Age-related loss of calcium binding proteins in rabbit hippocampus.

Using immunocytochemistry hippocampal levels of the calcium binding proteins calbindin 28K (CB) and parvalbumin (PV) was studied in young (1 month) to very old (60 month) Albino rabbits. Young (3 month) and senescent (30 month) Wistar rats were also examined to compare the distribution and age dependency of PV and CB in both species. The distribution of PV-ir is similar in the rabbit and rat hippocampus. Aging in both species yielded a small loss of PV-ir in axon terminals. The presence of CB-ir interneurons throughout the hippocampus, and the heavy investment of the dentate gyrus (DG) granular cells with CB-ir was also similar in both species. In rabbits, the number of CB-ir interneurons in the CA1, as well as the density of CB-ir in the DG decreased in the first year of life, and did not change between 12-48 months of age. A secondary reduction in the density of CB-ir in the DG was observed at ages beyond 48 months. A similar loss of CB-ir in the DG occurred in the rat. In the CA1, however, the density of CB-ir was similar in young and aged rats. Another remarkable finding was the total absence of CB-ir in CA1 pyramidal neurons of rabbits at any age. Thus, the distribution and age dependency of PV-ir in the hippocampus is similar in both species. The decline of CB-ir in the DG with advancing age is very prominent and may be related to an altered calcium homeostasis in these cells. However, the absence of CB-ir in the CA1 of rabbits makes a causal role for CB in the functional decline of CA1 pyramidal cells during aging unlikely.

Subicular efferents are organized mostly as parallel projections: a double-labeling, retrograde-tracing study in the rat.

To understand the functional relevance of the subiculum as a major distributor of hippocampally processed information, detailed information about its neuronal organization is necessary. A striking feature of the subiculum is that it can be divided into four different areas, each characterized by a specific set of efferent connections. To establish whether the different areas of the subiculum are similar with respect to the organization of the origin of their respective efferents, the double-fluorescence retrograde-tracing technique was used to study the degree of collateralization. Because CA1 gives rise to a major input to the subiculum but also projects to some of the targets reached by subicular projections, we compared the subicular degree of collateralization with that of CA1. Throughout CA1, the percentages of double-labeled cells were high, ranging from 17% to 39%. In contrast, the percentages of double-labeled cells in the subiculum were much lower, ranging from 0% to 12%, and no differences were noted between the four areas of the subiculum. This indicates that the four regions of the subiculum are organized in the same way with regard to the output connectivity. Because all four different regions of the subiculum share this paucity of collateralized projections, we conclude that subicular outputs generally originate as parallel projections. This characteristic organization is in line with a proposed function of the subiculum in information storage.

Age-dependent changes in the immunoreactivity for neurofilaments in rabbit hippocampus.

The distribution of the three subunits of neurofilaments was examined in the hippocampus of young adult rabbits (three months of age), employing a panel of six monoclonal antibodies. Thereafter, age-dependent and subunit-selective changes in neurofilament immunoreactivity in the ageing rabbit hippocampus were studied, using animals of one, three, six, 12, 24, 30, 36, 48, and 60 months. Principal cells, interneurons, axons, and various fibre systems were immunoreactive for all three subunits, although the localization and staining intensity of neurofilament immunoreactivity depended on the antibody used. Small cells immunopositive for the low subunit of neurofilament (presumably glial cells) were found abundantly in the hippocampal formation at one month, and (occasionally) at 30-36 months. Young rabbits (one to three months of age) had high numbers of interneurons stained for the high subunit of neurofilament in the stratum oriens/pyramidale. The number declined and plateaued to approximately 78% at six to 30 months, and further declined and plateaued to approximately 56% at 36-60 months. The first decline may reflect a process of maturation, while the latter decline most likely relates to ageing. Ageing pyramidal cells in 48-60 months animals revealed a slight increase for the low subunit of neurofilament, but no changes for the other subunits. Transient changes in neurofilament immunoreactivity were a striking observation in dentate gyrus granule cells during ageing. The staining intensity for the low subunit of neurofilament decreased gradually from one to 24-30 months until it was no longer detectable in these cells. The immunoreactivity then reappeared, most notably in granule cells lining the hilus, at the age of 36-48 months. By 60 months all granule cells were nearly immunonegative for this subunit. Axonal aberrations, immunoreactive for all three subunits, were found throughout the hippocampal formation. These aberrations first appeared in 24-month-old animals and increased in number and maximal size in older rabbits. The alterations in neurofilament immunoreactivity in the ageing hippocampus correlated with age-associated learning disabilities in the acquisition of a hippocampally-dependent learning task. The potential relevance of changes in the cytoskeletal profile of hippocampal neurons to age-associated learning and memory disabilities is discussed.

