Effects of prenatal protein malnutrition on mossy fibers of the hippocampal formation in rats of four age groups

The present study was undertaken to investigate the effect of prenatal protein deprivation on the postnatal development of the mossy fiber plexus of the hippocampal formation on postnatal (P) days 15, 30, 90, and 220. Although there is extensive information about the effects of malnutrition on cell body and dendrite morphology, little attention has been paid to axons or axon plexuses. The mossy fiber plexus represents the dentate gyrus granule cell axonal projection to areas CA4 and CA3 of the hippocampal formation and is readily demonstrated with Timm's heavy metal stain. With the use of this stain, the plexus was measured at 13 levels throughout the hippocampal complex. There was no effect of the diet on the anatomical distribution of the plexus. The current study, however, does show significant effects of prenatal protein malnutrition on postnatal development of the mossy fiber plexus that are age dependent. The prenatally malnourished rats show significant deficits in the total rostro-caudal extent and volume of the plexus on P15, P90, and P220, with the most marked dietary effect on P220. There was no significant diet effect on P30 in either extent or volume.

Stress-induced changes in brain-derived neurotrophic factor expression are attenuated in aged Fischer 344/N rats

Aging and stress can sometimes result in a decline in brain function. We addressed the question whether changes in the expression of neurotrophic factors, which are necessary for the survival and maintenance of neurons, might occur during aging and stress. Therefore, we used in situ hybridization to investigate the effects of aging and stress on neurotrophic factor expression in young (3-4 month) and old (24 month) male Fischer 344/N rats. The ability of acute immobilization stress (2 h) to modulate BDNF mRNA levels in old rats was significantly reduced both in the hippocampus (a smaller decrease in BDNF) and the PVN (a smaller increase in BDNF) compared to young rats. In contrast, the induction of nerve growth factor and neurotrophin 3 (NT-3) by stress was not influenced by age. The diminished BDNF responses to stress in aged rats may be relevant to difficulties in adaptation to stress encountered during old age.

Studies of dentate granule cell modulation: paired-pulse responses in freely moving rats at three ages

Dentate granule cell population responses to paired-pulse stimulations applied to the perforant pathway across a range of interpulse intervals (IPI) were examined in freely moving rats at 15, 30, and 90 days of age. The profile of the paired-pulse index (PPI), a measure of the type and degree of modulation of dentate granule cell excitability, was shown to change significantly as a function of age.

Long-term depression at the medial perforant path-granule cell synapse in developing rat dentate gyrus

Long-term depression (LTD) is a decrease in synaptic efficacy that may model the elimination of inappropriate synapses during brain development. LTD might therefore be expected to be prominent in the juvenile hippocampal dentate gyrus (DG), where the majority of neuronogenesis and excitatory synapse production and pruning occur in the first postnatal month. Thus far, however, LTD in immature DG remains unexplored. Low-frequency stimulus induced homosynaptic LTD was studied at the medial perforant path-granule cell synapse in rats 8-30 days of age. LTD was most consistent and was of greatest magnitude in the youngest animals, and was more robust in response to stimulation at 1 Hz than at 3 or 5 Hz. LTD was saturable by repetitive delivery of low-frequency stimulation, and reversible by tetanic stimulation that induced long-term potentiation (LTP). LTD of the field EPSP was not prevented by bath application of the NMDA receptor antagonist AP5, the mGluR antagonist MCPG, or the L-type voltage sensitive calcium channel antagonist nitrendipine. In whole cell recordings LTD induction was blocked by hyperpolarization of the postsynaptic neuron but not by calcium chelation with BAPTA. Calcium chelation blocked LTP and simultaneously unmasked tetanus induced LTD. These data demonstrate that LTD is prominent in immature DG, that LTP and LTD are complementary processes, and that LTD is likely to be induced postsynaptically because it is voltage dependent, although the mechanism of voltage dependence remains to be elucidated.

Developmental changes in membrane properties and postsynaptic currents of granule cells in rat dentate gyrus

1. Whole cell patch-clamp recordings were used to study dentate gyrus granule cells in hippocampal slices from juvenile rats (postnatal days 8-32). Membrane properties were measured with the use of current-clamp recordings and were correlated with the morphology of a subgroup of neurons filled with biocytin. The components of the postsynaptic currents (PSCs) induced by medial perforant path stimulation were characterized with the use of specific receptor antagonists in voltage-clamp recordings. 2. Granule cells located in the middle third of the superior blade of stratum granulosum from the rostral third of hippocampus were divided into three groups according to their input resistance (IR). Neurons with low IR (206 +/- 182 M omega, mean +/- SD) had hyperpolarized resting membrane potentials (-82 +/- 7 mV) and high-amplitude action potentials (108 +/- 23 mV). Neurons were high IR (1,259 +/- 204 M omega) had more depolarized resting membrane potentials (-54 +/- 6 mV) and lower-amplitude action potentials (71 +/- 10 mV). Neurons with intermediate IR (619 +/- 166 M omega) also had intermediate resting membrane potentials (-63 +/- 7 mV) and action potential amplitudes (86 +/- 14 mV). Low-IR neurons became increasingly prevalent with advancing postnatal age, but neurons from each group could be found throughout the entire period under study. 3. Morphological studies of low-IR neurons revealed an extensive dendritic arborization that traversed the entire molecular layer and was characteristic of mature granule cells. High-IR cells had smaller somata and short, simple dendritic arborization that incompletely penetrated the molecular layer and were classified as immature. Intermediate-IR cells had morphological features of intermediate maturity. 4. The initial phase of the PSC evoked at -80 mV was a fast inward current that was comparable with respect to latency to peak, latency to onset, and 10-90% rise time in neurons of all maturities held at -80 mV. This current was 6-cyano-7-nitroquinoxaline-2,3-dione sensitive. 5. The decay phases of PSCs at -80 mV varied with neuronal maturity. Mature neurons had monoexponential decays (tau = 8.9 +/- 3.6). Intermediate and immature neurons had prominent later inward currents that resulted in slower decays. In the case of the immature neurons, the inward current during the decay phase could be separated from the initial fast inward peak. The later inward currents in intermediate and immature neurons were bicuculline sensitive. 6. With the use of uniform ionic conditions of the extracellular and patch solutions, current-voltage relations and reversal potentials for pharmacologically isolated alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), N-methyl-D-aspartate (NMDA), and gamma-aminobutyric acid-A (GABAA) currents were comparable across all cell maturities. Calculated ratios for peak GABAA/NMDA/AMPA currents decreased significantly with maturation as follows: 9.4 +/- 2.9/1.4 +/- 0.5/1.0 for immature cells, 7.2 +/- 2.5/1.5 +/- 0.7O/1.0 for intermediate cells, and 2.0 +/- 1.2/0.9 +/- 0.4/1.0 for mature cells. 7. GABA current was mediated both by polysynaptic activation of interneurons and by direct activation of interneurons with monosynaptic input onto granule cells. The proportional contributions of mono- and polysynaptic GABA to total GABA were comparable across all cell maturities; latency to peak GABA current decreased with increasing cell maturity for both mono- and polysynaptic components. 8. We conclude that PSCs evoked in granule cells by medial perforant path activation in neurons of all maturities consist of both glutamatergic and GABAergic components. PSCS are dominated by GABAergic neurotransmission in immature granule cells, and the contribution of glutamatergic neurotransmission increases with neuronal maturation. The greater ratio of peak GABAA to glutamate currents and the longer time interval between their respective peaks combine to produce a distinctive PSC shape

Circuit-specific alterations of N-methyl-D-aspartate receptor subunit 1 in the dentate gyrus of aged monkeys

Age-associated memory impairment occurs frequently in primates. Based on the established importance of both the perforant path and N-methyl-D-aspartate (NMDA) receptors in memory formation, we investigated the glutamate receptor distribution and immunofluorescence intensity within the dentate gyrus of juvenile, adult, and aged macaque monkeys with the combined use of subunit-specific antibodies and quantitative confocal laser scanning microscopy. Here we demonstrate that aged monkeys, compared to adult monkeys, exhibit a 30.6% decrease in the ratio of NMDA receptor subunit 1 (NMDAR1) immunofluorescence intensity within the distal dendrites of the dentate gyrus granule cells, which receive the perforant path input from the entorhinal cortex, relative to the proximal dendrites, which receive an intrinsic excitatory input from the dentate hilus. The intradendritic alteration in NMDAR1 immunofluorescence occurs without a similar alteration of non-NMDA receptor subunits. Further analyses using synaptophysin as a reflection of total synaptic density and microtubule-associated protein 2 as a dendritic structural marker demonstrated no significant difference in staining intensity or area across the molecular layer in aged animals compared to the younger animals. These findings suggest that, in aged monkeys, a circuit-specific alteration in the intradendritic concentration of NMDAR1 occurs without concomitant gross structural changes in dendritic morphology or a significant change in the total synaptic density across the molecular layer. This alteration in the NMDA receptor-mediated input to the hippocampus from the entorhinal cortex may represent a molecular/cellular substrate for age-associated memory impairments.

