Neuron Segmentation With High-Level Biological Priors.

We present a novel approach to the problem of neuron segmentation in image volumes acquired by an electron microscopy. Existing methods, such as agglomerative or correlation clustering, rely solely on boundary evidence and have problems where such an evidence is lacking (e.g., incomplete staining) or ambiguous (e.g., co-located cell and mitochondria membranes). We investigate if these difficulties can be overcome by means of sparse region appearance cues that differentiate between pre- and postsynaptic neuron segments in mammalian neural tissue. We combine these cues with the traditional boundary evidence in the asymmetric multiway cut (AMWC) model, which simultaneously solves the partitioning and the semantic region labeling problems. We show that AMWC problems over superpixel graphs can be solved to global optimality with a cutting plane approach, and that the introduction of semantic class priors leads to significantly better segmentations.

Synapse formation in adult barrel cortex following naturalistic environmental enrichment.

Environmental enrichment paradigms in adult laboratory animals, consisting of physical, perceptual, and social stimulation, have been shown to affect synapse and cell morphology in sensory cortex and enhance learning ability, whereas enrichment, which is in harmony with the animal's natural habitat may have even greater implications for plasticity. Previous studies in our laboratory have shown that whisker stimulation induced the formation of synapses and spines in the corresponding barrel. In the present study adult C57/Bl6J female laboratory mice at 6 weeks of age were placed during 2 months in a protected enrichment enclosure in a forest clearing at the Chisti Les Biological Station, Tvier, Russia. We analyzed neuropil ultrastructure in the C2 barrel using serial-section electron microscopy on a total of eight mice (n=4 enriched, n=4 standard cagemate controls). Quantitative analyses of volumes of neuropil showed a significant increase in excitatory and inhibitory synapses on spines and excitatory synapses on dendritic shafts in the C2 barrel in the enriched group compared with standard cagemate controls. These results demonstrate that naturalistic experience alters the synaptic circuitry in layer IV of the somatosensory cortex, the first cortical relay of sensory information, leaving a lasting trace that may guide subsequent behavior.

Imaging of experience-dependent structural plasticity in the mouse neocortex in vivo.

The functionality of adult neocortical circuits can be altered by novel experiences or learning. This functional plasticity appears to rely on changes in the strength of neuronal connections that were established during development. Here we will describe some of our studies in which we have addressed whether structural changes, including the remodeling of axons and dendrites with synapse formation and elimination, could underlie experience-dependent plasticity in the adult neocortex. Using 2-photon laser-scanning microscopes and transgenic mice expressing GFP in a subset of pyramidal cells, we have observed that a small subset of dendritic spines continuously appear and disappear on a daily basis, whereas the majority of spines persists for months. Axonal boutons from different neuronal classes displayed similar behavior, although the extent of remodeling varied. Under baseline conditions, new spines in the barrel cortex were mostly transient and rarely survived for more than a week. However, when every other whisker was trimmed, the generation and loss of persistent spines was enhanced. Ultrastructural reconstruction of previously imaged spines and boutons showed that new spines slowly form synapses. New spines persisting for a few days always had synapses, whereas very young spines often lacked synapses. New synapses were predominantly found on large, multi-synapse boutons, suggesting that spine growth is followed by synapse formation, preferentially on existing boutons. Altogether our data indicate that novel sensory experience drives the stabilization of new spines on subclasses of cortical neurons and promotes the formation of new synapses. These synaptic changes likely underlie experience-dependent functional remodeling of specific neocortical circuits.

Barriers in the immature brain.

1. The term "blood-brain barrier" describes a range of mechanisms that control the exchange of molecules between the internal environment of the brain and the rest of the body. 2. The underlying morphological feature of these barriers is the presence of tight junctions which are present between cerebral endothelial cells and between choroid plexus epithelial cells. These junctions are present in blood vessels in fetal brain and are effective in restricting entry of proteins from blood into brain and cerebrospinal fluid. However, some features of the junctions appear to mature during brain development. 3. Although proteins do not penetrate into the extracellular space of the immature brain, they do penetrate into cerebrospinal fluid by a mechanism that is considered in the accompanying review (Dziegielewska et al., 2000). 4. In the immature brain there are additional morphological barriers at the interface between cerebrospinal fluid and brain tissue: strap junctions at the inner neuroependymal surface and these and other intercellular membrane specializations at the outer (piaarachnoid) surface. These barriers disappear later in development and are absent in the adult. 5. There is a decline in permeability to low molecular weight lipid-insoluble compounds during brain development which appears to be due mainly to a decrease in the intrinsic permeability of the blood-brain and blood-cerebrospinal fluid interfaces.

