Large variability in synaptic N-methyl-D-aspartate receptor density on interneurons and a comparison with pyramidal-cell spines in the rat hippocampus

Pyramidal cells receive input from several types of GABA-releasing interneurons and innervate them reciprocally. Glutamatergic activation of interneurons involves both alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-D-aspartate (NMDA) type glutamate receptors expressed in type I synapses, mostly on their dendritic shafts. On average, the synaptic AMPA receptor content is several times higher on interneurons than in the spines of pyramidal cells. To compare the NMDA receptor content of synapses, we used a quantitative postembedding immunogold technique on serial electron microscopic sections, and analysed the synapses on interneuron dendrites and pyramidal cell spines in the CA1 area. Because all NMDA receptors contain the obligatory NR1 subunit, receptor localisation was carried out using antibodies recognising all splice variants of the NR1 subunit. Four populations of synapse were examined: i). on spines of pyramidal cells in stratum (str.) radiatum and str. oriens; ii). on parvalbumin-positive interneuronal dendritic shafts in str. radiatum; iii). on randomly found dendritic shafts in str. oriens and iv). on somatostatin-positive interneuronal dendritic shafts and somata in str. oriens. On average, the size of the synapses on spines was about half of those on interneurons. The four populations of synapse significantly differed in labelling for the NR1 subunit. The median density of NR1 subunit labelling was highest on pyramidal cell spines. It was lowest in the synapses on parvalbumin-positive dendrites in str. radiatum, where more than half of these synapses were immunonegative. In str. oriens, synapses on interneurons had a high variability of receptor content; some dendrites were similar to those in str. radiatum, including the proximal synapses of somatostatin-positive cells, whereas others had immunoreactivity for the NR1 subunit similar to or higher than synapses on pyramidal cell spines.These results show that synaptic NMDA receptor density differs between pyramidal cells and interneurons. Some interneurons may have a high NMDA receptor content, whereas others, like some parvalbumin-expressing cells, a particularly low synaptic NMDA receptor content. Consequently, fast glutamatergic activation of interneurons is expected to show cell type-specific time course and state-dependent dynamics.     

Synaptic reorganization in the rabbit hippocampus after lesion of commissural afferents

Summary The degeneration of commissural afferents to the hippocampus in the rabbit was studied by using the Fink-Heimer degeneration method, electron microscopy, and the combined Golgi/EM technique. The stratum oriens (CA3) was selected for quantitative electron microscopic evaluation of postlesional changes since the degeneration of commissural fibers as seen in Fink-Heimer preparations was dense throughout the width of that layer. Accordingly, in electron micrographs of stratum oriens many electron-dense degenerating boutons were found after short survival times (3 and 6 days, respectively), most of them (96%) in synaptic contact with dendritic spines. In the fine structural analysis of Golgi-impregnated CA3 pyramidal cells, spines of basal dendrites were identified as postsynaptic elements of degenerating commissural afferents in stratum oriens.Three days after the lesion, the number of intact synapses/unit area was reduced in stratum oriens of CA3 to 64% of the control; 20% of the synapses were degenerating. Thus, part of the degenerated synapses had disappeared. Evidence is provided that phagocytosis of degenerated boutons still attached to fragments of dendritic spines played a role in this process.Seven weeks after the lesion, the number of intact synapses had returned to control level, suggesting reactive growth of synaptic structures. When the ratio of spine synapses versus shaft synapses was compared with controls, no change had occurred. Thus, after an initial loss of spine synapses after short survival times, new spines have been formed in parallel with ingrowth (sprouting) of neighbouring nonlesioned afferents.     

Synaptic modifications at the CA3–CA1 synapse after chronic AMPA receptor blockade in rat hippocampal slices

Maintenance of dendritic spines, the postsynaptic elements of most glutamatergic synapses in the central nervous system, requires continued activation of AMPA receptors. In organotypic hippocampal slice cultures, chronic blockade of AMPA receptors for 14 days induces a substantial loss of dendritic spines on CA1 pyramidal neurons. Here, using serial section electron microscopy, we show that loss of dendritic spines is paralleled by a significant reduction in synapse density. In contrast, we observed an increased number of asymmetric synapses onto the dendritic shaft, suggesting that spine retraction does not inevitably lead to synapse elimination. Functional analysis of the remaining synapses revealed that hippocampal circuitry compensates for the anatomical loss of synapses by increasing synaptic efficacy. Moreover, we found that the observed morphological and functional changes were associated with altered bidirectional synaptic plasticity. We conclude that continued activation of AMPA receptors is necessary for maintaining structure and function of central glutamatergic synapses.     

