Search

Central nervous system progenitor cells that are self-renewing in culture and also differentiate under controlled conditions are potentially useful for developmental studies and for cell-based therapies. We characterized growth and plasticity properties and gene expression in a rat mesencephalic cell line, AF5, that was immortalized with an N-terminal fragment of SV40 large T (T155g). For over 150 population doublings in culture, the growth rate of AF5 cells remained steady, the cells remained responsive to bFGF, and telomerase activity and telomere lengths were unchanged. While karyotype analyses revealed some chromosomal abnormalities, these were also unchanged over time; additionally, no mutations in p53 gene sequences were found, and wild-type p53 activation was normal. AF5 cells produced PDGF, TGFbeta1, TGFbeta2, and bFGF. Similar to primary progenitor cells, AF5 cells retained their plasticity in culture; they could be propagated in an undifferentiated state as "neurospheres" in serum-free media or as adherent cultures in serum-containing media, and they differentiated when allowed to become confluent. Adherent subconfluent actively growing cultures expressed a marker for immature neurons, nestin, while few cells expressed the mature neuronal cell marker betaIII-tubulin. Confluent cultures ceased growing, developed differentiated morphologies, contained few or no nestin-expressing cells, and acquired betaIII-tubulin expression. Global gene expression was examined using a 15,000 gene microarray, comparing exponential growth with and without bFGF stimulation, and the differentiated state. The AF5 cell line exhibited stable genetic and growth properties over extended periods of time, while retaining the ability to differentiate in vitro. These data suggest that the AF5 cell line may be useful as an in vitro model system for studies of neural differentiation.

Process extension was induced in cells of the N18-RE-105 neuroblastoma-retinal hybrid line by toxic agents, including glutamate and the p53-inducing anticancer agents adriamycin and etoposide. Both adriamycin and glutamate activated p53 as measured by a plasmid transfection assay. It was therefore hypothesized that SV40 large T antigen, which binds p53, would interfere with cellular differentiation. To test this hypothesis, the temperature-sensitive form of SV40 large T was transduced into N18-RE-105 cells by retroviral infection. SV40 large T-infected cells became de-differentiated, grew in tightly-packed colonies, lost expression of neurofilament, and lost the ability to differentiate in response to glutamate and adriamycin. The de-differentiating effect of SV40 large T antigen may be due to binding and inactivation of cellular proteins, such as p53, p107, p130, p300, and retinoblastoma protein, which are important in cellular growth and differentiation. It is suggested that p53 may play a role in cellular differentiation, perhaps under unusual circumstances involving stress or cytotoxicity.

As an alternative to primary fetal tissue, immortalized central nervous system (CNS)-derived cell lines are useful for in vitro CNS model systems and for gene manipulation with potential clinical use in neural transplantation. However, obtaining immortalized cells with a desired phenotype is unpredictable, because the molecular mechanisms of growth and differentiation of CNS cells are poorly understood. The SV40 large T antigen is commonly used to immortalize mammalian cells, but it interferes with multiple cell-cycle components, including p53, p300, and retinoblastoma protein, and usually produces cells with undifferentiated phenotypes. In order to increase the phenotypic repertoire of immortalized CNS cells and to address the molecular mechanisms underlying immortalization and differentiation, we constructed an expression vector containing a truncated SV40 large T gene that encodes only the amino-terminal 155 amino acids (T155), which lacks the p53-binding domain. Constructs were first transfected into a p53-temperature-sensitive cell line, T64-7B. Colonies expressing T155 proliferated at the growth-restrictive temperature. T155 was then transfected into primary cultures from embryonic day-14 rat mesencephalon. Two clonal cell lines were derived, AF-5 and AC-10, which co-expressed T155 and mature neuronal and astrocytic markers. Thus, the amino-terminal portion of SV40 large T is sufficient to: (1) overcome p53-mediated growth arrest despite the absence of a p53-binding region, and (2) immortalize primary CNS cells expressing mature markers while actively dividing. T155 and T155-transfectants may be useful for further studies of cell-cycle mechanisms and phenotyic expression in CNS cells or for further gene manipulation to produce cells with specific properties.

Transforming growth factor betas (TGFβs) are potent immunosuppressive molecules released in the brain after injury. We hypothesized that TGFβ levels in cerebrospinal fluid (CSF) of schizophrenic patients would be altered because TGFβ can influence neural cell adhesion moelecule (N-CAM) expression in vitro. The levels of TGFβ1 and β2 in CSF of patients with schizophrenia and normal controls measured by ELISA showed no differences. There was evidence that the stability of TGFβ in CSF may be altered in schizophrenia. For a limited sample, TGFβ1 and N-CAM concentrations were significantly correlated in normal patients (r = 0.98) but not in schizophrenics. The results do not support an active neurodegeneration or anti-inflammatory response in the central nervous system, which is reflected in the CSF of chronic schizophrenics. © 1997 American College of Neuropsychopharmacology [Neuropsychopharmacology 16:83–87,1997]

Medulloblastomas are poorly differentiated brain tumors believed to arise from primitive pleuripotential stem cells, and tend to express mixed neuronal and glial properties. In the present study, we examined immunohistochemical and neurotransmitter phenotypic properties in a newly established medulloblastoma cell line, MCD-1. MCD-1 cells were immortal, not contact-inhibited, but did not grow in soft agar. Immunohistochemical studies showed positive staining for neurofilament protein (NF), neuron-specific enolase (NSE), synaptophysin, MAP 2, tau, NCAM 180, vimentin, and S-100 protein. The cells expressed specific uptake of glutamate, serotonin, and choline, but not GABA or dopamine. A significant increase in process extension was seen in response to agents that enhance intracellular cyclic AMP, especially 3-isobutyl-1-methylxanthine (IBMX). Process formation induced by IBMX was associated with a decrease in cell proliferation as evidenced by a reduction in numbers of cells incorporating 5-bromo-2-deoxyuridine (BrdU). No increase in process extension was observed following exposure to NGF or retinoic acid. MCD-1 cells were shown to produce transforming growth factor beta (TGF beta), and were immunopositive for mutant p53. Transfection assays with the PG13-Luc reporter plasmid, which contains a p53-responsive enhancer element and a luciferase reporter gene, suggested MCD-1 cells are deficient in wild-type p53 and do not activate p53 on treatment with the anticancer agent adriamycin. The MCD-1 cell line is suggested to represent an abnormally differentiated cell type, which has some properties consistent with a multipotent neuronal phenotype while retaining some properties of immature cells of a glial lineage. The MCD-1 cell line can be used to provide a model of a medulloblastoma cell line that is resistant to growth-controlling and anticancer agents.

[3H]AMPA binding, a measure of the non-NMDA excitatory amino acid receptors, was measured in the frontal cortex, caudate nucleus, and nucleus accumbens of postmortem human brain tissue samples. In normal human frontal cortex, the binding data were best fit by a two-site model, with Kd values of 137 nM and 11.3 μM, and Bmax values of 2780 fmol/mg protein and 67.6 pmol/mg protein, respectively. Binding was linearly related to protein concentration and was strongly inhibited by glutamic acid and quisqualic acid. Binding was partially inhibited by kainic acid and glutamic acid diethyl ester and only slightly inhibited by N-methyl- d -aspartic acid. AMPA binding was not inhibited by neuroleptic drugs, in vitro. Freezing and storage did not result in a loss of AMPA binding, and there tended to be an increase in AMPA binding with extended freezer storage. When tissue frozen intact was compared to tissue frozen as a homogenate, the high-affinity binding parameters were unchanged, but there was an increase in the affinity and Bmax of the low-affinity site for the tissue frozen intact. Thus it appears that only the high-affinity site can be measured accurately in tissue frozen intact. AMPA binding was not significantly altered by premortem neuroleptic administration, age, postmortem delay, or by moderate durations of freezer storage. No differences in AMPA binding were found in psychotic subjects compared to normal controls. There was, however, a pronounced increase in total AMPA binding in the caudate nucleus in subjects that had committed suicide.

One therapeutic approach to stroke is the transplantation of cells capable of trophic support, reinnervation, and/or regeneration. Previously, we have described the use of novel truncated isoforms of SV40 large T antigen to generate unique cell lines from several primary rodent tissue types. Here we describe the generation of two cell lines, RTC3 and RTC4, derived from primary mesencephalic tissue using a fragment of mutant T antigen, T155c (cDNA) expressed from the RSV promoter. Both lines expressed the glial markers vimentin and S100β, but not the neuronal markers NeuN, MAP2, or β-III-tubulin. A screen for secreted trophic factors revealed substantially elevated levels of platelet-derived growth factor (PDGF) in RTC4, but not RTC3 cells. When transplanted into rat cortex, RTC4 cells survived for at least 22 days and expressed PDGF. Because PDGF has been reported to reduce ischemic injury, we examined the protective functions of RTC4 cells in an animal model of stroke. RTC4 or RTC3 cells, or vehicle, were injected into rat cortex 15–20 min prior to a 60-min middle cerebral artery ligation. Forty-eight hours later, animals were sacrificed and the stroke volume was assessed by triphenyl-tetrazolium chloride (TTC) staining. Compared to vehicle or RTC3 cells, transplanted RTC4 cells significantly reduced stroke volume. Overall, we generated a cell line with glial properties that produces PDGF and reduces ischemic injury in a rat model of stroke.

