The Effects of cAMP on Differentiation of Cultured Schwann Cells: Progression from an Early Phenotype (04 +) to a Myelin Phenotype (Po ÷, GFAP-, N-CAM-, NGF-Receptor-) Depends on Growth Inhibition

The present experiments were designed to clarify the relationship between cAMP elevation, proliferation and differentiation in Schwann cells. They were carried out on short-term cultures of cells obtained from neonatal rat sciatic nerves. It was found that the myelin-related phenotype was expressed in response to agents that elevate or mimic intraceUular cAMP (forskolin, cholera toxin, cAMP analogues), provided, cell division was absent. This phenotype included upregulation of the major myelin protein Po and downregulation of GFAP, N-CAM, A5E3, and NGF receptor. In contrast, when cells were cultured in conditions where cell division occurred, elevation of intraceUular cAMP produced an alternative response, characterized by DNA synthesis and absence of myelin-related differentiation. The cAMP mediated induction of an early Schwann cell antigen, 04, followed a different pattern since it was induced equally in dividing and nondividing cells. These observations are consistent with the proposal that during development of the rat sciatic nerve: (a) cAMP elevation, possibly induced by axon-associated factors, is a primary signal responsible for the induction of 04 expression in proliferating Schwann cells during the premyelination period; (b) subsequent withdrawal of cells associated with the larger axons from the cell cycle acts as a permissive secondary signal for induction of myelination, since in quiescent cells the ongoing cAMP elevation will trigger myelination. T hE adenyl cyclase-cAMP second messenger pathway has been implicated in the regulation of growth and development in several cell types, including Schwann cells, the major glial element in peripheral nerves. Thus, cAMP elevation stimulates DNA synthesis in Schwann cells cultured in the presence of serum or growth factors (9, 10, 14, 40, 42, 43, 51). In cultured Schwann cells it also triggers the surface expression of two lipids, galactocerebroside and 04 (sulfatide), both of which are found on myelin-forming and non-myelin-forming Schwann cells in adult nerves (20, 32, 50, 51). During development of the rat sciatic nerve, galactocerebroside appears at embryo day(E) 18-19 in cells destined to form myelin, and during the second to fifth postnatal week in non-myelin-forming cells. 04 on the other hand is an early Schwann cell differentiation marker appearing at El6. The effect of cAMP on induction of the major myelin glycoprotein Po is more controversial. Although it has been reported that elevation of cAMP does not induce Po in short-term cultured Schwann cells (49, 51), other studies show that cAMP induces expression of Po and also induces Po mRNA and myelin basic protein mRNA in long-term Schwann cell cultures that have previously been exposed to mitogenic stimulation (24). These in vitro studies suggest that cAMP might be an intracellular signal during several different stages of Schwann cell development. In synergy with other factors, it might drive Schwann cel ! proliferation in developing nerves; it could act as a signal triggering early Schwann cell differentiation, i.e., 04 appearance; or it could be a signal for the emergence of the myelin phenotype or for the final maturation of nonmyelin-forming cells. The proposal that cAMP elevation is involved in regulation of these disparate events during the in vivo development of the rat sciatic nerve raises at least three questions. First, does cAMP elevation unambiguously mimic myelin-related differentiation in rat Schwann cells? Second, if so, what is the timing mechanism that in vivo would allow cAMP elevation to induce 04 at El6 (32) but myelin-related differentiation, as exemplified by Po expression, 5-6 d later (around birth) (29.55). Third, is induction of the myelin phenotype by cAMP in cultured cells incompatible with stimulation of DNA synthesis, as predicted by the observation that in vivo myelinrelated differentiation does not occur in dividing ceils (56). To answer these questions we have investigated the relationship between induction of the myelin phenotype, induction of 04 expression and stimulation of DNA synthesis. We © The Rockefeller University Press, 0021-9525/91/02/457/11 $2.00 The Journal of Cell Biology, Volume 112, Number 3, February 1991457--467 457 on July 9, 2017 jcb.rress.org D ow nladed fom show for the first time that in short-term Schwann cell cultures elevation of intracellular cAMP not only induces expression of Po but, in the same