The nature and composition of the internal environment of the developing brain.

1. The fetal brain develops within its own environment, which is protected from free exchange of most molecules among its extracellular fluid, blood plasma, and cerebrospinal fluid (CSF) by a set of mechanisms described collectively as "brain barriers." 2. There are high concentrations of proteins in fetal CSF, which are due not to immaturity of the blood-CSF barrier (tight junctions between the epithelial cells of the choroid plexus), but to a specialized transcellular mechanism that specifically transfers some proteins across choroid plexus epithelial cells in the immature brain. 3. The proteins in CSF are excluded from the extracellular fluid of the immature brain by the presence of barriers at the CSF-brain interfaces on the inner and outer surfaces. These barriers are not present in the adult. 4. Some plasma proteins are present within the cells of the developing brain. Their presence may be explained by a combination of specific uptake from the CSF and synthesis in situ. 5. Information about the composition of the CSF (electrolytes as well as proteins) in the developing brain is of importance for the culture conditions used for experiments with fetal brain tissue in vitro, as neurons in the developing brain are exposed to relatively high concentrations of proteins only when they have cell surface membrane contact with CSF. 6. The developmental importance of high protein concentrations in CSF of the immature brain is not understood but may be involved in providing the physical force (colloid osmotic pressure) for expansion of the cerebral ventricles during brain development, as well as possibly having nutritive and specific cell development functions.

Development of motoneurons and primary sensory afferents in the thoracic and lumbar spinal cord of the South American opossum Monodelphis domestica.

The postnatal development of the primary sensory afferent projection to the thoracic (T4) and lumbar (L4) spinal cord of the marsupial species Monodelphis domestica was studied by using anterograde and retrograde neuronal tracers. Large numbers of primary afferents and motoneurons were labelled by application of the carbocyanine dye DiI into individual dorsal root ganglia (DRG) afferents in short-term organ cultures. Dorsal root axons had entered the cord at birth, but most primary afferent innervation of the grey matter and the establishment of cytoarchitectural lamination occurs postnatally. In addition to ipsilateral projections, some primary afferents that projected to the dorsal horn extended across the midline into the equivalent contralateral regions of the grey matter. Similarly, motoneuron dendrites occasionally extended across midline and into the contralateral grey matter. The first fibres innervating the spinal cord project to the ventral horn and formed increasingly complex terminal arbours in the motor columns between P1 and P7. After P5 many afferents were seen projecting to the dorsal horn, with the superficial dorsal horn being the last region of the spinal grey to be innervated. Histochemical labelling with the lectin Griffonia simplicifolia indicated that C fibre primary afferents had arborised in the superficial dorsal horn by P14. The sequence of primary afferent innervation is thus similar to that described in the rat, but this sequence occurs over a period of several weeks in Monodelphis, compared with several days in the rat.

Development of thalamocortical projections in the South American gray short-tailed opossum (Monodelphis domestica).