Estrogen and aging affect synaptic distribution of phosphorylated LIM kinase (pLIMK) in CA1 region of female rat hippocampus

17beta-Estradiol (E) increases axospinous synapse density in the hippocampal CA1 region of young female rats, but not in aged rats. This may be linked to age-related alterations in signaling pathways activated by synaptic estrogen receptor alpha (ER-alpha) that potentially regulate spine formation, such as LIM-kinase (LIMK), an actin depolymerizing factor/cofilin kinase. We hypothesized that, as with ER-alpha, phospho-LIM-kinase (pLIMK) may be less abundant or responsive to E in CA1 synapses of aged female rats. To address this, cellular and subcellular distribution of pLIMK-immunoreactivity (IR) in CA1 was analyzed by light and electron microscopy in young and aged female rats that were ovariectomized and treated with either vehicle or E. pLIMK-IR was found primarily in perikarya within the pyramidal cell layer and dendritic shafts and spines in stratum radiatum (SR). While pLIMK-IR was occasionally present in terminals, post-embedding quantitative analysis of SR showed that pLIMK had a predominant post-synaptic localization and was preferentially localized within the postsynaptic density (PSD). The percentage of pLIMK-labeled synapses increased (30%) with E treatment (P<0.02) in young animals, and decreased (43%) with age (P<0.002) regardless of treatment. The pattern of distribution of pLIMK-IR within dendritic spines and synapses was unaffected by age or E treatment, with the exception of an E-induced increase in the non-synaptic core of spines in young females. These data suggest that age-related synaptic alterations similar to those seen with ER-alpha occur with signaling molecules such as pLIMK, and support the hypothesis that age-related failure of E treatment to increase synapse number in CA1 may be due to changes in the molecular profile of axospinous synapses with respect to signaling pathways linked to formation of additional spines and synapses in response to E.     

Double dissociation between long-term depression and dendritic spine morphology in cerebellar Purkinje cells

Experiments in hippocampal area CA1 suggest that long-term potentiation could be associated with spine addition and enlargement, and long-term depression (LTD) with spine shrinkage and loss. Is this a general principle of synaptic plasticity? We used two-photon microscopy to measure dendritic spines in rat cerebellar Purkinje cells. Neither local synaptic induction of LTD nor global chemical induction of LTD changed spine number or size. Conversely, a manipulation that evoked persistent dendritic spine retraction did not alter parallel fiber-evoked excitatory postsynaptic currents.     

Mechanism of the distance-dependent scaling of Schaffer collateral synapses in rat CA1 pyramidal neurons

Schaffer collateral axons form excitatory synapses that are distributed across much of the dendritic arborization of hippocampal CA1 pyramidal neurons. Remarkably, AMPA-receptor-mediated miniature EPSP amplitudes at the soma are relatively independent of synapse location, despite widely different degrees of dendritic filtering. A progressive increase with distance in synaptic conductance is thought to produce this amplitude normalization. In this study we examined the mechanism(s) responsible for spatial scaling by making whole-cell recordings from the apical dendrites of CA1 pyramidal neurons. We found no evidence to suggest that there is any location dependence to the range of cleft glutamate concentrations found at Schaffer collateral synapses. Furthermore, we observed that release probability ( P _r), paired-pulse facilitation and the size of the readily releasable vesicular pool are not dependent on synapse location. Thus, there do not appear to be any changes in the fundamental presynaptic properties of Schaffer collateral synapses that could account for distance-dependent scaling. On the other hand, two-photon uncaging of 4-methoxy-7-nitroindolinyl-caged l -glutamate onto isolated dendritic spines shows that the number of postsynaptic AMPA receptors per spine increases with distance from the soma. We conclude, therefore, that the main synaptic mechanism involved in the production of distance-dependent scaling of Schaffer collateral synapses is an elevated postsynaptic AMPA receptor density.     