cells, triggers downregulation of several proteins normally downregulated during myelination in vivo, including GFAP, N-CAM, and ASE3, and also nerve growth factor (NGF) receptors (NGFR) ~ detected by the 217c antibody (8, 15, 18, 21, 23, 26, 27, 3t, 36). Significantly, we find that in short-term Schwann cell cultures, these changes only take place if the proliferative response to cAMP is prevented. In contrast, 04 expression is readily induced by cAMP in dividing cells. These observations, and earlier work on the regulation of galactocerebroside expression are consistent with the proposal that cAMP acts as a second messenger of an early and general Schwann cell differentiation signal. They also suggest that factors regulating Schwann cell proliferation control the transition from the premyelin to the myelin phenotype. Materials and Methods Schwann Cell Culture Sciatic nerves were dissected from early postnatal rats (5-7 d) and desheathed. 5-d rats were used in most experiments. The tissue was dissociated essentially by the method of Brockes et al. (4), and maintained in DMEM with added glutamine, insulin, and 10% calf serum. Cytosine arabinoside, 10 -5 M, was added after 24 h for 48 h to give cultures that were 95 % pure (4). On the fourth day cells were plated on to coverslips coated with polylysine, polylysine plus laminin, or extracellular matrix as required, at a density of between 2.5-20,000 cells/coverslip. Cells were left for 24 h before further treatment. In most experiments, Schwann cells were cultured on laminin but in some experiments, polylysine or coverslips coated with extracellular matrix from bovine corneal endothelial cells were used. Cells were cultured in defined medium consisting of a 1:1 mixture of DMEM and Ham's F12 medium supplemented with selenium (100/zg.ml-t), triiodothyronine (0.1 /zg.ml-l), transferrin (100 #g.ml-l), putrescine (16 /~g.ml-t), thyroxine (0.4/~g.ml-t), progesterone (60 ~g-ml-t), 30% BSA (Path-o-cyte, ICN Immunologicals, Lisle, IL) (0.3 mg.ml-~), dexamethasone (38 ng.ml-l), insulin (5 #g.mi-l) with or without the addition of NGF (10 U.mi-l), glucose (2 mg.ml-l), 10% calf serum or glial growth factor (GGF) at a concentration of 1.8 #g-ml -t. The GGF was a semipure fraction (GGF-CM) prepared by Dr. A. J. D. Goodearl (Ludwig Institute, London) as described elsewhere (47). Calf serum was added in concentrations ranging from zero to 10%. Cell Culture in the Presence of Cholera Toxin Cholera toxin (C'lx) (Sigma Chemical Co., St. Louis, MO) (150 ng.ml -x) was added to Sehwann cell cultures for 1 h at the start of experiments, and 24 hr later for an additional hour. In experiments lasting >4 d, CTx was added for 1 h at 96 and 168 h. As a control for the specificity of the CTx, the ~ subunit was used in some experiments at the same concentration and using a similar regime to that used for CTx. Cell Culture in the Presence of Forskolin In some experiments forskolin, a reversible activator of adenyl cyclase was added to cells cultured in media as described above. A 10 mM stock in alcohol was diluted in medium and added at concentrations varying from 0.01 to 200 #M. At the higher concentrations a control containing alcohol alone was added to the experiment. The forskolin was replaced in new medium every 24 h. Cell Culture in the Presence of cAMP Analogues In these experiments Schwann cells were placed 6n coverslips coated with 1. Abbreviations used in this paper: BrdU, bromodeoxyuridine; CTx, cholera toxin; GCF, glial growth factor; IBMX, 3-isobutyl-l-methylxanthine; NGFR, nerve growth factor receptor. polylysine, laminin, or extracellular matrix. They were cultured in media as described above, cAMP analogues, NS,2'-O-dibutyryladenosine 3':5'cyclic monophosphate (db-cAMP) (5 × 10 -4 M) plus 8-bromoadenosine 3':5'-cyclic monophosphate (8 bromo-cAMP) (5 × 10 -4 M) were added at the start of the experiment. At 24 h and 48 h, fresh medium containing the two analogues, each at a concentration of 5 × 10 -5 M, was added to the cultures. If experiments lasted >3 d, this dose was repeated every day until the end of the experiment. Similar concentrations of butyrate were used in control experiments. Cell Culture in Isobutylmethylxanthine A 50 mM stock solution of 3-isobutyM-methyl-xanthine (IBMX), an inhibitor of cAMP hydrolysis, (Sigma Chemical Co., St. Louis, MO) in 50% ethanol was diluted 1:5 in PBS. This solution (10 mM in 10% ethanol) was used to give a final dilution of 100 #M in Schwann cell cultures at the start of the experiment. At 24 and 48 h, fresh medium containing 100 I~M IBMX was added to the cultures. Control cultures contained 0.1% ethanol.