We determined the time-course and general pattern of thalamocortical development of Monodelphis domestica by tracing projections with carbocyanine dye in fixed postnatal brains between postnatal day 2 (P2) and P30. By P2, the first neurons have migrated to form the preplate of the lateral cortex and have sent out axons into the intermediate zone. By P3, fibers from the preplate of more dorsal cortex have entered the intermediate zone, and, by P5, they reach the primitive internal capsule. Crystal placements in the dorsal thalamus at P2-P3 reveal thalamic axons extending down through the diencephalon and growing out through the internal capsule among groups of back-labelled cells that already project into the thalamus. Thalamic axons arrive at the cortex after the arrival of cells of the true cortical plate has split the preplate into marginal zone and subplate. Axons from the ventral part of the dorsal thalamus reach the lateral cortex by P5: Dorsal thalamic fibers arrive at the extreme dorsal cortex by P9. The deeper layers of the cortex appear to mature relatively earlier in Monodelphis than in eutherian mammals, and the subplate becomes less distinct. Thalamic fibers and their side branches proceed into the cortex without an obvious period of waiting in the subplate, but they do not penetrate the dense cortical plate itself. Monodelphis could provide an excellent model species, because the development of its thalamocortical connections is entirely an extrauterine process: The period P0-P15 corresponds to that of E12-P0 in the rat.

Permeability of the developing and mature blood-brain barriers to theophylline in rats.

1. In the present study, the uptake of theophylline and L-glucose into the adult and neonatal rat brain has been investigated. Steady state cerebrospinal fluid (CSF) and brain concentrations of theophylline were reached within 1 h following a single intraperitoneal (i.p.) injection, whereas steady state CSF and brain concentrations of L-glucose were not approached until after 5 h. 2. Steady state brain:plasma and CSF:plasma concentration ratios for theophylline and L-glucose in neonatal rats were significantly higher than ratios in adult rats. Erythrocyte:plasma ratios for theophylline in neonatal rats were also significantly higher than ratios in adult rats. Steady state ratios for theophylline were significantly higher than those for L-glucose in both neonatal and adult rats. 3. Respiratory acidosis (pH 6.9-7.0) did not affect steady state CSF:plasma or brain:plasma ratios for theophylline in neonatal or adult rats. In contrast, steady state CSF:plasma and brain:plasma ratios for L-glucose were increased by respiratory acidosis. 4. The lower steady state CSF:plasma, brain:plasma and erythrocyte:plasma ratios for theophylline in adult rats are likely to be due to a higher concentration of plasma proteins in adult blood compared with neonates, with a greater retention of protein-bound (non-exchangeable) theophylline in adult blood, and are unlikely to be due to p-glycoprotein-mediated efflux of theophylline at the adult blood-brain barrier.

Development of walking, swimming and neuronal connections after complete spinal cord transection in the neonatal opossum, Monodelphis domestica.

Development of coordinated movements was quantitatively assessed in adult opossums (Monodelphis domestica) with thoracic spinal cords transected by (1) crushing 7-8 d after birth [postnatal days 7-8 (P7-P8)]; at 2-3 years of age, systematic behavioral tests (e.g., climbing, footprint analysis, and swimming) showed only minor differences between control (n = 5) and operated (n = 10) animals; and (2) cutting on P4-P6; at 1 month these opossums exhibited coordinated walking movements but were unable to right themselves from a supine position, unlike controls (n = 6). When tested at 2 or 6 months, they could right themselves and showed remarkable coordination, albeit with more differences from controls than after a crush. No animals with spinal cords that were crushed at P14-18 survived because of cannibalism by the mother. Morphological studies (n = 10) 3 months-3 years after crush at 1 week showed restoration of structural continuity and normal appearance at the lesion site. Animals with cut rather than crushed cords showed continuity but greater morphological deficits. That lesions were complete was demonstrated by examining morphology and nerve impulse conduction immediately after crushing or cutting the spinal cord in controls. After lumbar spinal cord injection of 10 kDa dextran amine, retrogradely labeled cells were found rostral to the lesion in hindbrain and midbrain nuclei. Conduction was restored across the site of the lesion. Thus complete spinal cord transection in neonatal Monodelphis was followed by development of coordinated movements and repair of the spinal cord, a process that included development of functional connections by axons that crossed the lesion.

Fetuin expression in the dorsal root ganglia and trigeminal ganglia of perinatal rats.

