E. G.; Peters, A., eds. Cerebral cortex, vol. 5. New York: Plenum Press; 1986:291–353. 18. Ivy, G. O.; Gould, III, H. J.; Killackey, H. P. Variability in the distribution of callosal projection neurons in the adult rat parietal cortex. Brain Res. 306:53–61; 1984. 19. Jensen, K. F.; Killackey, H. P. Terminal arbors of axons projecting to the somatosensory cortex of the adult rats. I. The normal morphology of specific thalamocortical afferents. J. Neurosci. 7:3529– 3543; 1987. 20. Johnson, J. I. Comparative development of somatic sensory cortex. In: Jones, E. G.; Peters, A., eds. Cerebral cortex vol. 8B. New York: Plenum Press; 1990:335–449. 21. Jones, E. G.; Powell, T. P. S. The commissural connexions of the somatic sensory cortex in the cat. J. Anat. 103:433–455; 1968. 22. Jones, E. G.; Powell, T. P. S. Connexions of the somatic sensory cortex of the rhesus monkey. II. Contralateral cortical connexions. Brain 92:717–730; 1969. 23. Koralek, K. A.; Jensen, K. F.; Killackey, H. P. Evidence for two complementary patterns of thalamic input to the rat somatosensory cortex. Brain Res. 463:346–351; 1988. 24. Koralek, K. A.; Olavarria, J.; Killackey, H. P. Areal and laminar organization of corticocortical projections in the rat somatosensory cortex. J. Comp. Neurol. 299:133–150; 1990. 25. Ledoux, M. S.; Whitworth, R. H.; Gould, H. G. Interhemispheric connections of the somatosensory cortex in the rabbit. J. Comp. Neurol. 258:145–157; 1987. 26. Lewis, J. W.; Olavarria, J. F. Two rules for callosal connectivity in striate cortex of the rat. J. Comp. Neurol. 361:119–137; 1995.
/ 2a3c 2367 Mp 346 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
CALLOSAL CONNECTIONS OF THE RAT SI 27. Manzoni, T.; Barbaresi, P.; Bellardinelli, E.; Caminiti, R. Callosal projections from the two body midlines. Exp. Brain Res. 39:1–9; 1980. 28. Mesulam, M. M. Tetramethyl benzidine for horseradish peroxidase neurohistochemistry: A non-carcinogenic blue reaction product with superior sensitivity for visualizing neuronal afferents and efferents. J. Histochem. Cytochem. 26:106–117; 1978. 29. Mountcastle, V. B. Central nervous mechanisms in mechanoreceptive sensibility. In: Darian–Smith, I., ed. Handbook of physiology. Sec I: Nervous system, vol. 3: Sensory processes, Part 2. Maryland: American Physiological Society; 1984:789–878. 30. Olavarria, J.; van Sluyters, R. C.; Killackey, H. P. Evidence for the complementary organization of callosal and thalamic connections within rat somatosensory cortex. Brain Res. 291:364 – 368; 1984. 31. Pidoux, B.; Verley, R. Projections on the cortical somatic I barrel subfield from ipsilateral vibrissae in adult rodents. Electroencephalogr. Clin. Neurophysiol. 46:715–726; 1979.
347 32. Saper, C. A simple inexpensive and reliable nanoliter syringe. Brain Res. Bull. 10:403–405; 1983. 33. Tracey, J. D.; Waite, P. M. E. Somatosensory system. In: Paxions, G., ed. Rat nervous system, 2nd ed. Sydney: Academic Press; 1995:689–704. 34. Wise, S. P.; Jones, E. G. The organization and postnatal development of the commissural projection of the rat somatic sensory cortex. J. Comp. Neurol. 168:313–344; 1976. 35. Wong–Riley, M. Changes in the visual system of monocularly sutured or enucleated cats demonstrable with cytochrome oxidase histochemistry. Brain Res. 171:11–28; 1979. 36. Woolsey, C. N.; Fairman, D. Contralateral, ipsilateral and bilateral representation of cutaneous receptors in somatic areas I and II of the cerebral cortex of pig, sheep and other mammals. Surgery 19:684– 702; 1946. 37. Yorke, C. H.; Caviness, V. S. Interhemispheric neocortical connections of the corpus callosum in the normal mouse. J. Comp. Neurol. 164:233–246; 1975.
