Interleukin-6 Regulation of the Human DNA Methyltransferase (HDNMT) Gene in Human Erythroleukemia Cells*

DNA interleukin regulates the methyltransferase promoter and resulting enzyme activity, which requires transcriptional activation by the Fli-1 transcription 5 m M EDTA, 25% glycerol, 0.5% Triton X-100, 1 m M dithiothreitol, 0.2 m M phenylmethylsulfonyl fluoride, 5 (cid:1) Ci of S -adenosyl- L -[ methyl - 3 H]methionine (12 Ci/mmol), 4 (cid:1) g of poly(dI-dC), and 200 (cid:1) g/ml bovine serum albumin and incubated at 37 °C for 2 h. Incorporated label was assessed by scintillation counting. Reverse Transcription-PCR Assays— cDNAs for each gene were pre- pared from TriZOL (Life Technologies, Inc.) extracted total RNAs. reverse transcription reactions were run on 2 (cid:1) g of total RNA. Following reverse transcription reactions, PCR reactions were run to the midpoint of each PCR fragment’s linear synthesis curve. Glyceraldehyde-3-phos-*

The transfer of a methyl group to the cytosine portion of the CpG dinucleotide by dnmt-1 permits or enables the binding of methyl-specific DNA-binding proteins to the methylated CpG site (1,2,4,5). The binding of methyl-specific proteins such as MeCP1 and MeCP2 to genetic regulatory elements represses transcription by blocking the binding of other positive acting transactivation factors (6). Methylcytosine-DNA-binding proteins can attract histone deacetylases to the site, which remodel chromatin into highly repressed states (7). Thus, DNA methylation can result in permanent epigenetic alteration of genes and is important in promoting or guiding the differentiation of cells and the establishment of tissue-specific gene expression patterns (8).
The inflammatory cytokine IL-6 1 is able to induce the maturation and differentiation of cells (9). Treatment of the human erythroleukemia cell line K562 with IL-6 induces the expression of megakaryocytic markers and the silencing of certain globin genes (10). Derived from an acute erythroblastic leukemia, K562 cells are multipotent in that they can be directed into two separate differentiation pathways (11). K562 cells express low levels of both erythrocytic-and megakaryocyticspecific genetic markers and can be induced to differentiate along one of these two major pathways depending upon the external stimuli applied to the cells (12,13). This ability suggests some form of epigenetic control over the differentiation process. The ETS family of transcription factors represent a large family of differentially expressed, positive and negative regulators of transcription and are involved in cell differentiation (3). Here we show that when K562 cells are induced to enter the megakaryocytic differentiation pathway by IL-6, an increase in Fli-1 expression occurs, which results in the transactivation of the human methyltransferase-1 gene expression.
Methylation Assay-Cell nuclear pellets were freeze-thawed three times and centrifuged to remove debris. Clarified lysates were mixed with an equal volume of Chelex-100 resin (50%v/v) to remove DNA and RNA from the sample. For each replicate, 5 g of the protein lysate was added to 200 l of an assay mixture consisting of 20 mM Tris-HCl, pH 7.4, 5 mM EDTA, 25% glycerol, 0.5% Triton X-100, 1 mM dithiothreitol, 0.2 mM phenylmethylsulfonyl fluoride, 5 Ci of S-adenosyl-L-[methyl-3 H]methionine (12 Ci/mmol), 4 g of poly(dI-dC), and 200 g/ml bovine serum albumin and incubated at 37°C for 2 h. Incorporated label was assessed by scintillation counting.