Fetuin, a fetal plasma glycoprotein, has been shown previously to be present in sub-populations of neurons in the developing central and peripheral nervous system. To gain a more complete description of the time course of the appearance of fetuin during neurogenesis we have examined fetuin immunoreactivity, and the presence of fetuin mRNA, in the developing rat trigeminal and dorsal root ganglia. Fetuin immunoreactivity and its mRNA were first seen at embryonic day 15 in the trigeminal ganglia, and at embryonic day 16 in dorsal root ganglia. In both trigeminal and dorsal root ganglion, fetuin appeared to be present up until around the time of birth, and then again between postnatal days 3 and 16. The results suggest that fetuin first appears at around the time that ganglion cell axons reach their central targets, which is also approximately when the cell-death period begins. The proportion of ganglion neurons that were fetuin immunoreactive at different ages was inversely related to the amount of cell death that is known to occur in these populations, thus it seems that fetuin is more likely to be associated not with dying cells, but with those that survive the cell-death period.

Albumin transfer across the choroid plexus of South American opossum (Monodelphis domestica).

1. Blood-cerebrospinal fluid (CSF) transfer of various exogenous albumins has been investigated in developing Monodelphis domestica (South American grey short-tailed opossum) and compared with the steady-state CSF: plasma ratios for endogenous (Monodelphis) albumin. Ratios for Monodelphis albumin and human albumin were similar and were the highest at postnatal day 5 (P5) (48.2 +/- 4.4 and 40.6 +/- 4.5%, respectively). The ratio for bovine albumin was similar to the steady-state ratio for Monodelphis albumin at P7-8 but became consistently lower than the Monodelphis albumin ratio at all other ages until P32-36 when all albumins tested attained a similar low ratio. The CSF:plasma ratio of chemically modified (succinylated) bovine albumin was always significantly lower than that of other albumins, except at the oldest age examined (P32-36). 2. Immunocytochemistry showed that within the brain, albumin was confined to the lumen and endothelial cells of blood vessels. In the choroid plexus only a small proportion (0.2-1.7% of the total cell number) of epithelial cells was positive for albumin, both endogenous and exogenous, at all ages studied (except the 3rd ventricle where cells were only positive from P8). The CSF was strongly positive for all albumins. The peak proportion of positive cells and of albumin concentrations in CSF occurred at P8. These findings suggest that the primary route for penetration of albumin into CSF is directly across the choroid plexus rather than via the brain. 3. Double-labelling immunocytochemistry revealed that the same epithelial cells contained both endogenous (Monodelphis) and exogenous (human) albumin. In contrast, for succinylated albumin, at P7 only about 35% (lateral ventricle) and 50% (4th ventricle) of Monodelphis albumin-positive cells were also positive for succinylated albumin, but by P30 this proportion increased to 90% at both sites. 4. Thus the developing choroid plexus distinguishes between different albumins. Chemical modification of albumin (succinylation) disrupts this mechanism. It is proposed that in older animals (P32-36) all of the albumin in the CSF is derived from plasma by diffusion (as in adult animals). At earlier stages of development, a proportion of the albumin in CSF also appears to be transferred from the plasma by diffusion with an additional component transferred by a mechanism that can distinguish between different species of albumin. The main route of entry of albumin to CSF seems likely to be via the choroid plexus epithelial cells.

TGF-beta receptor type II and fetuin in the developing sheep neocortex.

Fetuin shows a characteristic pattern of distribution in the developing neocortex in many mammalian species. Its expression is confined to early-appearing cortical-plate and later subplate neurons. A short 19 amino-acid sequence of fetuin shows a degree of homology to an 18 amino-acid sequence of the TGF-beta type II receptor (TbetaR-II) and in vitro fetuin binds to members of the TGF-beta family of cytokines. It has been suggested that fetuin is the biologically significant antagonist of these cytokines. We have compared, using immunocytochemistry, the distribution pattern of TbetaR-II and fetuin in the developing neocortex of foetal sheep. TbetaR-II immunoreactivity first appears at around 40 days of gestation in the fetal sheep (E40, term in sheep is 150 days from conception), localised in two discreet bands: one just outside the cortical plate in the inner part of the marginal zone and one deep in the cortical plate in what becomes the transient subplate zone. By E70-E80, TbetaR-II is prominent in a population of subplate cells, whereas, by E120 only small patches of TbetaR-II-positive cells are visible, principally in pyramidal cells in layer VI. The developmental sequence of the staining pattern for TbetaR-II in the neocortex is complementary to that for fetuin, rather than overlapping with it. Double-labelling of fetuin and TbetaR-II shows some cellular co-localisation, especially at E60, but most fetuin-positive cells are not immunoreactive for TbetaR-II. Thus, fetuin's proposed role as an antagonist of TGF-beta cytokines and mimic of TbetaR-II is not consistent with the observed distribution of these two molecules in the developing neocortex of the foetal sheep.

