Cultured oligodendrocytes mimic in vivo phenotypic characteristics: cell shape, expression of myelin-specific antigens, and membrane production

Steroid synthesis in rat brain cell cultures

Primary cultures derived from neonatal rat forebrains were established and cultured for several weeks. They grow entirely as glial cultures composed of oligodendrocytes and astrocytes. Glial cells undergo maturation and differentiation in culture. This was shown by measuring the oligodendroglial enzyme 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNPase), a specific marker for expression of oligodendrocyte differentiation. CNPase activity increased from days 10-21 of culture. Both cell types were characterized by indirect immunofluorescence staining using monoclonal antibodies to galactocerebroside (Gal C) and myelin basic protein (MBP) for oligodendrocytes, and glial fibrillary acidic protein (GFAP) for astrocytes. Using the above criteria, we measured about 60% oligodendrocytes and 40% astrocytes after 3 weeks of culture. Oligodendrocytes, expressing Gal C and MBP, were highly immunoreactive to monospecific polyclonal antibodies to the cytochrome P-450scc, enzyme involved in the synthesis of pregnenolone from cholesterol. After incubation of glial cultures with [3H]mevalonolactone in the presence of mevinoline and trilostane, biosynthesis of [3H]cholesterol, [3H]pregnenolone (P) and [3H]pregn-5-ene-3 beta, 20 alpha-diol (20-OHP) was demonstrated. Steroid biosynthesis was related to oligodendroglial differentiation, as the initial and rapid rate of increase in CNPase activity was found to occur at the same time as the onset of steroid synthesis. Both reached a maximum at 3 weeks of culture and remained stable for several weeks. Steroid synthesis was increased by dibutyryl cAMP (0.2 mM), as well as by dexamethasone (10 nM). When aminoglutethimide, a potent inhibitor of cytochrome P-450scc, was added during the incubation of cells with [3H]mevalonolactone, [3H]cholesterol accumulated in the cells. After the release of aminoglutethimide blockade, [3H]20-OHP was the major steroid produced and released in the culture medium. The demonstration of de novo steroid biosynthesis and of the cholesterol side-chain cleavage cytochrome P-450 in normal rat glial cells brings additional support to the concept of "neurosteroids".

Developmental profile and differential localization of mRNAs of myelin proteins (MBP and PLP) in oligodendrocytes in the brain and in culture

We applied the in situ hybridization technique to localize the mRNAs for two myelin proteins: proteolipid protein (PLP) and myelin basic protein (MBP). In the oligodendrocyte in primary culture, PLP mRNA was located exclusively in the cell body throughout development. However, MBP mRNA was first located in the cell body and was rapidly distributed to the processes but not to the membranous sheets formed from the tips and lengths of the processes. Expression of PLP and MBP genes progressed in the caudo-cranial direction in the brain as far as we examined it in the tissue sections up to the 30th postnatal day: mRNAs of both genes were first detected in the pons and the medulla oblongata on the 3rd postnatal day and then in the cerebellum and the anterior part of the brain. PLP mRNA was located exclusively in the cell body throughout development. The number of PLP mRNA-positive cells reached a plateau in the posterior part of the brain on the 18th postnatal day, whereas it continued to increase in the anterior part of the brain by the 30th day. MBP mRNA was first expressed in the cell body, but later, it was found along the myelin sheath.

Differential expression of galactocerebroside, myelin basic protein, and 2',3'-cyclic nucleotide 3'-phosphohydrolase during development of oligodendrocytes in vitro

This paper describes the differential expression and localization of myelin components within membrane sheets produced by oligodendrocytes in vitro. In double-labeling experiments using antibodies to the myelin antigens 2',3'-cyclic nucleotide 3'-phosphohydrolase (CNP) and galactocerebroside (GC), the two antigens were coexpressed in at least 95% of oligodendrocytes at all ages examined. A small population of relatively undifferentiated cells expressed one antigen before the other. Within the membrane sheets produced by the cultured cells, CNP and GC are distributed differently. CNP is highly concentrated in cell bodies, in a network of processes extending from the cell body into the sheets, and around the perimeter of the sheets. CNP staining cannot be detected in some areas within the body of the sheet. When present, it is of low intensity. Under our labeling conditions, GC staining is found throughout the membrane sheets, except in the network of veins which are CNP+. GC and myelin basic protein (MBP) staining are seen in similar membrane domains even though GC is a surface component while MBP resides on the cytoplasmic face. Both the timing and localization of CNP immunostaining show that CNP is as early a marker for oligodendrocytes as GC, and support the idea that CNP may play a structural role in the myelin membrane. Double-labeling studies with GC and CNP antibodies also show that the true shape of a cell and the extent of its development are not always revealed by a single antigen. The differential distribution of antigens within membrane sheets illustrates that they contain areas of structural specialization that may reflect the situation in intact myelin.

