Hippocampus and Parahippocampal Region

The hippocampus is part of the cerebral cortex. It is closely linked to a restricted number of related cortical areas, which are collectively referred to as the parahippocampal region. Both the hippocampus and parahippocampal region are reciprocally connected with a wide variety of higher order association cortices representing all sensory domains. From this recently emerged perspective it comes as no surprise that scientific conjectures about the functions of the (para)hippocampal network emphasize an important role in higher order cognitive functions, specifically learning and memory processes. The (para)hippocampal structures also have strong functional links with subcortical structures that researchers believe play a role in the selection of behavior in favor of either survival of the individual or of the species.

General Features of the (Para)Hippocampus

In order to describe any feature of a structure, either morphological or connectional, one needs to have a nomenclature. It is inevitable that opinions about nomenclature differ, so for the purposes of this essay differences in numbers of principal layers have been selected as the major criterion. All fields of the hippocampus exhibit a characteristic three-layered appearance. The parahippocampal region, in contrast to the hippocampus, comprises cortical regions, which show more than three, generally five or six, layers. The hippocampus consists of three major subdivisions: the dentate gyrus, the Ammon's horn (fields CA1, CA2, and CA3), and the subiculum. Note that some authors distinguish a prosubiculum from the subiculum proper. Also, readers may find texts in which reference is made to the subicular complex, consisting of (pro)subiculum, presubiculum, and parasubiculum. Based on connectional arguments as well as the criterion chosen for this discussion (number of layers), it is preferred to consider the presubiculum and parasubiculum as functionally different from the subiculum.

In all species the parahippocampal region comprises five main regions: the pre- and parasubiculum, the perirhinal and entorhinal cortices, and a fifth area that in primates is commonly referred to as the parahippocampal cortex. In nonprimates, the most likely homologue for the latter area is the so-called postrhinal cortex. Different researchers have subdivided each of these five regions in a variety of different ways.

All fields of the (para)hippocampus can be easily appreciated as longitudinal strips of cortex, neatly aligned one next to the other, beginning at the dentate gyrus on one end and the most outside portion of the parahippocampal region, in effect the perirhinal field, at the other end. The precise orientation and curvature of this entire cortical structure and thus its overall position in the brain may vary between different species. In species with a clearly developed temporal lobe (humans and monkeys, for example), the hippocampus is more ventrally and anteriorly located; in contrast, the rat's hippocampus looks more like a c-shaped structure positioned in the caudal third of the hemisphere. However, such difference in position does not alter the major characteristics and the topological relations between the hippocampus and the parahippocampal structures. Nor does it influence the fact that the most anterior/ventral portion of the hippocampus has a close spatial relationship with the amygdala.

Wiring of the (Para)Hippocampus

Most detailed information about the connectivity of the system comes from anatomical tracing studies in the rat, although for some pathways relevant detailed information has been collected in other species as well. Overall, the connectivity is rather conservative, such that a general non-species-specific description may suffice. Moreover, this "conserved" connectional organization does allow making inferences concerning the overall organization of the human (para)hippocampus. The hippocampus has two major pathways through which is connected to the rest of the brain. The first pathway makes use of the parahippocampal connectivity and predominantly mediates the connections with the cortex. The second pathway, which mainly, but not exclusively, links the hippocampus to subcortical structures, makes use of the fornix.

The Parahippocampal-Cortical System

The famous Spanish anatomist Ramón y Cajal, who provided one of the most detailed and earliest descriptions of the (para)hippocampal system, emphasized the so-called trisynaptic circuit, comprising an exclusive unidirectional pathway from the entorhinal cortex to the distal dendrites of the granular (simple pyramidal) cells in the dentate gyrus, which, in turn, give rise to the mossy fiber pathway to the proximal part of the apical dendrites of CA3 pyramidal cells. These CA3 neurons finally convey their output by way of the Schaffer axon collaterals to the apical and basal dendrites of pyramidal cells in CA1. The trisynatic circuit was once thought to be organized in restricted planes perpendicular to the longitudinal axis of the hippocampus, such that the structure as a whole was considered as a series of repetitive circuits, so-called lamellae, stacked together along the long axis. This proposal has stimulated functional analysis of the network using isolated brain slices containing these lamellar circuits. However, it also emphasized the relative isolation of the trisynaptic circuitry, which contrast to the notion that the hippocampal trisynaptic circuit has to be part of a larger neural system in order to function. Added details emphasize the overall longitudinal connectivity as an integral feature of the trisynaptic organization. Moreover, the entorhinal cortex is a complex cortical structure in itself that, in addition, has strong reciprocal connections with widespread association cortices, largely mediated by its neighboring parahippocampal fields. Finally, the subiculum has been added on a crucial position within this circuit.