Ultrastructural evidence that androgen receptors are located at extranuclear sites in the rat hippocampal formation

Like estrogens in female rats, androgens can affect dendritic spine density in the CA1 subfield of the male rat hippocampus [J Neurosci 23:1588 (2003)]. Previous light microscopic studies have shown that androgen receptors (ARs) are present in the nuclei of CA1 pyramidal cells. However, androgens may also exert their effects through rapid non-genomic mechanisms, possibly by binding to membranes. Thus, to investigate whether ARs are at potential extranuclear sites of ARs, antibodies to ARs were localized by light and electron microscopy in the male rat hippocampal formation. By light microscopy, AR immunoreactivity (-ir) was found in CA1 pyramidal cell nuclei and in disperse, punctate processes that were most dense in the pyramidal cell layer. Additionally, diffuse AR-ir was found in the mossy fiber pathway. Ultrastructural analysis revealed AR-ir at several extranuclear sites in all hippocampal subregions. AR-ir was found in dendritic spines, many arising from pyramidal and granule cell dendrites. AR-ir was associated with clusters of small, synaptic vesicles within preterminal axons and axon terminals. Labeled preterminal axons were most prominent in stratum lucidum of the CA3 region. AR-containing terminals formed asymmetric synapses or did not form synaptic junctions in the plane of section analyzed. AR-ir also was detected in astrocytic profiles, many of which apposed terminals synapsing on unlabeled dendritic spines or formed gap junctions with other AR-labeled or unlabeled astrocytes. Collectively, these results suggest that ARs may serve as both a genomic and non-genomic transducer of androgen action in the hippocampal formation.     

Development of synapses on pyramidal and multipolar non-pyramidal neurons in the visual cortex of rabbits. A combined Golgi-electron microscope study

A combined Golgi-electron microscope method was used to study the ultrastructural maturation of synapses on identified pyramidal and multipolar non-pyramidal neurons in the visual cortex of young and adult rabbits. In samples of 10 (time of eye opening), 14, 20 day old and 7 month old animals, fully impregnated pyramidal neurons within the layers II-V and multipolar non-pyramidal neurons mainly located in lower layer III and layer IV was studied. We found that synapses in 10 and 14 day old animals were occasionally immature in appearance. They were characterized by either a poorly defined postsynaptic band or equal rims of pre- and postsynaptic electron-dense material and could therefore not be classified as Gray type I or II. The distinction between both types of synapses was easier at day 20 and in the adults when the postsynaptic band of the asymmetrical (type I) synapses had become remarkably thicker. In pyramidal neurons the cytoplasmic organelles increased in number during development. Although a few symmetrical synapses were present on dendritic spines of pyramidal neurons in 14 and 20 day old animals, all pyramidal neurons exhibited the same types of synapses on specific sites of their neuronal surface. They received exclusively type II synapses on their somata, type I synapses on their dendritic spines and both types of synapses on their dendritic shafts. However, in the adult animals the frequency of occurrence of type II synapses, especially on basal dendritic shafts, had increased. In some cases only type II and no type I synapses were present. A striking finding in all young and adult animals was that synapses at the borderline between somata and apical dendritic shafts as well as on dendritic spines were frequently complex or interrupted. The characteristic ultrastructural features of adult spine-free and sparsely spiny multipolar non-pyramidal neurons e.g. the many cytoplasmic organelles and type I and II synapses on somata and on dendrites were already present at day 10. After day 10 the number of organelles and synapses increased prominently and in adult animals the different types of synapses on dendrites were located at relatively short intervals of about 4 microns. In contrast with the dendritic shafts of pyramidal neurons many asymmetrical synapses were observed on dendritic shafts of the non-pyramidal neurons analysed in the adult animals. Furthermore, it appeared that the number of synapses on these non-pyramidal neurons is about twice that on pyramidal neurons in day 20 old animals and about four times in adult animals.(ABSTRACT TRUNCATED AT 400 WORDS)     

