Number and regional distribution of GAD65 mRNA-expressing interneurons in the rat hippocampal formation.

In rodent models for neuropsychiatric disorders reduced number of hippocampal interneurons have been reported, but the total number of GABAergic neurons in the normal rat hippocampus is yet unknown. We used in situ hybridization method to label the 65 isoform of glutamic acid decarboxylase (GAD65) and counted the number of GAD65 mRNA-expressing neurons along the entire septo-temporal axis of the hippocampus. We found that 2/3 of the interneurons were in Ammon's horn (61,590) and 1/3 in the dentate gyrus (28,000). We observed the following numbers in Ammon's horn: CA3 area 33,400, CA2 area 4,190, CA1 area 24,000 and in the dentate gyrus: 6,000 in the molecular and 9,000 in the granule cell layers and 13,000 in the hilus. GAD65 mRNA-expressing neurons were significantly more numerous in dorsal than in ventral hippocampus. The ratio between interneurons and principal cells was lowest in the granule cell layer (0.9%) and highest in hilus (21%). In Ammon's horn this ratio was constant being 13% in CA3 and 8% in CA1-2 areas. In the entire hippocampal formation, the interneuron/principal cell ratio was 6%, with a significant difference between Ammon's horn (9.5%) and the dentate gyrus (2.8%) including the hilus. Such low ratios could suggest that even a limited loss of GABAergic neurons in the hippocampus may have a considerable functional impact.

Parathyroid hormone 2 receptor and its endogenous ligand tuberoinfundibular peptide of 39 residues are concentrated in endocrine, viscerosensory and auditory brain regions in macaque and human.

Parathyroid hormone receptor 2 (PTH2R) and its ligand, tuberoinfundibular peptide of 39 residues (TIP39) constitute a neuromodulator system implicated in endocrine and nociceptive regulation. We now describe the presence and distribution of the PTH2R and TIP39 in the brain of primates using a range of tissues and ages from macaque and human brain. In situ hybridization histochemistry of TIP39 mRNA, studied in young macaque brain, due to its possible decline beyond late postnatal ages, was present only in the thalamic subparafascicular area and the pontine medial paralemniscal nucleus. In contrast, in situ hybridization histochemistry in macaque identified high levels of PTH2R expression in the central amygdaloid nucleus, medial preoptic area, hypothalamic paraventricular and periventricular nuclei, medial geniculate, and the pontine tegmentum. PTH2R mRNA was also detected in several human brain areas by RT-PCR. The distribution of PTH2R-immunoreactive fibers in human, determined by immunocytochemistry, was similar to that in rodents, including dense fiber networks in the medial preoptic area, hypothalamic paraventricular, periventricular and infundibular (arcuate) nuclei, lateral hypothalamic area, median eminence, thalamic paraventricular nucleus, periaqueductal gray, lateral parabrachial nucleus, nucleus of the solitary tract, sensory trigeminal nuclei, medullary dorsal reticular nucleus, and dorsal horn of the spinal cord. Co-localization suggested that PTH2R fibers are glutamatergic, and that TIP39 may directly influence hypophysiotropic somatostatin containing and indirectly influence corticotropin releasing-hormone containing neurons. The results demonstrate that TIP39 and the PTH2R are expressed in the brain of primates in locations that suggest involvement in regulation of fear, anxiety, reproductive behaviors, release of pituitary hormones, and nociception.

Morphological analysis of the connective tissue reaction in linear hypertrophic scars treated with intralesional steroid or silicone-gel sheeting. A light and electron microscopic study.

