|Cancer||Cell death||Cell cycle||Cytoskeleton||Exo/endocytosis||Differentiation||Division||Organelles||Signalling||Stem cells||Trafficking|
Cell Biology International (2004) 28, 273279 (Printed in Great Britain)
Differentiation of chromaffin cells elicited by ELF MF modifies gene expression pattern
Tatiana Olivares‑Bañuelosa, Luz Navarrob, Alicia Gonzáleza and Rene Drucker‑Colı́na*
aDepartamento de Neurociencias, Instituto de Fisiologı́a Celular, Universidad Nacional Autónoma de México, Apartado Postal 70-600, México, D.F. 04510, Mexico
bDepartamento de Fisiologı́a, Facultad de Medicina, Universidad Nacional Autónoma de México, Apartado Postal 70-250, México, D.F. 04510, Mexico
Chromaffin cells exposed to extremely low frequency magnetic fields (ELF MF, 60 Hz, 0.7 mT) differentiate into sympathetic neuron-like cells. This complex process must involve both qualitative and quantitative variations in gene expression. This study looks at whether ELF MF treatment provokes changes in the global transcription profile of chromaffin cells, using the RT-Differential Display method. When the gene expression patterns of experimental groups (nerve growth factor (NGF) and ELF MF) were compared to those receiving no treatment, at least 53 transcripts showing differential expression were detected. Eight RT–PCR products, corresponding to six genes, were re-amplified, sequenced and compared with the rat gene bank. Sequence analysis showed that these genes most likely encode: phosphoglucomutase-1, neurofibromatosis-2 interacting protein, microtubule associated protein-2, thiamine pyrophosphokinase, and two unidentified hypothetical proteins (RNOR02022103 and ROR01044577), and that the presumed regulatory regions of these genes contained CTCT-clusters, which are thought to be required for electromagnetic field-dependent gene expression.
Keywords: ELF MF, Differential display, Chromaffin cells, NGF.
*Corresponding author. Tel.: +52-5-550-6662; fax: +52-5-550-0904
Chromaffin cells have been used extensively as models to study differentiation. Several reports have shown that such cells can change from an endocrine phenotype to a sympathetic neuron-like phenotype when stimulated with either NGF or ELF MF (Black et al., 1986; Drucker-Colı́n et al., 1994; Unsicker et al., 1978). We have recently demonstrated that ELF MF neuron-like differentiated cells from Wistar neonatal rats predominantly form dopaminergic cells (Verdugo-Dı́az et al., 1998) and develop neurites with a large number of tyrosine hydroxylase-positive vesicles, as well as augmented neurofilaments (Feria-Velasco et al., 1998).
It has also been shown that this differentiation process largely depends on
The mechanisms by which magnetic fields induce functional modifications are not fully understood. Concern about the potential health effects of ELF MF is due in part to the fact that some epidemiological studies suggest an association between certain neoplasms, especially childhood leukemia, and exposure to such fields (Shahidain et al., 2001). With regard to the modification of gene expression by exposure to ELF MF, it has been found that magnetic fields increase the binding of transcriptional activators to their cognate sequences, and several hypotheses have been put forward to explain how this is achieved (Blank and Goodman, 2001; Goodman and Blank, 2002).
The present study attempts to identify genes that could be involved in ELF MF-induced chromaffin cell differentiation. To this end, we used the mRNA differential display method. This experimental approach allowed the analysis of a large number of expressed transcripts, thus we were able to show the existence of specific patterns present exclusively in ELF MF treated cells. In order to gain a better insight into the selectivity of ELF MF elicited patterns, we also analyzed the expression profile of NGF treated cells.
2 Materials and methods
2.1 Chromaffin cell cultures
Adrenal medulla cells from 80 neonatal rats (Wistar, 1–3 days old) were obtained and homogeneous chromaffin cell pellets were prepared from them. Enzymatic dissociation was carried out using 40 mg/ml collagenase (Worthington, Type I) and 15 mg/ml deoxyribonuclease I (Sigma, Type II) in a calcium-free Spinner's Saline Solution (Sigma, product No. H2387) for 45 min at 37 °C with continuous stirring (Unsicker et al., 1978). The tissue was mechanically dispersed with a Pasteur pipette and the cell suspension was centrifuged for10 min at 900 rpm at 20 °C. Both chromaffin cell viability and purity were 95% (as determined by the trypan blue exclusion method, counting cells with a Neubauer chamber). The cell suspension was plated over fifteen 35mm Petri dishes (1×106cells per dish) and divided into three experimental groups (5 per group). The culture medium was Dulbecco's modified Eagle (DMEM, Gibco) supplemented with 10% fetal bovine serum (Gibco), 4.5 mg/ml insulin (Sigma), 100 U/ml penicillin (Sigma), 100 mg/ml streptomycin (Sigma), and 2.5 mg/ml fungizone (Gibco).