/ 2a3c 2367 Mp 347 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
PII S0361-9230(97)00018-X
Regional Differences of Callosal Connections in the Granular Zones of the Primary Somatosensory Cortex in Rats T. HAYAMA1 AND H. OGAWA Department of Physiology, Kumamoto University School of Medicine, Honjo 2-2-1, Kumamoto 860, Japan [Received 20 July 1996; Revised 16 December 1996; Accepted 23 December 1996] ABSTRACT: The primary somatosensory cortex (SI) in rats is cytoarchitectonically divided into three zones: the granular, peri-, and dysgranular zones. To examine callosal connections in the granular zone that bears representation of the body somatosensory map, the distribution of lectin-conjugated horseradish peroxidase labels was explored in the right SI after single or multiple injections into the granular zone of the left SI. After injections in the upper and lower jaw regions, many labeled cell bodies and dense terminal labeling were found in the regions homotopical to the injection sites. Both kinds of labels were densely seen in layers II–III and V, less densely in layer VI. Densely labeled terminals were also observed in layer I. In layer IV, many terminals and a few cell bodies were labeled in the septa, while labeling inside the barrels was sparse or absent. After injections in other regions, i.e., those representing the facial whiskers, fore- and hindlimbs, or trunk in the granular zone, labeled callosal cell bodies and terminals were sparse or absent, except in the septa of the posteromedial barrel subfield representing the facial whiskers. The results clearly show that the density of callosal connections in the granular zone differs in different subfields, and that at least the jaw regions in the granular zones of both hemispheres are directly interconnected, in contrast to the previous assumption that only the dysgranular zone mediates information transmission between the granular zones of both sides. Q 1997 Elsevier Science Inc.
or their absence in all subfields of the SI body map have been described only in rats. The rat SI is cytoarchitectonically divided into three zones: the granular (GZ), peri- (PGZ), and dys- (DGZ) granular zones [7]. The GZ is characterized by barrels in a well-developed layer IV, which receive inputs from the principal thalamic relay [8,19]; the other two zones have a less-developed layer IV and receive thalamic inputs from the posterior thalamic nucleus [23]. It has been repeatedly reported that the callosal cell bodies and terminals are confined to the DGZ, and were found to be sparse in the GZ in which the somatotopical body map was depicted [1,30,34]. However, some callosal interconnections were found in the septa between the barrels representing the facial vibrissae [30]. Bilateral projections of the inferior dental nerve [14] and facial vibrissae [31] to the subfields of the GZ in rat SI have also been reported. Furthermore, the ipsilateral projections from the vibrissae were shown to be mediated through callosal fibers [31], suggesting the existence of callosal connections in the GZ. On this basis, we reexamined the callosal connections of the GZ in the rat SI with tract tracing, and disclosed differences in the density of the callosal connections in the different SI subfields. MATERIALS AND METHODS
KEY WORDS: Barrel, Septum, WGA-HRP, Cytochrome oxidase histochemistry.
Animals Animals were handled in strict adherence to the guidelines for animal treatment issued by our institution and by the Physiological Society of Japan. Eighteen Sprague–Dawley albino rats (200–350 g), male or female, were anesthetized with amobarbital sodium (80 mg/kg) and mounted on a stereotaxic instrument. Part of the parieto-temporal bone was removed with a dental drill to expose the SI on the left side.
INTRODUCTION In many species of mammals, the primary somatosensory cortex (SI) has a precise somatotopical map where, in general, contralateral body surface and bilateral face are represented [29,36]. The corpus callosum connects the SI cortices of both hemispheres. The relationship of callosal connections to subfields in the SI body map has been studied in many mammalian species, such as monkeys [4,22], rabbits [25], rats [1,30,34], mice [37], squirrels [13], cats [3,11,21,27], and raccoons [11,15]. In general, the midline or axial portions of the body have been reported to have callosal connections; for other body portions, some species differences have been reported. Sparse callosal connections 1
Tracer Injection Wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP; ca 4%; Toyobo) was dissolved in a 0.1 M KCl and 0.05 M Tris-buffer (pH 8.0) solution. A glass micropipette, with a tip diameter of 10–20 mm, containing the tracer, was positioned in appropriate SI subregions with the aid of the SI somatosensory
To whom requests for reprints should be addressed.
