Search

Using median nerve injury and immunocytochemical methods, we examined the temporal changes in neuropeptide Y (NPY) expression in the cuneate nucleus (CN) in rats following median nerve transection. Under normal circumstances, neuropeptide Y-immunoreactive (NPY-IR) fibers was not detectable in the CN. A few NPY-IR fibers were observed in the ipsilateral CN 5 days after the median nerve transection, and peaked at 4 weeks. Thereafter, they were gradually returned to nearly control level after 16 weeks. Quantitative evaluation showed that the mean percentage of area occupied by NPY-IR fibers in entire and three subregions of the CN at 4 weeks were significantly higher than that at other post-operated time points, respectively. The present ultrastructural observations in the middle region of CN showed that the significantly increased NPY immunoreactivity was confined only in the myelinated axons and terminals but not detected in the dendrites, somata, and glial cells. The NPY-IR terminals made axodendritic synaptic contacts with unlabeled elements. The present results indicate that the time course of the increase of NPY immunoreactivity is similar to c-Fos expression as described in a previous study. It is speculated that the increased NPY in the CN after axotomy may affect the excitability of postsynaptic cuneate neurons, however, the functional interaction between NPY and c-Fos-IR neurons needs to be further studied.

By histochemical and immunocytochemical techniques, this study aimed to determine the possible involvement of apoptosis in regulating the microglial distribution in fetal rat brain. While microglial cells were labeled with the isolectin Griffonia simplicifolia (GSA I-B4), apoptotic cells were detected by using terminal transferase-mediated dUTP nick end-labeling (TUNEL). TUNEL-labeled cells occurred mainly in the dorsal midline along its rostral-caudal axis of the brain where lectin-labeled microglia were also observed. Occasional TUNEL-labeled cells were observed in the intermediate zone lateral to the striatum (IZS) where lectin-labeled microglia were common from embryonic day 16 (E16) onwards. Some of lectin-labeled microglia showing different morphological forms ingested TUNEL-labeled bodies. In contrast, lectin-labeled microglia showing signs of apoptosis appeared to be lacking. These results clearly demonstrated that lectin-labeled microglia were distributed in areas with and without the occurrence of a large concentration of TUNEL-labeled cells. Our studies suggest that microglia in fetal rat brain will undergo differentiation and activation rather than apoptotic death to govern their population.

The present study examined the effect of maternal administration of dexamethasone (DEX) on amoeboid microglial cells (AMC) in fetal rats extending from 16 to 20 days postconception (E16 to E20). After an intraperitoneal injection of DEX into pregnant rats at E10, the external morphology and distribution of immunolabelled AMC as detected with OX-42 and ED1 monoclonal antibodies remained unaltered when compared with those of the controls. The major effect of dexamethasone was on microglial cell population. Thus, with OX-42 and ED1, the numbers of immunolabelled AMC in the intermediate zone lateral to the striatum (IZS) of DEX-treated fetuses which remained relatively unchanged at E16 were significantly reduced at E18. However, OX-42 labelled cells showed an unexpected increase in number at E20 following DEX treatment. Microglial response to DEX was also analyzed in sections stained with the isolectin, GSA I-B4, which specifically binds alpha-D-galactosyl glycoproteins on microglia. The number of GSA I-B4 labelled AMC was significantly increased at E16, declined at E18 and remained constant thereafter in DEX-treated rats when compared with that of the controls. A major finding after DEX treatment was the wider occurrence of AMC double labelled with anti-BrdU antibody and GSA I-B4 or OX-42 at E16 compared with those in the controls suggesting that the initial increase of GSA I-B4 labelled AMC may be attributed to their proliferation. The drastic reduction of OX-42 and ED1 positive microglial cells notably at E18 may be due to the downregulation of surface antigens as a result of possible suppressive action of dexamethasone. On the basis of present findings, it is concluded that the antigenic expressions of fetal AMC may be modulated by DEX administrated maternally. Such however appeared to be extremely selective as reflected by the varied expression for certain immune molecules at different stages of brain development. This information would be useful in potential use of glucocorticoids in prenatal therapy of brain pathology via maternal circulation.

In the course of the development of cerebral endothelia, the lectin labelling of alpha-D-galactosyl residues on the luminal front of the endothelial cells was gradually increased but that on the abluminal surface was concomitantly attenuated. In the latter, the isolectin (GSA I-B4) decoration was totally diminished with the maturation of the brain. Associated with this change, the end feet of astrocytes resting on the endothelia became lectin positive. Numerous cytoplasmic vacuoles, Golgi apparatus, vesicles and tubule-like structures were strongly labelled in the cerebral endothelia in the early stages of development but only sparse vacuoles and vesicles remained lectin-stained in the mature cerebral endothelia. Labelled pericytes closely associated with the wall of blood vessels, displayed a lectin binding pattern similar to that of the developing cerebral endothelia. Following the blood-brain barrier damage induced by a cold lesion, the altered cerebral endothelia that were surrounded by densely-stained macrophages/microglia or pericytes remained unstained with the isolectin. Present results suggest that during blood-brain barrier maturation, the brain endothelia develop a functional polarity whereby their membrane glycoprotein is down regulated. The membrane glycoprotein, however, underwent remodeling when the blood-brain barrier was disrupted.