The linear hypertrophic scar has become the most common type of pathologic scarring. Silicone-gel sheeting is the first line therapy while intralesional steroid is the second. A light and electron microscopic analysis was carried out to reveal differences in tissue reaction following the two different treatments. Two groups of 12 patients each were treated for 4 months. For the first group, diluted Triamcinolone acetonide was injected until an inactive state was achieved. The other group of patients was treated with silicone-gel sheeting. The scars were examined every two weeks and their appearance documented. After reaching the expected therapeutic response, inactive scars were removed. The excised scars were evaluated through light microscopic histopathology and electron microscopy. The light and electron microscopic observations revealed marked differences following treatments. The activity of fibroblasts and the numbers of collagen fibers forming bundles decreased and the orientation of the collagen fibers was more variable in the treated scars. The amount of elastic fibers increased after both steroid and silicone-gel sheeting treatment. Vascularization was also slightly changed, with more capillaries and fewer pre-capillary arteries detected in the treated scars. Both treatments resulted in the same decrease in score but steroid treatment was more rapid in onset. We suggest that the two different treatments work through different mechanisms, although the final functional outcome is similar.

Mossy cells and different subpopulations of pyramidal neurons are immunoreactive for cocaine- and amphetamine-regulated transcript peptide in the hippocampal formation of non-human primates and tree shrew (Tupaia belangeri).

Cocaine- and amphetamine-regulated transcript peptide mRNA was discovered in the rat striatum following cocaine and amphetamine administration. Since both psychostimulants elicit memory-related effects, localization of cocaine- and amphetamine-regulated transcript peptide in the hippocampal formation may have functional importance. Previous studies demonstrated different cellular localizations of cocaine- and amphetamine-regulated transcript peptide in humans and in rodents. Mossy cells were cocaine- and amphetamine-regulated transcript-positive in the human dentate gyrus, whereas granule cells contained this peptide in the rat. In the present study, the localization of cocaine- and amphetamine-regulated transcript peptide was examined using immunohistochemistry in the hippocampal formation of the rhesus monkey (Macaca mulatta), the common marmoset monkey (Callithrix jacchus) and in the tree shrew (Tupaia belangeri). In these species principal neurons of the hippocampal formation were cocaine- and amphetamine-regulated transcript-immunoreactive. In both monkeys and tree shrews, mossy cells of the hilus were cocaine- and amphetamine-regulated transcript-positive whereas granule cells of the dentate gyrus were cocaine- and amphetamine-regulated transcript-negative. The dense cocaine- and amphetamine-regulated transcript-immunoreactive axonal plexus of the associational pathway outlined the inner one-third of the dentate molecular layer. In the hippocampus of the tree shrew and marmoset monkey, a subset of CA3 pyramidal cells were cocaine- and amphetamine-regulated transcript-immunoreactive. In the marmoset monkey, cocaine- and amphetamine-regulated transcript labeling was found only in layer V pyramidal cells of the entorhinal cortex, while in the rhesus monkey, pyramidal cells of layers II and III were cocaine- and amphetamine-regulated transcript-immunopositive. Our results show that cocaine- and amphetamine-regulated transcript positive neurons in the dentate gyrus of non-human primates are similar to that of the human. Furthermore, in the hippocampal formation of the tree shrew similar cocaine- and amphetamine-regulated transcript-immunoreactive cell-types were observed as in monkeys, supporting their evolutionary relationship with primates. Mossy cells and granule cells are members of a mutual excitatory intrahippocampal circuitry, therefore cocaine- and amphetamine-regulated transcript-immunoreactivity of these neurons in primates and rodents suggests that psychostimulants cocaine and amphetamine may induce memory-related effects at different points of the same excitatory circuitry in the hippocampal formation.

Novel calretinin and reelin expressing neuronal population includes Cajal-Retzius-type cells in the neocortex of adult pigs.

