Antiproliferative effect of the garlic compound S-allyl cysteine on human neuroblastoma cells in vitro.

A variety of compounds derived from garlic bulbs have been shown in animal systems to possess anticancer properties. However, little information is available regarding the effectiveness of garlic in the prevention or treatment of human cancers. In the current study, we have assessed the ability of S-allyl cysteine (SAC), a derivative of aged garlic extract, to affect the proliferation and differentiation of LA-N-5 human neuroblastoma cells in vitro. Time-and dose-dependent inhibition of cell grow was observed in cultures treated with SAC for at least 2 days, with a half-maximal response at approximately 600 micrograms/ml. SAC treatment was unable to induce differentiation in neuroblastoma cells as assessed by morphological, biochemical and molecular markers. In addition, SAC was unable to potentiate the effects of retinoic acid and 8-bromo-cyclic AMP, agents known to promote differentiation of LA-N-5 cells. Our results indicate that SAC can inhibit human neuroblastoma cell growth in vitro. However, the apparent inability of this compound to induce differentiation may limit its therapeutic potential.

Retinoids increase perinatal spinal cord neuronal survival and astroglial differentiation.

In this report we demonstrate that retinol and retinoic acid (RA) increase the survival and morphological differentiation of rat spinal cord neurons in vitro. Micromolar amounts of retinol and RA increased the number of surviving neurons by 2- to 3-fold and affected neuritic density resulting in increased secondary and tertiary processes compared to untreated sister cultures. A marked morphological differentiation of the astrocyte population in conjunction with an antiproliferative effect in the presence of retinoids were apparent. These trophic effects occurred mainly after 5 days in vitro, a time that corresponds to the time of birth in vivo. Retinoic acid exerted a direct trophic effect on spinal cord neurons in the absence of glial cells while retinol lost its effectiveness. Metabolic labeling suggested that retinol is converted to the biologically active RA within astrocytes but not in neurons. Taken together, our results have demonstrated direct trophic effects of RA on spinal cord neurons and have suggested another role for astrocytes in the maintenance of normal neural physiology by regulating RA concentrations through the oxidation of retinol.

[Interferon stimulates the expression of cholinergic properties in human spinal cord neurons in culture]. / L'interféron stimule l'expression des propriétés cholinergiques des neurones de moelle épinière humaine en culture.

A preparation of dissociated monolayer cultures from embryonic human spinal cord has been developed and characterized (Kato, Touzeau, Bertrand, Bader, 1985) as a model system for the study of amyotrophic lateral sclerosis (Touzeau and Kato, 1986). The cultures contain cholinergic and GABAergic neurons, astrocytes and fibroblasts. We have recently found that gamma-interferon (IFN) can increase the choline acetyltransferase (CAT) activity without altering the level of glutamic acid decarboxylase (GAD) or the neuronal survival; an antibody to IFN can prevent these effects. Gamma-IFN appears to mediate these effects via the non-neuronal cells since in the absence of non-neuronal cells, gamma-IFN has no effect on the cholinergic properties. The non-neuronal cells alone have no CAT or GAD activity. Astrocytes may be responsible for these changes since gamma-IFN increases the development of GFAP immunoreactivity in cultures of 6-7 week old spinal cord cells and it causes no visible change in the Thy-1 immunoreactivity of the fibroblasts. Thus we propose that IFN acts on non-neuronal cells, possibly the astrocytes, which in turn stimulate neuronal cholinergic traits either by means of a diffusible factor or via cell-cell contact. These studies could be relevant in understanding the effects of the immune system on the nervous system and also in the search for new drugs which act specifically on cholinergic neurons.

Differential susceptibilities of spinal cord neurons to retinoic acid-induced survival and differentiation.

In a previous study, we demonstrated trophic effects of vitamin A and its active metabolite, retinoic acid (RA), on perinatal rat spinal cord neurons and astrocytes in vitro. We now report that RA increases the survival of cholinergic neurons without affecting that of GABAergic neurons. These results were supported by measured levels of acetylcholinesterase (AChE), choline acetyltransferase (ChAT), and glutamic acid decarboxylase (GAD) activities, key enzymes of acetylcholine and gamma-aminobutyric acid metabolism, respectively, which showed RA-induced increases in AChE and ChAT levels but no elevations of GAD activity. In contrast to these phenotype-specific effects, most neurons showed RA-induced increases in neuritic outgrowth, density, and silver impregnation. Taken together, these results demonstrate neurotransmitter-specific and generalized effects of RA on developing CNS neurons.

Effects of interferon-gamma and its interaction with retinoic acid on human neuroblastoma differentiation.

Differentiation-promoting effects of interferon-gamma (IFN-gamma), both alone and in combination with retinoic acid (RA), were studied on the human neuroblastoma cell line, LA-N-5. The results show that IFN-gamma inhibited the growth and induced morphological differentiation in a dose- and time-dependent manner with measurable effects appearing at 20-40 IU/ml after 3 to 4 days of treatment in vitro. Acetylcholinesterase activity, used as a biochemical index of neuroblastoma differentiation, increased up to 2.5-fold in the presence of IFN-gamma with a half maximal concentration of approximately 100 IU/ml. Concomitantly, modest IFN-induced increases (less than or equal to 2-fold) in choline acetyltransferase (ChAT) and tyrosine hydroxylase (TH) activities were seen. Combination treatment of cells with IFN-gamma and RA resulted in synergistic effects on morphological differentiation, growth inhibition and induction of ChAT. Reversal of IFN-gamma's ability to influence neuroblastoma cell growth as well as potentiate the anti-tumor effects of RA was obtained in the presence of an antibody against the IFN-gamma receptor, implying receptor-mediated physiological events. Taken together, these data confirm the differentiating effects of IFN-gamma on human neuroblastoma cells and suggest that combination therapy with RA may be beneficial in the treatment of this disease.

Early reduction in insulin-like growth factor gene expression in diabetic nerve.

Diabetic neuropathy is a common and disabling complication of diabetes mellitus whose pathogenesis remains unknown. Insulin-like growth factors (IGFs) have been recently implicated in the development and maintenance of the peripheral nervous system, and circulating IGF levels are decreased in experimental and clinical diabetes. Therefore, we tested the hypothesis that IGF gene expression is reduced in peripheral nerves early after the onset of diabetes. Sciatic nerves from nondiabetic and streptozotocin-treated rats were removed 5-7 days after the induction of diabetes. RNA was isolated and analyzed by Northern and slot blots. IGF-I mRNA content was significantly decreased per milligram wet weight nerve (P < 0.025) as well as per poly(A)+ RNA (P < 0.01) in diabetic vs nondiabetic nerves. Likewise, the amount of IGF-II mRNA was significantly decreased per milligram wet weight nerve (P < 0.01) as well as per poly(A)+ RNA (P < 0.005). These effects were selective because histone 3.3 mRNA content, as well as poly(A)+ mRNA content, per milligram nerve were unchanged. Insulin treatment partially prevented this decline in IGF-I and IGF-II mRNA levels. The diminished IGF mRNA content is one of the earliest biochemical abnormalities to be observed in the diabetic nerve, supporting the hypothesis that a reduction in IGF activity in diabetic nerves precedes and contributes to the development of neuropathy.

Interferon induces astrocyte maturation causing an increase in cholinergic properties of cultured human spinal cord cells.

It has been well established that interferon-gamma (IFN-gamma) can modify the immune status of cells in the central nervous system (CNS) by inducing major histocompatibility antigens. Furthermore, it has been shown that endogenous IFN can be produced in the brain following viral infection and a form of IFN-alpha/beta can be produced by astrocytes in culture. Here we show that IFN can induce astrocyte maturation and alter neurotransmitter properties in cultured CNS neurons at a given developmental stage. IFN causes a dose-dependent increase in choline acetyltransferase activity and glial fibrillary acidic protein (GFAP) immunoreactivity in cultures of human embryonic spinal cord neurons. The GABAergic activity and the Thy1 immunoreactivity remain unchanged. IFN-gamma does not act directly on the neurons but via the nonneuronal cells, probably the astrocytes, which in turn stimulate the cholinergic traits. These studies could be important for demonstrating an action of the immune system on glial cell maturation and on the neurotransmitter phenotype expression in CNS neurons.

Retinoic acid up-regulates nuclear retinoic acid receptor-alpha expression in human neuroblastoma cells.

Retinoic acid (RA) nuclear receptors (RARs) are thought to mediate the cellular and molecular effects of RA on a wide variety of tissues. In most cell types, RAR alpha expression remains relatively constant following exposure to RA, while that of RAR beta is rapidly induced. In this study, we show that in human neuroblastoma, a cell type exceptionally sensitive to RA-induced differentiation, RAR alpha as well as RAR beta is markedly up-regulated by RA treatment. This effect was consistent in all 5 neuroblastoma cell lines tested and was reflected in a 2- to 5-fold increase in receptor mRNA levels as assessed by Northern-blot analysis. Using LA-N-5 human neuroblastoma cells, we found that receptor up-regulation occurred in a time- and dose-dependent fashion with increases in both RAR alpha and beta mRNA detectable 1-2 hr after the addition of RA. These inductions were not abrogated by cycloheximide, indicating that protein synthesis was not required for the RA responses. Nuclear run-off experiments combined with Northern-blot analysis of RAR alpha stability directly demonstrated that the up-regulation of RAR alpha mRNA levels reflected an increased rate of transcription without changes in message half-life. These findings, showing direct activation by RA of RAR alpha gene transcription in human neuroblastoma cells, suggest differences in the overall regulation of this receptor from that found in most other RA-inducible tissue.

Uptake of circulating insulin-like growth factor-I into the cerebrospinal fluid of normal and diabetic rats and normalization of IGF-II mRNA content in diabetic rat brain.

