Intrinsic connections of the macaque monkey hippocampal formation: I. Dentate gyrus

We have carried out a detailed analysis of the intrinsic connectivity of the Macaca fascicularis monkey hippocampal formation. Here we report findings on the topographical organization of the major connections of the dentate gyrus. Localized anterograde tracer injections were made at various rostrocaudal levels of the dentate gyrus, and we investigated the three-dimensional organization of the mossy fibers, the associational projection, and the local projections. The mossy fibers travel throughout the transverse extent of CA3 at the level of the cells of origin. Once the mossy fibers reach the distal portion of CA3, they change course and travel for 3-5 mm rostrally. The associational projection, originating from cells in the polymorphic layer, terminates in the inner one-third of the molecular layer. The associational projection, though modest at the level of origin, travels both rostrally and caudally from the injection site for as much as 80% of the rostrocaudal extent of the dentate gyrus. The caudally directed projection is typically more extensive and denser than the rostrally directed projection. Cells in the polymorphic layer originate local projections that terminate in the outer two-thirds of the molecular layer. These projections are densest at the level of the cells of origin but also extend several millimeters rostrocaudally. Overall, the topographic organization of the intrinsic connections of the monkey dentate gyrus is largely similar to that of the rat. Such extensive longitudinal connections have the potential for integrating information across much of the rostrocaudal extent of the dentate gyrus.     

Mixed electrical-chemical transmission between hippocampal mossy fibers and pyramidal cells

Morphological and electrophysiological studies have shown that granule cell axons, the mossy fibers (MFs), establish gap junctions and therefore electrical communication among them. That granule cells express gap junctional proteins in their axons suggests the possibility that their terminals also express them. If this were to be the case, mixed electrical-chemical communication could be supported, as MF terminals normally use glutamate for fast communication with their target cells. Here we present electrophysiological studies in the rat and modeling studies consistent with this hypothesis. We show that MF activation produced fast spikelets followed by excitatory postsynaptic potentials in pyramidal cells (PCs), which, unlike the spikelets, underwent frequency potentiation and were strongly depressed by activation of metabotropic glutamate receptors, as expected from transmission of MF origin. The spikelets, which persisted during blockade of chemical transmission, were potentiated by dopamine and suppressed by the gap junction blocker carbenoxolone. The various waveforms evoked by MF stimulation were replicated in a multi-compartment model of a PC by brief current-pulse injections into the proximal apical dendritic compartment, where MFs are known to contact PCs. Mixed electrical and glutamatergic communication between granule cells and some PCs in CA3 may ensure the activation of sets of PCs, bypassing the strong action of concurrent feed-forward inhibition that granule cells activate. Importantly, MF-to-PC electrical coupling may allow bidirectional, possibly graded, communication that can be faster than chemical synapses and subject to different forms of modulation.     

Increased density of hippocampal kainate receptors but normal density of NMDA and AMPA receptors in a rat model of prenatal protein malnutrition

The postnatal development of excitatory amino acid receptor types including kainate, N-methyl-D-aspartate (NMDA), and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) was assessed in the hippocampus, entorhinal cortex, and adjacent neocortex in normal and prenatally protein malnourished rats ages 15, 30, 90, and 220 postnatal days by quantitative autoradiography. Tritiated ligands used to measure binding site density were (3)[H]kainate, (3)[H]MK-801, and (3)[H]AMPA, respectively. Kainate receptors showed statistically significant increases in binding density in stratum lucidum of CA3 (hippocampal mossy fiber zone) in 90- and 220-day-old malnourished rats compared with age- and sex-matched controls but not in 15- or 30-day-old malnourished rats. Compared with previous anatomic studies, these results are mostly in agreement with a significantly decreased hippocampal mossy fiber plexus in 15-, 90-, and 220-day-old rats but not in 30-day-old rats. These results suggested that the increased density of postsynaptic kainate receptors located mainly on proximal apical dendrites of CA3 pyramidal cells may be compensatory to decreased glutamate release due to the reduction in mossy fiber plexus. In contrast, the density of putative NMDA and AMPA receptors quantified in prenatally malnourished rats was comparable to the density quantified in age- and sex-matched control rats, as were all three receptor types in entorhinal cortex and adjacent neocortex. Thus, the selectivity of the compensation of (3)[H]kainate-labeled mossy fiber plexus in adult but not in early postnatal developing malnourished rats may help ensure continued breeding and survival of the species under otherwise adverse environmental conditions.     