Immortalized central nervous system (CNS) cell lines are useful as in vitro models for innumerable purposes such as elucidating biochemical pathways, studies of effects of drugs, and ultimately, such cells may also be useful for neural transplantation. The SV40 large T (LT) oncoprotein, commonly used for immortalization, interacts with several cell cycle regulatory factors, including binding and inactivating p53 and retinoblastoma family cell-cycle regulators. In an attempt to define the minimal requirements of SV40 T antigen for immortalizing cells of CNS origin, we constructed T155c, encoding the N-terminal 155 amino acids of LT. The p53 binding region is known to reside in the C-terminal region of LT. An additional series of mutants was produced to further narrow the molecular targets for immortalization, and plasmid vectors were constructed for each. In a p53 temperature sensitive cell line model, T64-7B, expression of T155c and all constructs having mutations outside of the first 82 amino acids were capable of overriding cell-cycle block at the non-permissive growth temperature. Several cell lines were produced from fetal rat mesencephalic and cerebral cortical cultures using the T155c construct. The E107K construct contained a mutation in the Rb binding region, but was nonetheless capable of overcoming cell cycle block in T64-7B cell and immortalizing primary cultured cells. Cells immortalized with T155c were often highly dependent on the presence of bFGF for growth. Telomerase activity, telomere length, growth rates, and integrity of the p53 gene in cells immortalized with T155c did not change over 100 population doublings in culture, indicating that cells immortalized with T155c were generally stable during long periods of continuous culture.

We examined in the present study alterations in cerebral blood flow (CBF) and blood-brain barrier (BBB) permeability following intrastriatal transplantation of mouse bone marrow stromal cells (BMSCs) or saline infusion in adult Sprague-Dawley rats. Laser Doppler revealed that transplanted animals exhibited near normal cerebral blood flow (CBF, 150 perfusion units) at a much earlier period post-transplantation (day 4) compared to animals that received saline infusion (day 12) (p's

We monitored alterations in cerebral blood flow (CBF) and blood-brain barrier (BBB) permeability following middle cerebral artery occlusion (MCAo) and intrastriatal transplantation of mouse bone marrow stromal cells (BMSCs) or saline infusion in adult Sprague-Dawley rats. Laser Doppler and Evans Blue assay revealed that BMSC grafts dose-dependently restored CBF and BBB to near normal levels at a much earlier period (Days 4-5 post-MCAo) in transplanted stroke animals compared to stroke animals that received saline infusion (Days 11-14 post-MCAo). Xenografted BMSCs survived in the absence of immunosuppression, and elevated levels of transforming growth factor-beta superfamily of neurotrophic factors were detected in transplanted stroke animals. These data suggest that early restoration of CBF and BBB following transplantation of BMSCs could mediate the reported functional outcomes in stroke animals.

Three promoters, cellular polypeptide chain elongation factor 1 alpha (EF1), cytomegalovirus (CMV), and Rous sarcoma virus (RSV) were examined for stable transgene expression in mouse embryonic stem (ES) cells and their progeny during dopaminergic neural differentiation. In undifferentiated ES cells the EF1 promoter was highly effective, while CMV had moderate activity. After 3 months in culture, expression of humanized renilla green fluorescent protein (hrGFP) was unchanged for the EF1 promoter and decreased for CMV. At the nestin-positive stage of differentiation, hrGFP and nestin were colocalized in about 20% of cells for EF1, in contrast to 80% of cells for the CMV promoter. In tyrosine hydroxylase (TH)-positive neurons neither the EF1 nor CMV promoter were effective. The RSV promoter was inactive in undifferentiated, nestin-positive, and TH-positive cells. Thus, EF1 and CMV are effective promoters for transgene expression in undifferentiated ES cells and nestin-positive neural precursors.