Networks of the hippocampal memory system of the rat. The pivotal role of the subiculum.

The hippocampal system, consisting of the hippocampus, subiculum, and adjacent parahippocampal region, is known to play an important role in learning and memory processes. It is also known that the originally proposed trisynaptic circuit is a simplified representation of the organization of this system. In this paper, we present evidence, both anatomically and electrophysiologically, for the existence of direct and indirect parallel pathways through the hippocampal memory system arising from the perirhinal and postrhinal cortex. These pathways form nested loops. The subiculum occupies a central position within these loops. In the subiculum, both "raw" and highly processed information will converge. Therefore, we propose that the subiculum occupies a pivotal position in the hippocampal memory system, both as recipient and comparator of signals and as a distributor of processed information.

Anatomical organization of the parahippocampal-hippocampal network.

The anatomical organization of the parahippocampal-hippocampal network indicates that it consists of different parallel circuits. Considering the topographical distribution of sensory cortical inputs, the hypothesis is that the major parallel circuits carry functionally different information. These functionally different parallel routes reach different portions of the hippocampal network along the longitudinal axis of all fields as well as along the perpendicularly oriented transverse axis of CA1 and the subiculum. In the remaining fields of the hippocampal formation, that is, the dentate gyrus and CA2/CA3, separation along the transverse axis is not present. By contrast, here the functionally different pathways converge onto the same neuronal population. The entorhinal cortex holds a pivotal position among the cortices that make up the parahippocampal region. By way of the networks of the superficial and deep layers, it mediates, respectively, the input and output streams of the hippocampal formation. Moreover, the intrinsic entorhinal network, particularly the interconnections between the deep and superficial layers, may mediate the comparison of hippocampal input and output signals. As such, the entorhinal cortex may form part of a novelty detection network. In addition, the organization of the entorhinal-hippocampal network may facilitate the holding of information. Finally, the terminal organization of the presubicular input to the medial entorhinal cortex indicates that the interactions between the deep and superficial entorhinal layers may be influenced by this input.

Parallel input to the hippocampal memory system through peri- and postrhinal cortices.

In the rat, the rhinal cortices consist of the perirhinal, postrhinal and entorhinal cortices. The perirhinal and postrhinal cortices, which serve as major input sources to the entorhinal cortex, receive functionally different types of information. In this study we looked at the projections from the perirhinal and postrhinal cortices to the different parts of the entorhinal cortices using an anterograde tracing technique. Our results show that the perirhinal cortex preferentially projects to the lateral entorhinal cortex, whereas the postrhinal cortex mainly sends fibers to the medial entorhinal cortex. Since the lateral and medial entorhinal cortices are differentially connected with the hippocampus, we suggest that functionally different types of information are processed in parallel in the hippocampal memory system.

Differential distribution of barrel or visual cortex. Evoked responses along the rostro-caudal axis of the peri- and postrhinal cortices.

The functional connections between the barrel cortex and visual cortex on the one hand and the perirhinal (PER) and postrhinal (POR) cortices on the other hand were investigated in the rat. Stimulation of the barrel cortex evoked field potentials throughout the longitudinal extent of both PER and POR. In contrast, visual cortex stimulation evoked responses only in the caudal portion of PER as well as in POR. Therefore, the information from the visual cortex on the way to the hippocampus is transferred preferentially by a relay in POR, whereas somatosensory information is transferred via both PER and POR. Moreover, stimulation of both cortical regions elicited firing of multiple units; however, unit activity was more commonly found in POR than in PER. We conclude that the transfer of somatosensory and visual information to the hippocampal formation is preferentially mediated by parallel channels through PER and POR respectively. Although the information transfer through these channels does overlap to some extent, each channel appears to have specific properties.

Perirhinal cortex input to the hippocampus in the rat: evidence for parallel pathways, both direct and indirect. A combined physiological and anatomical study.

The possibility of a direct projection from the perirhinal cortex (PER) to areas CA1 and subiculum (SUB) in the hippocampus has been suggested on the basis of tracer studies, but this projection has not unequivocally been supported by physiological studies. The demonstration of such a functional pathway might be important to understand the functioning of the hippocampal memory system. Here we present physiological and further anatomical evidence for such a connection between PER and the hippocampus. Electrical stimulation of PER in vivo evoked field potentials (EFPs) at the border area of CA1/SUB, consisting of a short latency and a longer latency component. Current source density analysis revealed that the sink of the short latency component was situated in the molecular layer of area CA1/SUB, while the longer latency component had its sink in the outer molecular layer of the dentate gyrus (DG). Anterograde tracer injections in PER showed labelled fibres in the border area of CA1/SUB, but anatomical evidence for a projection of PER to DG was not found. When synaptic transmission in the entorhinal cortex was partly blocked, the amplitude of the longer latency component of the recorded EFPs in the hippocampus was decreased while the short latency component was not affected, which suggests that the indirect pathway originating in PER is mediated through a synaptic relay in the entorhinal cortex. From the present results we conclude that information originating in PER reaches area CA1/SUB by parallel, direct and indirect, routes. The existence of this parallel organization appears to form an essential feature for the proper function of the medial temporal lobe memory system.