Neurons in entorhinal layers II and III, which are the main recipients of the extensive cortical inputs, give rise to the already mentioned input to the hippocampus, the so-called perforant pathway. This name is derived from the traditional descriptions by Ramón y Cajal, who noted that fibers from the entorhinal cortex perforate the pyramidal-cell layer of the subiculum to gain access to the dentate gyrus molecular layer. The perforant pathway harbors two different projection systems. Layer II cells distribute their axons to most, if not all of the dentate gyrus, as originally described, but also to CA3. Cells in one particular subdivision of the entorhinal cortex, generally referred to as lateral entorhinal cortex, distribute their axons exclusively to the most distal portions of the dendrites of dentate and CA3 cells. The other subdivision, referred to as medial entorhinal cortex, sends its projection to the middle portions of the apical dendrite. This routing results in a specific laminar termination, such that each neuron receives both pathways but on different segment of its apical dendrite. The second pathway, which attracted more experimental attention only in the late twentieth century, originates from layer III cells and distributes to CA1 and the subiculum. In contrast to the laminar pattern as described for the lateral and medial layer II components, axons of layer III cells target only re stricted groups of the available neurons in CA1 and the subiculum. This targeting results in an organization such that the lateral entorhinal cortex disseminates its output only to the neurons clustered around the CA1-subiculum border, whereas the medial entorhinal cortex selectively innervates CA1 neurons close to the border with CA2/CA3 and subiculum neurons, which are close to the border with the presubiculum.

CA1 neurons project to the subiculum and, by doing so, add a fourth synapse to the originally defined trisynaptic circuit. These projections, which are again almost entirely unidirectional, show an interesting topographical organization along the transverse axis, which is reminiscent of that of the entorhinal terminal distribution. Neurons in CA1, close to the border with the subiculum, are connected to subicular neurons, which are similarly close to that border; these two connected populations are thus most likely innervated by entorhinal inputs from the lateral entorhinal cortex. In contrast, CA1 neurons closer to the border with CA3 will distribute their axons to subicular neurons close to the border with the presubiculum; these two connected populations thus most likely receive medial entorhinal inputs.

The lateral and medial entorhinal cortex may process functionally different types of information. The main cortical input of the lateral entorhinal cortex originates from the perirhinal cortex and olfactory cortices, whereas the medial entorhinal cortex receives strong inputs from the presubiculum as well as from the parahippocampal or postrhinal cortex. In view of the aforementioned anatomical organization researchers have suggested that in the hippocampus these two functionally different input streams converge at the level of the dentate gyrus and CA3, whereas they are kept more or less separate at the level of CA1 and the subiculum. This connectional differentiation taken in conjunction with the overall differences in intrinsic wiring between, for example, CA3 and CA1 suggests that the (para)hippocampal system harbors two systems that may be related to different memory processes.

Both CA1 and the subiculum constitute the major output structures of the hippocampus. They distribute strong projections back to the entorhinal cortex, mainly to its deep layers. The overall topographical organization of this projection is in register with that of the perforant pathway, such that a particular portion of the entorhinal cortex, projecting to restricted populations of neurons in CA1 and the subiculum, receive a return projection originating from these same neuronal groups in CA1 and the subiculum. This striking reciprocity, taken in conjunction with the aforementioned overall intrinsic hippocampal organization, may have important functional implications that researchers do not yet fully understand.

The Fornix

The fornix is a major fiber bundle that connects the hippocampus to the hypothalamus, in particular the mammillary bodies. The fornix originates from CA1 and subiculum, although parahippocampal cortices, in particular the pre- and parasubiculum and, to a much lesser extent, the entorhinal cortex, contribute fibers. On its way to the mammillary bodies, the fornix also issues fibers to the septal complex, the ventral striatum, and the amygdala. The fornix also carries the fibers from CA1 and subiculum targeting parts of the prefrontal cortex. The fornix is not a pure output pathway since projections from the septal complex to the hippocampus and in part to the entorhinal cortex travel by way of the fornix. These septal afferents provide the (para)hippocampus with most of its cholinergic inputs. The fornix also forms one of the input routes for the noradrenergic, serotonergic, and dopaminergic innervation. Additional aminergic fibers enter the parahippocampal region through a ventral route. Finally, the commissural connections between the left and right hippocampi also partially travel by way of the fornix.

The hippocampal connection to the mammillary bodies is part of the traditionally described limbic or Papez circuit, which includes the mammillo-thalamic tract connecting the mammillary bodies to the anterior complex of the thalamus, which in turn project to large portions of the limbic cortex, including anterior and posterior cingulate cortex, and pre- and parasubiculum. All these structures, in turn, provide input to the hippocampus, either directly, or indirectly by way of the entorhinal cortex.

Conclusion

The (para)hippocampus may be viewed as a cortical system with bidirectional connections with almost all of the multimodal associational domains of the cerebral cortex as well as a number of subcortical structures reportedly involved in motivation and selection of appropriate behaviors. Therefore, the (para)hippocampus may be at the crossroad of the cognitive and the affective side of behavior. Most functionally oriented research unfortunately addresses either of those sides, whereas a combined analysis of both is rare.

Bibliography

Amaral, D. G., and Witter, M. P. (1995). Hippocampal formation. In G. Paxinos, ed., The rat nervous system. San Diego, CA: Academic Press.

Burwell, R. D. (2000). The parahippocampal region: Corticocortical connectivity. Annals of the New York Academy of Sciences 911, 25-42.

Ramón y Cajal, S. (1911). Histologie du Système Nerveux de l'Homme et des Vertebrés. Maloine, Paris

Witter, M. P., Wouterlood, F. G., Naber, P. A., and van Haeften, T. (2000). Anatomical organization of the parahippocampalhippocampal network. Annals of the New York Academy of Sciences 911, 1-24.

Larry W.Swanson

Revised byMenno P.Witter