| Abstract: | To understand the cellular basis of learning and memory, the neurophysiology of the hippocampus has been largely examined in thin transverse slice preparations. However, the synaptic architecture along the longitudinal septo-temporal axis perpendicular to the transverse projections in CA1 is largely unknown, despite its potential significance for understanding the information processing carried out by the hippocampus. Here, using a battery of powerful techniques, including 3D digital holography and focal glutamate uncaging, voltage-sensitive dye, two-photon imaging, electrophysiology, and immunohistochemistry, we show that CA1 pyramidal neurons are connected to one another in an associational and well-organized fashion along the longitudinal axis of the hippocampus. Such CA1 longitudinal connections mediate reliable signal transfer among the pyramidal cells and express significant synaptic plasticity. These results illustrate a need to reconceptualize hippocampal CA1 network function to include not only processing in the transverse plane, but also operations made possible by the longitudinal network. Our data will thus provide an essential basis for future computational modeling studies on information processing operations carried out in the full 3D hippocampal network that underlies its complex cognitive functions.The hippocampus is widely used to study functional connectivity of the brain with the hope that principles that operate within its relatively simple architecture may be extended to more complex cortical structures. Its manageable number of cell types also provides an attractive opportunity to examine fundamental issues in neuroscience such as the relationship between network circuitry and function. For example, considerable effort has been devoted toward elucidating the circuitry supporting episodic memory—a property closely linked to the hippocampus. CA3 pyramidal neurons form extensive recurrent connections with each other (1). Such connections are able to learn to associate components of an input pattern with each other (2), which, in turn, has greatly influenced thinking on the mechanisms of memory formation and recall (3). Under appropriate conditions, computer simulations reveal that recurrent neural networks have the capacity to learn temporal sequences and to carry out pattern completion (4, 5). Interestingly, although they are quite near to the CA3 region, CA1 pyramidal neurons reportedly form remarkably few associational connections (6, 7). This distinctive difference in network architecture might suggest that, although area CA1 could serve to decode the output of CA3, it would not possess the intrinsic ability for autoassociational computations. This idea would imply that the ability of CA1 to carry out independent information processing operations may be more limited than that of CA3. However, even after removal of all input from area CA3, CA1 pyramidal neurons still have the capacity to transform location-modulated signals from the entorhinal cortex into accurate spatial firing patterns (8). In addition, deficits in temporal sequence learning are more severe after selective lesions to CA1 than to CA3 (9). Finally, CA1 is more closely linked to memory of temporal order of visual objects and especially over long intervals (10). Thus, area CA1 appears to have a greater ability for intrinsic information processing than would be expected based on current understanding of its circuitry. Intrinsic processing could represent autoassociational computations through direct excitatory synaptic contacts among the CA1 pyramidal cells, but as noted, there is little evidence for such connectivity within CA1. This puzzle led us to reexamine the apparent sparseness of associational synaptic connections between CA1 pyramidal neurons using experimental techniques that were not previously available for this investigation.The “trisynaptic circuits” (dentate gyrus: CA3–CA1) oriented transversely to the hippocampal long axis, the basis of the “lamellar hypothesis” (11), has greatly influenced thinking about the structure-function relationships of this structure. This hypothesis suggests that the hippocampus is organized as a stack of parallel, trisynaptic circuits. Although this view has been challenged by the observation of fibers running across lamellae, especially in dentate gyrus and CA3 area (12, 13), the hypothesis supported an explosion in the use of the transverse slice for electrophysiological studies of the hippocampus. However, axons oriented along the longitudinal axis are unavoidably severed in the preparation of the transverse slice, meaning that these studies are heavily weighted in favor of conclusions based on fibers traveling within the transverse plane. We used a whole hippocampus preparation, as well as longitudinal and transverse slice preparations, to obtain a more accurate picture of synaptic connections among CA1 pyramidal neurons in three dimensions. Remarkably, we found prominent associational connectivity along the longitudinal axis. Furthermore, synapses of the longitudinal network possess the capacity for synaptic plasticity that includes a novel memory mechanism we recently described called dendritic hold and read (DHR) (14). These findings may help to explain the intrinsic ability of area CA1 to process information transfer and provide novel data that will lead to more realistic models of hippocampal function in three dimensions. |
