The serotonergic innervation of the cerebral cortex in man and its changes in focal cortical dysplasia.

We present the morphology and the laminar distribution of the serotonin (5-hydroxytryptamine, 5-HT) innervation of the cerebral cortex of patients who underwent cortical resection for partial seizures. The limits of the resections were established by stereoelectroencephalography. The 5-HT innervation was mapped by using an antiserum anti-5-HT. Two patients had cryptogenic epilepsies and two others had seizures related to focal cortical dysplasia. 5-HT immunoreactive axons were morphologically heterogeneous and projected diffusely to the cerebral cortex with regional-specific densities. Two types of terminal axon were demonstrated. Type I had large and spherical (intensely immunoreactive) varicosities and was distributed sparsely with a characteristic predominance in the molecular layer. Type II had fine and pleiomorphic varicosities (granular or fusiform) and was distributed through all cortical layers. The distribution of the 5-HT innervation varied according to the different architectonic areas investigated. The granular cortical areas characterized by a highly developed layer IV (primary somatosensory, primary visual and prefrontal cortices) had the highest density of 5-HT-ir fibers distributed from layer I to layer V. The agranular primary motor cortex had the lowest density with fibers preferentially seen in layers I, IIIa and V-VI. The orbital cortex with a poorly defined layer IV had an intermediate density with a laminar repartition predominant in the supragranular layers. In patients with cryptogenic epilepsies, the brain epileptogenic tissue was histologically normal as well as the serotonergic innervation. In contrast, in patients with focal cortical dysplasia, the dysplastic epileptogenic tissue was characterized by a serotonergic hyperinnervation. In agreement with previous data in primates, we give morphological evidence for two morphologically distinct serotonergic subsystems and for regional specific densities in the human cerebral cortex. Moreover, we previously reported an altered pattern of the catecholaminergic innervation in the same dysplasia areas. All these results provide evidence that this development epileptogenic lesion involves several sets of neurons which may contribute to epileptogenic activity.