2.2 Experimental groups
Three chromaffin cell groups were prepared: (1) control group, chromaffin cells without any stimulation, (2) NGF group, cells cultured in the presence of 50 ng/ml NGF added to the culture every three days, and (3) ELF MF group, cells exposed to ELF MF stimulus, 0.7 mT, 60 Hz. The first two groups were placed in a separate incubator to prevent exposure to ELF MF. Cells were grown for 7 days in a mixture of 5% CO
2.3 ELF MF treatment
The cultures exposed to ELF MF were placed in an incubator with acrylic supports and two pairs of Helmholtz coils, as previously described (Drucker-Colı́n et al., 1994). Each coil consisted of 1000 turns of enameled copper wire (22 mm diameter) contained in plastic boxes (10.5×10.5×3.5 cm). The coils generated a vertical magnetic field over the dishes (resistance of 10 ohms and inductance of 83.23 mH) in the form of a sinusoidal wave with a frequency of 60 Hz and amplitude of 0.7 mT. The magnetic field background was 0.44 μT. Petri dishes were placed between two coils at a distance of 21 cm and the magnetic field intensity over each dish was measured with an Electromagnetic Field Radiation Tester (Lutron, EMF-827). Chromaffin cells were exposed to ELF MF (as described by Drucker-Colı́n et al., 1994) for 4 h a day for 7 days (2 h in the morning and 2 h in the afternoon), beginning on the second day of culture. As a control, we alternated the use of the incubators (switching the Helmholtz coils) so that the three groups were incubated under exactly the same conditions.
2.4 Total RNA isolation
RNA isolation was carried out following the manufacturer's protocol (TRIzol, Gibco BRL). RNA samples were suspended in 20 μl RNAse-free water (DEPC). Total RNA was analyzed on denaturing (1×MOPS and 6.6% formaldehyde, Gibco BRL) 1% agarose gels (Gibco BRL) and quantified at 260 nm.
2.5 Differential display
A reverse transcription (RT) reaction was carried out with 4 μg total RNA from each group (n=3), using an anchored primer (T
2.6 Northern blot
Twenty μg total RNA obtained from each chromaffin cell group was run on a denaturing (1×MOPS and 6.6% formaldehyde, Gibco BRL) 1% agarose gel of 11×14 cm. The gel was run for 2 h at 80 mV with 1×MOPS and transferred to a nylon membrane (Gene Screen plus, Dupont). Prehybridization was carried out for 2 h with 6×SSC (Standard Saline Citrate, Gibco BRL), 5×Denhardt's, 50% formamide plus 85 μg/ml salmon sperm DNA (Gibco BRL). 100 ng of the re-amplified cDNA samples were labeled with 50 μC α32P-dCTP (NEN) using a random primer labeling kit (Life Technologies). The membrane was hybridized overnight at 42 °C, with the previous pre-hybridization solution plus 1×106cpm/ ml labeled probe. The membrane was washed once in 2×SSC plus 0.1% SDS at room temperature, and twice with 0.1×SSC plus 0.1% SDS at 42 °C (Fourney et al., 1987). The membrane was exposed to X-ray film (Kodak), which was developed after 2 days.
2.7 DNA sequence
The differential bands were cloned in T-vector, amplified in pBSSK+plasmid (Invitrogen) and sequenced in an automatic sequencer from Applied Biosystems (Perkin Elmer), ABI PRISM™ 310 version 3.2, using the T7 primer (GTA ATA CGA CTC ACT ATA GGG).