RGD-containing peptides inhibit the synthesis of myelin-like membrane by cultured oligodendrocytes

A synthetic peptide derived from the fibronectin cell-binding domain, GRGDSP, inhibits the adhesion of rat oligodendrocytes to a number of substrates. However, while GRGDSP inhibited the adhesion of cells in a short term adhesion assay, the presence of the peptide did not prevent cells from adhering and thriving in longer term culture. The morphological characteristics of individual cells cultured with 0.1 mg/ml GRGDSP were similar to untreated cultures; small rounded cell bodies radiating numerous fine processes. Peptide-treated cultures were inhibited in their ability to produce myelin specific components. The characteristic developmental peak in sulfolipid synthesis which occurs both in vivo and in vitro was completely inhibited when cells were cultured with GRGDSP. In addition, the synthesis of myelin basic protein was inhibited. Ultrastructurally, cells treated with GRGDSP showed a greatly reduced number of multilamellar myelin-like membrane figures than cells grown without peptide or those grown with GRADSP. Cultured oligodendrocytes did not become sensitive to inhibition of sulfolipid synthesis by GRGDSP until a period immediately preceding the peak in sulfolipid biosynthesis. The effects of pretreatment with peptide for 5 d before this time were completely reversible. Pretreatment which extended into the time of peak myelin synthesis resulted in permanent impairment in the cell's ability to synthesize sulfolipid. The oligodendrocyte's ability to synthesize a myelin-like membrane in culture is, in part, inherent since it occurs in the absence of neurons. The present results indicate that myelin membrane production is also subject to external control since it appears that occupancy of an RGD-dependent cell surface receptor during a critical period of in vitro development is required for the oligodendrocyte to produce myelin-like membrane.

Expression of galactocerebroside in developing normal and jimpy oligodendrocytes in situ

Intense and specific immunostaining of oligodendrocytes in vivo has been obtained for the first time using antibodies to galactocerebroside. We have examined the differentiation of oligodendrocytes in normal mice and then compared their differentiation to the myelin-deficient mouse jimpy, using immunoperoxidase, immunogold and immunofluorescence labelling techniques. We also compared staining for galactocerebroside with staining obtained using antibodies to myelin basic protein, carbonic anhydrase II, 2', 3'-cyclic nucleotide 3'-phosphohydrolase and proteolipid protein. The results of this comparative study confirm previous tissue culture studies and show that galactocerebroside is specific for oligodendrocytes in situ. As in tissue culture, galactocerebroside is one of the earliest oligodendrocyte markers to be expressed, making it an important marker for studying the differentiation of this cell type. The shape of oligodendrocytes in situ changes distinctly with time, shifting from an early stellate form with numerous spidery processes to a cell with a few processes radiating from the perikaryon. These morphological changes are observed for both normal and jimpy mice and they parallel those described in vitro. Oligodendrocytes in jimpy mice express most myelin markers, but the staining within the cells is generally less intense than in normal oligodendrocytes and the antigens are restricted to the cell body and processes without being incorporated into myelin sheaths. Quantification of the number of oligodendrocytes stained for galactocerebroside in normal and jimpy mice show that their number is not reduced in the corpus callosum and cerebellum during the first 2 weeks postnatal. This finding shows that many cells in jimpy mice which were considered to be unclassifiable by the application of morphological criteria have, in fact, differentiated to the stage where they are galactocerebroside-positive.

Glial conditioned medium enables jimpy oligodendrocytes to express properties of normal oligodendrocytes: production of myelin antigens and membranes

Cultures of cerebral cells from 2-day-old jimpy (jp) and normal animals were immunocytochemically stained with antibodies to galactocerebroside, 2', 3'-cyclic nucleotide 3'-phosphohydrolase and proteolipid protein (PLP). As the normal cultures mature, oligodendrocytes express these markers in cell bodies, in processes and in expansive membrane sheets produced along the lengths of cell processes. In contrast, oligodendrocytes from jp cultures do not produce large membrane sheets and their total numbers are reduced. Only rarely do jp oligodendrocytes exhibit immunostaining for PLP. Thus, jp oligodendrocytes grown in vitro mimic deficiencies of jp CNS in situ. In order to examine the effect of environment and cell-cell interactions on expression of the jp mutation we grew jp cerebral cells in the presence of medium conditioned for short periods by normal cerebral cells. Under these conditions jp oligodendroglia appeared nearly normal by immunostaining criteria. Their numbers were increased; they were able to produce and maintain membrane sheets; and some cells expressed PLP. These results show that jp oligodendrocytes have the capacity to express certain normal phenotypic parameters, and when given an appropriate environment they do so. The effects of normal conditioned medium may be due to a secreted factor, possibly produced by the astrocytes, which constitute the vast majority of cells in the conditioning cultures. The dramatic effect of normal glial conditioned medium on jp oligodendrocytes suggests that the steps leading to hypomyelination in jp are exceedingly complex and may involve glial-glial interactions.

Production and characterization of high titer antibodies to galactocerebroside

High titer antibodies primarily of the IgG class were produced against galactocerebroside (GalC) by including keyhole limpet hemocyanin (KLH) and supplemental M. tuberculosis in the adjuvant mixture used for immunization of rabbits. Antibody titers were determined by an ELISA in which microtiter wells were coated with liposomes containing lecithin, cholesterol and GalC. The antibodies showed reactivity with GalC and psychosine, but not glucocerebroside, sulfatide, mixed gangliosides or asialo GM1. Specificity was further demonstrated by absorption of antibodies with GalC. Binding was inhibited by galactose, but only at high concentrations. Further, the antibodies did not bind to any brain proteins on immunoblots, indicating lack of reactivity with glycoproteins which might contain a terminal galactose. Antibodies to GalC are directed against different determinants than those reacting with peanut agglutinin since the lectin will not react with GalC, and the antibodies will not react with asialo GM1. The antibodies raised to GalC by this method show specific staining for oligodendroglia in culture. Peanut agglutinin binds intensely to process-bearing GalC+ oligodendroglia, but very poorly to the membrane sheets elaborated by oligodendroglia after longer times in culture. Other process-bearing GalC-, GFAP- cells were also stained with peanut agglutinin; these cells may represent glial precursors.