Cajal-Retzius cells and their secreted product reelin are essential for the lamination of the cerebral cortex. In all species studied to date Cajal-Retzius cells form a transient neuronal population that almost completely disappears from the neocortex postnatally. Recently, in the hippocampal formation of adult domestic pig, we have found a large calretinin- and reelin-immunoreactive cell population that morphologically corresponded to Cajal-Retzius cells. In the present study, we examined calretinin- and reelin-immunoreactive neurons in layer I of the prefrontal, temporal, parietal and occipital neocortical areas of newborn, young adult and adult domestic pigs. Large numbers of bipolar or fusiform calretinin-positive cells were found in the upper half of layer I in all examined age groups. The morphology of these neurons resembled that of the Cajal-Retzius cells. Layer I was occupied by a dense calretinin-positive axonal plexus that was similar to the previously described axons of Cajal-Retzius cells in other species. In a similar location, where calretinin-positive cells occurred in layer I, large numbers of reelin-immunoreactive cells were found in all examined age groups. In addition, reelin colocalized with calretinin in layer I neurons. The number of calretinin and reelin-positive neurons decreased from 1 day to one year, but calretinin-positive Cajal-Retzius-type cells still comprised a remarkable large population in 12-month-old animals. Correlated light and electron microscopic examination of calretinin-labeled Cajal-Retzius-type cells indicated that these cells are integrated in the synaptic circuitry of the neocortex. Our results suggest that Cajal-Retzius cells do not disappear inevitably from the mature neocortex in all mammalian species. The function of this cell type is not known, but late persisting Cajal-Retzius-type cells in the domestic pig provide an opportunity to study their neuronal connections and the possible role of reelin in plasticity and regeneration of neocortex.

NADPH-diaphorase positive neurons of the rat hippocampal formation: regional distribution, total number and colocalization with calcium binding proteins.

The present study aimed to asses the total number and distribution of the NADPH-diaphorase-positive non-pyramidal neurons in Ammon's horn and dentate gyrus of rat hippocampal formation. Cell bodies were counted according to the "disector" principle. The total numbers varied from 27 000 to 32 400. In all strains, approximately one third of the NADPH-diaphorase-reactive non-principal cells were found in the dentate gyrus and the remaining two thirds were within the Ammon's horn. Analysis of the dorsoventral differences revealed that approximately 70% of NADPH-diaphorase-positive cells were in the dorsal and 30% in the ventral hippocampus. Distribution of NADPH-diaphorase-reactive cells in the different layers of the dentate gyrus and Ammon's horn was similar in all strains. Double-labelling studies revealed colocalization of NADPH-diaphorase with calretinin, but none with calbindin or parvalbumin. NADPH-diaphorase-positive neurons appear to form the largest chemically identified subpopulation of the GABAergic inhibitory cell population of the hippocampal formation.

Cocaine- and amphetamine-regulated transcript peptide (CART) is a selective marker of rat granule cells and of human mossy cells in the hippocampal dentate gyrus.

Cocaine- and amphetamine-regulated transcript (CART) peptide immunocytochemistry was used to reveal cellular localization in the dentate gyrus and in Ammon's horn of the rat and human hippocampal formations. In the rat dentate gyrus, only granule cells were labeled, whereas in humans, only mossy cells of the hilar region expressed CART peptide immunoreactivity. In the rat, CART-positive granule cells were located at the molecular layer border of the granule cell layer and had no features that would distinguish them from other granule cells. The mossy fiber bundle was labeled in the hilus as well as along the entire CA3 area of Ammon's horn. In the human, CART-immunoreactive mossy cells displayed the characteristic thorny excrescences both on their somata and their main dendrites. Axon collaterals of mossy cells could be seen in the hilus and the main axons formed a dense band in the inner molecular layer of the dentate gyrus, suggesting that mossy cells are the principal source of the associational pathway. Granule cells of the dentate gyrus and pyramidal neurons of the human hippocampal formation were devoid of CART peptide immunoreactivity. A few labeled non-pyramidal cells and a large group of strongly immunostained axons of unknown origin were present in all layers of CA1-3. Granule cells are the main excitatory cell population of the dentate gyrus while mossy cells are in a key position in controlling activity of granule cells. The specific location of CART peptide in the dentate granule cells of rodents and in the mossy cells of the human hippocampus may indicate involvement of neuronal circuitry of the dentate gyrus in the memory-related effects of cocaine and amphetamine. Independently of its functional role, CART peptide can be used as a specific marker of human mossy cells and of the dentate associational pathway. The sensitivity of CART peptide to postmortem autolysis may restrict the use of this marker in surgically removed hippocampi or in human brains removed and fixed shortly after death.