341 / 2a3c 2367 Mp 341 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
342
HAYAMA AND OGAWA
map of Chapin and Lin [6]. The pipette was inserted through a tiny incision in the dura, roughly perpendicular to the cortical surface, to a cortical depth of 0.5–0.7 mm. The tracer solution was injected by a pressure-injection system [32] in a total of 49 sites in the 18 rats. The volume of injected solution was 10–20 nl (45 sites) or 40–200 nl (4 sites). Five rats received a single injection, and the remaining 13 rats received multiple injections at two to six sites in SI. Histology After a survival period of 24–28 h, the animals were deeply anesthetized by an overdose of amobarbital sodium and perfused through the heart with ca. 50 ml of a cold phosphate buffer (0.1 M, pH 7.3) followed by 200–300 ml of a cold phosphate-buffered 10% formalin solution. After removal from the skull, the brains were kept for 4–12 h in a cold phosphate buffer containing 20% sucrose without postfixation. The left cerebral hemispheres receiving tracer injections were removed, flattened between two slide glasses, and soaked in the sucrose solution for an additional 4–24 h. The flattened tissue was cut tangentially (40–50 mm) on a freezing microtome. The right cerebral hemispheres were processed in the same way for 12 rats, and cut coronally (50 mm) for the remaining 6 rats. For both tangential and coronal preparations, alternate brain sections were treated for peroxidase activity following Mesulam’s protocol [28] or for cytochrome oxidase (CO) activity [35]. Coronal sections processed for HRP histochemistry were subsequently counterstained with thionin. Brain sections were examined after a conventional dehydration procedure. Data Analysis Patch-like distribution patterns of CO activity, distribution of the reaction products of HRP histochemistry, and distribution of many brain vessels were projected and traced on paper. HRP distribution was further examined with a light microscope using bright-field or polarized optics. The locus of the injection centers was superimposed onto CO activity in a flattened cortex, using the loci of blood vessels as reference landmarks. The diameter of the injection sites tended to be larger with more superficial brain sections. To compare the range of different injections, the diameters of the injection sites were measured at the level of layer IV on a brain section prepared for HRP activity. Because of the different shrinkage rate between brain tissues prepared for HRP reaction and those prepared for CO activity, the measured values were normalized to the size on the preparations for CO activity, using rates of distance between blood vessels in the neighboring brain preparations. Individual injections were pooled into ‘‘large’’ ( ¢1.5 mm), ‘‘medium’’ ( ¢1.0 mm, and õ1.5 mm), and ‘‘small’’ ( õ1.0 mm) groups based on the normalized diameter. Unavoidably, the injection sites included areas in the septa, often the PGZs surrounding the GZs, and sometimes the DGZs intercalating the GZs and regions outside the SI. Each injection site was classified with a body region in the SI map on which a larger area of injection sites was located. After tracer injections in the left SI, cell bodies and terminals were labeled in the homotopical regions in the right hemisphere. Label density in the right SI was classified into three groups: dense (clearly visible with the naked eye), moderate (clearly recognizable with a microscope), and sparse (quite sparse at the microscopic examination). Transported HRP was recognized in other portions in the ipsi- and contralateral cortices, i.e., motor regions and regions lateral to the SI, which probably included the secondary somatosensory area, the parietal ventral somato-
sensory area, and the parietal rhinal area, as defined by Fabri and Burton [12]. However, we did not describe them in this article. RESULTS Somatosensory Body Map in SI The body map of the rat SI has been repeatedly identified in the tangential plane of layer IV, the granular layer in the cortex [6,12,24]. A typical tangential distribution of CO activity in layer IV of a flattened cortex is shown in Fig. 1, labeled after Chapin and Lin [6]. Patches of high CO activity, corresponding to the barrels in each subfield, were evident in the whisker and upper and lower jaw regions, and less evident in the forelimb and hindlimb regions. The trunk region displayed a lower CO activity and its boundaries were, therefore, less evident. Difference in the Density of Callosal Connections Among the Subfields Table 1 summarizes the experimental cases. All of the cases were divided into two groups: (1) those receiving small to moderate tracer injections (cases 1–13 in Table 1), and (2) those receiving large ones (cases 14–18). In the 13 animals of group 1, 36 injection sites covered all subfields of the GZ in SI. As shown in Table 1, injections into the lower jaw (LJ) and upper jaw (UJ) regions produced dense labeling in the homotopical portions in the GZ of the opposite hemisphere. However, after injections into the other regions, dense labeling was not observed, clearly showing the regional difference of callosal connections in the GZ in SI. Two rats received injections in the LJ region (cases 1 and 2). In case 1, the animal received a medium-sized injection in the central portion and a small one in the caudomedial portion. In the contralateral SI, we found dense labeling in the central portion of the LJ region
FIG. 1. SI somatosensory map revealed by the distribution of cytochrome oxidase activity in a tangentially sectioned right hemisphere. Terms indicating regions in the map are derived from Chapin and Lin [6]. FBP, front buccal pad; FL, forelimb; HL, hindlimb; LJ, lower jaw; N, nose; RV, rostral small vibrissae; T, trunk; W, facial whisker. The term upper jaw in the present study includes FBP, N, and RV. Arrows r and l point to rostral and lateral directions. Scale bar, 1 mm.