Evidence for a direct projection from the postrhinal cortex to the subiculum in the rat.

Behavioral data indicate that three of the areas which form the parahippocampal region in the rat, i.e., the entorhinal, perirhinal, and postrhinal cortices, have different, although related functions that also differ from those of the hippocampal formation. These functional differences might be related to differences in connectivity, on the one hand with parts of the association cortex, and on the other with the hippocampal formation. In a previous study, we showed the existence of both a direct and an indirect projection from the perirhinal cortex to areas CA1 and subiculum of the hippocampus. Here we present the result of a second study, demonstrating a similarly organized projection from the postrhinal cortex to the subiculum, comprising both a direct and an indirect route. Electrical stimulation of the postrhinal cortex in vivo evoked field potentials throughout the subiculum and the dentate gyrus. Current source density analysis in both the subiculum and dentate gyrus revealed the presence of sink-source pairs, indicative of a synaptic termination. Based on comparison with the sink-source pairs found after stimulation of the medial entorhinal cortex, we conclude that the connection between the postrhinal cortex and the dentate gyrus most likely is formed by a polysynaptic pathway mediated via the medial entorhinal cortex, while the pathway from the postrhinal cortex to the subiculum is likely monosynaptic. In order to substantiate these findings, we carried out several tracer experiments. Anterograde tracer injections in the postrhinal cortex resulted in labeled fibers in limited parts of the subiculum, but no anatomical evidence for a projection of the postrhinal cortex to the dentate gyrus was found. Additional retrograde tracer injections in the subiculum also showed evidence for a direct postrhinal-to-subiculum projection with a strong topological organization. Based on these combined anatomical and electrophysiological data, we conclude that the postrhinal cortex indeed can reach the subiculum via both a direct and an indirect pathway.

Reciprocal connections between the entorhinal cortex and hippocampal fields CA1 and the subiculum are in register with the projections from CA1 to the subiculum.

The topology of the connections between the entorhinal cortex (EC), area CA1, and the subiculum is characterized by selective and restricted origin and termination along the transverse or proximodistal axis of CA1 and the subiculum. In the present study, we analyzed whether neurons in CA1 and the subiculum that receive EC projections are interconnected and give rise to return projections to EC, such that they terminate deep in the area of origin of the EC-to-CA1/subiculum projections. Both for the lateral and medial subdivision of EC, the projections to CA1/subiculum, as well as the projections from CA1 to the subiculum and back to EC, are rather divergent. Interestingly, we only rarely observed evidence for the presence of "reentry loops," i.e., cells in layer III of EC giving rise to projections to interconnected neurons in CA1 and the subiculum, while the targeted CA1 neurons also projected back to the deep layers of the area of origin of the pathway in EC. We conclude that although fibers originating from a restricted part of EC distribute extensively in a divergent way along the longitudinal axis of CA1 and the subiculum, only restricted portions of the latter two areas, receiving inputs from the same entorhinal area, are interconnected. Moreover, only a small percentage of the CA1 neurons that project to the correspondingly innervated subicular neurons give rise to projections that return to the deep layers of the originating part of EC. The present findings are taken to indicate that the EC-hippocampal circuitry functionally comprises many parallel-organized specific "reentry loops."

Cortico-hippocampal communication by way of parallel parahippocampal-subicular pathways.

The hippocampal memory system, consisting of the hippocampal formation and the adjacent parahippocampal region, is known to play an important role in learning and memory processes. In recent years, evidence from a variety of experimental approaches indicates that each of the constituting fields of the hippocampal memory system may serve functionally different, yet complementary roles. Understanding the anatomical organization of cortico-parahippocampal-hippocampal connectivity may lead to a further understanding of these potential functional differences. In the present paper we present the two main conclusions of experiments in which we studied the anatomical organization of the hippocampal memory system of the rat in detail, with a focus on the pivotal position of the entorhinal cortex. We first conclude that the simple traditional view of the entorhinal cortex as simply the input and output structure of the hippocampal formation needs to be modified. Second, our data indicate the existence of two parallel pathways through the hippocampal memory system, arising from the perirhinal and postrhinal cortex. These two parallel pathways may be involved in separately processing functionally different types of sensory information. This second proposition will be subsequently evaluated on the basis of series of electrophysiological studies we carried out in rats and some preliminary functional brain imaging studies in humans.