The development of the inferior olivary complex in preweanling opossums

Summary Pre- and postsynaptic elements within the developing inferior olive (IO) of both control and experimental opossums were examined via electron microscopy. Electron dense boutons identified di-/mesencephalic, cerebellar and spinal afferents within the IO of 8–71 day old animals, which survived 4–48 hours following either midbrain hemisections or spinal transections.During its initial stage of development (3–22 days) the neuropil of the IO is segregated into fields of small diameter neurites or flocculent profiles. Within the fields of flocculent profiles, synaptic interactions are established, which are both infrequent and immature. Although some flocculent profiles are presynaptic, most are postsynaptic and emanate from olivary somata and dendrites. Synaptic contacts also occur with olivary somata, dendritic shafts, spines and dendritic varicosities. Clear round vesicles (crv's; 40 m) predominate within all boutons, normal ones as well as those which degenerate after di-/mesencephalic, cerebellar and spinal lesions; however, larger (70 m) dense cored vesicles (dcv's) are occasionally observed within some boutons. Degenerating terminals from all three sources primarily contact flocculent profiles and dendritic shafts.As the opossum matures (42 days) dramatic increases occur in the number and complexity of both pre- and postsynaptic elements. Marked variations are observed in the matrix density of dendritic shafts. Although all terminal boutons predominantly contain crv's, the number of dcv's within the population of presynaptic elements increases markedly. Concurently, olivary neurons are profusely studded with spines. Simple dendritic spines and spiny appendages as well as dendritic shafts are the most frequent postsynaptic structures within the principal nucleus (PO). Olivary somata and their spines, however, are postsynaptic to degenerating de-/mesencephalic afferents within the PO. Flocculent profiles, which persist within the accessory nuclei, and dendritic shafts are postsynaptic to degenerating spinal boutons.By 70 days of age synaptic contacts appear more mature and more nearly approximate those seen in the adult (King 1980). Few somatic contacts, opaque dendrites, dendritic varicosities, and flocculent profiles are evident within the PO. Dendritic shafts and spines are the principal postsynaptic structures. Many di-/mesencephalic and cerebellar afferents synapse within maturing synaptic clusters on spines between which a rare gap junction is observed. Other di-/mesencephalic and cerebellar endings in the PO as well as spinal endings in the accessory nuclei are presynaptic to dendritic shafts and spines external to synaptic clusters. This predilection for contacting more specific loci on olivary neurons provides good evidence for synaptic remodeling.As the olivary nuclei develop further, the incidence of gap junctions increases and pleomorphic vesicles appear within boutons. The glial investment of neuronal elements, including synaptic clusters, also becomes more extensive.In conclusion, early di-/mesencephalic, cerebellar and spinal synaptic contacts appear qualitatively uniform in their synaptic features and postsynaptic interactions. As olivary development proceeds, however, the distinguishing synaptic features of the nuclear complex become more apparent. Synaptic remodeling occurs as some midbrain and cerebellar terminals are localized within synaptic clusters. The ultrastructural features characteristic of the adult IO are finally achieved by 80 days of age.This research was supported by N.I.H. Research Grant NS-08798     

Nitric oxide-containing pyramidal neurons of the subiculum innervate the CA1 area

Neurons and axon terminals containing neuron-specific nitric oxide synthase (nNOS) were examined in the rat subiculum and CA1 area of Ammon's horn. In the subiculum, a large subpopulation of the pyramidal neurons and non-pyramidal cells are immunoreactive for nNOS, whereas in the neighbouring CA1 area of Ammon's horn only non-pyramidal neurons are labelled with the antibody against nNOS. In the pyramidal layer of the subiculum, nNOS-positive axon terminals form both asymmetric and symmetric synapses. In the adjacent CA1 area the nNOS-positive terminals that form symmetric synapses are found in all layers, whereas those terminals that form asymmetric synapses are only in strata radiatum and oriens, but not in stratum lacunosum-moleculare. In both the subiculum and CA1 area, labelled terminals make symmetric synapses only on dendritic shafts, whereas asymmetric synapses are exclusively on dendritic spines. Previous observations demonstrated that all nNOS-positive non-pyramidal cells are GABAergic local circuit neurons, which form exclusively symmetric synapses. We suggest that nNOS-immunoreactive pyramidal cells of the subiculum may innervate neighbouring subicular pyramidal cells and, to a smaller extent, pyramidal cells of the adjacent CA1 area, forming a backward projection between the subicular and hippocampal principal neurons. Electronic Publication     

Synaptic organization of intracellularly stained CA3 pyramidal neurons in slice cultures of rat hippocampus