3.1 Low frequency magnetic fields modify gene expression in chromaffin cells
Gene expression in differentiated chromaffin cells was analyzed using the differential display method, with four primers (Ap1, Ap5, Ap6 and Ap7) showing at least 53 differential bands between groups (see Fig. 1-A). Arrows indicate the bands chosen for further analysis: Ap1-7, Ap5-14, Ap5-15, Ap5-19, Ap6-7, Ap7-1, Ap7-7 and Ap7-13 (they were named according to the primer that was used). Sequencing of specific fragments and comparison with the rat genome sequence, accessed through the Ensembl Data Base (www.ensembl.org on 4 December 2003), revealed that six RT PCR products were down-regulated by ELF MF, as follows: Ap1-7 corresponded to a predicted protein RNOR02022103 of unknown function, Ap5-15 and Ap7-1 to phosphoglucomutase-1, Ap5-19 to neurofibromatosis 2-interacting protein, and Ap5-14 and Ap7-7 to microtubule-associated protein 2. Two up-regulated transcripts were chosen: Ap6-7 corresponded tothiamine pyrophosphokinase, and Ap7-13 to RNOR01044577 hypothetical protein. Northern analysis of total chromaffin cell RNA, probed with Ap5-15 and Ap1-7 cDNA fragments, validated the observations obtained (Fig. 1-B) from differential display.
(A) Representative silver-stained polyacrylamide gel of differential expressed transcripts from chromaffin cells. The cDNA was amplified with an anchorage primer (T
The expression patterns from the 53 transcripts are presented in Fig. 2, showing that down-regulation is more prevalent in ELF MF and NGF-treated cells than in controls (Fig. 2-A, B, C and D). With respect to the up-regulated genes, Fig. 2-B shows that there is a conspicuous group of eight transcripts whose expression is particularly increased in ELF MF treated cells. Conversely, Fig. 2-A and C show a group of up-regulated transcripts that is peculiar to NGF treated cells. Reproducibility of results is evident from the standard deviation that was obtained, indicating that neural differentiation is accompanied by changes in geneexpression.
Expression patterns of ELF MF chromaffin cell transcripts according to the amplification primer. (A) Ap1; (B) Ap5; (C) Ap6 and (D) Ap7. White bars, control; grey bars, NGF and black bars, ELF MF. Mean data from three experiments: standard deviation is shown.
3.2 Analysis of the 5′ regulatory region of ELF MF-regulated genes
In order to determine the presumed regulatory regions, known as Electromagnetic Response Elements (Lin et al., 2001), of the six genes whose transcripts were regulated by ELF MF treatment, we analyzed a 1000 bp sequence located in the 5′ region up-stream of the ATG coding sequence of each reported gene. As Fig. 3shows, the presumed regulatory region of the six analyzed genes contained CTCT sequences. The RNOR02022103 presumed regulatory region contains 12 CTCT boxes within the first 500 bp, and 18 boxes in the 1000 bp. The 1000 bp representing the regulatory region of the rest of the identified genes carry a variable number of CTCT motifs, from 8 to 10 (Fig. 3), and in most cases a cluster of motifs is found within the first 500 bp. As a control, we determined the number of CTCT boxes contained in the regulatory region of the constitutively expressed actin and Histone-2 genes; in these cases, only 2 to 3 boxes were found in the first 500 bp region.
Promoter regions of genes identified by differential display. Black boxes represent CTCT sequences, white boxes represent ATG.
Chromaffin cells constitute a model of cellular differentiation that has been widely studied (Drucker-Colı́net al., 1994; Unsicker et al., 1978). It has been shown that ELF MF treatment elicits a process leading to the differentiation of chromaffin cells into dopaminergic neuron-like cells (Verdugo-Dı́az et al., 1998) presenting an increased number of neurofilaments (Feria-Velascoet al., 1998) in neonatal Wistar rats. This trans-differentiation process is complex and implies important modifications to gene expression are occurring, as was confirmed by this study.
For example, Fig. 1-A and Fig. 2show sharp differences in the transcriptional profile of the cells, due to modifications in 53 transcript expression patterns. The fact that 30 transcripts had decreased expression indicates that down-regulation is prevalent in both ELF MF and NGF-treated cells. This suggests that the differentiation process elicited by these two factors causes changes in cellular programs, turning off expression of complete sets of genes. Similar results have been reported by Angelastro et al. (2000)with PC12 NGF treated cells. Other genes are over-expressed, so the combination of the two may be responsible for the process.
