Components of the plasma membrane of growing axons. I. Size and distribution of intramembrane particles.

The plasmalemma of mature and growing olfactory axons of the bullfrog has been studied by freeze-fracture. Intramembrane particles (IMPs) of mature olfactory axons are found to be uniformly distributed along the shaft. However, during growth, a decreasing gradient of IMP density is evident along the somatofugal axis. The size histograms of axolemmal IMPs from different segments of growing nerve reveal regional differences in the particle composition. The distribution of each individual size class of particles along the growing nerve forms a decreasing gradient in the somatofugal direction; the slope of these gradients varies directly with particle diameter. These size-dependent density gradients are consistent with a process of lateral diffusion of membrane components that are inserted proximally into the plasma membrane. The membrane composition of the growth cone, however, appears to be independent of these diffusion gradients; it displays a mosaic pattern of discrete domains of high and low particle densities. The relative IMP profiles of these growth cone regions are similar to one another but contain higher densities of large IMPs than the neighboring axonal shaft. The shifting distributions of intramembrane particles that characterize the sprouting neuron give new insights into cellular processes that may underlie the establishment of the functional polarity of the neuron and into the dynamics of axolemmal maturation.

Components of the plasma membrane of growing axons. II. Diffusion of membrane protein complexes.

Intramembrane particles (IMPs) of the plasmalemma of mature, synapsing neurons are evenly distributed along the axon shaft. In contrast, IMPs of growing olfactory axons form density gradients: IMP density decreases with increasing distance from the perikarya, with a slope that depends upon IMP size (Small, R., and K. H. Pfenninger, 1984, J. Cell Biol., 98: 1422-1433). These IMP density gradients resemble Gaussian tails, but they are much more accurately described by the equations formulated for diffusion in a system with a moving boundary (a Stefan Problem), using constants that are dependent upon IMP size. The resulting model predicts a shallow, nearly linear IMP density profile at early stages of growth. Later, this profile becomes gradually transformed into a steep nonlinear gradient as axon elongation proceeds. This prediction is borne out by the experimental evidence. The diffusion coefficients calculated from this model range from 0.5 to 1.8 X 10(-7) cm2/s for IMPs between 14.8 and 3.6 nm, respectively. These diffusion coefficients are linearly dependent upon the inverse IMP diameter in accordance with the Stokes-Einstein relationship. The measured viscosity is approximately 7 centipoise. Our findings indicate (a) that most IMPs in growing axons reach distal locations by lateral diffusion in the plasma membrane, (b) that IMPs--or complexes of integral membrane proteins--can diffuse at considerably higher rates than previously reported for iso-concentration systems, and (c) that the laws of diffusion determined for macroscopic systems are applicable to the submicroscopic membrane system.

Components of the plasma membrane of growing axons. III. Saxitoxin binding to sodium channels.

The density of sodium channels was measured in growing and mature axons of the olfactory nerve of the bullfrog, using as a probe the drug saxitoxin (STX). The toxin binds to control nerves from adult animals in a saturable manner with a dissociation constant of approximately 23 nM at 4 degrees C and a capacity of 72 fmol/mg wet weight, equivalent to about five sites per square micrometer of axolemma. In growing nerves, obtained from adult frogs 4-5 wk following removal of the original nerve, the STX-binding capacity per wet weight of tissue is markedly reduced, to approximately 25% of control values, and appears to decrease in the proximodistal direction. STX-binding data, expressed as STX/mg wet weight, was converted to STX/micron 2 of axolemma using stereologically derived values of membrane area per milligram wet weight of nerve. The axolemmal content (area/mg wet weight) of all regions of growing nerve is substantially decreased compared to controls, but increases in the proximodistal direction by 60%. These changes in axolemmal area result in calculated STX receptor densities (per unit axolemmal area) which, in distal regions, are approximately at the level of the mature nerve and, in proximal regions, are actually increased above controls by 50 to 70%. Upon comparing the axolemmal density of intramembrane particles, reported in the companion paper, with the calculated density of STX receptors in both mature and growing nerves, we find a correlation between STX receptors and intramembrane particles with diameters of 11.5-14.0 nm. The growing axon's gradient of sodium channels and the shift from this gradient to a uniform distribution in the mature axon suggest (a) that sodium channels are inserted into the perikaryal plasmalemma and diffuse from there into the growing axolemma, and (b) that the axolemma undergoes functional maturation during growth.

Early recovery of function after olfactory tract section correlated with reinnervation of olfactory tubercle.

Behavioral recovery and cortical reinnervation after early olfactory tract section were assessed in the infant golden hamster (Mesocricetus auratus). Hamster pups show strong thermotaxis at birth which declines abruptly after postnatal day (P) 8 in normal pups. Unilateral olfactory bulbectomy on P5 causes persistent thermotaxis through the second postnatal week. In this study, the bulb's output pathway, the lateral olfactory tract, was unilaterally severed on P5 and pup thermotaxis was tested through P15. Complete tract section, like bulbectomy, prolonged thermal responding beyond P8. In contrast to bulbectomy, however, some tract-sectioned pups showed recovery before P15 while others continued to show persistent thermotaxis throughout testing. The olfactory bulb projection was examined 10 days after tract section in order to determine whether recovery and persistent thermotaxis were associated with different patterns of cortical innervation. Eleven pups with complete transections showed recovery during the second week. In 10 of these pups, olfactory bulb fibers had penetrated the damaged region after surgery to reinnervate the olfactory tubercle. Three of these pups also exhibited some reinnervation of piriform cortex. The lesions of pups showing persistent thermotaxis were more severe, extending bilaterally or into deep cortical layers, and olfactory fibers had failed to reinnervate caudal terminal fields. All pups with olfactory tract sections showed extensive sprouting rostral to the cut, regardless of their behavioral profile. In no case had postlesion growth innervated the entorhinal or amygdaloid areas. Inhibition of thermotaxis was associated with reinnervation of the olfactory tubercle rather than more rostral, lateral or caudal olfactory cortex. We conclude that regrowth of olfactory tract fibers caudal to early transection is rapid and has functional consequences for early behavioral development.

Membrane specializations of neuritic growth cones in vivo: a quantitative IMP analysis.

The internal structure of the membranes of axonal and Schwann cell growth processes was examined by freeze-fracture in the growing olfactory nerve, a simple in vivo system consisting of a homogeneous neuronal population. Excision of the mature nerve of adult bullfrogs provides well-synchronized primary neuritic outgrowth that is highly enriched in growth cones at its distalmost segment. The extreme uniformity of olfactory axons in terms of their diameter and their intramembrane particle (IMP) composition permits clear identification of the cellular origin of the growth cone structures observed in replicas. In vivo, growth cones of the olfactory nerve appear as irregularly shaped enlargements of the distal tip; filopodia are only infrequently exposed by the fracture plane. Axonal and Schwann cell growth cones are distinguished by 1) the larger size of the Schwann cell growth cone and the smaller diameter of its attached processes, and 2) the distinct differences in IMP composition of Schwann cell and axonal growth cones and cell processes. Schwann cell growth cones display a uniformly high IMP density on their P-face leaflet, with the exception of circumscribed moundlike protrusions that are relatively free of IMPs. In contrast, axonal growth cones display sharp regional variation in IMP density on their P-face: broad regions almost devoid of IMPs are interspersed with zones of high IMP density. Cytotic profiles occur within high IMP density zones located, most often, at the base of the axonal growth cone. A comparison of IMP size histograms of both high and low-density regions of axonal growth cones and that of the neighboring distal shaft of the axon indicates a strict partitioning of membrane components between these two regions. The IMP profile of the axonal growth cone, notable for its relative enrichment in large-diameter particles, suggests that IMP components of the growth cone are delivered to the distal tip by a mechanism that is distinct from the lateral diffusion process described for particles of the growing axon's shaft [cf. Small and Pfenninger, 1984]. The IMP profile of the concave P-face leaflet of the internal vesicles found clustered at the base of the growth cone is more similar in composition to the profile of the neuronal shaft than that of the growth cone.

Response of Müller cells to growth factors alters with time in culture.

We have developed an explant culture technique, using the retinae of newborn guinea pigs, that reliably yields cultures of Müller cells showing uniform morphology and phenotype. Since the guinea pig retina is avascular and lacks astrocytes, Müller cells are the only glial cell-type and the only vimentin-positive population present. Virtually all passaged cells (greater than 98%) contain vimentin-positive intermediate filaments and no glial fibrillary acidic protein (GFAP) has been detected using a range of GFAP antibodies known to label astrocytes in the guinea pig optic nerve. Most vimentin-positive cells were also labeled with an antibody to carbonic anhydrase II, an enzyme which in the retina is specific for Müller cells. Proliferating Müller cells were identified within the inner nuclear layer of retinal fragments as early as 2 days in culture using bromodeoxyuridine (BrdU) and vimentin double labeling. Cultured Müller cells change their growth characteristics with successive passaging. The length of the cell cycle increases from 25.4 h for cells at first passage, to 66.7 h for cells at fourth passage. Altered responses to mitogens were also observed with passaging. First-passage cultures responded to basic fibroblast growth factor (bFGF) but not to several other factors tested including interleukin-2 (IL-2). In contrast, older cultures were highly responsive to IL-2 but showed a minimal response to bFGF. The altered responsiveness to mitogens observed in vitro may be relevant to changes in growth control of Müller cells in the developing and mature retina. The guinea pig retina provides an ideal mammalian tissue for generating Müller cell cultures that are free of astrocytes, endothelial cells, and pericytes, the most frequent contaminants of retinal glial cultures. The monolayers obtained show a high degree of homogeneity and are well suited for studies of Müller cell function.