Reverse Transcription-PCR Assays-cDNAs for each gene were prepared from TriZOL (Life Technologies, Inc.) extracted total RNAs. reverse transcription reactions were run on 2 g of total RNA. Following reverse transcription reactions, PCR reactions were run to the midpoint of each PCR fragment's linear synthesis curve. Glyceraldehyde-3-phos-* This work was supported in whole or in part with Federal funds from the NCI, National Institutes of Health, under Contract NO1-CO-56000 and sponsored in part by the NCI, Department of Health and Human Services, under a contract with SAIC. The content of this publication does not necessarily reflect the views or policies of the Department of Health and human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the United States Government. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

RESULTS AND DISCUSSION
We examined the effect of IL-6 treatment on methyltransferase activity by using K562 cells in a rested state in RPMI 1640 medium supplemented with 0.05% FBS for ϳ48 h. The cells were rinsed twice in serum-free RPMI 1640 and then treated with IL-6 (at 100 ng/ml). After 8-h incubation, the cells were harvested, and methylation activity assays were performed as described previously (14) to determine the relative levels of activity following IL-6 treatment. Lanes 1 and 2 in Fig.  1 show the results obtained from control reactions utilizing only cell lysates with no poly(dI-dC)⅐poly(dI-dC) substrate added and poly(dI-dC)⅐poly(dI-dC) substrate with no cell lysates added, respectively. Lane 3 represents the basal level of methylation activity obtained from rested K562 cells, while lane 4 shows a 3.2-fold increase in activity following treatment with IL-6. Based on these observations, treatment with IL-6 appears to increase overall methylation activity.
To determine whether treatment with IL-6 activates the hdnmt-1 promoter, we generated a series of deletion constructs as shown in Fig. 2A. The constructs were sequenced and used to transfect K562 cells, which were rested prior to stimulation with IL-6 as described above. Gradual deletion of increasing amounts of the wild-type promoter as shown in lane 1 (⌬MT1), lane 2 (⌬MT2) (Ϫ1214 to ϩ71 bp), lane 3 (⌬MT3) (-815 to ϩ71 bp), and lane 4 (⌬MT4) (Ϫ474 to ϩ71) did not abrogate the IL-6induced activity. The results shown in Fig. 2B, lane 5, indicate that IL-6-induced promoter activity is localized to the ⌬MT5 segment (Ϫ243 to ϩ71 bp), which encodes several potential ETS family recognition sites. Fig. 2C, lane 1, shows the results of transfecting wild-type ⌬MT5 and then stimulating with IL-6. Unstimulated wild-type ⌬MT5 reporter levels are represented in lane 2. Fig. 2C, lanes 3 and 4, show the activity levels of a ⌬MT5 triple-mutant reporter stimulated with IL-6 and unstimulated, respectively. The ETS site-mutated ⌬MT5 reporter shows a markedly suppressed response, as the loss of the three Fli-1 binding sites in ⌬MT5 abrogated the IL-6-mediated response.
To determine whether IL-6 induced increased expression of hdnmt-1 and Fli-1 mRNA, K562 cells were incubated in RPMI 1640 medium supplemented with 0.05% FBS for ϳ48 h. The cells were rinsed twice in serum-free RPMI 1640 and then treated with IL-6 at a concentration of 100 ng/ml. Treated cells were collected at 0, 1, 2, 4, 6, 8, 12, and 24 h post-treatment with IL-6. Total cellular RNA was prepared at each time point and stored at Ϫ70°C. Ultraviolet spectroscopy was used to quantitate equally each RNA sample, and final working dilutions for each time point were rechecked following initial dilution to a concentration of 100 ng/l. The adjusted total RNA preparations were then used to create cDNA, on which PCR reactions were performed. Using primers specific for hdnmt-1 and Fli-1, PCR was performed for each time point to determine the relative expression level of each gene. The temporal expression pattern of hdnmt-1 is shown in Fig. 3A. The expression of hdnmt-1 begins to appear at 6 h post-treatment, reaching a peak between 8 and 12 h. At 24 h, hdnmt-1 mRNA is still expressed, but the level is considerably diminished. The double-banded PCR product seen for the hdnmt-1 is due to an intronic insertion, and the PCR primers were chosen to amplify this region to determine whether both possible gene products are affected equally by IL-6 (15). In K562 cells, no difference in expression levels between the two possible hdnmt-1 mRNA products were noted.