Several kidney cell lines were investigated for their ability to produce glial cell line-derived neurotrophic factor (GDNF). Cell line-conditioned medium was analyzed using ELISA and two cell lines were identified which produce GDNF in physiologically active concentrations. ELISA analyses revealed that conditioned medium from these two cell lines also contained PDGF, bFGF, TGFβ1 and TGFβ2. Both of these cell lines were then transplanted into the striatal penumbra of rats, 1 h following middle cerebral artery occlusion. Behavioral testing revealed that both cell lines reduced the deficit associated with cerebral ischemia and reduced the infarct volume relative to controls. Reduction of infarct volume was likely achieved by the action of GDNF and/or other growth factors produced by the cells.

The M213-20 and M213-1L cell lines were immortalized from rat striatum using the tsA58 allele of the SV40 large T antigen, contain the GAD enzyme, and produce GABA (Giordano et al., 1994, Exp. Neurol. 124:395-400). Cell lines that produce large amounts of GABA may be useful for transplantation into the brain in conditions such as Huntington's disease or epilepsy, where localized application of GABA may be of therapeutic value. We have explored the potential use of the pREP10 plasmid vector, which replicates episomally, to increase GAD expression and GABA production in M213-20 and M213-1L cells. Human GAD(67) cDNA was transfected into M213-20 and M213-1L, and subclones were isolated with hygromycin selection. Immunochemical studies showed increased GAD(67) expression compared to the parent M213-20 and M213-1L cell lines. Staining for the EBNA antigen and Southern blots demonstrated that the pREP10 plasmid was stably maintained in the cells for at least 12-15 months in culture. Several clones were isolated in which GABA concentrations were increased by as much as 4-fold (M213-1L) or 44-fold (M213-20) compared to the parent cell lines or 12-fold (M213-1L) and 94-fold (M213-20) greater than rat striatal tissue (1.678 +/- 0.4 micromol/g prot). The ability of these cells to continue to produce large amounts of GABA while being maintained in culture for extended periods suggests that similar methods might be used with human cell lines to produce cells that can be transplanted into the brain to deliver GABA for therapeutic purposes.

In vitro growth properties of three CNS-derived cell lines were compared under a variety of culture conditions. The M213-20 and J30a cell lines were each derived from embryonic CNS culture with the temperature-sensitive (ts) allele of SV40 large T antigen, tsA58, while the A7 cell line was immortalized using wild-type SV40 large T antigen. Cells immortalized with tsA58 SV40 large T proliferate at the permissive temperature, 33 degrees C, while growth is expected to be suppressed at the nonpermissive temperature, 39.5 degrees C. Both the M213-20 and J30a cell lines were capable of proliferating at 39.5 degrees C continuously for up to 6 mo. All three cell lines showed no appreciable differences in growth rates related to temperature over a 7-day period in either serum-containing or defined serum-free media. The percentage of cells in S-phase of the cell cycle did not decrease or was elevated at 39.5 degrees C for all three cell lines. After 3 wk at 39.5 degrees C, the three cell lines also showed positive immunostaining using two monoclonal antibodies reacting with different epitopes of SV40 large T antigen. Double strand DNA sequence analyses of a 300 base pair (bp) fragment of the large T gene from each cell line, which included the ts locus, revealed mutations in both the J30a and M213-20 cell lines. The J30a cell line ts mutation had reverted to wild type, and two additional loci with bp substitutions with predicted amino acid changes were also found. While the ts mutation of the M213-20 cells was retained, an additional bp substitution with a predicted amino acid change was found. The A7 cell line sequence was identical to the reference wild-type sequence. These findings suggest that (a) nucleic acid sequences in the temperature-sensitive region of the tsA58 allele of SV40 large T are not necessarily stable, and (b) temperature sensitivity of cell lines immortalized with tsA58 is not necessarily retained.