Pyramidal cells of regio inferior in slice cultures of the rat hippocampus were impaled and intracellularly stained with horseradish peroxidase. A correlated light- and electron-microscopic analysis was then performed to study the properties of these neurons under culture conditions with particular emphasis on input synapses onto these cells. Like pyramidal cells in situ, CA3 pyramidal neurons in slice cultures had a triangular cell body with an apical stem dendrite emerging from it. Several basal dendrites and the axon arose from the basal pole of the cell body. The peripheral thin branches of both apical and basal dendrites were covered with small spines, whereas proximal thick dendritic segments and portions of the cell body exhibited large spines or excrescences. The axon gave off numerous fine varicose collaterals which projected to stratum radiatum of CA1 (Schaffer collaterals), to the alveus and to the hilar region. In one case a collateral could be followed to stratum moleculare of the fascia dentata. Electron-microscopic analysis of the injected pyramidal neurons revealed that their cell bodies, dendritic shafts and spines formed synaptic contacts with presynaptic terminals. Mossy fiber endings were identified by their large size and their numerous clear synaptic vesicles with some dense-core vesicles intermingled, and were observed to form synaptic contacts on the large spines or excrescences. Since extrinsic afferents degenerate in slice cultures, the numerous synaptic boutons on the identified pyramidal neurons probably arise from axons of intrinsic neurons that have sprouted in response to deafferentation. This assumption is supported by the finding that collaterals of the injected neurons formed abundant synaptic contacts on dendritic shafts and spines of other cells. These results suggest that, although pyramidal cells under culture conditions retain a remarkable number of their normal characteristics, considerable synaptic reorganization does take place.     

Synaptic plasticity in the oedematous human cerebral cortex

Synaptic plastic changes are fundamental events which occur spontaneously during development, maturity and aging processes or can be induced by injury or trauma. To examine lesion-induced synaptic plasticity, cortical biopsies were taken from the frontal, parietal, temporal and occipital cortex of living patients during neurosurgical treatment of brain trauma, brain tumours and vascular malformations, and processed for transmission electron microscopy. Enlargement of both pre- and postsynaptic endings, irregularly shaped, lobulated, stellate and bifurcated presynaptic endings and conformational changes of dendritic spines were observed. Numerous flat, curved and invaginated axodendritic and axospinous asymmetric synapses were distinguished and a smaller proportion of axodendritic and axosomatic symmetric synapses. Activated or sensitized synapses showed numerous frontline spheroid synaptic vesicles, prominent dense presynaptic dense projections and increased length of synaptic membrane complex. Perforated synapses, multiple synapses and serial synapses were also found evincing synaptic splitting and formation of new synaptic connections. The overall images suggest increased number of excitatory circuits, which were correlated with the tonico-clonic convulsion or post-traumatic seizures observed in some patients. Numerous coated vesicles were observed in pre- and postsynaptic structures. Increased number of polyribosomes were found in the dendritic shafts. The dilated spine apparatus, the coated vesicles and the increased number of polyribosomes seem to represent a system for synthesis, transport and storage of synaptic proteins for the formation of new synapses. Coexisting synaptic plasticity and synaptic degeneration were observed in the patients under study. Dendritic and astrocyte synapse-like junctions were also characterized.     

Total number and distribution of inhibitory and excitatory synapses on hippocampal CA1 pyramidal cells

The integrative properties of neurons depend strongly on the number, proportions and distribution of excitatory and inhibitory synaptic inputs they receive. In this study the three-dimensional geometry of dendritic trees and the density of symmetrical and asymmetrical synapses on different cellular compartments of rat hippocampal CA1 area pyramidal cells was measured to calculate the total number and distribution of excitatory and inhibitory inputs on a single cell.A single pyramidal cell has approximately 12,000 microm dendrites and receives around 30,000 excitatory and 1700 inhibitory inputs, of which 40 % are concentrated in the perisomatic region and 20 % on dendrites in the stratum lacunosum-moleculare. The pre- and post-synaptic features suggest that CA1 pyramidal cell dendrites are heterogeneous. Strata radiatum and oriens dendrites are similar and differ from stratum lacunosum-moleculare dendrites. Proximal apical and basal strata radiatum and oriens dendrites are spine-free or sparsely spiny. Distal strata radiatum and oriens dendrites (forming 68.5 % of the pyramidal cells' dendritic tree) are densely spiny; their excitatory inputs terminate exclusively on dendritic spines, while inhibitory inputs target only dendritic shafts. The proportion of inhibitory inputs on distal spiny strata radiatum and oriens dendrites is low ( approximately 3 %). In contrast, proximal dendritic segments receive mostly (70-100 %) inhibitory inputs. Only inhibitory inputs innervate the somata (77-103 per cell) and axon initial segments. Dendrites in the stratum lacunosum-moleculare possess moderate to small amounts of spines. Excitatory synapses on stratum lacunosum-moleculare dendrites are larger than the synapses in other layers, are frequently perforated ( approximately 40 %) and can be located on dendritic shafts. Inhibitory inputs, whose percentage is relatively high ( approximately 14-17 %), also terminate on dendritic spines.Our results indicate that: (i) the highly convergent excitation arriving onto the distal dendrites of pyramidal cells is primarily controlled by proximally located inhibition; (ii) the organization of excitatory and inhibitory inputs in layers receiving Schaffer collateral input (radiatum/oriens) versus perforant path input (lacunosum-moleculare) is significantly different.     