/ 2a3c 2367 Mp 342 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
CALLOSAL CONNECTIONS OF THE RAT SI
343 TABLE 1
SUMMARY OF ALL EXPERIMENTAL CASES Injection Sites in Left SI
Callosal Label in Right SI
Case No.
s
Summary of all 18 experimental cases. Circles show the number, site, and diameter of tracer injections in the left SI. Injections were classified by diameter into small ( õ1.0 mm), medium ( ¢1.0 mm and õ1.5 mm), and large ( ¢1.5 mm), denoted by different-sized circles. Star size corresponds to the density of callosal labels in the GZ homotopical to the injection sites; large and small filled stars denote dense labeling visible with the naked eye and moderate labeling recognizable with a microscope. Small unfilled stars denote sparse labeling or none at all. LJ, lower jaw; UJ, upper jaw; W, facial whisker; T, trunk; FL, forelimb; HL, hindlimb. ‘‘Others’’ means the DGZs or SII. The injections, areas of which did not include any portions of the GZ, were classified with ‘‘Others.’’ Letters ‘‘t’’ and ‘‘c’’ in the column ‘‘s’’ indicate that the right hemisphere was sectioned tangentially or coronally. The amount of tracer in each injection was ca. 200 nl for one of the two sites of case 17 and for case 18; it was ca. 50 nl for cases 11 and 14, and ca. 15 nl for other injection sites. Survival time of the animal was 36–48 h for cases 16, 17, and 18, and 24–29 h for other cases.
but no labeling in the caudomedial portion. In case 2, after an injection into the central portion of the LJ region, moderate density of labeling was found in the homotopical portion of the contralateral SI. Four rats received injections in the UJ region (cases 2, 3, 4, and 5); the eight injection sites covered whole UJ region except for a medial portion of the nose region. In the contralateral SI, we saw dense labeling in the region homotopical to the injection sites. The five animals in group 2 received a total of 13 injections. The difference in the density of callosal connections in the granular subfields was similar to that found in group 1, except for the UJ region. In cases 15 and 16, three injection sites were located on a medial portion of the nose region, into which no injections were made in the animals of group 1. Sparse labeling, or none at all, was observed in the homotopical portion of the contralateral SI. In case 18, the injection site covered the whole area of the LJ region. In the contralateral hemisphere, dense labeling was found
in the whole LJ area except for the caudomedial portion where the labeling was less dense. Labeling Patterns in Each Subfield LJ and UJ subfields. Callosal labeling was found in the tangential preparations for the LJ (cases 1, 2, and 18) and UJ regions (cases 2 and 4), and in the coronal preparations for the LJ (case 14) and UJ regions (cases 3, 5, and 17). In all four tangential preparations, a similar labeling pattern was observed in both regions except for the LJ region of case 2. Labeled cell bodies and terminals were found in both supra- and infragranular layers, and were denser in the former layers. In layer IV, the granular layer, honeycomb-like label patterns were seen; these labels were located in the septa because the pattern was complementary to the patch-like patterns of CO activity in a neighboring tangential section. In the septa many terminals and a few cell bodies were
/ 2a3c 2367 Mp 343 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
344
HAYAMA AND OGAWA
labeled, whereas labels inside the barrels were quite sparse. In the LJ region of case 2 we saw labels only in the supragranular layer. Figure 2 shows an example of the tangential distribution of labels in the UJ region of the right hemisphere after multiple tracer injections into the homotopical region (case 4). In the coronal preparations (cases 3, 5, 14, and 17), we observed labeled axons traveling through the corpus callosum and the external capsule, which entered the cortex at the site homotopical to the injection site in the opposite hemisphere. In the cortex, labeled cell bodies and terminals were seen in almost all layers. They were dense in layers II–III and in both the superficial and deep portions of layer V, and less dense in layer VI. Labeled terminals were also densely distributed in layer I. In layer IV, a label pattern reflecting the barrel-septum organization was clearly found in caudal portions of the UJ region, i.e., the regions of rostral small vibrissae and lateral portions of the nose region. In the septa, many terminals and a few cell bodies were labeled; inside the barrels, labels were sparse or absent. In