Evidence for migration of oligodendrocyte--type-2 astrocyte progenitor cells into the developing rat optic nerve.

Formation of myelinated tracts in central nervous system (CNS) regions such as the optic nerve seems to depend on two glial cell types, both of which derive from a common progenitor cell. This oligodendrocyte--type-2 astrocyte (O-2A) progenitor cell gives rise to oligodendrocytes, which produce internodal myelin sheaths, and to type-2 astrocytes, which extend fine processes in the region of the nodal axolemma. The optic nerve also contains a third glial cell, the type-1 astrocyte, which derives from a separate precursor. These three glial cells develop in a fixed sequence over a two-week period: type-1 astrocytes appear at embryonic day 16 (E16), oligodendrocytes at the day of birth (E21 or postnatal day P0), and type-2 astrocytes between P8 and P10. Type-1 astrocytes secrete a potent mitogen which causes expansion of the O-2A progenitor cell population in vitro. Here, we report that dividing O-2A progenitor cells are highly motile and seem to migrate from the brain into the optic nerve, beginning at its chiasmal end. Our results indicate that long-distance migration along the neural axis is characteristic only of progenitors of the O-2A lineage and may serve to distribute these cells to regions of the CNS that will become myelinated. These results also suggest that the intrinsic neuroepithelial cells of the optic stalk may be even more restricted than previously thought, giving rise only to type-1 astrocytes.

Functional properties of retinal Müller cells following transplantation to the anterior eye chamber.

Two types of glial cells occur in the retina, Müller cells and astrocytes. These cells share several structural features such as extending endfeet onto blood vessels of the retina. Retinal vessels express a tight blood-retinal barrier which is comparable to the blood-brain barrier (BBB) of the CNS. While astrocytes have been implicated in the induction of the BBB, the role of Müller cells in the blood-retinal barrier is unknown. To determine if Müller cells are capable of influencing vascular permeability, we have prepared Müller cells that are free of astrocytes and transplanted them to a peripheral target, the anterior eye chamber. Müller cells were identified 2 weeks to 3 months after injection and were predominantly localized within the connective tissue of the ciliary body. The Müller cells occurred as dense clusters of cells closely associated with ciliary blood vessels. The ciliary vessels adjacent to Müller cells were freely permeable to circulating horseradish peroxidase (HRP), suggesting that Müller cells did not induce tight barrier properties from these leaky peripheral vessels. In contrast, cortical astrocytes injected into the anterior eye chamber preferentially formed a monolayer on the anterior surface of the iris, a region known to contain blood vessels that are impermeable to circulating tracers (e.g., Raviola, Exp Eye Res [Suppl] 25:27, 1977). Müller cells were rarely associated with the iris and the few cells that were present were located deep within the iris stroma rather than on the surface. The behaviour of guinea pig Müller cells transplanted to the anterior eye chamber contrasts sharply with that of cortical astrocytes in terms of: 1) the ocular compartment to which Müller cells migrate; 2) the tissue invasiveness of the cells; and 3) the degree of permeability of blood vessels adjacent to transplanted cells. The results of this study emphasize the functional distinctness of the two types of retinal glia and suggest that Müller cells from guinea pig retina may not be active in modifying the permeability properties of peripheral blood vessels, a function that has been suggested for astrocytes.

1H NMR study of cerebral development in the rat.

1H NMR spectroscopy of brain extracts was used to investigate the metabolic changes that take place during development of the neonatal rat brain. Data were obtained over the range 1-21 days. The concentration of N-acetylaspartate rose by a factor of 9 during this period, the most rapid rise occurring after day 9. The total creatine concentration rose from days 1-21, with a large increase between days 1 and 5. Taurine concentration rose until day 5, then fell from days 5-21. The concentration of choline-containing compounds fell during the 21 day period. The results are discussed in relation to brain development and conventional biochemical data. A major conclusion in relation to spectroscopy of children is that interpretation of changes seen in disease will require adequate data from age-matched controls.

Brain metabolites as 1H NMR markers of neuronal and glial disorders.

1H NMR spectroscopy of human brain in vivo can be used to detect a number of cerebral metabolites including N-acetylaspartate, creatine + phosphocreatine and choline-containing compounds. We have used 1H NMR spectroscopy to analyse these signals in (i) biopsy material from both normal human brain and astrocytomas, and (ii) primary astrocyte cultures. On the basis of this analysis, we conclude that in vivo 1H NMR spectroscopy could play an important clinical role in the non-invasive assessment of neuronal degeneration and proliferation of non-neuronal cells.