Estradiol increases spine density and NMDA-dependent Ca2+ transients in spines of CA1 pyramidal neurons from hippocampal slices

To investigate the physiological consequences of the increase in spine density induced by estradiol in pyramidal neurons of the hippocampus, we performed simultaneous whole cell recordings and Ca2+ imaging in CA1 neuron spines and dendrites in hippocampal slices. Four- to eight-days in vitro slice cultures were exposed to 17beta-estradiol (EST) for an additional 4- to 8-day period, and spine density was assessed by confocal microscopy of DiI-labeled CA1 pyramidal neurons. Spine density was doubled in both apical and basal dendrites of the CA1 region in EST-treated slices; consistently, a reduction in cell input resistance was observed in EST-treated CA1 neurons. Double immunofluorescence staining of presynaptic (synaptophysin) and postsynaptic (alpha-subunit of CaMKII) proteins showed an increase in synaptic density after EST treatment. The slopes of the input/output curves of both alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-D-aspartate (NMDA) postsynaptic currents were steeper in EST-treated CA1 neurons, consistent with the observed increase in synapse density. To characterize NMDA-dependent synaptic currents and dendritic Ca2+ transients during Schaffer collaterals stimulation, neurons were maintained at +40 mV in the presence of nimodipine, picrotoxin, and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). No differences in resting spine or dendritic Ca2+ levels were observed between control and EST-treated CA1 neurons. Intracellular Ca2+ transients during afferent stimulation exhibited a faster slope and reached higher levels in spines than in adjacent dendrites. Peak Ca2+ levels were larger in both spines and dendrites of EST-treated CA1 neurons. Ca2+ gradients between spine heads and dendrites during afferent stimulation were also larger in EST-treated neurons. Both spine and dendritic Ca2+ transients during afferent stimulation were reversibly blocked by D, L-2-amino-5-phosphonovaleric acid (D,L-APV). The increase in spine density and the enhanced NMDA-dependent Ca2+ signals in spines and dendrites induced by EST may underlie a threshold reduction for induction of NMDA-dependent synaptic plasticity in the hippocampus.     

HNK-1 (human natural killer-1) glyco-epitope is essential for normal spine morphogenesis in developing hippocampal neurons

The human natural killer-1 (HNK-1) glyco-epitope possesses a unique structural feature, a sulfated glucuronic acid attached to lactosamine on the non-reducing termini of glycans. The expression of HNK-1 is temporally and spatially regulated by glucuronyltransferase (GlcAT-P) in the brain. Our previous report showed that mice lacking GlcAT-P almost completely lost HNK-1 expression in the brain and exhibited reduced long-term potentiation (LTP) at hippocampal CA1 synapses. GlcAT-P-deficient mice also showed impaired hippocampus-dependent spatial learning. Although HNK-1 plays an essential role in synaptic plasticity and memory formation, it remains unclear how HNK-1 regulates these functions. In this study, we showed that loss of the HNK-1 epitope resulted in an increase of filopodium-like immature spines and a decrease of mushroom-like mature spines in both the early postnatal mouse hippocampus and cultured hippocampal neurons. However, HNK-1 had no influence on spine density or filopodium formation. Immunofluorescence staining revealed that loss of HNK-1 altered the distribution of postsynaptic proteins such as α-amino-3-hydroxy-5-methylisoxazolepropionate (AMPA)-type glutamate receptor subunit GluR2 and PSD-95 from spine heads onto dendritic shafts without affecting synapse formation, resulting in an increase of shaft synapses in cultured GlcAT-P-deficient neurons. GluR2, a major HNK-1 carrier glycoprotein in postsynaptic density, has the ability to promote spine morphogenesis. Overexpression of GluR2 promoted spine growth in both wild-type and GlcAT-P-deficient neurons, but the increase in GlcAT-P-deficient neurons was lower than that in wild-type neurons. This is the first evidence that HNK-1 is a key factor for normal dendritic spine maturation and is involved in the distribution of postsynaptic proteins.     