Axosomatic synapses on granule cells are preserved in human non-infiltrating tumour or lesion-related and mesial temporal sclerotic epilepsy, but markedly reduced in tumour-infiltrated dentate gyrus with or without epilepsy.

Granule cells of the human hippocampal dentate gyrus were examined. In controls, granule cells displayed somatic spines and cell nuclei with small infoldings. In addition, the cytoplasm of human granule cells always displayed lipofuscin. Subsurface cisterns of endoplasmic reticulum were frequently observed in the human granule cells. Two types of axosomatic synapses were found; most frequently symmetric and less frequently asymmetric. Many of the axosomatic synapses were isolated by glial processes in tumour or lesion-related epileptic patients, but the ultrastructural characteristics of granule cells were not different from those of the control patients. Large bundles of reactive astroglial fibres appeared regularly in all layers of the dentate gyrus. In tumour infiltrated hippocampi, glial processes dominated the neuropil and the number of perisomatic synapses was markedly reduced. Reduction in the number of perisomatic synapses did not correlate with severity and duration of seizures but did correlate with the malignancy of the tumour. It is suggested that reduction of perisomatic inhibition may not be a characteristic of granule cells in the epileptic human dentate gyrus.

Mitochondrial DNA4977 deletion in brain of newborns died after intensive care.

Mitochondrial DNA (mtDNA) deletion affecting 4977 base pairs (mtDNA4977), the most common mtDNA mutation in humans, was analysed in brain specimens (frontal, temporal, and cerebellar cortices, caudate nucleus, thalamus, and hippocampus) and in other tissues (blood clot, liver, kidney, heart, and muscle) taken at autopsy of deceased neonates. mtDNA4977 deletion determined by polymerase chain reaction (PCR) could be demonstrated in each neonatal sample, however, quantity of mtDNA4977 deletion was less in the newborn samples than in those of the elderlies. Results obtained suggest that contrary to certain data mtDNA4977 deletion can be present in neonates. The mtDNA4977 deletion could be generated by perinatal hypoxia or temporary oxygen oversaturations during the intensive care of the neonates, as the mtDNA is sensitive to oxidative damage. In combination with other factors an additional causative role of mtDNA4977 deletion reported here cannot be ruled out in development of cerebral palsy or mental retardation of unknown origin often seen in neonates underwent neonatal intensive care procedures.

Neonatal anandamide treatment results in prolonged mitochondrial damage in the vanilloid receptor type 1-immunoreactive B-type neurons of the rat trigeminal ganglion.

Capsaicin acting on the vanilloid type 1 receptor (VR1) excites a subset of primary sensory neurons. Systemic capsaicin treatment of adult or neonatal rats results in selective damage of the B-type neurons in the rat sensory ganglia by causing a long-lasting mitochondrial lesion that has been described in detail in previous studies. The endocannabinoid, anandamide, exhibits an agonist effect on VR1 receptors. The physiological role of anandamide as a VR1 agonist is still uncertain. This study addresses whether high doses of anandamide induce similar ultrastructural changes to those described for capsaicin. The effect of neonatally administered anandamide (1 mg/kg) on neurons of the trigeminal ganglia and the hippocampal formation was examined in the light and electron microscope from the first day after injections to the 20th week after treatment. Anandamide was found to cause mitochondrial damage of the B-type neurons of trigeminal ganglia similar to what has been described for capsaicin. The time course of damage was also comparable. In addition to the cells of the trigeminal ganglia, B-type cells of dorsal root ganglia were also damaged. A-type neurons and satellite glial cells were not affected either in the trigeminal or in the dorsal root ganglia. In the hippocampal formation, where a subpopulation of local circuit neurons is known to contain cannabinoid type 1 (CB1) but not VR1 receptors, anandamide did not cause morphological changes of mitochondria either in the dentate gyrus or in Ammon's horn. At 3 weeks of age, all VR1-immunoreactive neurons in the trigeminal ganglia of animals treated neonatally with anandamide displayed swollen mitochondria. The results suggest that anandamide, at pharmacologically relevant doses, acts on the VR1 receptor and causes prolonged and selective mitochondrial damage of B-type sensory neurons, as has previously been described for capsaicin.

