Morphological and functional differentiation in BE(2)-M17 human neuroblastoma cells by treatment with Trans-retinoic acid

Abstract

Background: Immortalized neuronal cell lines can be induced to differentiate into more mature neurons by adding specific compounds or growth factors to the culture medium. This property makes neuronal cell lines attractive as in vitro cell models to study neuronal functions and neurotoxicity. The clonal human neuroblastoma BE(2)-M17 cell line is known to differentiate into a more prominent neuronal cell type by treatment with trans-retinoic acid. However, there is a lack of information on the morphological and functional aspects of these differentiated cells.

Results: We studied the effects of trans-retinoic acid treatment on (a) some differentiation marker proteins, (b) types of voltage-gated calcium (Ca2+) channels and (c) Ca2+-dependent neurotransmitter ([3H] glycine) release in cultured BE(2)-M17 cells. Cells treated with 10 μM trans-retinoic acid (RA) for 72 hrs exhibited marked changes in morphology to include neurite extensions; presence of P/Q, N and T-type voltage-gated Ca2+ channels; and expression of neuron specific enolase (NSE), synaptosomal-associated protein 25 (SNAP-25), nicotinic acetylcholine receptor α7 (nAChR-α7) and other neuronal markers. Moreover, retinoic acid treated cells had a significant increase in evoked Ca2+-dependent neurotransmitter release capacity. In toxicity studies of the toxic gas, phosgene (CG), that differentiation of M17 cells with RA was required to see the changes in intracellular free Ca2+ concentrations following exposure to CG.

Conclusion: Taken together, retinoic acid treated cells had improved morphological features as well as neuronal characteristics and functions; thus, these retinoic acid differentiated BE(2)-M17 cells may serve as a better neuronal model to study neurobiology and/or neurotoxicity.

Figure 1

Morphology of M17 cells without and with RA differentiation. M17 neuroblastoma cells were grown on cover slips and treated with or without 10 μM RA for 72 hours to induce differentiation. Cells were fixed, stained, and light microscopy images were taken. (A) Undifferentiated cells, (B) Differentiated cells. 40X magnification; Scale bar is approximate.

Figure 2

Progressive development of neuronal morphologies induced by RA treatment. M17 neuroblastoma cells were grown on cover slips. Cells were fixed, stained, and immunofluorescent images were taken (63X). β3-tubulin (red), synapsin-1/2 (green), and nuclei (blue). Undifferentiated M17 cells (A), development of radial glia-like morphology 72 h after RA addition (B), neurite extension and network formation (C, D).

Figure 3

Growth cone organization 120 h after RA treatment. M17 neuroblastoma cells were grown on cover slips. Cells were fixed, stained, and immunofluorescent images were taken (63X). Synapsin-1/2 (green), β3-tubulin (red) and nuclei (blue). β3-tubulin expression is predominantly localized to the neurite body, whereas synapsin-1/2 accumulates within the growth cone.

Figure 4

Expression of Neuron Specific Proteins in M17 cell cultures. M17 neuroblastoma cells were grown and treated with or without 10 μM RA for 72 hours to induce differentiation. The cells were washed and solubilized in sample buffer and analyzed by Western blotting for (A) SNAP-25, (B) synapsin, and (C) vimentin. Either β-actin or GAPDH was used as a house keeping protein marker to show equal protein loading of gels. The relative amount of each marker protein was quantified by densitometric analysis using Image J program (NIH public domain program, http://rsbweb.nih.gov/ij/index.html). Differences in marker proteins in differentiated vs. undifferentiated cells were assessed in terms of % normalized optical density in differentiated cells compared to undifferentiated (no RA) controls as shown in the bottom panel (D). Normalized optical density of undifferentiated control was 100% for each type of marker protein. **p<0.01, *p<0.05, using the student t test. n=6.

Figure 5

Expression of markers specific for cholinergic neurons. M17 cells were grown without or with 10 μM RA for 72 hours to induce differentiation. The cells were washed and solubilized in sample buffer and analyzed by Western blotting for (A) choline acetyltransferase (ChAT), (B) M1 muscarinic acetylcholine receptor (mAcR), and (C) nicotinic acetylcholine receptor α 7 (nAcR a-7). Either β-actin or GAPDH was used as a house keeping protein marker to show equal protein loading of gels. The relative amount of each marker protein was quantified by densitometric analysis using Image J program (NIH public domain program, http://rsbweb.nih.gov/ij/index.html). Differences in marker proteins in differentiated vs. undifferentiated cells were assessed in terms of % normalized optical density in differentiated cells compared to undifferentiated (no RA) controls as shown in the bottom panel (D). Normalized optical density of undifferentiated control was 100% for each type of marker protein. ND = not detected and as such could not be quantified. For M1 mAcR, the difference between undifferentiated vs differentiated was not statistically significant (student t test). n=4.

Figure 6

Effect of 10 μM RA on K⁺-evoked [³H]-glycine release in M17 cells. M17 neuroblastoma cells were grown and treated with or without 10 μM RA for 72 hours to induce differentiation. The treated M17 cells were then incubated with 2μCi/mL of [³H] glycine for 30 min and then stimulated with 80 mM KCl. The % glycine release/Total release was then calculated. ** Significantly different from control on corresponding day after Student’s t-test (p<0.05).

Figure 7

Effect of 10 μM RA on the expression of functional voltage-sensitive. Ca²⁺ channels in M17 cells. M17 cells were treated with 1 mCi/ml ³H-Valine 24 hours prior to each experiment. M17 cells were either (A) undifferentiated cells stimulated for 4 minutes with KCL, or differentiated cells (B) stimulated for 4 minutes with KCl, (C) KCl + 10 μM NNC 55-0396/KCl, (D) KCl + 1 mM w conotoxin, GVIA, or (E) KCl + 300 nM w agatoxin IVA. Each of these KCl solutions contained 1 mCi/ml of ⁴⁵Ca²⁺. The ratio of ⁴⁵Ca²⁺/³H was then used to calculate the percentage difference of Ca²⁺ channel activity. % of control was calculated by dividing the experimental ratio of ⁴⁵Ca²⁺/³H by the ratio of ⁴⁵Ca²⁺/³H generated with 5.9 mM KCl alone. n=4 *p<0.05 when compared to 5.9 mM KCl. **p<0.01 when compared to 5.9 mM KCl.

Figure 8

The effects of a Ca²⁺ionophore and phosgene (CG) on intracellular Ca²⁺ changes in differentiated vs. undifferentiated M17 cells. M17 cells were cultured in Transwell inserts without and with RA differentiation to 80% confluency. They were then exposed to 5 μM Ca²⁺ ionophore, A23187 or 0 and 16 ppm CG. Intracellular free Ca²⁺ levels were monitored using Fluo-4 Ca²⁺ indicator assay. (A) Change in intracellular Ca²⁺ following A23187 exposure (N=3); (B) Change in intracellular Ca²⁺ following either 0 ppm (air) or 16 ppm CG exposure for 8 min (N=13). *p<0.05 when compared to treated (A23187 or CG) cells.