Effect of prenatal protein deprivation on postnatal granule cell generation in the hippocampal dentate gyrus

The effect of prenatal malnutrition, produced by protein deprivation, on postnatal neurogenesis of granule cells in the fascia dentata of the rat hippocampal formation was examined by injecting tritiated thymidine on P8 and P15 and sacrificing the pups on P30, or by injecting on P30 and sacrificing on P90. The number of labeled granule cells was significantly decreased in prenatally malnourished rats injected on P8, and unaffected in those injected on P15. In contrast, the number of labeled granule cells in prenatally malnourished rats was significantly increased in animals injected in P30. The study shows that prenatal malnutrition significantly alters the postnatal pattern of granule cell neurogenesis in rat hippocampal formation and that the effect persists despite nutritional rehabilitation at birth.

Diet-induced alterations in the ontogeny of long-term potentiation

The ability of prenatally malnourished rats to establish and maintain long-term potentiation (LTP) of the perforant path/dentate granule cell synapse was examined in freely moving rats at 15, 30, and 90 days of age. Measures of the population EPSP slope and population spike amplitude (PSA) were calculated from dentate field potential recordings obtained prior to and at various times following tetanization of the perforant pathway. Significant enhancement of both population EPSP slope and PSA measures was obtained from all animals of both malnourished and well-nourished diet groups at 15 days of age. However, the magnitude of enhancement obtained from 15-day-old prenatally malnourished animals was significantly less than that of age-matched, well-nourished controls. At 30 days of age, PSA measures obtained from approximately 50% of prenatally malnourished 30-day-old rats showed no significant effect of tetanization, while measures obtained from the remaining 50% of these animals did not differ significantly from controls. EPSP slope measures for this age group followed much the same pattern, i.e., malnourished animals showing no significant enhancement of PSA measures exhibited only slight increases in EPSP slope beginning 1 h after tetanization and returned to baseline by 18 h post-tetanization. EPSP slope measures obtained from PSA-enhanced malnourished animals did not differ significantly from controls. At 90 days of age, PSA measures obtained from 50% of malnourished animals declined from pretetanization levels immediately following tetanization. Three hours after tetanization, however, this measure had increased to a level which did not differ significantly from that of the control group. PSA measures obtained from the remaining 50% of 90-day-old malnourished animals showed initial and sustained enhancement which did not differ significantly from those obtained from well-nourished age-matched controls. These results indicate that gestational protein malnutrition significantly affects the magnitude of tetanization-induced enhancement of dentate granule cell response in preweanling rats (15-day-old animals) and significantly alters the time-course and magnitude of potentiation in approximately half of prenatally malnourished animals tested at 30 and 90 days of age. Given the primarily postnatal development of the dentate granule cells, these results may reflect malnutrition-induced delays in the neurogenesis and functional development of granule cells previously reported by our group. Most striking is the fact that significant impairments in LTP establishment were obtained from prenatally malnourished animals at 90 days of age, implying that dietary rehabilitation commencing at birth is an intervention strategy incapable of ameliorating the effects of the gestational insult.

Regulation of neuronal birth, migration and death in the rat dentate gyrus

The granule cell population of the rat dentate gyrus forms over an extended period which begins during gestation and continues into adulthood. During the embryonic period, the postnatal period and in adulthood, granule cells proliferate, migrate and degenerate. We have found that granule cell production is dependent on the levels of circulating adrenal steroids and NMDA receptor-mediated excitatory input throughout life. In general, increases in adrenal steroid levels or NMDA receptor activation diminish the rate of cell proliferation whereas decreases in adrenal steroid levels or NMDA receptor activation increase the rate of cell production. This paper describes the regulation of granule cell proliferation, migration and survival by adrenal steroids and excitatory input and presents evidence that these factors may affect dentate gyrus-mediated behaviors.