Neonatal capsaicin treatment results in prolonged mitochondrial damage and delayed cell death of B cells in the rat trigeminal ganglia.

Capsaicin acts on the vanilloid receptor subtype 1, a noxious heat-gated cation channel located on a major subgroup of nociceptive primary afferent neurons. Following the systemic capsaicin treatment of neonatal rats, the loss of B-type sensory neurons in trigeminal ganglion of adult rats with chemoanalgesia and abolition of neurogenic inflammation was investigated. Our quantitative morphometric analysis revealed that in the trigeminal ganglion of neonatal rats treated with 50 mg/kg s.c. capsaicin, the total number of neurons, morphology of B-type cells and cell-size histograms did not differ from that of the controls 1 or 5 days after treatment. These observations indicate that early cell death does not play a significant part in the loss of B-type cells, which in our sample was 39.4% on the 19th day. However under the electron microscope pronounced selective mitochondrial swelling with disorganized cristae was observed in B-type neurons at 1-20 weeks after capsaicin treatment. Daily treatment with nerve growth factor (NGF, 10 x 100 microg/kg s.c.), started 1 day after capsaicin injection, prevented the loss of B-type cells but did not counteract the development of long-lasting mitochondrial damage. After NGF treatment, partial restitution of chemonociception to capsaicin instillation into the eye occurred but capsaicin-induced inhibition of neurogenic plasma extravasation in the hindpaw evoked by topical application of mustard oil remained unaltered. We conclude, that capsaicin treatment in neonatal rats, as in the adults, destroys terminal parts of the sensory neurons supplied by vanilloid receptors and induces long-lasting mitochondrial swelling in the soma. We hypothesize that loss of NGF uptake results in delayed cell death of B-type neurons in neonates.

Age-related mitochondrial damage in the B-type cells of the rat trigeminal ganglia.

Aging is associated with signs of sensory impairment and neurological symptoms. Advancing age is characterized by increased thresholds of thermal, tactile and vibratory sensations. One important cause of the sensory disturbances has been stated to be the loss of neurons. Decreases have been observed in the number of peripheral nerve fibers and in the number of neurons in the spinal ganglia of rats. In the present study, the cytoplasmic organelles of the neurons of the trigeminal ganglia were examined in young and senescent rats in order to reveal the cause of cell loss during aging. Mitochondrial alterations, swelling and loss of internal cristae were observed from 23 week of age in the B-type neurons of the trigeminal ganglia. Other cytoplasmic elements were intact. Mitochondrial damage was never seen in A-type neurons and satellite glial cells. It was concluded that the ultrastructural changes in the mitochondria of the B-type cells may contribute to the nervous disturbances that occur in senescent individuals. The diminution of mitochondrial damage and the protection of B-type neurons through the use of nerve growth factors may prevent the sensory impairment late in life.

Cell formation in the human hippocampal formation from mid-gestation to the late postnatal period.