the LJ and rostral UJ regions, the label pattern of the barrel-septum organization was not clear. The barrel-septum structure was also less evident in CO preparations, probably because the axes of the barrels were not parallel to the coronal plane in rostral portions of the cortex. Figure 3 shows laminar distributions of callosal cell bodies and terminals in two experiments in which single and multiple injections were made into the LJ (case 14) and UJ (case 5) regions, respectively. Other subfields. Moderate labeling of callosal cell bodies and terminals was found in the whisker region in the tangential preparations (cases 2, 6, 15, and 16). Labeling patterns reflected the barrel-septum organization in layer IV. Sparsely and moderately labeled cell bodies and terminals were confined to the septa in honeycomb-like pattern with denser labels between barrel rows than between barrel numbers. The labels were continuous with those in the supra- and infragranular layers, where the honeycomb-like label patterns were less evident or absent. In the trunk, forelimb, and hindlimb regions, labels were quite sparse or completely lacking. In two cases, labels slightly invaded the supragranular space of the GZ from the surrounding PGZ or DGZ in the hindlimb (case 16) and trunk (case 9) regions. DISCUSSION Regional Differences of Callosal Connections in the Granular SI Previous researchers [1,30,34], tracing terminal degeneration or transport of HRP and amino acids, concluded that the GZs in the rat SI do not have callosal inputs across all layers, while the DGZs intercalating or surrounding the GZs [1,30,34], and the septa between the barrels [30], both have dense callosal connections. Ivy et al. [18] reported a wide distribution of callosal cell bodies in layer Va in the rat GZ. However, the authors [18] did not specifically mention where those neurons were distributed in the SI somatosensory map. The present study, using neuronal tracing with WGA-HRP, disclosed regional differences in the density of the callosal connections in the GZ of SI: the connections were dense in the UJ and LJ regions, and were sparse in the other regions, as summarized in Fig. 4. In the jaw regions, the connections were homotopic and reciprocal. The callosal cell bodies and terminal fields were distributed in almost all of the cortical layers, but mainly in layers II–III and V, and moderately in layer VI and in the septa of layer IV. In layer I, dense callosal terminals, but not callosal cell bodies, were found. Discrepancies between the previous and the present results may be due to differences in the employed neuronal tracing methods: Wise and Jones [34] and Akers and Killackey [1] concluded
FIG. 2. Tangential distribution of anterograde and retrograde labels in the UJ region of the primary somatosensory cortex in the right hemisphere after multiple injections of WGA-HRP into the homotopical region of the left hemisphere ( case 4 ) . ( A ) Tangential distribution of barrels in the rostrolateral portion of the SI revealed by cytochrome oxidase histochemistry at the level of layer IV. ( B ) Label distribution in the supragranular layer, 50 mm superficial to the section shown in A. ( C ) Label distribution at the level of layer IV, 50 mm deeper than in ( A ) . ( D ) Label distribution in the infragranular layer, 200 mm deeper than in ( A ) . Photographs B, C, and D were taken with polarized optics. For abbreviations, see legend to Fig. 1. Arrowheads in A to D show corresponding blood vessels. Scales, 1 mm. The scale in B applies to C and D.
that the GZ had an acallosal nature using a degeneration method, which is less sensitive than WGA-HRP tracing. This may also account for their previous negative finding of callosal connections in the septa [1,34], which were instead detected in the pres-
/ 2a3c 2367 Mp 344 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
CALLOSAL CONNECTIONS OF THE RAT SI
345
FIG. 3. Laminar distribution of callosal cell bodies of origin and terminals. The upper figure in A shows the injection sites of two experiments (cases 5 and 14); the lower figure sketches the rostrocaudal levels of the coronal sections, shown in B and C, of the opposite hemisphere. (B) Coronal label distributions in the ‘‘b’’ level of the lower figure in A after tracer injections in the UJ jaw region shown by ‘‘b’’ in the upper figure in A. (C) Coronal label distributions in the ‘‘c’’ level of the lower figure in A after tracer injections in the LJ region shown by ‘‘c’’ in the upper figure in A. Consecutive sections rostral to C had denser labels in the supragranular layer.