Remodelling of synaptic morphology but unchanged synaptic density during late phase long-term potentiation (LTP): a serial section electron micrograph study in the dentate gyrus in the anaesthetised rat

In anaesthetised rats, long-term potentiation (LTP) was induced unilaterally in the dentate gyrus by tetanic stimulation of the perforant path. Animals were killed 6 h after LTP induction and dendritic spines and synapses in tetanised and untetanised (contralateral) hippocampal tissue from the middle molecular layer (MML) were examined in the electron microscope using stereological analysis. Three-dimensional reconstructions were also used for the first time in LTP studies in vivo, with up to 130 ultrathin serial sections analysed per MML dendritic segment. A volume sampling procedure revealed no significant changes in hippocampal volume after LTP and an unbiased counting method demonstrated no significant changes in synapse density in potentiated compared with control tissue. In the potentiated hemisphere, there were changes in the proportion of different spine types and their synaptic contacts. We found an increase in the percentage of synapses on thin dendritic spines, a decrease in synapses on both stubby spines and dendritic shafts, but no change in the proportion of synapses on mushroom spines. Analysis of three-dimensional reconstructions of thin and mushroom spines following LTP induction revealed a significant increase in their volume and area. We also found an increase in volume and area of unperforated (macular) and perforated (segmented) postsynaptic densities. Our data demonstrate that whilst there is no change in synapse density 6 h after the induction of LTP in vivo, there is a considerable restructuring of pre-existing synapses, with shaft and stubby spines transforming to thin dendritic spines, and mushroom spines changing only in shape and volume.     

Differential plasma membrane distribution of metabotropic glutamate receptors mGluR1α, mGluR2 and mGluR5, relative to neurotransmitter release sites

Two group I metabotropic glutamate receptor subtypes, mGluR1 and mGluR5, have been reported to occur in highest concentration in an annulus surrounding the edge of the postsynaptic membrane specialisation. In order to determine whether such a distribution is uniform amongst postsynaptic mGluRs, their distribution was compared quantitatively by a pre-embedding silver-intensified immunogold technique at electron microscopic level in hippocampal pyramidal cells (mGluR5), cerebellar Purkinje cells (mGluR1α) and Golgi cells (mGluR2). The results show that mGluR1α, mGluR5 and mGluR2 each have a distinct distribution in relation to the glutamatergic synaptic junctions. On dendritic spines, mGluR1α and mGluR5 showed the highest receptor density in a perisynaptic annulus (defined as within 60 nm of the edge of the synapse) followed by a decreasing extrasynaptic (60–900 nm) receptor level, but the gradient of decrease and the proportion of the perisynaptic pool (mGluR1α, 50%; vs mGluR5, 25%) were different for the two receptors. The distributions of mGluR1α and mGluR5 also differed significantly from simulated random distributions. In contrast, mGluR2 was not closely associated with glutamatergic synapses in the dendritic plasma membrane of cerebellar Golgi cells and its distribution relative to synapses is not different from simulated random distribution in the membrane. The somatic membrane, the axon and the synaptic boutons of the GABAergic Golgi cells also contained immunoreactive mGluR2 that is not associated with synaptic specialisations. In the hippocampal CA1 area the distribution of immunoparticles for mGluR5 on individual spines was established using serial sections. The results indicate that dendritic spines of pyramidal cells are heterogeneous with respect to the ratio of perisynaptic to extrasynaptic mGluR5 pools and about half of the immunopositive spines lack the perisynaptic pool. The quantitative comparison of receptor distributions demonstrates that mGluR1α and mGluR5, but not mGluR2, are highly compartmentalised in different plasma membrane domains. The unique distribution of each mGluR subtype may reflect requirements for different transduction and effector mechanisms between cell types and different domains of the same cell, and suggests that the precise placement of receptors is a crucial factor contributing to neuronal communication.     