Transforming growth factor (TGF)beta plays a role in injury repair in sites surrounding brain injury. The present study tested the hypothesis that TGFbeta1 and TGFbeta2 levels in the postmortem CSF of patients with neurodegenerative disorders would be elevated compared to those in normal subjects. Free TGFbeta1 and total TGFbeta2 were measured by ELISA in postmortem ventricular cerebrospinal fluid (vCSF) of patients with Parkinson's disease (n = 30), Alzheimer's disease (n = 30), multiple sclerosis (n = 15), and schizophrenia (n = 12) and of normal controls (n = 16). In addition, albumin, IgG, and total protein in vCSF were measured. Both TGFbeta1 and TGFbeta2 were significantly different between groups (P0.002 and P0.001, respectively). Parkinson's disease vCSF showed significant increases in both TGFbeta1 (P = 0.015) and TGFbeta2 (P = 0.012) compared to normal controls. There was a trend for TGFbeta2 to be elevated in Alzheimer's disease and multiple sclerosis vCSFs, which failed to achieve significance. There were no differences between controls and schizophrenics in TGFbeta1 or TGFbeta2. Alzheimer's disease vCSF showed a significant decrease in protein compared to all other groups, which was not related to blood-brain barrier permeability, age, or autolysis differences. Evidence is presented suggesting that some TGFbeta1 may leak into the vCSF from plasma. Autopsy vCSF levels of TGFbeta isoforms were found to be distinctly different from those reported for human serum, especially for TGFbeta2, which is undetectable in plasma. These results indicate that further in vivo studies of TGFbeta2 in the CSF of Parkinson's disease patients are warranted to determine the relationship between clinical status, medication, and TGFbeta2 concentrations.

Cultured cerebellar granule cells express phospholipase C-coupled muscarinic cholinergic, histaminergic, alpha 1-adrenergic, and serotonergic receptors. In an attempt to study desensitization of these neurotransmitter receptors, cells were prestimulated with saturating concentrations of carbachol, histamine, norepinephrine, or serotonin during the labeling of cells with myo-[3H]inositol and then rechallenged with various receptor agonists for their ability to elicit accumulation of [3H]inositol monophosphate in the presence of lithium. Prestimulation with each of these receptor agonists was found to cause a time-dependent desensitization to subsequent stimulation with the desensitizing agonist. Thus, prestimulation for 0.5, 4, and 18 h decreased carbachol response to 87 +/- 4, 52 +/- 2, and 40 +/- 1% of the control, respectively; histamine response to 37 +/- 2, 24 +/- 2, and 18 +/- 2%, respectively; norepinephrine response to 55 +/- 5, 14 +/- 1, and 10 +/- 1%, respectively; and serotonin response to 36 +/- 1, 18 +/- 1, and 9 +/- 2%, respectively. In all cases, the responses mediated by receptors which were not prestimulated remained virtually unchanged, thus indicating homologous desensitization. Dose-response studies indicate that the desensitization was associated with a major reduction in the maximal extent of agonist-induced responses. The basal accumulation was markedly enhanced following 0.5- and 4-h prestimulation, but returned to near normal after 18-h pretreatment. Biologically active phorbol ester, 4 beta-phorbol 12-myristate 13-acetate, rapidly attenuated basal phospholipase C activity, as well as the responses mediated by carbachol, histamine, norepinephrine, and serotonin, suggesting that activation and translocation of protein kinase C might play a role in the desensitization of phospholipase C-coupled receptors.(ABSTRACT TRUNCATED AT 250 WORDS)

Phosphoinositide hydrolysis was studied in neurohybrid NCB-20 cells prelabeled with myo-[3H]inositol. Among nearly 20 neurotransmitters and neuromodulators examined, only bradykinin, carbachol, and histamine significantly increased the accumulation of [3H]inositol monophosphate (IP1) in the presence of lithium. The EC50 of bradykinin was 20 nM and the saturating concentration was approximately 1 microM. The bradykinin response was robust (10-fold) and was potently and selectively blocked by a bradykinin antagonist, B 4881 [D-Arg-(Hyp3, Thi, D-Phe)-bradykinin], with a Ki of 10 nM. This effect of bradykinin appeared to be additive to that mediated by activation of muscarinic cholinergic and histamine H1 receptors. The accumulation induced by bradykinin or carbachol was dependent on the presence of calcium in the incubation medium; less than twofold stimulation was observed in the absence of exogenous calcium. Bradykinin-induced [3H]IP1 accumulation required high concentration of lithium to elicit its maximal stimulation; the concentration of lithium required for half maximal effect was about 13 mM, similar to the value reported previously for carbachol-induced accumulation in the same cell line. In contrast, using related neurohybrid NG108-15 cells, bradykinin-induced [3H]IP1 accumulation was found to require much less lithium. IN the presence of lithium, bradykinin also evoked a transient increase in the production of [3H]-inositol bis- and trisphosphate. Basal and bradykinin-induced phosphoinositide breakdown was inhibited by 4 beta-phorbol 12,13-dibutyrate, but was unaffected by the biologically inactive 4 beta-phorbol. Pretreatment of cells with pertussis toxin induced only about 30% loss of the bradykinin-induced [3H]IP1 accumulation, without affecting basal activity.(ABSTRACT TRUNCATED AT 250 WORDS)