Differential expression and localization of the phosphorylated and nonphosphorylated neurofilaments during the early postnatal development of rat hippocampus

Neurofilament (NF) proteins are expressed in most mature neurons in the central nervous system. Although they play a crucial role in neuronal growth, organization, shape, and plasticity, their expression pattern and cellular distribution in the developing hippocampus remain unknown. In the present study, we have used Western blotting and immunocytochemistry to study the low- (NF-L), medium- (NF-M), and high- (NF-H) molecular-weight NF proteins; phosphorylated epitopes of NF-M and NF-H; and a nonphosphorylated epitope of NF-H in the early postnatal (through P1-P21) development of the rat hippocampus. During the first postnatal week, NF-M was the most abundantly expressed NF, followed by NF-L, whereas the expression of NF-H was very low. Through P7-P14, the expression of NF-H increased dramatically and later began to plateau, as also occurred in the expression of NF-M and NF-L. At P1, no NF-M immunopositive cell bodies were detected, but cell processes in the CA1-CA3 fields were faintly immunopositive for NF-M and for the phosphorylated epitopes of NF-M and NF-H. At P7, CA3 pyramidal neurons were strongly immunopositive for NF-L and NF-H, but not for NF-M. The axons of granule cells, the mossy fibers (MFs), were NF-L and NF-M positive through P7-P21 but were NF-H immunonegative at all ages. Although they stained strongly for the phosphorylated NF-M and NF-H at P7, the staining intensity sharply decreased at P14 and remained so at P21. The cell bodies of CA1 pyramidal neurons and granule cells remained immunonegative against all five antibodies in all age groups. Our results show a different time course in the expression and differential cell type and cellular localization of the NF proteins in the developing hippocampus. These developmental changes could be of importance in determining the reactivity of hippocampal neurons in pathological conditions in the immature hippocampus.     

Disruption of hippocampal CA3 network: effects on episodic-like memory processing in C57BL/6J mice

Lesion studies have demonstrated the prominent role of the hippocampus in spatial and contextual learning. To better understand how contextual information is processed in the CA3 region during learning, we focused on the CA3 autoassociative network hypothesis. We took advantage of a particularity of the mossy fibre (MF) synapses, i.e. their high zinc concentration, to reversibly disrupt the afferent MF pathway by microinfusions of an intracellular (DEDTC) or an extracellular (CaEDTA) zinc chelator into the CA3 area of the dorsal hippocampus of mice. Disruption of the CA3 network significantly impaired the acquisition and the consolidation of contextual fear conditioning, whereas contextual retrieval was unaffected. These results also suggest a heterogeneity between the cognitive processes underlying spatial and contextual memory that might be linked to the specific involvement of free zinc in contextual information processing.     

Structural plasticity of hippocampal mossy fiber synapses as revealed by high-pressure freezing