ent study. Olavarria et al. [30] also concluded that the GZ had a generally acallosal nature by observing the tangential distribution of callosal labels in flattened cortex preparations with CO and HRP histochemistry, a method similar to ours. However, close inspection of their Fig. 1 reveals that the callosal connections are denser in the UJ and LJ regions than in other subfields of the GZ. However, these authors [30] stressed the generally acallosal nature of the GZ and did not specifically mention regional differences among its subfields.
Pathways Linking the SI of Both Hemispheres Akers and Killackey [1] proposed a scheme of somatosensory information flow between the rat SI of both hemispheres based on findings that the DGZ was connected with callosal fibers while the GZ was not [1,34]. Sensory messages conveyed from the specific thalamic nuclei to layer IV of the GZ in SI [8,19] should have been, therefore, transmitted indirectly through the DGZs to the contralateral SI. Chapin et al. [7], examining the connections of cytoarchitectonic SI subfields, disclosed the information flow from the GZ to the DGZ through the PGZ. The present finding of dense callosal connections in the jaw regions of the GZ indicates that such a scheme [6,33] cannot be applied to all the subfields of the GZ, and that another pathway of interhemispheric interaction, at least for the jaw regions, complements the indirect pathway through the DGZ. Somatosensory information, transmitted from the thalamus to the barrel portions of layer IV [8,19], can thus be conveyed to the supra-, and infragranular layers and to the septum area [2,10], and can then be directly transferred to the GZ of the homotopical SI in the opposite hemisphere. The Midline Rule for Callosal Connections
FIG. 4. Regional difference in the density of reciprocal callosal connections in the granular zone of the primary somatosensory cortex. Shaded regions had denser connections than other regions. For abbreviations, see legend to Fig. 1.
Studies on the visual system have revealed that the corpus callosum connects border regions between 17 and 18 areas of two cerebral hemispheres, which represent the visual field lying close to the vertical meridian [5,9,11,16,26]. These callosal connections have been considered to have the function of fusing images of the right and left visual hemifields, known as the ‘‘midline rule’’ or ‘‘midline fusion theory’’ of callosal organization of the sensory system [17]. In the somatosensory system, the rela-
/ 2a3c 2367 Mp 345 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
346
HAYAMA AND OGAWA
tionship of callosal connections to the subfields of the somatosensory body map in SI has been examined in many animal species such as monkeys [4,22], rabbits [25], rats [1,30,34], mice [37], squirrels [13], cats [3,11,21,27], and raccoons [11,15]. All these species, except rats, were reported to have callosal connections at least in regions representing the midline or axial portions of the face and trunk. The callosal connection of the somatosensory system has also been explained on the basis of the ‘‘midline rule’’ as a neuronal substrate for providing a continuous somatosensory image of the right and left sides [13,17,21,25,27,37]. Because the generally acallosal nature of the GZs in the rat SI has been stressed [1,30,34], rats have hitherto represented the only exception to the midline rule [20]. However, the dense callosal connections in the jaw regions detected in the present study indicate that the midline rule is applicable to rats as well. Callosal connections were found also in the present study to be sparsely distributed in the whole trunk region of the rat SI, inconsistent with the midline rule. However, before drawing definite conclusions, some technical problems should be taken into account. The trunk region is relatively small and its boundaries are not clearly defined due to the much lower CO activity than in other subfields of the GZ. Previous researchers have drawn its boundaries differently, in particular those between the proximal portions of the forelimb and the ventral axis portion of the trunk [6,12,24]. In the present study, we could detect a slight invasion of callosal labels from the surrounding regions into the supragranular layer of the trunk region. Hence, any definite conclusions must be drawn only after experiments in which HRP and CO histochemistry are combined with electrophysiological mapping to disclose the configuration of the trunk region. Callosal Fibers May Mediate Ipsilateral Somatosensory Potential Pidoux and Verley [31] recorded somatosensory field potentials in the rat SI by mechanically stimulating bilateral facial vibrissae, and they concluded, by making unilateral lesions in SI, that the field potentials evoked ipsilaterally were mediated by callosal fibers. We previously reported bilateral projections to the LJ region, a subfield of the GZ of the rat SI from the inferior dental nerve [14], although pathways for the ipsilateral projections remain to be elucidated. The LJ region was shown in the present study to have dense callosal connections, while the posteromedial barrel subfield representing the facial vibrissae had sparse to moderate callosal connections ([30] and the present study). Because such sparse to moderate connections could be sufficient to mediate the field potentials evoked by stimulating the ipsilateral vibrissae, the dense callosal connections in the LJ region could represent a substrate for the pathway of SI field potentials evoked from the ipsilateral inferior dental nerve. ACKNOWLEDGEMENTS
The authors wish to express their thanks to Dr. A. Rosen for corrections of the English text.