Evidence for the elaboration of multiple axons by developing dentate granule cells

A sample of 185 Neurobiotin-filled dentate granule cells has been collected from rats aged P14 to P120 in a study of the effects of serotonin (5-hydroxytryptamine, 5-HT) depletion on granule cell development. A small number (5.9%) of these neurons exhibit more than one axon. These neurons have morphologies consistent with that described for dentate granule cells. One axon typically arises from the soma while additional axons take origin from either dendrites or the soma. Both axons may be mossy fibers or one axon may be a mossy fiber and the second assume a morphology and distribution similar to that described for interneurons. These latter neurons therefore exhibit a mixed phenotype by having a granule cell morphology and an interneuronal axon type. These data suggest that some granule cells give rise to multiple mossy fibers or express a mixed axonal phenotype during maturation.

Developmental variation of the permeability to Ca2+ of AMPA receptors in presumed hilar glial precursor cells

Glial cells in the hilus of the dentate gyrus of the rat were investigated using the patch-clamp technique in acute slices of the hippocampal formation. According to their voltage-gated current patterns, two classes of glial cells--putative astrocytes and presumed glial precursor cells--were apparent. The glutamate receptor agonists kainate, glutamate, and alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) evoked inward currents at a holding potential of -70 mV in astrocytes and presumed glial precursor cells. Inward currents could also be induced in nucleated patches, indicating a direct action on glial receptors. In presumed hilar glial precursor cells, 6,7-dinitroquinoxaline-2,3-dione (DNQX; 10 microM) blocked the kainate-induced current, while it was partially inhibited by Zn2+ (2 mM) and Evans Blue (10 microM). Cyclothiazide (100 microM), in contrast, potentiated this current, indicating the presence of AMPA receptors. In 90% of the presumed glial precursor cells the excitatory amino-acid-evoked current voltage (I/V) relations were linear or outwardly rectifying and reversed close to 0 mV, which is characteristic for non-specific cation channels. To determine the permeability to Ca2+, I/V relations were determined in a Na(+)-free solution containing 40 mM Ca2+ and showed reversal potentials of a wide variation ranging from -63 mV to +1 mV with corresponding PCa/PCs permeability ratios of between 0.09 and 2.10. Statistical analysis revealed that the permeability to Ca2+ significantly decreased with an advance in age (r = -0.596; n = 21; P < 0.01). These data suggest that the Ca2+ influx mediated by the activation of AMPA receptors expressed in presumed hilar glial precursor cells is dependent on the developmental stage.

Polysialylated neural cell adhesion molecule expression by neurons and astroglial processes in the rat dentate gyrus declines dramatically with increasing age

The expression of polysialylated neurons in the dentate gyrus of the hippocampal formation of young (postnatal day 40), mature (postnatal day 80) and aged (postnatal day 540) male Wistar rats has been investigated by immunohistochemical techniques employing a monoclonal antibody specific for neural cell adhesion molecule-linked alpha 2,8 polysialic acid. A strong immunoreactivity was found on the cell bodies, dendrites and axons of granule-like neuronal cells at the border between the hilar region and the granule cell layer of the young rat. In the mature animal the number of immunoreactive neurons declined dramatically and were virtually absent in the aged group. Using an alternative fixation procedure, glial fibrillary acidic protein-positive and polysialylated astroglia processes were found in close proximity to the dendrites of the polysialylated granule-like cells. The number of astroglial processes traversing the granule cell layer showed a similar age-dependent decline to that observed with the polysialylated neurons. Glial fibrillary acidic protein-positive and polysialylated stellate astroglia were present throughout the hippocampal formation, but did not show the marked age-dependent decline observed with the astroglial processes in the granule cell layer. The neuronal dendrites and astroglial processes exhibited a strict numerical ratio in the young and mature animal and, in double immunofluorescence studies with anti-polysialic acid and anti-glial fibrillary acidic protein, the astroglial processes exhibited apparent points of cell and/or dendritic contact. These findings suggest that loss of polysialylated astroglial processes precedes the decline in polysialylated dentate neurons.