Despite recent progress in fluorescence microscopy techniques, electron microscopy (EM) is still superior in the simultaneous analysis of all tissue components at high resolution. However, it is unclear to what extent conventional fixation for EM using aldehydes results in tissue alteration. Here we made an attempt to minimize tissue alteration by using rapid high-pressure freezing (HPF) of hippocampal slice cultures. We used this approach to monitor fine-structural changes at hippocampal mossy fiber synapses associated with chemically induced long-term potentiation (LTP). Synaptic plasticity in LTP has been known to involve structural changes at synapses including reorganization of the actin cytoskeleton and de novo formation of spines. While LTP-induced formation and growth of postsynaptic spines have been reported, little is known about associated structural changes in presynaptic boutons. Mossy fiber synapses are assumed to exhibit presynaptic LTP expression and are easily identified by EM. In slice cultures from wildtype mice, we found that chemical LTP increased the length of the presynaptic membrane of mossy fiber boutons, associated with a de novo formation of small spines and an increase in the number of active zones. Of note, these changes were not observed in slice cultures from Munc13-1 knockout mutants exhibiting defective vesicle priming. These findings show that activation of hippocampal mossy fibers induces pre- and postsynaptic structural changes at mossy fiber synapses that can be monitored by EM.     

Ontogeny of dynorphin-like immunoreactivity in the hippocampal formation of the rat

Light microscopic immunocytochemical techniques were used to analyze the ontogeny of dynorphin (A)-like immunoreactivity (DLI) in the hippocampal formation of the Sprague-Dawley rat. For comparison purposes, alternate sections of the same brains were processed for the localization of methionine enkephalin-like immunoreactivity (ELI). DLI was first detectable in CA3a stratum lucidum and the suprapyramidal hilus on postnatal day (P) 6. On P7, DLI was evenly present throughout the full extent of the mossy fiber system. From P8 to P19, DLI progressively increased in intensity and could be localized in the fine axons and spherical swellings. The mossy fiber system and occasional perikarya superficial to stratum granulosum were the only hippocampal elements that exhibited DLI. In corroboration with earlier results, stratum lucidum ELI was first detected within large spherical bouton-like swellings on P13. From these data it is concluded that DLI appears in morphologically immature mossy fibers soon after they reach their target fields. In contrast, enkephalin is first detected within morphologically elaborated mossy fiber boutons well after the establishment of functional synapses.     

Ultrastructural heterogeneity of enkephalin-containing terminals in the rat hippocampal formation

Opioid peptides, including leu-enkephalin (LE), are important neuromodulators in the hippocampal formation where they may play a role in learning and memory as well as epileptogenesis. We examined the cellular substrates that underlie the function of LE in each lamina of the rat hippocampal formation by immunocytochemistry at the electron microscopic level in single section analysis. LE-like immunoreactivity (LE-LI) was primarily associated with large dense-core vesicles (80–100 nm), usually found in axons and axon terminals, but was also observed in perikarya and occasionally in dendrites. The morphology and synaptic associations of LE-LI-containing terminals were strikingly distinct in each region of the hippocampal formation. In the molecular layer of the dentate gyrus, terminals with LE-LI were typically small (0.6 μm) and formed primarily asymmetric (excitatory type) synapses on single dendritic spines, which is consistent with the presence of LE in the lateral perforant path. In the hilus of the dentate gyrus, twd types of LE-containing terminals were present: (1) small round terminals that were heterogeneous in size (0.4–1 μm) and in type of contact formed and (2) larger (3–5 μm) terminals exhibiting the characteristic morphology of mossy fiber boutons that formed asymmetric synapses on spines. This variation in morphology and the type of contact suggests LE may have a heterogeneous influence on diverse hilar interneurons. In the CA3 region of the hippocampus, LE-LI was localized to large mossy fiber boutons (3–7 μm) that formed multiple asymmetric synapses on complex spiny dendritic processes and often formed puncta adherentia with the shafts of large CA3 pyramidal cell dendrites, indicating that this peptide may be directly released onto pyramidal cells. At the border of stratum radiatum and lacunosum moleculare in the CA1 region of the hippocampus, LE-labeled terminals averaged 0.8 μm in diameter and often formed symmetric (inhibitory type) synapses on dendritic shafts, which is consistent with a role in disinhibition. In conclusion, these heterogeneous cellular interactions indicate that LE has diverse functional roles and mechanisms of action within each lamina of the hippocampal formation and may directly and indirectly modulate hippocampal cell activity. © 1995 Wiley-Liss, Inc.     