In the present study cell formation was studied in the human hippocampal formation from the 24th gestational week until the end of the first postnatal year. Proliferating cells were detected with the monoclonal antibody MIB-1. The cytoarchitectonic layers of Ammon's horn are formed before the 24th gestational week. In harmony with this observation, cell proliferation in the hippocampal ventricular zone is minimal after the 24th week. In addition, local cell multiplication in Ammon's horn is occasional and the proliferating cells are glial or endothelial cells. In contrast, cell formation continues in the hilar region of the dentate gyrus even after birth. Immature cells accumulate in the hilus, and at the border between the hilus and the granule cell layer throughout the first eight postnatal months. The subgranular zone of the dentate gyrus becomes a cell sparse area at about the 11th postnatal month, indicating that immature cells from the hilus have already migrated to the granule cell layer and differentiated into granule cells. There is an increase in glial cell proliferation both in Ammon's horn and the dentate gyrus at the 11.5th postnatal month suggesting the onset of myelination by the end of the first year. Our findings indicate that most pyramidal neurons of Ammon's horn are generated in the first half of pregnancy and no pyramidal neurons are formed after the 24th gestational week. In contrast, granule cells of the dentate gyrus proliferate in a decreasing rate during the second half of pregnancy and after birth. Proliferating neuronal precursors occur in a low percentage in the dentate gyrus of 3-, 5- and 11.5-month-old children.

Granule cells are the main source of excitatory input to a subpopulation of GABAergic hippocampal neurons as revealed by electron microscopic double staining for zinc histochemistry and parvalbumin immunocytochemistry.

Immunocytochemistry was combined with a recent modification of Timm's method to evaluate semiquantitatively the mossy fiber innervation of dendrites and somata of parvalbumin-containing neurons of the hilus of the dentate gyrus and the CA3 area of Ammon's horn. Using this electron microscopic double staining technique, it was found that (1) the overwhelming majority (95%) of terminals forming asymmetric synapses with parvalbumin-positive dendrites in the dentate hilus, and the strata pyramidale and lucidum of the CA3 area of Ammon's horn, originated from granule cells; (2) two-thirds of the asymmetric axosomatic terminals of parvalbumin-positive neurons contained zinc; and (3) no zinc-containing axon terminals formed synapses with somata or main dendritic shafts of the granule cells.

Effect of neonatal dentate gyrus lesion on allothetic and idiothetic navigation in rats.

Goal-directed navigation is believed to be the combined product of idiothetic and allothetic orientation. Although both navigation systems require the hippocampal formation, it is probable that different circuits implement them. Examination of Long-Evans rats with dentate gyrus lesions induced by neonatal X-ray irradiation may show the dissociation of these two components of navigation. Two recently developed place avoidance tasks on a rotating circular arena were used to test this hypothesis. In the first test, the position of the punished area is stable in the room frame but is permanently changing on the surface of the arena. This task requires the rat to use allothetic orientation and to disregard idiothetic orientation. In the second test, the prohibited area is fixed in the coordinate system of the arena and the experiment is conducted in complete darkness, forcing the rat to rely exclusively on idiothesis supported by substratal cues. The results suggest that the dentate gyrus lesion interferes less with idiothetic orientation than with allothetic orientation. In addition, an attempt was made to control the number of developing granule cells by exact timing of a single high dose of perinatal irradiation, and to measure the ensuing behavioral deficits. Rats irradiated at 6, 18, or 24 h after birth were tested as adults in the Morris water maze. Irradiated animals showed significant, but highly variable, learning deficit, but histological examination indicated that the granule cell loss did not correlate with the degree of behavioral impairment.

Cell formation in the cortical layers of the developing human cerebellum.

Cell proliferation has been studied in the human cerebellar cortex between the 24th gestational week and the 12th postnatal month. Intensive cell formation has been found in the external granular layer (EGL) of the human cerebellum, where the highest cell proliferation rate occurs between the 28th and 34th gestational weeks. This is followed by a gradual decrease that lasts up to the eighth postnatal month. As late in development as the fifth postnatal month, still 30% of cells of the EGL are labeled with the monoclonal antibody Ki-67, which is specific for dividing cells. The width of the EGL remained unchanged from the 28th gestational week to the end of the first postnatal month, when it starts to decrease and completely disappears by the 11th postnatal month. Large number of Ki-67 labeled cells occurs in the internal granular layer (IGL) between the 24th and 28th gestational weeks. From the 36th week onwards, the labeling index is less than 1%, although a few labeled cells have always been found in this layer even in the late postnatal period. Labeled cells are distributed in the entire width of the IGL. However, from the 34th gestational week, almost all labeled cells are found among and directly below the Purkinje cells. Their position, the nuclear features, and their occasionally stained cell processes suggest that those are Bergmann glial cells. There are few Ki-67 labeled cells in the molecular layer (ML) and in the white matter (WM) of the cerebellum throughout the examined period. It is likely that most of these are glial cells. Pyknotic index has been found to be small in all layers of the cerebellum during the examined period.

The use of a sodium tungstate developer markedly improves the electron microscopic localization of zinc by the Timm method.

The Timm's sulfide-silver method is frequently used for the demonstration of the mossy fiber bundle or sprouted mossy fibers in the normal or epileptic hippocampal dentate gyrus. Under the light microscope the results are excellent, but the ultrastructure is considerably impaired and the silver grains produced are too large as compared to the sizes of intra-synaptic structures. The present study was meant to test a series of physical developers containing, instead of gum arabic, sodium tungstate as protective colloid. One of them left the ultrastructure fairly intact and produced small, round silver grains, making it possible to precisely locate zinc in mossy terminals. With this method, it could be demonstrated that zinc is contained inside synaptic vesicles in the resting axon terminals of granule cells. As a consequence of prolonged sodium sulfide perfusion, zinc is released from synaptic vesicles and enters the synaptic cleft.

Muscle carnitine acetyltransferase and carnitine deficiency in a case of mitochondrial encephalomyopathy.

Profound decrease of the carnitine acetyltransferase activity (0.08 U/g wet weight; 1.67% of control) and carnitine deficiency (total carnitine was 230 nmol/g wet weight in the patient vs 2730 in the controls) was detected in the skeletal muscle of a female paediatric patient. She died of her illness, which included cerebellar symptoms and slight muscle spasticity affecting mainly the lower extremities, at 1 year of age. Histological examination of the autopsy specimens revealed a selective Purkinje cell degeneration in the cerebellum: the cells had abnormal position, were shrunken and decreased in number, and displayed abnormal dendritic trees and fragmented, disorganized axons. Electron microscopy revealed mitochondrial abnormalities in skeletal and cardiac muscle and also in the Purkinje cells. Deletions of the mitochondrial DNA were detected in the muscle in heteroplasmic form (up to 7%). Mainly the ND4-ND4L region was affected, as evidenced by the PCR; however, other regions of the mitochondrial genome also showed deletions of varying size and extent, suggesting multiple deletions of the mitochondrial DNA.

Electrophysiological characteristics and morphological properties of dentate granule--and CA3 pyramidal cells in slices cut from neonatally irradiated rats.

Neonatal irradiation reduces the dentate granule cells by 60-80%, and consequently the mossy fiber projection toward the CA3 and hilar areas decreases. The number of hilar cells diminishes. Thorny excrescences on the dendrites of the CA3 pyramidal cells get smaller both in number (from 20-30 per neuron in normal to 1-6 per neuron after irradiation) and in size. In spite of these morphological changes functional efficacy of the mossy-fiber projection to CA3 pyramidal cells remains sufficient to generate monosynaptic action potentials when stimulated electrically. Inhibitory circuits activated by mossy fiber volleys seem to be unaffected by irradiation. Main biophysical properties of CA3 pyramidal and surviving granule cells remain within the normal range. Further work should determine if efficacy of the mossy fiber projection increases to compensate for the substantial decrease of presynaptic input, or the power of transmission far exceeds the level needed to fire postsynaptic cells in normal rats.