The metabolic pathway involved in each stimulus is different, as shown by the comparison of the expression profiles of ELF MF and NGF-treated cells in Fig. 2. Although in both cases down-regulation was predominant, it appears that a set of up-regulated genes was conspicuously increased in ELF MF treated cells (Fig. 2-B). Conversely, Fig. 2-A and C show a group of up-regulated transcripts peculiar to NGF treated cells. Taken together, these results indicate that ELF MF treatment produces specific modifications in the gene expression profile, which are different to those obtained with NGF. Identification of the genes whose regulation is specifically increased by ELF MF treatment will surely shed light on the nature of the signal transduction pathway needed for ELF MF-dependent differentiation. It is worth mentioning that the ELF MF treated cells, as a control, were cultured in an incubator that had been used for control and NGF treated cells in a previous experiment.
Modification of gene expression after ELF MF treatment has been reported in other systems. For example, leukemic and normal T-lymphocytes modified the expression of c-myc, c-jun and c-fos in sea urchins, although no change in expression pattern was reported (Balcer-Kubiczek et al., 2000; Lyle et al., 1991).
It is thought that exposure to ELF MF induces a stress response (Lin et al., 1999). Of the six genes that we were able to identify, the only one that has been related to stress is that encoding phosphoglucomutase type 1. This enzyme participates in galactose metabolism, converting galactose to glucose-1-phosphate, and it has been reported that the yeast gene encoding phosphoglucomutase-1 reduces its expression after stress treatments (Hirata et al., 2003). These results suggest that ELF MF chromaffin differentiated cells could change the expression profile of genes whose products are related to a stress response.
That low frequency electromagnetic fields induce increased gene expression through the action of CTCT binding sites has previously been proposed (Lin et al., 2001). It has also been shown that a 900 bp region containing eight CTCT sequences within the fragment was capable of promoting ELF MF-induced expression of CAT or luciferase reporter constructs. Furthermore, it was found that the presence of even one CTCT site was sufficient to sustain a 1.5-fold increase in CAT response (Lin et al., 2001). Thus, it has been concluded that these elements play a crucial role in ELF MF-induced gene expression. Considering these propositions, analysis of a 1000 bp zone, located at 5′ region of the ATG from the full coding sequence, was performed to determine the CTCT sequences therein.
As Fig. 3shows, the presumed regulatory regions of the six analyzed genes contained CTCT sequences within their first 1000 bp. The identified genes of phosphoglucomutase-1, microtubule-associated protein 2, thiamine pyrophosphokinase, RNOR01044577 hypothetical protein, and neurofibromatosis 2 interacting protein carry a variable number of CTCT motifs, from 8 to 10, in the 1000 bp (Fig. 3).
However, the control genes that are not regulated by ELF MF (histone 2 and actin) contained a similar amount and distribution of CTCT sequences to some of the ELF MF-dependent genes. Thus, a more careful analysis will have to be performed in order to establish the role of the CTCT boxes. Most interesting was the finding that two genes showed a cluster of motifs within the first 500 bp. In particular, the RNOR02022103 predicted promoter sequence contained 18 CTCT boxes in 1000 bp (12 boxes in 500 bp). This number far exceeds the one found for c-myc promoter (eight boxes in 900 bp) (Lin et al., 1999). In contrast, the regulatory regions of the constitutively expressed actin and histone 2 genes have only three CTCT boxes in the first 500 bp region.
Thus, it can be speculated that the organization of the promoter region of ELF MF regulated genes containing clusters of CTCT boxes could be essential for their induced expression. However, other sequences that play a role in the ELF MF transcriptional response must also be present, since, as mentioned earlier, not all the studied genes show CTCT clusters. Although the presence of the CTCT motif has been associated with increased ELF MF-dependent expression, our results show that RNOR02022103 responds to ELF MF treatment with a 40-fold decreased expression compared to control cells. Thus, it seems plausible that CTCT elements could participate in either up- or down-regulation responses. Similarly, the CTCT motifs present in the phosphoglucomutase gene could participate in its diminished expression after ELF MF treatment.
Taken together, our results indicate: (i) that ELF MF treatment elicits a peculiar transcriptional response, qualitatively different to that elicited by NGF, and (ii) that there is an increased density of CTCT elements in ELF MF down-regulated promoters. This observation supports the proposition that the large repulsive forces generated by ELF MF could cause a physical modification of the DNA structure of the region (Blank and Goodman, 2001), which could either favor or hinder transcription.