Morphological and molecular markers of mouse area CA2 along the proximodistal and dorsoventral hippocampal axes

Hippocampal area CA2 is a molecularly and functionally distinct region of the hippocampus that has classically been defined as the area with large pyramidal neurons lacking input from the dentate gyrus and the thorny excrescences (TEs) characteristic of CA3 neurons. A modern definition of CA2, however, makes use of the expression of several molecular markers that distinguish it from neighboring CA3 and CA1. Using immunohistochemistry, we sought to characterize the staining patterns of commonly used CA2 markers along the dorsal–ventral hippocampal axis and determine how these markers align along the proximodistal axis. We used a region of CA2 that stained for both Regulator of G-protein Signaling 14 (RGS14) and Purkinje Cell Protein 4 (PCP4; “double-labeled zone” [DLZ]) as a reference. Here, we report that certain commonly used CA2 molecular markers may be better suited for drawing distinct boundaries between CA2/3 and CA2/1. For example, RGS14+ and STEP+ neurons showed minimal to no extension into area CA1 while areas stained with VGluT2 and Wisteria Floribunda agglutinin were consistently smaller than the DLZ/CA2 borders by ~100 μ on the CA1 or CA3 sides respectively. In addition, these patterns are dependent on position along the dorsal–ventral hippocampal axis such that PCP4 labeling often extended beyond the distal border of the DLZ into CA1. Finally, we found that, consistent with previous findings, mossy fibers innervate a subset of RGS14 positive neurons (~65%–70%) and that mossy fiber bouton number and relative size in CA2 are less than that of boutons in CA3. Unexpectedly, we did find evidence of some complex spines on apical dendrites in CA2, though much fewer in number than in CA3. Our results indicate that certain molecular markers may be better suited than others when defining the proximal and distal borders of area CA2 and that the presence or absence of complex spines alone may not be suitable as a distinguishing feature differentiating CA3 from CA2 neurons.     

Intracerebroventricular kainic acid administration in adult rat alters hippocampal calbindin and non-phosphorylated neurofilament expression

Calbindin and non-phosphorylated neurofilament proteins were assessed in hippocampus following a unilateral intracerebroventricular kainic acid injection at 4, 26, and 60 days post-lesion, using immunocytochemical expression. The density of calbindin-positive non- pyramidal neurons throughout the hippocarnpus showed no significant alteration at 4 days post-lesion, a significant decrease at 26 days post-lesion, and a partial recovery at 60 days post-lesion. In addition, calbindin immunoreactivity was dramatically reduced at 26 days post-lesion in the CA1 pyramidal and dentate granule cell layers and the mossy fibers, bilaterally. Although not significant statistically, most of these reductions showed signs of reversal at 60 days post-lesion except the CA1 pyramidal cell layer where the dramatic reductions persisted. Neurofilaments were also altered throughout the post-lesion period, particularly in abnormal expression of non-phosphorylated neurofilament proteins in mossy fibers. The apparent return of calbindin immunoreactivity in non-pyramidal neurons by 60 days post-lesion suggests that recovery from the lesion may involve remaining neuronal elements which either become reactivated with time or have the capability to express normal levels of calbindin with re-innervation. On the other hand, prolonged calbindin reductions in superficial CA1 pyramidal cells suggest sustained down-regulation of calbindin expression owing to persistent reductions in the activity of these neurons. The temporal correlation of the expression of non-phosphorylated neurofilamenta in mossy fibers with their sprouting response following target loss suggests a potential role for non-phosphorylated neurofilaments in neuronal plasticity involving axonal sprouting. Alternatively, it may also suggest that injury- induced neurofilament modifications are either conducive or permissive for axonal sprouting. © 1995 Wiley-Liss, Inc.     