The ETS family transcription factor, Fli-1, which is known to be expressed in megakaryocytic lineages as a mediator of differentiation (16), begins to be expressed at ϳ4 h post-treatment (Fig. 3B) and continues to increase throughout the sampling period. The expression pattern of the prototypic ETS family member, Ets-1, did not show a response to IL-6 when cDNA from the rested K562 cells was analyzed in parallel reactions with hdnmt-1 and Fli-1, and Ets-1 did not produce a discernable band when the reactions were analyzed at the midpoint of the amplification curve (data not shown.) Fig. 3C shows equivalent expression levels of the GAPDH gene cDNA control at each time point.
Analysis of the ⌬MT5 promoter elements reveals three potential Fli-1 binding sites at Ϫ194, Ϫ170, and Ϫ60 base pairs (17). A series of singular and multiple point mutations of the ⌬MT5 constructs, shown in Fig. 4a, were co-transfected into COS-1 cells with pSG5Fli-1 expression plasmid to determine the authenticity of each potential ETS binding site. Fig. 4b shows the strongest activation with all three potential Fli-1 binding sites left intact. The intact promoter construct These results provide evidence of a novel mechanism of IL-6 cytokine-mediated alteration, via the Fli-1 transcription factor, of methyltransferase gene expression. Previously, it was shown that IL-6 activation of the immediate-early gene junB occurred through an ETS family protein, in cooperation with a CREB-ATF factor (18). An analogous situation exists in fos-transformed cells, in which the expression of DNA methyltransferase is three times that of normal levels (19). Thus, it has been proposed that fos transformation is mediated by increased methyltranferase expression. Therefore, by analogy, even slight alterations in methyltransferase expression resulting from chronic exposure to IL-6 could, over time, result in abnormal patterns of cellular DNA methylation, similar to those caused by transformation of fos. Indeed, the importance of dnmt-1 activity in the establishment and propagation of neoplastic growth has emerged as an important diagnostic factor (20). Methylated CpG dinucleotides are susceptible to spontaneous deamination of 5-methylcytosine to uracil and are believed to be responsible for approximately one-third of C-T transition mutations found in human genetic diseases and tumors (21). Hypermethylation of tumor suppressor genes such as p53 (22), retinoblastoma (Rb) (23), and p16 ink (24) occur in many different tumors types, serving to promote tumor growth by rendering these genes inactive. The effects of promiscuous methylation of important tumor suppressor and cell cycle regulatory genes potentially resulting from prolonged exposure to FIG. 2. A, hdnmt-1 promoter constructs in pGL-3 Basic luciferase reporter vector. Methyltransferase I (HDNMT-1) promoter region spanning nucleotides 248 -1950 from the published complete genomic clone sequence. Nucleotides encoding restriction endonuclease sites for NheI and HindIII were included in the 5Ј ends of PCR primers used to generate various promoter deletion constructs. Primers used to construct promoter deletions are described under "Experimental Procedures." B, activation of hdnmt-1 promoter by IL-6 stimulation of K562 cells. Cells were transfected with 2 g of plasmid by Fugene-6 reagent and allowed to incubate overnight. Cells were then rinsed in PBS, pH 7.4, and resuspended in resting medium as described. After 48 h, IL-6 was added, and cells were harvested the following morning (16 h  inflammatory cytokines are probably cumulative in nature, remaining latent until sufficient insult to the cell permits it to transform into a neoplastic growth (25).
We demonstrate here that IL-6, an inflammatory cytokine capable of mediating cellular differentiation, is capable of increasing DNA methytransferase expression and activity. The data suggest that one of the normal molecular consequences of the biological activity of IL-6 may be mediated by DNA methyltransferase activity modifying gene expression. Additionally, IL-6 has been implicated in numerous cancer models, including multiple myeloma and prostate carcinoma (26,27). However, in a few cell lines, IL-6 has shown inhibitory effects on tumorigenesis (28) and anti-inflammatory capacity (29). Notwithstanding these pleiotropic characteristics, chronic exposure of cells to inflammatory cytokines such as IL-6 may have serious consequences by altering the normal levels, or time of expression of many genes, by up-regulating methyltransferase, whose expression is normally tightly controlled in a cell cycle-dependent manner (30,31). Dysregulation of DNA methyltransferase may result in the methylation of important tumor suppressor and cell cycle regulatory genes eventually initiating or enhancing neoplastic growth (32,33).