The termination of callosal fibres in the auditory cortex of the rat. A combined Golgi--electron microscope and degeneration study

When the corpus callosum of the rat is sectioned, the callosal fibres in the cerebral cortex undergo degeneration. In the auditory cortex (area 41) the degenerating axon terminals form asymmetric synapses, and the vast majority of them synapse with dendritic spines. Some other synapse with the shafts of both spiny and smooth dendrites, and a few with the perikarya of non-pyramidal cells. The degenerating axon terminals are contained principally within layer II/III, in which they aggregate in patches. Using a technique in which neurons within the cortex are Golgi-impregnated, then gold-toned and examined in the electron microscope, it has been shown that the dendritic spines of pyramidal neurons with cell bodies in different layers receive the degenerating callosal afferents. The spines arise from the main apical dendritic shafts and their branches, from the dendrites of the apical tufts, and in some cases from the basal dendrites of the pyramidal neurons. The shafts of some pyramidal cell apical dendrites also form asymmetric synapses with callosal afferents. Since we have encountered no spiny non-pyramidal neurons in Golgi preparations of rat auditory cortex, and because other types of non-pyramidal cells have few dendritic spines, it is concluded that practically all of the dendritic spines synapsing with callosal afferents originate from pyramidal neurons.     

Tyrosine hydroxylase-immunoreactive boutons in synaptic contact with identified striatonigral neurons, with particular reference to dendritic spines

Tyrosine hydroxylase-immunoreactive fibres in the rat neostriatum were studied in the electron microscope in order to determine the nature of the contacts they make with other neural elements. The larger varicose parts of such fibres contained relatively few vesicles and rarely displayed synaptic membrane specializations; however, thinner parts of axons (0.1-0.4 micron) contained many vesicles and had symmetrical membrane specializations, indicative of en passant type synapses. By far the most common postsynaptic targets of tyrosine hydroxylase-immunoreactive boutons were dendritic spines and shafts, although neuronal cell bodies and axon initial segments also received such input. Six striatonigral neurons in the ventral striatum were identified by retrograde labelling with horseradish peroxidase and their dendritic processes were revealed by Golgi impregnation using the section-Golgi procedure. The same sections were also developed to reveal tyrosine hydroxylase immunoreactivity and so we were able to study immunoreactive boutons in contact with the Golgi-impregnated striatonigral neurons. Each of the 280 immunoreactive boutons examined in the electron microscope displayed symmetrical synaptic membrane specializations: 59% of the boutons were in synaptic contact with the dendritic spines, 35% with the dendritic shafts and 6% with the cell bodies of striatonigral neurons. The dendritic spines of striatonigral neurons that received input from immunoreactive boutons invariably also received input, usually more distally, from unstained boutons that formed asymmetrical synaptic specializations. A study of 87 spines along the dendrites of an identified striatonigral neuron showed that the most common type of synaptic input was from an individual unstained bouton making asymmetrical synaptic contact (53%), while 39% of the spines received one asymmetrical synapse and one symmetrical immunoreactive synapse. It is proposed that the spatial distribution of presumed dopaminergic terminals in synaptic contact with different parts of striatonigral neurons has important functional implications. Those synapses on the cell body and proximal dendritic shafts might mediate a relatively non-selective inhibition. In contrast, the major dopaminergic input that occurs on the necks of dendritic spines is likely to be highly selective since it could prevent the excitatory input to the same spines from reaching the dendritic shaft. One of the main functions of dopamine released from nigrostriatal fibres might thus be to alter the pattern of firing of striatal output neurons by regulating their input.     

Characterization of input synapses on intracellularly stained neurons in hippocampal slices: an HRP/EM study

Summary This study describes the fine structure of input synapses on identified neurons in slices of the guinea pig hippocampus. For morphological identification, granule cells of the fascia dentata and pyramidal neurons of regio inferior of the hippocampus were impaled and intracellularly stained with horse-radish peroxidase (HRP). Input synapses on the HRP-stained neurons were identified in the electron microscope by the location of the synapses in inner or outer zones of the dentate molecular layer, as in the case of the synaptic contacts on injected granule cells, or by unique fine structural characteristics, as in the case of the giant mossy fiber boutons on CA3 pyramidal cells. As in tissue fixed in situ by transcardial perfusion, a large number of terminals arising from the different afferents in inner and outer zones of the dentate molecular layer were well preserved and formed synaptic contacts with small spines, large complex spines, and dendritic shafts of the HRP-filled granule cells. Mossy fiber synapses on the stained CA3 neurons were densely filled with clear vesicles, contained a few dense-core vesicles, and formed synaptic contacts with large spines or excrescences. Occasionally electrondense degenerating boutons were also found impinging on the stained dendrites and spines. The significance of the present findings for electrophysiological and pharmacological studies on brain slices is discussed.