Conduction latency along CA3 hippocampal axons from rat

Relatively few physiological studies have been carried out on intrahippocampal axons. We have recorded compound potentials from fiber groups and the activity of individual axons at 22-25 degrees C to characterize the conduction in subsets of the broad fan-shaped CA3 pyramidal axonal tree, including the Schaffer collaterals and longitudinal branches. The same wide axonal branching was indicated by antidromic activation of individual CA3 pyramidal cells. The average compound action potential latency from the CA3 to the CA1 area (Schaffer collaterals) increased by 4.16 +/- 0.06 ms/mm separation between the stimulation and registration electrodes. The impulses spread 31% faster in the 45-degree oblique temporal than in the transverse direction across CA1. The latency of the longitudinal axons in the CA3 area increased by 6.19 +/- 0.19 ms/mm. More impressive than these direction-dependent differences in latency were the large differences between individual axons running in the same direction. For both the longitudinal axons and the Schaffer collaterals, there was a broad distribution of antidromic latencies for a given distance between the stimulation and recording points. Typically, the fastest impulses arrived in half the time of the slowest. The distribution of compound action potential latencies between two points in the tissue could be made narrower by surgical restriction of the thickness and width of the preparation. By comparison, the cerebellar parallel fibers showed a narrower distribution of their latencies than the Schaffer collaterals. Because the cerebellar fibers run more straight than Schaffer collaterals, this suggested that some of the latency differences of the latter were due to differences in the path length of the axons. One consequence of our findings is that synchronous firing of neighboring CA3 pyramidal cells does not necessarily give synchronous inputs to common target CA1 neurons.     

Revisiting the role of the hippocampal mossy fiber synapse

The mossy fiber pathway has long been considered to provide the major source of excitatory input to pyramidal cells of hippocampal area CA3. In this review we describe anatomical and physiological properties of this pathway that challenge this view. We argue that the mossy fiber pathway does not provide the main input to CA3 pyramidal cells, and that the short‐term plasticity and amplitude variance of mossy fiber synapses may be more important features than their long‐term plasticity or absolute input strength. Hippocampus 2001;11:408–417. © 2001 Wiley‐Liss, Inc.     

Cytoarchitectonic distribution of zinc in the hippocampus of man and the rat

Zinc was measured in whole hippocampus and in hippocampal sub-regions by stable-isotope dilution mass spectrometry. In both man and the rat, the most zinc (102–145 ppm, dry weight) was found in the hilar region, the least (27–37) in the fimbria. The amount of zinc directly associated with mossy-fiber axons was estimated to be approximately 8% of the total zinc in the hippocampus, and the concentration of mossy-fiber zinc was estimated at 220–300 μM. Methodological and theoretical implications of the quantitative findings were discussed.     

The distribution of cholecystokinin-like immunoreactivity in the hippocampal formation of the guinea pig: Localization in the mossy fibers

Immunocytochemical techniques were used to localize cholecystokinin octapeptide (CCK-8)-like immunoreactivity in the hippocampal formation of the guinea pig. As in the rat, CCK immunoreactive perikarya are most dense in and around the stratum pyramidale, within the superficial cell layer of the subiculum, and within the polymorph zone of the hilus. Immunoreactive axons are observed within and loosely surrounding the stratum pyramidale, within the stratum lacunosum moleculare, and diffusely distributed across the subiculum. In contrast to the rat, the mossy fiber system also exhibited significant CCK immunoreactivity. The latter system has previously been demonstrated to contain enkephalin-like immunoreactivity in the guinea pig. The present results suggest, therefore, that the enkephalin-like and CCK-like substances either coexist within the mossy fiber boutons or are present within separate subpopulations of the mossy fibers.     

Effects of morphine and naloxone on hippocampal CA3 field potentials following systemic administration in the freely-moving rat

The effect of IV morphine, 2, 6 and 15 mg/kg, on hilar-evoked CA3 field potentials was studied to determine if this area would be more sensitive to mu-type opiate agonists than the CA1 or dentate regions. In addition, the effect of IV naloxone, 2 and 25 mg/kg, on the same responses was studied to determine if endogenous opiates reported to be present in the mossy fibers are released by electrical stimulation of this pathway. Neither morphine nor naloxone had an effect on CA3 field potentials at any dose used. The CA1 region of the hippocampus is the area most sensitive to morphine, and this effect of morphine correlates best, anatomically, with the localization of mu-receptors identified by the binding of dihydromorphine. Physiological release of endogenous opiates from the hippocampus remains to be shown.     