Repair and recovery following spinal cord injury in a neonatal marsupial (Monodelphis domestica).

1. Repair and recovery following spinal cord injury (complete spinal cord crush) has been studied in vitro in neonatal opossum (Monodelphis domestica), fetal rat and in vivo in neonatal opossum. 2. Crush injury of the cultured spinal cord of isolated entire central nervous system (CNS) of neonatal opossum (P4-10) or fetal rats (E15-E16) was followed by profuse growth of fibres and recovery of conduction of impulses through the crush. Previous studies of injured immature mammalian spinal cord have described fibre growth occurring only around the lesion, unless implanted with fetal CNS. 3. The period during which successful growth occurred in response to a crush is developmentally regulated. No such growth was obtained after P12 in spinal cords crushed in vitro at the level of C7-8. 4. In vivo, in the neonatal (P4-8) marsupial opossum, growth of fibres through, and restoration of, impulse conduction across the crush was apparent 1-2 weeks after injury. With longer periods of time after crushing a considerable degree of normal locomotor function developed. 5. By the time the operated animals reached adulthood, the morphological structure of the spinal cord, both in the region of the crush and on either side of the site of the lesion, appeared grossly normal. 6. The results are discussed in relation to the eventual longterm possibility of devising effective treatments for patients with spinal cord injuries.

The nature of the decrease in blood-cerebrospinal fluid barrier exchange during postnatal brain development in the rat.

1. The blood-cerebrospinal fluid (CSF) exchange of a wide range of passively transported lipid insoluble compounds (0.43-5.4 nm molecular radius) has been investigated in rats at different stages of postnatal development (2 days old to adult). A novel 'litter-based' model for investigating blood-CSF barrier exchange in immature animals is described. 2. At each age investigated there was a clear inverse correlation between molecular radius and blood-CSF barrier exchange, in addition to an overall decrease in blood-CSF barrier exchange with increasing age. 3. The decrease in blood-CSF barrier exchange with age was not consistent with a reduction in pore diameters, nor does it appear to be due to an increase in the CSF sink effect with age. It seems likely to be due to a relative decrease in the number of a population of large diameter pores.

Development of connections by axons growing through injured spinal cord of neonatal opossum in culture.

The ability of neurites to grow through a lesion and form synaptic connections has been analyzed in a developing mammalian spinal cord in vitro. After isolation of the entire central nervous system (CNS) of the newly born South American opossum (Monodelphis domestica) the spinal cord was crushed. Outgrowth through and beyond the lesion was observed in living preparations for 2-5 days by staining axons with carbocyanine dyes. The structure of the acute crush and the growing neurites was examined by light and electron microscopy in tissue fixed immediately after the crush had been made. All axons had been severed and the site was filled with debris and amorphous vesicular structures. By 3 days after injury, numerous labelled neurites had grown into the lesion; by 4 days, many had extended several millimetres beyond it. At this time normal axonal profiles were apparent in electron micrographs of the crush site. Although fewer axons grew across the lesion than had been severed by the crush, the amplitudes of compound action potential volleys conducted across the crush in injured preparations were comparable with those recorded from uninjured spinal cords. Physiological experiments made with raised concentrations of extracellular magnesium in the culture fluid indicated that growing axons had formed synaptic connections. Thus, delayed major peaks of the response were abolished while the small component corresponding to through conduction remained unaffected by magnesium. These experiments demonstrate the development of synaptic interactions by the growing neurites and confirm the far greater powers of repair in neonatal mammals compared to adults. They set the stage for comparing molecular mechanisms involved in development and regeneration of the mammalian CNS.