|
|
||||||||
Address correspondence to Yih-Horng Shiao, Ph.D., Building 538, Room 205, NCI-Frederick, Frederick, MD 21702, USA; tel 301 846 1246; fax 301 846 5946; e-mail shiao{at}mail.ncifcrf.gov.
| Abstract |
|---|
|
|
|---|
B, and ATF-1 binding sites in both the -5969 to -1331 and -1331 to +530 regions. In contrast, two (CA)n repeats (-5642 to -5582 and -2078 to -2031), a family of B2 (-5428 to -5247) and B1 (-4559 to -4420) short interspersed repeated elements, and an (AT)n repeat (-5243 to -5230) were identified only in the -5969 to -1331 sequence. The results suggest that up-regulation of murine MTH1 expression by nickel(II) is controlled by the repeat sequences, potential candidates for nickel-responsive elements. (received 12 October 2000; accepted 18 October 2000)
Keywords: nickel, MTH1 gene, promoter, responsive element, gene expression
| Introduction |
|---|
|
|
|---|
Differential expression of several genes has been observed in cells treated with nickel compounds [1115]. Thus far, all known nickel effects on gene expression seem to be indirect and modulated by transcription factors. For example, nickel(II)-induced increase of ATF-1 transcription factor negatively regulates thrombospondin 1 expression [12], and the induction of Cap43 expression by nickel(II) treatment is dependent on the presence of HIF-1 transcription factor [14]. Nickel(II)-mediated DNA methylation, another indirect nickel control, has been also shown to silence expression of the gpt gene [16]. In contrast, elements modulating gene expression which would be regulated by nickel(II) directly have not yet been found. Although the metal-responsive element of the metallothionein-IIA gene and the heat shock element of the hsp70 gene respond directly to several metals, nickel(II) appears to have no effect on these two elements [17].
The responsiveness of the MTH1 gene to nickel-associated ROS or, perhaps, to nickel itself has not been established. To test the hypothesis that nickel(II) has a potency to regulate MTH1 expression, and to determine the direct or indirect mechanism of this regulation, different regions of the upstream MTH1 gene were examined to localize segment(s) responsive to nickel(II)-dependent expression control. Homologous gene sequence(s) and transcription factor binding consensus sequence(s) were analyzed using computer programs.
| Materials and Methods |
|---|
|
|
|---|
Plasmids and DNA transfection. Plasmids carrying the MTH1 upstream sequences (pHI103: nucleotides -5969 to +530; pHI104: nucleotides -1331 to +530; and pHI108: nucleotides -151 to +530) and the chloramphenicol acetyltransferase (CAT) reporter gene, described previously [18], were used as transfectants. NIH 3T3 mouse fibroblast cells were cultured in 12.5 x 8.5 cm 6-well plates (35 mm in diameter x 18 mm deep wells) with 3 ml of Dulbeccos Modified Eagles medium (DMEM) supplemented with 10% fetal bovine serum per well, in a humidified atmosphere of 5% CO2 at 37°C. Three hundred thousand cells were seeded in a single well. After 24-hr incubation, when the cells reached approximately 50% confluence, Lipofectamine Plus reagent (Life Technologies, Gaithersburg, MD) was used to transfect the cells with 2 µg of plasmid according to the manufacturers instructions.
Nickel(II) acetate treatment. After DNA transfection for 3 to 5 hr, the reagent was replaced with 3.65 ml pre-warmed DMEM containing 13% fetal bovine serum and the cells were incubated at 37°C for another 20 hr. Aliquots of 10 mM or 100 mM nickel(II) acetate stock solution in water were then added to the cells to make the final nickel concentrations in DMEM ranging from 0 to 1000 µM. No attenuation in the cell growth, or viable cell loss below 5%, were observed up to 200 µM Ni(II). Above that concentration, the number of detached (floating) cells increased gradually from approximately 10% at 400 µM to more than 60% at 1000 µM Ni(II) in 20 hr, indicating a concurrent increase in the metal toxicity. After 20-hr incubation with nickel(II), the attached cells were analyzed for CAT protein as described below.
Quantification of CAT protein expression. The culture medium was removed, the cells were rinsed twice with 2 ml portions of PBS, and finally lysed in 225 µl of 1x CAT Lysis Buffer (Boehringer Mannheim, Indian-apolis, IN) with shaking for 20 min at room temperature. Aliquots of 2.5 µl of the lysate were transferred into a 96-well microplate for total protein determination by the bicinchoninic acid method, using a kit (Pierce, Rockford, IL). The remaining lysate was used for CAT protein measurements with the CAT ELISA immunoassay kit (Boehringer Mannheim, Indian-apolis, IN), according to the producers protocol. The relative CAT protein expression was calculated as the ratio of CAT concentration to total protein.