Adult learning and remodeling of hippocampal mossy fibers: Unheralded participant in circuitry for long‐lasting spatial memory

Two articles in this issue concerning the overexpression of GAP‐43 on mossy fiber growth are related to the plasticity of these axons in relation to learning and memory. © 2009 Wiley‐Liss, Inc.     

Rapid induction by kainic acid of both axonal growth and F1/GAP-43 protein in the adult rat hippocampal granule cells

Hippocampal granule cells do not normally express the axonal growth and plasticity-associated protein F1/GAP-43 in the adult rat. Using three different methods that lead to hypersynchronous activity in limbic circuits, expression of F1/GAP-43 mRNA can be induced in granule cells which is followed by sprouting in mossy fibers, the axons of granule cells. F1/GAP-43 mRNA expression in granule cells was induced in the temporal, but not septal, hippocampus beginning at 12 hours after kainic acid (KA) subcutaneous injection (10 mg/kg). Beginning 2 days after KA treatment, mossy fiber sprouts restricted to the temporal hippocampus were observed in the supragranular layer. In the same animal we also observed that levels of protein F1/GAP-43 immunoreactivity in this layer apparently increased at this same 2 day time point and same ventral hippocampal location. F1/GAP-43 protein levels and mossy fiber sprouting showed an increase up to 10 days after KA treatment. Sprouting was at a maximum at 40 days, the longest time point studied. These events parallel axonal regeneration with one critical difference: granule cell axons are not damaged by kainate. The rapid onset of axonal growth in the adult is striking and occurs earlier than reported previously (2 days vs. 12 days). Such growth closely associated with elevated levels of protein F1/GAP-43 may occur as a result of a) reactive synaptogenesis caused by the availability of post-synaptic surface on granule cell dendrites at the supragranular layer, b) Hebbian co-activation of the post-synaptic granule cells and their presynaptic afferents, and c) loss of target-derived inhibitory growth factor. © 1996 Wiley-Liss, Inc.     

Excitation-inhibition balance in the CA3 network--neuronal specificity and activity-dependent plasticity

Activation of the axons of the granule cells, the mossy fibers, excites pyramidal cells and interneurons in the CA3 area, which, in turn, inhibit pyramidal cells. The integration of the various inputs that converge onto CA3 cells has been studied by pharmacological dissection of either the excitatory or inhibitory components. This strategy has the disadvantage of partially isolating the recorded cell from the network, ignoring the sources and the impact of concurrent inputs. To overcome this limitation, we dissociated excitatory and inhibitory synaptic conductances by mathematical extraction techniques, and analysed the dynamics of the integration of excitatory and inhibitory inputs in pyramidal cells and stratum lucidum interneurons (Sl-Ints) of CA3. We have uncovered a shunting mechanism that decreases the responsiveness of CA3 output cells to mossy fiber input after a period of enhanced excitability. The activation of the dentate gyrus (DG) after applying a kindling-like protocol in vitro, or after producing one or several seizures in vivo, results in a graded and reversible increase of inhibitory conductances in pyramidal cells, while in Sl-Ints, an increase of excitatory conductances occurs. Thus, interneurons reach more depolarized membrane potentials on DG activation yielding a high excitatory postsynaptic potential-spike coupling, while the contrary occurs in pyramidal cells. This effective activation of feedforward inhibition is synergized by the emergence of direct DG-mediated inhibition on pyramidal cells. These factors force the synaptic conductance to peak at a potential value close to resting membrane potential, thus producing shunt inhibition and decreasing the responsiveness of CA3 output cells to mossy fiber input.