REFERENCES 1. Akers, R. M.; Killackey, H. P. Organization of corticocortical connections in the parietal cortex of the rat. J. Comp. Neurol. 181:513– 538; 1978. 2. Armstrong–James, M.; Fox, K.; Das–Gupta, A. Flow of excitation within rat barrel cortex on striking a single vibrissa. J. Neurophysiol. 68:1345–1358; 1992.
3. Barbaresi, P.; Fabri, M.; Conti, F.; Manzoni, T. D-[3H]Aspartate retrograde labelling of callosal and association neurones of somatosensory areas I and II of cats. J. Comp. Neurol. 263:159–178; 1987. 4. Beck, P. D.; Kaas, J. H. Interhemispheric connections in neonatal owl monkeys (Aotus trivirgatus) and Galagos ( Galago crassicaudatus). Brain Res. 651:57–75; 1994. 5. Berlucchi, G.; Gazzaniga, M. S.; Rizzolatti, G. Microelectrode analysis of transfer of visual information by the corpus callosum. Arch. Ital. Biol. 105:583–596; 1967. 6. Chapin, J. K.; Lin, C.-S. The somatic sensory cortex of the rat. In: Kolb, B.; Tees, C. R., eds. The cerebral cortex of the rat. Cambridge, MA: MIT Press; 1990:341–380. 7. Chapin, J. K.; Sadeq, M.; Guise, J. L. U. Corticocortical connections within the primary somatosensory cortex of the rat. J. Comp. Neurol. 263:326–346; 1987. 8. Chmielowska, J.; Carvell, G.; Simons, D. J. Spatial organization of thalamocortical and corticothalamic projection systems in the rat SmI barrel cortex. J. Comp. Neurol. 285:325–339; 1989. 9. Choudhury, B. P.; Whitteridge, D.; Wilson, M. E. The function of the callosal connections of the visual cortex. Q. J. Exp. Physiol. 50:214–219; 1965. 10. Diamond, M. E.; Armstrong–James, M.; Ebner, F. Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus. J. Comp. Neurol. 318:462–476; 1992. 11. Ebner, F. F.; Myers, R. E. Distribution of corpus callosum and anterior commissure in cat and raccoon. J. Comp. Neurol. 124:353– 366; 1965. 12. Fabri, M.; Burton, H. Ipsilateral cortical connections of primary somatic sensory cortex in rats. J. Comp. Neurol. 311:405–423; 1991. 13. Gould, H. J.; Kaas, J. H. The distribution of commissural terminations in somatosensory areas I and II of the grey squirrel. J. Comp. Neurol. 196:489–504; 1981. 14. Hayama, T.; Hashimoto, K.; Ogawa, H. Projection of the inferior dental nerve to the primary somatosensory cortex in rats. Neurosci. Lett. 164:13–16; 1993. 15. Herron, P.; Johnson, J. I. Organization of intracortical and commissural connections in somatosensory cortical areas I and II in the raccoon. J. Comp. Neurol. 257:359–371; 1987. 16. Hubel, D. H.; Wiesel, T. N. Cortical and callosal connections concerned with the vertical meridian of visual fields in the cat. J. Neurophysiol. 30:1561–1573; 1967. 17. Innocenti, G. M. General organization of callosal connections in the cerebral cortex. In: Jones, E. G.; Peters, A., eds. Cerebral cortex, vol. 5. New York: Plenum Press; 1986:291–353. 18. Ivy, G. O.; Gould, III, H. J.; Killackey, H. P. Variability in the distribution of callosal projection neurons in the adult rat parietal cortex. Brain Res. 306:53–61; 1984. 19. Jensen, K. F.; Killackey, H. P. Terminal arbors of axons projecting to the somatosensory cortex of the adult rats. I. The normal morphology of specific thalamocortical afferents. J. Neurosci. 7:3529– 3543; 1987. 