Moderate copper deprivation during gestation and lactation affects dentate gyrus and hippocampal maturation in immature male rats

The hippocampal formation (HF) is involved in higher brain functions including learning and declarative memory. The possibility that dietary copper has a role in the morphological development, and therefore the function of the HF, has received little attention. A rat model of tiered copper deficiency, initiated during gestation, was employed to determine the susceptibility of the HF, regions of which develop postnatally, to copper deficiency. At postnatal 23, pups whose dams had received either 1.8 or 1.4 mg Cu/kg diet during both gestation and lactation, compared with offspring of a group that had received 4.3 mg Cu/kg diet during both periods had, significantly more cell nuclei in the infrapyrimidal arm of the dentate gyrus. Offspring of rats fed 1.4 mg Cu/kg diet, but not those fed 1.8 mg/kg, compared with those fed 4.3 mg/kg, exhibited smaller, shorter, and narrower cell nuclei in the infrapyrimidal and suprapyrimidal arms of the dentate gyrus and smaller cell nuclei in region CA3c of the hippocampus. A fourth group (gestation, 1.8 mg Cu/kg diet; lactation, 0.9 mg Cu/kg diet) exhibited alterations less marked than those exhibited by the group fed 1.4 mg Cu/kg diet. All alterations in the groups fed low copper diets were consistent with slowed cell nuclear maturation. The findings indicate that copper is required for maturation of the dentate gyrus and hippocampus. Also, copper supplied at or below 1.8 mg/kg is insufficient for morphological maturation of the dentate gyrus and hippocampus. Because the HF is important for higher brain functions, further research is needed to determine whether the copper deficiency-induced alterations in dentate gyrus and hippocampus development are transient or permanent.

The process of reinnervation in the dentate gyrus of adult rats: gene expression by neurons during the period of lesion-induced growth

Neurons in the hippocampal dentate gyrus are extensively reinnervated following the destruction of their normal inputs from the ipsilateral entorhinal cortex (EC). The present study evaluates gene expression by dentate granule neurons and the neurons giving rise to the sprouting connections during the period of synapse growth. Adult male rats were prepared for in situ hybridization at 2, 4, 6, 8, 10, 12, 14, 20, and 30 days following unilateral EC lesions. Sections were hybridized using 35S-labeled cRNA probes for mRNAs that encode proteins thought to be important for neuronal structure and/or synapse function, including (1) mRNAs that are normally present in dendrites--the mRNAs for the high molecular weight microtubule-associated protein 2 (MAP2) and the alpha-subunit of calcium/calmodulin-dependent protein kinase II (CAMII kinase), (2) mRNAs that are upregulated in neurons that are regenerating their axons (T alpha 1 tubulin and F1/GAP43) and (3) mRNAs for proteins that are the principal constituents of neurofilaments and microtubules (the low molecular weight neurofilament protein NF68 and beta-tubulin). Although there were small changes in the levels of labeling for the mRNAs that are normally present in dendrites, there were no dramatic increases in the levels of any of the mRNAs either in dentate granule cells or in neurons giving rise to the reinnervating fibers at any postlesion interval. These results indicate that neurons in mature animals can substantially remodel their synaptic terminals and their dendrites in the absence of large-scale changes in gene expression (at least as measured by steady-state mRNA levels at various time points).

Evidence for physiological growth of hippocampal mossy fiber collaterals in the guinea pig during puberty and adulthood

By means of Timm's procedure and computer-assisted morphometry, the left and right hippocampi of 69 hybrid guinea pigs from nine age levels (P5, P10, P20, P40, P80, P160, P320, and P610, and P1100) were analyzed for postnatal growth of recurrent hippocampal mossy fiber collaterals (RMFC) terminating below, within, and above the dentate granule cell layer. Postnatal growth of RMFCs showed, in both sexes, a first peak at P40, with stainable mossy fiber boutons covering the cell bodies of large neurones, some of which were reminiscent of basket cells. No significant changes of the density of mossy fiber collaterals were noticed from P40 to P160. At P320 a remarkable expansion of RMFCs was noted in a few animals, and by P610 all animals showed highly proliferated RMFCs which densely covered cell bodies and dendrites of target cells. The oldest group (P1100) showed an equal or slightly lowered density of RMFCs. We conclude that the growth of recurrent mossy fiber collaterals occurs in two spurts. The first completes just before sexual maturity. The second spurt occurs in the mid-life period, between P160 and P610.