This work was partially supported by CONACyT 119311 to TNOB and 25122-M to RDC, and by Fideicomiso UNAM to RDC. We also want to thank Juan Carlos Calcino, Rafael Goudea, Diana Millan-Aldaco, Marcela Palomero-Rivero, Alexander de Luna and Victor Anaya for their technical assistance, and Ma. Teresa Torres-Peralta for typing the manuscript.
Balcer-Kubiczek EK, Harrison, GH, Davis, CC, Haas, ML, Koffman, BH. Expression analysis of human HL60 cells exposed to 60 Hz square- or sine-wave magnetic fields. Radiat Res 2000:153:670-8
Blank M, Goodman, R. . Czerske E, Al-Barazi H, Casamento J, Davis C, Elson E, Ning J, et al. Comparison of the effect of elf fields on c-myc oncogene expression in normal and transformed human cells. Bioelectromagnetic Society, Thirteenth Annual Meeting, Salt Lake City, UT, 1991, B-2–14.
Drucker-Colı́n R, Verdugo-Dı́az, L, Méndez, M, Carrillo-Ruı́z, J, Morgado-Valle, C, Hernández-Cruz, A. Comparison between low frequency magnetic field stimulation and nerve growth factor treatment of cultured chromaffin cells, on neurite growth, noradrenaline release, excitable properties, and grafting in nigrostriatal lesioned rats. Mol Cell Neurosci 1994:5:485-98
Feria-Velasco A, Castillo-Medina, S, Verdugo-Dı́az, L, Castellanos, E, Orozco-Suárez, S, Sánchez-Gómez, C. Neuronal differentiation of chromaffin cells in vitro, induced by extremely low frequency magnetic fields or nerve growth factor: a histological and ultrastructural comparative study. J Neurosci Res 1998:53:569-82
Goodman R, Chizmadzhev, Y, Shirley-Henderson, A. Electromagnetic fields and cells. J Cell Biochem 1993:51:436-41
Jacobson J. Influence of electromagnetism on genes and associated structures. Isr J Med Sci 1994:30:245-8
Morgado-Valle C, Verdugo-Dı́az, L, Garcı́a, D, Morales-Orozco, C, Drucker-Colı́n, R. The role of voltage-gated Ca2+channels in neurite growth of cultured chromaffin cells induced by extremely low frequency (ELF) magnetic field stimulation. Cell Tissue Res 1997:632:1-14
Peng L, Bauer, D, Averboukh, L, Warthoe, P, Rohrwild, M, Muller, H. Analysis of altered gene expression by differential display. Methods Enzymol 1995:254:304-21
Phillips J. Effects of electromagnetic field exposure on gene transcription. J Cell Biochem 1993:51:381-6
Phillips J. . Phillips JL, Haggren W, Adey WR. Effects of 60 Hz magnetic field exposure of specific gene transcription in CEM-CM3 human T-lymphoblastoid cells. Bioelectromagnetic Society, Thirteenth Annual Meeting, Salt Lake City, UT, 1991, B-4–3.
Shahidain R, Mullins, RD, Sisken, JE. Calcium spiking activity and baseline calcium levels in ROS 17/2.8 cells exposed to extremely low frequency electromagnetic fields (ELF EMF). Int J Radiat Biol 2001:77:2:241-8
Takahashi K, Kaneko, I, Date, M, Fukada, E. Effect of pulsing electromagnetic fields on DNA synthesis in mammalian cells in culture. Experimentia 1986:42:185-6
Unsicker K, Krisch, B, Otten, U, Thoenen, H. Nerve growth factor-induced fiber outgrowth from isolated rat adrenal chromaffin cell: Impairment by glucocorticoids. Proc Natl Acad Sci U S A 1978:81:2242-6
Verdugo-Dı́az L, Palomero, M, Drucker-Colin, R. Differentiation of chromaffin cells by extremely low frequency magnetic fields changes ratios of catecholamine type messenger. Bioelectrochem& Bioenerg 1998:46:297-300
Received 14 August 2003/11 December 2003; accepted 20 January 2004doi:10.1016/j.cellbi.2004.01.002