Subplate neuron ablation alters neurotrophin expression and ocular dominance column formation -- Data Supplement - Supplemental Results -- Proceedings of the National Academy of Sciences

Supplementary material for Lein et al. (1999) Proc. Natl. Acad. Sci. USA 96 (23), 13491-13495.

Results
Specificity of Cell Death After Kainate Injections.
Ghosh and Shatz (1, 2) have extensively characterized the consequences of kainate injections in cats during the first postnatal week. They have shown that injections of kainic acid into the subplate reliably lead to (i) loss of large cells in the subplate/white matter with preservation of normal laminar cytoarchitecture in the overlying cortex shortly after kainate injection, as demonstrated by using standard histochemical methods (Nissl stain); (ii) loss of immunohistochemical markers for neurons (microtubule-associated protein-2) in the subplate with the preservation of glial cell markers (glial fibrillary acidic protein and glial-specific vimentin); (iii) preservation of layer 4 neurons, the target layer for thalamocortical axons (demonstrated by ³H-thymidine birthdating); (iv) the survival of thalamic neurons whose axons project to the site of kainate injection (identified both by the retrograde transport of latex microspheres coinjected with kainate and by transneuronal transport of ³H-proline); and (v) the survival of other subcortical cell types, such as cholinergic afferents to visual cortex and astrocytes (identified by immunostaining with antibodies against choline-acetyltransferase and glial fibrillary acidic protein). These data indicate that this manipulation selectively ablates subplate neurons, while leaving cortical plate and subcortical structures such as thalamic and basal ganglion neurons intact.

We have further investigated the selectivity of subplate neuron ablation by assessing directly the extent of neuronal damage immediately after kainic acid injection with in situ end labeling (ISEL), a sensitive and early signal of cell death (3). On postnatal day 3 (P3), 24 hr after kainate injection, there are many labeled cells in both the lower subplate (white matter) and upper subplate (located at the base of the cortical plate; supplemental Fig. 4 A and B). In contrast, cortical neurons are largely unaffected, with the exception of some cells in deep layer 6, just above the upper subplate. After the neurogenesis of subplate neurons, layer 6 neurons are the next population of cortical neurons to be born (4). Hence at P7, of the neurons that will form the adult cortical plate, layer 6 neurons are the most mature. Thus, it is not surprising that some of the earliest generated neurons of layer 6 (those nearest the upper subplate) are also affected by kainate administration. Most importantly, however, the ISEL staining rarely labels layer 4 neurons, the major targets of thalamocortical axons. These ISEL data are direct confirmation that the effect of the kainate injection is restricted almost completely to the subplate region and spares layer 4 neurons.

To demonstrate that kainate administration ablates subplate neurons, we directly examined subplate neurons after kainate injection using the technique of cellular birthdating. 5-Bromodeoxyuridine (BrdU) administered on embryonic day 24 specifically labels many subplate neurons undergoing their final mitotic division (4). At P28, BrdU-labeled cells are present in the subplate in normal animals (supplemental Fig. 4E); however, at P28 after kainate injection on P7, very few labeled cells remain in the region surrounding the injection sites (supplemental Fig. 4F). Thus, kainate injection into the subplate during the first postnatal week leads to the nearly complete loss of subplate neurons.

Seizures Are a Secondary Result of Subplate Neuron Ablations.
The fact that systemic administration of kainic acid in mature animals leads to seizures, as well as to changes in all of the genes assayed in this study, raises the concern that the effects seen here on gene expression could be because of immediate and direct effects of kainic acid on cortical or subcortical neurons. We believe that both seizures and changes in gene expression are a secondary consequence of subplate neuron ablation on the basis of several points. First, neither seizures nor changes in gene expression nor abnormal ocular dominance columns (1, 2) are observed in control animals in which either saline was injected into the subplate (n = four animals) or in which kainic acid was injected directly into layer 4 rather than the subplate (n = three animals), consistent with the well known observation that immature cortical neurons are not susceptible to the excitatory effects of kainic acid (5). Second, seizures are unlikely to occur as a result of damage to subcortical structures such as thalamus and basal ganglia. These structures are intact after kainate injections (1, 2). (See also Specificity of Cell Death After Kainate Injections above.) The small volume of kainate injected, along with the coinjection of fluorescent latex microspheres, restricts the kainate to the posterior part of the visual cortex and makes it unlikely that kainate can diffuse from the injection sites in primary visual cortex to subcortical structures. Third, important differences exist between our observations and those made after kainic acid administration in the adult. Whereas kainate-induced seizures in the adult occur within 1 hr of administration (6-8), the seizures observed here are delayed in onset, occurring days after the immediate excitatory effects have dissipated. Moreover, the increase in cortical brain-derived neurotrophic factor (BDNF) mRNA levels after subplate neuron ablation is similarly delayed: no change is detected 24 hr postinjection (n = two animals), but change is evident after 4 days (n = two animals). In contrast, after kainic acid-induced seizures in adult animals, BDNF mRNA is elevated within 1-2 hr (6-8). Furthermore, kainate-induced seizures in the adult have a global effect on gene expression across many cortical areas (6-8), whereas the changes in gene expression seen here are restricted specifically to cortical regions neighboring the injection sites. Finally, it is unlikely that the observed seizures alone are responsible for the disruption of ocular dominance columns, because another method of altering cortical excitability with penicillin-induced seizures does not alter ocular dominance column plasticity at similar developmental times (9). These observations suggest that cortical excitability, gene expression changes, and ocular dominance column formation have been altered secondarily, as a consequence of subplate neuron removal and not as an immediate consequence of kainic acid acting directly on cortical or subcortical neurons. Correlation of GAD Up-Regulation with Seizures.
The up-regulation of GAD levels concurrently with seizure activity may seem paradoxical; however, there is a large literature correlating seizures with up-regulation of phenotypic markers for GABAergic neurons (for example, see ref. 10). Activity-dependent GAD expression may be a compensatory mechanism to counterbalance excessive cortical activity levels (11).

1. Ghosh, A. & Shatz, C. J. (1992) Science 255, 1441-1443.

2. Ghosh, A. & Shatz, C. J. (1994) J. Neurosci. 14, 3862-3880.

3. Gavrieli, Y., Sherman, Y. & Ben-Sasson, S. A. (1992) J. Cell Biol. 119, 493-501.

4. Luskin, M. B. & Shatz, C. J. (1985) J. Neurosci. 5, 1062-1075.

5. Coyle, J. T., Bird, S. J., Evans, R. H., Gulley, R. L., Nadler, J. V., Nicklas, W. J. & Olney, J. W. (1981) Neurosci. Res. Bull. 19, 331-427.

6. Garcia, M. L., Garcia, V. B., Isackson, P. J. & Windebank, A. J. (1997) NeuroReport 8, 1445-1449.

7. Kornblum, H. I., Sankar, R., Shin, D. H., Wasterlain, C. G. & Gall, C. M. (1997) Mol. Brain Res. 44, 219-228.

8. Dugich-Djordjevic, M. M., Tocco, G., Lapchak, P. A., Pasinetti, G. M., Najm, I., Baudry, M. & Hefti, F. (1992) Neuroscience 47, 303-315.

9. Videen, T. O., Daw, N. W. & Collins, R. C. (1986) Brain Res. 371, 1-8.

10. Liang, F. & Jones, E. G. (1997) J. Neurosci. 17, 2168-2180.

11. Sloviter, R. S., Dichter, M. A., Rachinsky, T. L., Dean, E., Goodman, J. H., Sollas, A. L. & Martin, D. L. (1996) J. Comp. Neurol. 373, 593-618.