Brain injury has been prevented recently by systemic administration of human insulin-like growth factor-I (hIGF-I). It is widely believed that protein neurotrophic factors do not enter the brain from blood, and the mechanism by which circulating hIGF-I may be neuroprotective is uncertain. This investigation tested the hypothesis that hIGF-I is taken up into cerebrospinal fluid (CSF) from the circulation. (125)I-hIGF-I was injected subcutaneously into rats. The (125)I-IGF-I recovered from CSF and plasma were indistinguishable in size from authentic (125)I-hIGF-I on SDS-PAGE. An ELISA was used that detected immunoreactive hIGF-I, but not rat IGF-I, rat IGF-II, human IGF-II, or insulin. Osmotic minipumps were implanted for constant subcutaneous infusion of various hIGF-I doses. Uptake into CSF reached a plateau at plasma concentrations above approximately 150 ng/ml hIGF-I; the plateau was consistent with carrier-mediated uptake. The plasma, but not CSF, hIGF-I level was significantly reduced in streptozotocin diabetic vs. nondiabetic rats, and uptake of hIGF-I into CSF was nonlinear with respect to plasma hIGF-I concentrations. Nonlinear uptake excluded leakage or transmembrane diffusion of IGF-I from blood into CSF as a dominant route for entry, but the site and mechanism of uptake remain to be established. The IGF-II mRNA content per milligram brain (P < 0.02) as well as per poly(A)(+) RNA (P < 0.05) was significantly increased towards normal in diabetic rats treated by subcutaneous administration of hIGF-I vs. vehicle. This effect of circulating hIGF-I may have been due to regulation of IGF-II gene expression in the choroid plexus and leptomeninges, structures at least in part outside of the blood-central nervous system barrier. These data support the hypothesis that circulating IGF-I supports the brain indirectly through regulation of IGF-II gene expression as well as by uptake into the CSF.

Insulin-like growth factor (IGF) gene expression is reduced in neural tissues and liver from rats with non-insulin-dependent diabetes mellitus, and IGF treatment ameliorates diabetic neuropathy.

Neural disturbances are observed in the peripheral and central nervous systems of patients with insulin-dependent diabetes mellitus (IDDM) and non-IDDM (NIDDM). Insulin-like growth factors (IGFs) are neurotrophic growth factors that can support nerve regeneration and neuronal survival in the types of neurons known to be afflicted in diabetes. We tested the hypotheses that IGF gene expression is reduced in neural tissues and liver of spontaneously diabetic obese Zucker (fa/fa) rats and that IGF treatment can prevent neuropathy. There was a significant early reduction in IGF-II mRNA content as measured per mg of wet tissue or per poly(A)+ RNA in sciatic nerves, spinal cord and brain from spontaneously diabetic obese (fa/fa) vs. nondiabetic lean (+/+) adult rats. In addition, IGF-I mRNA content was reduced in liver but not nerve or spinal cord of NIDDM rats. Pain/pressure thresholds were abnormal (hyperalgesia) in diabetic (fa/fa) vs. nondiabetic (+/+) rats, and subcutaneous infusion of IGF-II restored thresholds toward normal. The low dose of IGF-II that prevented hyperalgesia in contrast had no effect on hyperglycemia or obesity. These data suggest that IGF treatment may provide rational therapy for diabetic neuropathy and that therapy may be effective even in patients unable to adequately control their hyperglycemia.

Brain insulin-like growth factor-II mRNA content is reduced in insulin-dependent and non-insulin-dependent diabetes mellitus.

Diabetic encephalopathy, characterized by structural, electrophysiological, neurochemical, and cognitive abnormalities, is observed in insulin-dependent diabetes mellitus (IDDM) and non-IDDM (NIDDM). Identification of early biochemical lesions potentially may provide clues pointing to its pathogenesis. Insulin-like growth factors (IGFs) are neurotrophic factors that recently have been implicated in the pathogenesis of diabetic neuropathy. Because IGF-II is the predominant IGF in adult brain, we tested the hypothesis that IGF-II gene expression is decreased in the CNS in both IDDM and NIDDM. Brain and spinal cord were isolated from streptozotocin-diabetic rats, a model of IDDM with weight loss and impaired insulin production. IGF-II mRNA content was measured by northern and slot blots. After 2 weeks of diabetes, IGF-II mRNA content per milligram of tissue wet weight, as well as per unit of poly(A)+ RNA, declined significantly (p < or = 0.05) in brain and spinal cord. Insulin replacement therapy partially restored IGF-II mRNA levels in brain, cortex, medulla, and spinal cord. The obese, hyperinsulinemic, and spontaneously diabetic (fa/fa) Zucker rat was used as a model of NIDDM. Brain weight (p < 0.025) and IGF-II mRNA contents (p < 0.01) were significantly decreased in (fa/fa) versus lean nondiabetic (+ /?) rats. Therefore, the decline in IGF-II mRNA levels in diabetic brain was independent of the type of diabetes, the direction of change in body weight, and the insulinemic state. We speculate that this early biochemical lesion may contribute to the development of diabetic encephalopathy.