20. Johnson, J. I. Comparative development of somatic sensory cortex. In: Jones, E. G.; Peters, A., eds. Cerebral cortex vol. 8B. New York: Plenum Press; 1990:335–449. 21. Jones, E. G.; Powell, T. P. S. The commissural connexions of the somatic sensory cortex in the cat. J. Anat. 103:433–455; 1968. 22. Jones, E. G.; Powell, T. P. S. Connexions of the somatic sensory cortex of the rhesus monkey. II. Contralateral cortical connexions. Brain 92:717–730; 1969. 23. Koralek, K. A.; Jensen, K. F.; Killackey, H. P. Evidence for two complementary patterns of thalamic input to the rat somatosensory cortex. Brain Res. 463:346–351; 1988. 24. Koralek, K. A.; Olavarria, J.; Killackey, H. P. Areal and laminar organization of corticocortical projections in the rat somatosensory cortex. J. Comp. Neurol. 299:133–150; 1990. 25. Ledoux, M. S.; Whitworth, R. H.; Gould, H. G. Interhemispheric connections of the somatosensory cortex in the rabbit. J. Comp. Neurol. 258:145–157; 1987. 26. Lewis, J. W.; Olavarria, J. F. Two rules for callosal connectivity in striate cortex of the rat. J. Comp. Neurol. 361:119–137; 1995.
/ 2a3c 2367 Mp 346 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367
CALLOSAL CONNECTIONS OF THE RAT SI 27. Manzoni, T.; Barbaresi, P.; Bellardinelli, E.; Caminiti, R. Callosal projections from the two body midlines. Exp. Brain Res. 39:1–9; 1980. 28. Mesulam, M. M. Tetramethyl benzidine for horseradish peroxidase neurohistochemistry: A non-carcinogenic blue reaction product with superior sensitivity for visualizing neuronal afferents and efferents. J. Histochem. Cytochem. 26:106–117; 1978. 29. Mountcastle, V. B. Central nervous mechanisms in mechanoreceptive sensibility. In: Darian–Smith, I., ed. Handbook of physiology. Sec I: Nervous system, vol. 3: Sensory processes, Part 2. Maryland: American Physiological Society; 1984:789–878. 30. Olavarria, J.; van Sluyters, R. C.; Killackey, H. P. Evidence for the complementary organization of callosal and thalamic connections within rat somatosensory cortex. Brain Res. 291:364 – 368; 1984. 31. Pidoux, B.; Verley, R. Projections on the cortical somatic I barrel subfield from ipsilateral vibrissae in adult rodents. Electroencephalogr. Clin. Neurophysiol. 46:715–726; 1979.
347 32. Saper, C. A simple inexpensive and reliable nanoliter syringe. Brain Res. Bull. 10:403–405; 1983. 33. Tracey, J. D.; Waite, P. M. E. Somatosensory system. In: Paxions, G., ed. Rat nervous system, 2nd ed. Sydney: Academic Press; 1995:689–704. 34. Wise, S. P.; Jones, E. G. The organization and postnatal development of the commissural projection of the rat somatic sensory cortex. J. Comp. Neurol. 168:313–344; 1976. 35. Wong–Riley, M. Changes in the visual system of monocularly sutured or enucleated cats demonstrable with cytochrome oxidase histochemistry. Brain Res. 171:11–28; 1979. 36. Woolsey, C. N.; Fairman, D. Contralateral, ipsilateral and bilateral representation of cutaneous receptors in somatic areas I and II of the cerebral cortex of pig, sheep and other mammals. Surgery 19:684– 702; 1946. 37. Yorke, C. H.; Caviness, V. S. Interhemispheric neocortical connections of the corpus callosum in the normal mouse. J. Comp. Neurol. 164:233–246; 1975.
/ 2a3c 2367 Mp 347 Monday Jun 02 07:18 PM EL–BRB (v. 43, no. 4) 2367