Statistical analysis. Each treatment was repeated at least 3 times. The t-test was used to determine the difference of CAT protein level among treatments. Analysis of trend was carried out using linear regression statistics. Differences between means or slopes with p values <0.05 were considered statistically significant.
DNA sequencing and computer analysis. The ABI Prism Dye Terminator Cycle Sequencing Kit and ABI 373 automatic DNA sequencer (Perkin-Elmer Applied Biosystems, Foster City, CA) were used to determine DNA sequence. Potential homologous sequences in the MTH1 upstream region were searched using the Blast program available from the National Center for Biotechnology Information web site (http://www.ncbi.nlm.nih.gov) [19]. In the Advanced Blast, the default settings were applied except that the Filter function was inactivated to include sequences with low complexity and the Alignment was increased to maximize printout of the matched sequences. The Blast 2 Sequences program was used to determine similarity and identity between two sequences [20]. Transcription factor binding consensus sequences were examined using the Transfac program (version 4.0) at its web site (http://transfac.gbf.de) [21]. A match >80% to Genbank or Transfac databases was considered to be a homologous or consensus sequence.
| Results |
|---|
|
|
|---|
|
|
|
B, and ATF-1 binding motifs, were observed numerously in both the -5969 to -1331 and -1331 to +530 regions. In Blast analysis, 2 (CA)n repeats (-5642 to -5582 and -2078 to -2031), a family of B2 (-5428 to -5247) and B1 (-4559 to -4420) short interspersed repeated elements (SINE), and a (AT)n repeat (-5243 to -5230) were detected in the -5969 to -1331 sequence, and the 5 repeat sequences distinguished clearly the region from the -1331 to +530 sequence (Fig. 3| Discussion |
|---|
|
|
|---|
Since cellular nickel(II) is capable of initiating redox reactions that lead to generation of ROS, it is possible that the observed response of pHI103 plasmid to this metal was triggered by oxidative stress. It has been shown, indeed, that nickel(II) increases cellular levels of the ATF-1 transcription factor [12] and regulates DNA binding of AP-1 and NR-
B transcription factors [22,23], which are sensitive to redox regulation [reviewed in ref 3]. However, in our transiently transfected cells, the positive response of the MTH1 -5969 to -1331 region to nickel(II) appears not to be associated solely with the Ap1, NF-
B, or ATF-1 because binding motifs of these transcription factors are present also in the nickel-unresponsive 1331 to +530 regions. Presence of the FJH1 homologue in antisense direction of the MTH1 upstream region is interesting. Further study is needed to determine if the FJH1 homologue is transcriptionally active and contributes to MTH1 expression regulation.
Other possibilities are that nickel(II) could regulate MTH1 expression by modulating cytosine methylation/demethylation processes, or by reacting directly with selected DNA base sequence(s) in the MTH1 -5969 to -1331, thus altering conformation of certain DNA region(s) and, in turn, gene expression. It has been shown that nickel(II) can regulate gene expression by affecting DNA methylation at CpG sites in the promoter region [16]. However, in the present study we did not detect any CpG island in the MTH1 promoter region (up to -5969). Therefore, the different nickel(II) responses between the pHI103 plasmid and the pHI104 and pHI108 plasmids are most likely not associated with global DNA methylation status, although the role of individual site is not excluded.
The remaining possibility of a direct nickel(II) action on DNA, mentioned above, that would be indicative of the presence of nickel(II)-responsive element(s) in pHI103 plasmid was explored by computer analysis. The resulting finding of many repeat sequences in the MTH1 -5659 to -1331 region was quite intriguing. The B1 and B2 elements found in pHI103, but not in the other two plasmids, belong to the SINE family in rodents, corresponding to Alu repeats in humans [24]. They function as retrotransposons transposed in the genome through RNA polymerase III-transcribed RNA intermediates during evolution [24]. Up-regulation of the B1 and B2 RNA levels has been observed in cells under various stress conditions [25,26]. Previously, we also observed differential expression of the B1- and B2-like RNA in nickel(II)-treated Chinese hamster ovary (CHO) cells [13; and unpublished data]. Double-stranded Alu RNA has been shown to enhance translation by inactivating PKR kinase and subsequently inhibiting phosphorylation of eIF2-alpha, a translation suppressor if phosphorylated [27]. It has been also reported that flanking of B2 elements in either sense or antisense orientation can influence gene expression [28]. These observations would provide at least a partial explanation of the observed nickel(II)-dependent increase of CAT reporter protein in pHI103-transfected cells. Possible association of the B2 element with nickel(II)-dependent MTH1 gene expression in the current study is interesting and needs to be investigated further, including elucidation of the underlying chemistry of nickel(II) interactions with these elements.