The present study examined the expression of different antigens in amoeboid microglial cells (AMC) in fetal rat brain extending from 12 to 20 d postconception (E12-E20) using a panel of monoclonal antibodies which recognised the major histocompatibility complex (MHC) class I (OX-18) and class II (OX-6) antigens, leucocyte common antigen (OX-1), CD4 receptor (OX-35), complement type 3 receptor (OX-42) or macrophage antigens of unknown function (ED1 and ED2). Of the above-mentioned antigens, ED1 and ED2-labelled AMC were observed in the neuroepithelia as early as embryonic day 12 (E12); other antigens were not detected at this stage. At E14, except for MHC class I antigen, all other antigens were expressed by AMC distributed predominantly in the developing white matter. At E16, AMC in the intermediate zone lateral to the striatum were endowed with all the above-mentioned antigens including MHC class I. At E18, the immunoreactivities of AMC stained with OX-6, OX-18, OX-35 and OX-42 antigens were noticeably reduced when compared with those cells at E16. At E20, amoeboid microglial cells exhibited full complement of antigen expression similar to those cells at E16; some of the labelled cells emitted a variable number of cytoplasmic processes. It is suggested that the successive and differential expression of various macrophage related antigens on AMC in fetal brain is related to the specific requirement of local environment in different stages of development.

The present study describes the ultrastructural localization and labelling pattern of lectin in different microglial cell phenotypes in the postnatal rat brain using the isolectin, GSA I-B4. The nascent round and amoeboid microglial cells (round cells and cells displaying short processes) were labelled at their cytoplasmic membrane and the membrane of the subplasmalemmal vacuoles. In the course of their transformation into ramified forms with age, dense lectin labelling was observed successively at different sites in the differentiating cells. The most striking feature was the staining of the Golgi saccules on the trans face, the trans tubular network and associated vesicles and vacuoles in the 'intermediate' ramified microglia (ramified cells bearing thick and long processes and those with thin and long processes). The vacuoles with accumulated reaction products were closely associated with many microtubules extending into the cytoplasmic processes. At the surface, the lectin-labelled vacuoles and vesicles appeared to fuse with the membrane and their contents communicated with the exterior. In the advanced or most differentiated ramified microglial cells (cells bearing attenuated processes), the lectin staining at all the above mentioned sites became diminished. In conclusion, in the transformation of the round microglia into their ramified derivatives, the glycoconjugates at the cytoplasmic membrane are progressively reduced. It is postulated from this study that the down-regulation of the glycoconjugates of the microglial plasma membrane is due primarily to their internalization during endocytosis. This process would trigger a de novo galactosyl protein synthesis and/or modification at the trans Golgi saccules and trans tubular network probably in an attempt to degrade the internalized membrane glycoproteins or to replenish the consumption of the membrane glycoconjugates.

Following horseradish peroxidase injection into the cervical and lumbosacral cords of adult albino rats, labeled neurons were seen in the first cervical segment, brain stem, and cerebellar and diencephalic nuclei. A new pathway, the faciospinal projection, originating in the medial portion of the rostral part of the facial nucleus, was traced. Another new pathway, the olivospinal pathway, is probably also present. Our results for neurons projecting to the spinal cord (spinal-projecting neurons) from the nucleus ambiuus, dorsal motor nucleus of the vagus, superior vestibular nucleus and nucleus f, nucleus Darshevch, nucleus Rolleri, nucleus prepositus hypoglossi, and nucleus of the posterior commissure have been reported before in other mammals but not in rats. Projections from the following regions are in general agreement with previous results in rats, but show significant topographical differences: the first cervical segment; nuclei gracilis, cuneatus, and cuneatus lateralis; the midline and lateral reticular nuclear complex; the trigeminal nuclear complex (spinal, principal, and mesencephalic); nucleus of the tractus solitarius; the medial, lateral, and descending vestibular nuclei, nuclei coeruleus and subcoeruleus; superior colliculus; interstitial nucleus of Cajal, and the deep cerebellar nuclei. The distribution of labeled neurons in the nucleus parabrachialis, nucleus tegmentolaterodorsalis, nucleus Kölliker-Fuse, nucleus Edinger-Westphal, and the hypothalamic nuclear complex confirmed that of previous reports in rats. With the exception of a few nuclear groups which project primarily to either lumbosacral (e.g., the paraventricular nucleus of the hypothalamus) or cervical segments (e.g., the facial motor nucleus and the superior colliculus) most of the other nuclear groups project to both the lumbar and cervical levels. There is no distinct somatotopy in the neuronal groups projecting to both cervical and lumbosacral levels. With only a few exceptions (e.g., the superior vestibular nucleus) most of the spinal-projecting neurons are bilaterally distributed, some with contralateral and others with ipsilateral predominance.

According to results from the horseradish peroxidase (HRP) method of tracing neuronal connections, the spinal cords of neonatal and immature rats receive a large number of descending projections from the first cervical cord segment, various brain-stem nuclei, and deep cerebellar and diencephalic nuclei. All these projections are present at birth, though at this age some of them are not fully established. Thus, only a few cells in the trigeminospinal, solitariospinal, tectospinal, and cerebellospinal groups were labeled after HRP injection in the lumbosacral or cervical cord segments in neonatal animals. They were clearly labeled in older, immature animals. The labeled neurons in other descending pathways appeared to be equal in density in neonatal, immature, and adult rats. This visual impression was stregthened by the counts of neurons in the interstitial nucleus of Cajal, which showed no significant difference in the number of labeled neurons in the three age groups. However, counts of labeled cells in the lateral vestibular nucleus and nucleus of the posterior commissure showed that there is a steady rise in the number of labeled neurons as the animals increase in age.