Direct interaction of nickel(II) and other metal ions with several synthetic double- and single-stranded DNAs, including poly d(A-C).d(G-T) and poly d(A-T), has been investigated previously [2931]. This interaction promotes various types of conformational and structural transitions in DNA, including ultimately the switch from B to Z form [31]. Nickel(II) bound at the A-T base pairs is especially effective in stabilizing the syn conformation of the deoxyadenine residue and thus the Z conformation of DNA [30]. Presence of the (CA)n and (AT)n repeats in the MTH1 promoter region suggests that nickel(II)-dependent up-regulation of the CAT reporter protein in pHI103-transfected cells might be mediated by the direct interaction of nickel(II) with these repeat sequences and the subsequent DNA transitions. The B to Z transition was first observed in vitro for relatively high (
100 µM) concentrations of the reagents [30]. However, as reported by Rosetto and Nieboer [31], the metal-induced effects on DNA structure are not limited to the B to Z transition. The structural changes develop gradually and are often multiphasic [biphasic for nickel(II)] with increasing metal concentration. The first detectable conformational alterations become apparent at much lower concentrations (about 10 µM) of the metal, and the final result depends more on the DNA:metal molar ratio and binding affinity than on the absolute concentration itself. Also, high local intra-cellular concentrations of nickel(II) seem to be feasible, especially in cells phagocytizing and solubilizing internally particulate nickel compounds [32]. Controlling the interconversion between the different structural forms of DNA may relax or condense the double-stranded DNA and in this way directly affect gene expression. Although the (CA)n repeat is abundant in mammalian genome, accessibility of the sequences to nickel(II) may be dependent on the topography during chromatin unwinding. Our concept of nickel(II) targeting the (CA)n and (AT)n repeats in DNA is consistent with the finding by Kawanishi et al [33] that nickel(II) mediates oxidative DNA cleavage preferably at cytosines and thymines (followed by guanines). This process requires generation of ROS site-specifically, ie, most likely by nickel(II) bound at these residues.
The relevance of the present finding to the mechanisms of nickel-induced carcinogenesis is, at this moment, difficult to ascertain. Further studies are necessary to dissect the MTH1 promoter regions and to confirm in different models the potential involvement of the repeat sequences in gene expression regulation by nickel(II) (and possibly other carcinogenic metals), reported here for the first time, and to pinpoint the most responsive one(s). Such a direct action of nickel(II) on gene expression was observed in this study only for low nickel(II) concentrations, which might not have induced a detectable oxidative stress [34]. This, however, does not exclude the possibility that at higher cellular nickel(II) concentrations [32], ROS could also contribute to up-regulation of MTH1 expression through redox-sensitive transcription factors, as observed by others in human cells exposed to hydrogen peroxide [10]. The latter response was limited to certain low concentrations of H202; highest concentrations tend to be inhibitory.
In conclusion, the presence of repeat sequences, including B1, B2, (CA)n, and (AT)n, in the MTH1 upstream region coincides with the nickel(II)-dependent response of gene expression. The results suggest that regulation of the MTH1 gene may be mediated in part by direct interaction of nickel(II) with these repeat sequences, putative nickel(II)-responsive elements.
| Acknowledgements |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
T and A
C substitutions. J Biol Chem 1992;267:166172.This article has been cited by other articles:
![]() |
T. Josse, H. Mokrani-Benhelli, R. Benferhat, E. Shestakova, Z. Mansuroglu, H. Kakanakou, A. Billecocq, M. Bouloy, and E. Bonnefoy Association of the interferon-{beta} gene with pericentromeric heterochromatin is dynamically regulated during virus infection through a YY1-dependent mechanism Nucleic Acids Res., January 28, 2012; (2012) gks050v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Pang, M. Bartlam, Q. Zeng, H. Miyatake, T. Hisano, K. Miki, L.-L. Wong, G. F. Gao, and Z. Rao Crystal Structure of Human Pirin: AN IRON-BINDING NUCLEAR PROTEIN AND TRANSCRIPTION COFACTOR J. Biol. Chem., January 9, 2004; 279(2): 1491 - 1498. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |