Identification of methylated CpG motifs as inhibitors of the immune stimulatory CpG motifs

Abstract

The unmethylated CpG motifs within E. coli DNA (EC) cause immune stimulation. In contrast, mammalian DNA such as calf thymus (CT) DNA had been thought to be immunologically inert. In this article, we demonstrate that CT DNA unexpectedly specifically inhibits the immune activation by EC but not that by endotoxin. This inhibitory effect was mediated in the signaling pathway activated by EC since CT DNA markedly inhibited the CpG-induced nuclear translocation of the transcription factors, NF-κB and AP-1. In addition, CT DNA significantly inhibited the synergistic immune activation by EC and endotoxin. The mechanism of the inhibition by CT DNA probably did not involve the inhibition of the cellular uptake of EC. Using a CpG-depleted plasmid, we demonstrated that CpG methylation played an important role in the inhibition by CT DNA. Compared with unmethylated plasmid DNA, CpG-methylated DNA inhibited the immune activation by EC to the same extent as did CT DNA. Importantly, the inhibitory effect of CT DNA was also observed in vivo. Our results suggest that methylated DNA may be applied to alleviate the unwanted immune stimulation and inflammation in systemic inflammatory response syndrome and in gene therapy with plasmid DNA.

Introduction

Bacterial DNA stimulates B cell activation and immunoglobulin secretion,1 activates natural killer (NK) cells, induces interferon γ (IFNγ) secretion and confers resistance to tumors in vivo.2,3 In contrast, vertebrate DNA had been considered to be immunologically inert. Bacterial genomes and vertebrate genomes are different in the frequency and methylation status of CpG dinucleotides. CpG dinucleotides in bacterial genomes generally occur at a frequency of about 1/16 as predicted by random base utilization and are unmethylated. In contrast, CpG dinucleotides in vertebrate DNA are suppressed to a frequency of only 1/50–60, almost all of which are methylated.4

It is believed that unmethylated CpGs in the context of certain sequences in bacterial DNAs are responsible for their immune stimulatory effects.5 For example, CpG dinucleotides are important in eliciting immune responses by DNA vaccines. If the CpG dinucleotides are deleted or methylated, the immune responses are significantly reduced.5,6,7 However, the same CpG dinucleotides within plasmids are responsible for inflammation in vivo.8 The administration of plasmid DNA complexed with liposome in vivo induces the production of a large amount of inflammatory cytokines, which can be responsible for toxic effects.9,10 Instillation of bacterial DNA intratracheally elicits inflammatory responses in the mouse lower respiratory tract.11 Injection of bacterial DNA containing CpG motifs into knee joints induced a transient arthritis in mice.12 Furthermore, bacterial DNA potentiated the toxicity of lipopolysaccharide (LPS) and might play a role in the systemic inflammatory response syndrome.13,14,15 The inflammatory responses elicited by CpG DNA are a major obstacle in the safe application of plasmid DNA to gene therapy.9,16,17,18

Although the immune stimulatory effects of CpG DNA have been extensively studied, there are only limited reports regarding how to inhibit the immune stimulation induced by CpG DNA. Recent reports showed that DNA sequences such as calf thymus (CT) DNA, human placenta DNA and synthetic phosphodiester or phosphorothioate oligonucleotides containing poly Gs were able to inhibit the immune activation by bacterial DNA.19,20 The inhibition of DNA uptake through macrophage scavenger receptor was responsible for their inhibitory effects. Dexamethasone and antimalarial compounds such as chloroquine were shown to be able to inhibit the immune activation by CpG DNA.17,18 In this article, we investigated whether certain elements of the immunologically inert vertebrate DNA could also inhibit the immune activation by bacterial DNA. We found that calf thymus DNA inhibited the immune activation by E. coli DNA (EC) and that CpG methylation played an important role in the inhibitory effect by blocking the activation of NF-κB and AP-1 without affecting the EC uptake.

Results

Inhibition of E.coli DNA-induced cytokine production by CT DNA

Bacterial DNA has been shown to possess immune stimulatory effects due to the high frequency of unmethylated CpG dinucleotides in the genome. By contrast, mammalian genomic DNA has been thought to be immunologically inert because it contains a lower frequency of CpG dinucleotides, most of which are methylated. To determine whether the presence of CT DNA would affect the immune stimulation induced by EC, we mixed CT DNA and EC at a 1:1 ratio and added them to spleen cell or RAW 264 cell cultures. After incubation for the indicated time, cytokines in the supernatants were determined by ELISA. Since IFNγ and TNFα induced by plasmid DNA were shown to inactivate gene expression and since neutralizing Abs against these cytokines prolonged the duration of gene expression,16 IFNγ and TNFα were used as indicators to show whether immune activation by EC was suppressed. As shown in Figure 1a, CT DNA can significantly inhibit IFNγ induction by EC. By contrast, the plasmid pGT, in which the CpG motifs were largely depleted by mutation (see Materials and methods section) but were not methylated, was not immune stimulatory by itself under these conditions, and only minimally inhibited IFNγ secretion. Similarly, CT DNA could inhibit TNFα secretion induced by EC while pGT did not under these experimental conditions (Figure 1b).

Figure 1
figure1

Calf thymus DNA inhibits the stimulatory effect of EC. Balb/c spleen cells (5 × 106/ml) (a) and RAW264 cells (6 × 104/ml) (b) were stimulated with EC (4 μg/ml) in the presence or absence of other different DNAs at 4 μg/ml. After 24-h (spleen cells) or 5-h (RAW264 cells) incubation at 37°C, the concentration of IFNγ or TNFα in the supernatant was measured by ELISA. CT DNA, calf thymus DNA; CT DNA (pstI), calf thymus DNA digested with restriction enzyme PstI; none, medium alone. The experiment was repeated at least three times with similar results.

Since CT DNA is a large molecule, it is possible that it can physically interfere with the interaction between spleen cells and the stimulatory EC. To explore this possibility, we digested CT DNA with restriction enzyme PstI into fragments of average size of about 4 kb and mixed the purified DNA mixture with EC. As shown in Figure 1, digested CT DNA is also inhibitory, suggesting that it is not simply the size of CT DNA that is responsible for the inhibition. In addition, the inhibition of EC by CT DNA or CT DNA digested by PstI was dose dependent as shown in Figure 2. Although the EC had levels of endotoxin too low to account for the observed stimulation of cytokine production, some experiments were performed using spleen cells from C3H/HeJ mice to exclude any role of LPS. When spleen cells were stimulated with increasing doses of EC, their secretion of IFNγ was correspondingly increased and the addition of a fixed 4 μg/ml of CT DNA is still partially inhibitory even when 32 μg/ml of EC were added (Table 1). In order to determine the structural requirements for inhibition, we tested other nucleic acids for any effect on EC-induced cytokine production. Interestingly, another vertebrate DNA, salmon sperm DNA exerted a similar inhibition on the immune activation by EC. In contrast, yeast tRNA and denatured CT DNA were not inhibitory (Table 2).

Figure 2
figure2

Inhibition by calf thymus DNA is dose dependent. Balb/c (a) or C3H/HeJ spleen cells (b) (5 × 106/ml) were stimulated with EC (4 μg/ml) in the presence of CT DNA or CT DNA digested with PstI at different concentrations. After 24-h incubation at 37°C, IFNγ concentration in the supernatant was measured with ELISA. CT, CT DNA; pstI, CT DNA digested with PstI; none, medium alone. The experiment was repeated three times with similar results.

Table 1 The inhibitory effect of CT DNA on the immune activation by EC
Table 2 Different effects of nucleic acids on the immune activation by EC

Failure of CT DNA to inhibit the uptake of stimulatory DNA

There is a possibility that CT DNA could inhibit the immune activation by EC through competitive inhibition of the uptake of EC by spleen cells, which is believed to be the first step in immune stimulation by EC.5 Indeed, previous studies have suggested that binding of oligos to the scavenger receptors is required for their immune stimulatory activities and that scavenger receptor competitors such as poly-Gs and fucoidan are inhibitory on the immune activation by EC and CpG ODN.19,20,21 Several experiments were performed to explore the possibility of uptake inhibition in more detail. First of all, EC and CT DNA were mixed at a 1:1 ratio starting from 4 μg/ml and the mixture was serially diluted. Spleen cells were incubated with the DNA mixture for 24 h and the IFNγ concentration in the supernatants was measured. CT DNA at concentration as low as 0.125 μg/ml was still inhibitory (Figure 3), suggesting that the mechanism of inhibition may not simply be the inhibition of the spleen cell uptake of EC by CT DNA. Secondly, when CT DNA was added to spleen cell culture half an hour after EC, inhibition was still observed although not as strong as when they were added simultaneously (Table 3). Thirdly, we injected mice with 2 μg (dissolved in 50 μl normal saline) pCMV-luc, a mammalian expression vector encoding luciferase in the presence or absence of CT DNA intramuscularly at a 1:1 ratio. Three days after injection, the quantity of luciferase expression in mouse muscle was determined. As shown in Figure 4, there was no difference in luciferase expression between mice injected with pCMV-luc alone or together with CT DNA, suggesting that CT DNA did not interfere with the uptake of plasmid pCMV-luc. To assess further whether CT DNA could inhibit the uptake of EC, we labeled EC with fluorescein-12-dUTP and incubated spleen cells with 4 μg/ml of fluorescein-labeled EC with or without 4 μg/ml of CT DNA. Two hours after incubation, the cells were analyzed with FACScan. As shown in Figure 5, the uptake of fluorescein-labeled EC was not affected by the presence of CT DNA.

Figure 3
figure3

Inhibition by calf thymus DNA is still effective at low concentration. Balb/c spleen cells (5 × 106/ml) were stimulated with serially diluted EC in the presence of calf thymus DNA at the same concentration to maintain a 1:1 ratio between the DNAs. After 24-h incubation at 37°C, IFNγ concentration in the supernatant was measured with ELISA. EC 4, EC at 4 μg/ml; EC2, EC at 2 μg/ml; EC1, EC at 1 μg/ml and so forth. EC+CT4, EC and calf thymus DNA both at 4 μg/ml; EC+CT2, EC and calf thymus DNA both at 2 μg/ml and so forth. The experiment was repeated twice with similar results.

Table 3 The inhibitory effect of CT DNA when added after EC
Figure 4
figure4

CT DNA does not affect gene expression by muscle cells. Luciferase assay was tested with TA muscles removed 3 days after injection of 2 μg pCMV-luc with or without 2 μg CT DNA as described in Materials and methods. Each group consists of 12 mice.

Figure 5
figure5

CT DNA does not affect the uptake of EC DNA. Balb/c spleen cells were incubated with fluorescein-labeled EC DNA (4 μg/ml) in the presence or absence of CT DNA (4 μg/ml) at 37°C for 2 h. The cells were then washed and analyzed with FACScan and Flowjo software.

The inhibition of CT DNA on the activation of NF-κB and AP-1 by E. coli DNA

Since the inhibitory effect by CT DNA was less likely to occur extracellularly, we next examined whether CT DNA could inhibit the immune activation by EC at the transcriptional level. Transcription factors NF-κB and AP-1 were recently shown to be activated by CpG DNA.22,23,24,25 The activation of these transcription factors is thought to mediate cytokine induction by CpG DNA. In order to understand the mechanism of the inhibitory activity of CT DNA on spleen cell stimulation by bacterial DNA, we determined the effect of CT DNA on the nuclear translocation of NF-κB and AP-1 in murine spleen cells. As expected, both LPS and EC induced the nuclear translocation of NF-κB and AP-1 6 h after stimulation. While CT DNA itself little affected the activation of NF-κB and AP-1, it inhibited the nuclear translocation of NF-κB and AP-1 induced by EC. In contrast, CT DNA did not or only minimally inhibited that induced by LPS (Figure 6). Competition experiments confirmed the specificity of NF-κB and AP-1 bands, while supershift experiments showed that CT-DNA preferentially affected p50/p65 heterodimer nuclear translocation and c-Jun/ c-Fos induction (data not shown).

Figure 6
figure6

CT DNA inhibits the activation of NF-κB and AP-1 by EC. Murine spleen cells were cultured in the presence or absence of different stimulators as described in Materials and methods. Before stimulation (T0) or 6 h after stimulation, nuclear extracts of spleen cells were mixed with 32P-labeled probes for NF-κB and AP-1 and run on polyacrylamide gels. The gels were dried and exposed to X-ray films.

The role of CpG methylation in mediating the inhibition of E. coli DNA

There are two major differences between EC and CT DNA. First, EC contains a higher frequency of CpG dinucleotides than CT DNA because of CpG suppression in the latter. Second, most CpGs in CT DNA are methylated. Since the unmethylated plasmid pGT also contains less CpGs but is not inhibitory (Figure 1), we hypothesized that CpG methylation might play a role in the inhibition of immune activation by EC. To investigate this possibility, we methylated pGT with SssI methylase and repeated the above inhibition experiment. As shown in Figure 7a, pGT itself could not efficiently inhibit the stimulatory effect of EC, whereas the CpG-methylated pGT significantly inhibited IFNγ secretion induced by EC. To demonstrate that the methylation at the specific C of CpG was responsible for the inhibition, we methylated pGT with MspI methylase, which methylated the external C of CCGG. As shown in Figure 7b, while CpG-methylated pGT was inhibitory, pGT methylated by MspI methylase was not, indicating that the inhibition requires specific CpG methylation and that the inhibition of the uptake of bacterial DNA may not be the mechanism for the inhibition of EC immunostimulation by CpG-methylated plasmid. In this experiment, spleen cells from C3H/HeJ mice were used to exclude any effect of LPS.

Figure 7
figure7

CpG methylation plays an important role in the inhibition of the stimulatory effect of EC. Balb/c spleen cells (a) or C3H/HeJ spleen cells (b) (5 × 106/ml) were stimulated with 4 μg/ml EC in the presence of other different DNAs at 4 μg/ml. After 24-h incubation at 37°C, IFNγ concentration in the supernatant was measured by ELISA. Me-pGT, pGT methylated by SssI methylase; pGT (MspI), pGT methylated with MspI methylase; none, medium alone. The experiment was repeated three times with similar results.

The inhibition of CT DNA on the synergistic immune stimulation by LPS and EC

During gram-negative bacterial infection, both LPS and bacterial DNA may be released and can affect the host immune system. We found that there was a synergy between EC and LPS in stimulating the immune response (Figure 8). Since CT DNA was shown to be able to inhibit the stimulatory effect by EC, we were interested to know whether CT DNA could also inhibit the synergy. As shown in Figure 8, CT DNA inhibited the immune stimulation by EC but not that by LPS. In addition, CT DNA could also abolish the synergy between LPS and EC.

Figure 8
figure8

Calf thymus DNA inhibits the synergistic effect by EC and LPS. Balb/c spleen cells (5 × 106/ml) were stimulated with EC (4 μg/ml) in the presence or absence of CT DNA (4 μg/ml) or LPS (80 ng/ml). After 24-h incubation at 37°C, IFNγ concentration in the supernatant was measured by ELISA. CT DNA, calf thymus DNA; none, medium alone.

The inhibitory effect of CT DNA on the immune stimulation by EC in vivo

To determine whether CT DNA could also inhibit the stimulatory effect of EC in vivo, we injected i.p. EC with or without CT DNA into Balb/c mice and measured cytokine secretion in the blood. Since a single i.p. injection of a mouse with 100 μg EC was not able to induce a significant increase in TNFα production, we injected mice two times at a 4-h interval. One hour after the last injection, mice were bled and TNFα concentration in the sera was determined by ELISA. As shown in Figure 9, the injection of EC alone stimulated strong release of TNFα in the circulation. The addition of CT DNA significantly inhibited TNFα secretion (Student's t test, P < 0.01).

Figure 9
figure9

Calf thymus DNA inhibits the stimulatory effect of EC in vivo. Balb/c mice (three in one group) were injected i.p. with EC (100 μg per mouse), EC (100 μg per mouse) plus calf thymus DNA (100 μg per mouse) or PBS. Four hours later, the same injections were repeated as before. One hour after the last injection, mice were bled and the concentration of TNFα in the serum was measured by ELISA. The experiment was repeated three times with similar results. The difference between the concentrations of TNFα in the sera of mice treated with or without CT DNA is statistically significant (Student's t test, P < 0.01).

Discussion

CpG stimulatory motifs are reported to play an important role in the immunogenicity of DNA vaccines.6,7 However, when plasmid DNA is exploited in gene therapy, the immune activation by these motifs can give rise to serious adverse effects in the host. For example, the presence of CpG motifs in gene therapy vectors can elicit cytokine secretion in vivo.26 When plasmid DNA complexed with liposomes was delivered into the lung or intravenously, it induced an influx of leukocytes and proinflammatory cytokines such as IFNγ, IL-12 and TNFα. These inflammatory responses reportedly resulted in the toxic effects.9,16 The cytokines induced by plasmid DNA may also be responsible for the inhibition of gene expression. It has been reported that TNFα and IFNγ inhibit CMV promoter activity and thus may attenuate gene transcription.27,28,29,30 For example, TNFα can markedly inhibit luciferase expression in mouse lung endothelial cells.16,17 Administration of antibodies against IFNγ and TNFα enhanced and prolonged gene expression in vivo.16 These studies demonstrated the desirability of finding new ways to reduce or eliminate CpG-induced immune stimulation. Currently, there are several possible ways to inhibit the immune activation induced by plasmid DNA. For example, methylation of CpGs in a gene therapy vector can abolish immune responses but it also severely reduces the promoter activity and thus decreases gene expression.8,31 In addition, even when the protein-coding sequences are methylated with the promoter intact, the expression can still be significantly impaired.32 Mutation of CpG motifs can reduce the immune activation by plasmid DNA to about 50%.18 However, not all of the 200 or more CpG motifs in a typical plasmid can be mutated without deleterious effects: changing the sequence close to the plasmid replication origin can make plasmid replication inside bacteria far less efficient.18 This would hamper the large-scale production of plasmid DNA. Unfortunately, insertion of the previously described neutralizing motifs into plasmid DNA was not potent enough to inhibit the strong stimulation by plasmid DNA.4,19,33 Interestingly, the administration of antimalarial drugs or dexamethasone can also significantly inhibit the immune activation by plasmid DNA and enhance gene expression.17,18 However, since the adverse effects of these drugs remain a concern in clinical application, it is desirable to find additional inhibitory elements to suppress the immune activation and enhance gene expression.

In this article, we focused on a new avenue to inhibit cytokine production induced by EC. Our data suggest that CpG methylation is responsible for the inhibitory effects of CT DNA. This inhibition is CpG specific. Only when CpG is methylated can the plasmid DNA exert its inhibition on the immune stimulation by EC, whereas the methylation of the external C of CCGG had little effect. In addition, CT DNA, while inhibitory on the immune activation by bacterial DNA, had no effect on that by LPS, suggesting the specific inhibition of bacterial DNA. Importantly, CT DNA was also effective in inhibiting immune activation by EC in vivo (Figure 9). Our results indicate that vertebrate DNA may have therapeutic effects in the systemic inflammatory response syndrome where the inhibition of the immune activation by bacterial DNA can abolish its synergy with LPS.

Our experiments demonstrated that salmon sperm DNA exerted a similar inhibitory effect to CT DNA (Table 2), suggesting that the inhibitory effect may be shared by other vertebrate DNAs.

The mechanism of the inhibition of bacterial DNA by CT DNA is not clear but presumably differs from the previously reported inhibitory effects of certain viral DNAs since those were unmethylated.4 CT DNA was shown to inhibit the activation of NF-κB and AP-1 by EC but not that by LPS (Figure 6), and these activated transcription factors are implicated in the induction of cytokines by CpG DNA.22,25 It could be that the methylated DNA specifically targets the signaling pathways activated by CpG DNA. The activation of NF-κB by CpG DNA occurred through the generation of reactive oxygen species,22 while that of AP-1 happened via the activation of the p38 and c-Jun NH2-terminal kinase MAPK.23 Interestingly, inhibition was highly dependent on the double-stranded DNA structure since neither single-stranded DNA nor RNA was inhibitory (Table 2). Further study is required to elucidate how CT DNA inhibits the immune stimulation signaling pathways upstream of the transcriptional factors AP-1 and NF-κB.

Four lines of evidence argue that the competitive inhibition of the uptake of bacterial DNA probably is not the mechanism by which CT DNA inhibits the immune activation by EC. First, serial dilution of both CT DNA and bacterial DNA did not affect the inhibitory effects of CT DNA (Figure 3). Second, compared with the inhibitory CpG-methylated pGT, the MspI-methylated pGT was unable to inhibit the immune activation of bacterial DNA (Figure 7). Third, when CT DNA was co-injected with pCMV-luc i.m. into mouse, the luciferase expression by muscle cells was minimally affected, suggesting that CT DNA did not inhibit the cellular uptake of pCMV-luc and luciferase expression (Figure 4). Finally, CT DNA did not affect the uptake of fluorescein-labeled EC as analyzed by flow cytometry (Figure 5).

We believe that the inability to inhibit the uptake of plasmid DNA and the strong inhibitory effect on the immune activation by CpG DNA make the vertebrate DNA or CpG-methylated DNA a candidate for the suppression of immune stimulation by gene therapy vectors. The co-administration of vertebrate DNA or CpG-methylated DNA with gene therapy vectors may greatly dampen the inflammatory responses elicited by the plasmid without affecting the gene expression. Further experiments will be required to determine the in vivo utility of CpG-methylated DNA for improving the safety and efficacy of gene therapy.

Materials and methods

Animals and in vivo procedures

Female Balb/c and C3H/HeJ mice of 6 to 12 weeks old were purchased from Jackson Laboratories (Bar Harbor, ME, USA). All mice were held under specific pathogen-free conditions at all times. For i.p. injection, the DNAs were dissolved in 200 μl PBS. Balb/c mice were injected i.p. with 100 μg EC in the presence or absence of 100 μg CT DNA. Four hours later, mice were injected with the same reagents. One hour later, each mouse was bled under anesthetization with 0.3 ml Avertin. The blood was put on ice for 1 h and centrifuged at 10 000 g for 10 min to pellet the red blood cells. For i.m. injection, the DNAs were dissolved in 50 μl normal saline. Balb/c mice were injected directly into tibialis anterior (TA) muscles with 2 μg pCMV-luc in the presence or absence of 2 μg CT DNA.

Cell lines and spleen cells

RAW264 cells and murine spleen cells were grown in RPMI medium consisting of 10% fetal calf serum (Hyclone), 2 mM L-glutamine (GIBCO, BRL, Life Technologies), 2 × 10−5M 2-mercaptoethanol (Kodak, Rochester, NY, USA), 100 U/ml penicillin G (Sigma Chemical, St Louis, MO, USA) and 100 μg/ml streptomycin sulfate (Sigma Chemical). Spleen cells were stimulated with EC at indicated concentrations in the presence or absence of other different DNAs for 24 h at 37°C in a 5% CO2 atmosphere. In some experiments, C3H/HeJ spleen cells were used to exclude the possible interference from LPS. Our experiments have shown that bacterial DNA elicits similar immune activation in both C3H/HeJ and Balb/c mice.

DNAs and reagents

E. coli (strain B) DNA, LPS (from Salmonella typhimurium), salmon sperm DNA, yeast tRNA and CT DNA were purchased from Sigma Chemical. Plasmid pGT was derived from plasmid pUK21,4 in which over 70 CpG were mutated and was thus less stimulatory. All DNAs were purified with phenol/chloroform. Plasmid pCMV-luc encoding firefly luciferase was described before.34 pCMV-luc was purified on Qiagen anion-exchange chromatography columns and dissolved in PBS (Qiagen, Chatsworth, CA, USA). The endotoxin level in the DNAs was less than 1.7 ng/mg as assayed with Limulus Amebocyte Lysate QCL-1000 (BioWhittaker, Walkersville, MD, USA). Gey's solution consists of 0.0005% Phenol Red Solution (GIBCO, BRL, Life Technologies), 0.83% ammonium chloride (Sigma Chemical) and 0.1% potassium bicarbonate (Sigma Chemical) in water. Avertin consists of 2% 2,2,2-tribromoethanol (Aldrich Chemical Company, Milwaukee, WI, USA) and 2% tert-amyl alcohol (Aldrich Chemical Company) in water. Cell lysis buffer consists of 10 mM Hepes (pH 7.5), 10 mM KCl, 0.1 mM EDTA, and 0.1 mM EGTA. Nuclear extraction buffer consists of 20 mM Hepes (pH 7.5), 50 mM NaF, 440 mM NaCl, 20% glycerol, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 1 mM PMSF, 30 mg/ml leupeptin, 50 mg/ml aprotinin, 5 mg/ml antipain and 5 mg/ml pepstatin.

Labeling of E. coli DNA

E. coli DNA was boiled for 10 min and quickly chilled in ice for 5 min. The denatured DNA was then labeled with fluorescein-12-dUTP using BioProbe Random Primed DNA Labeling System (Enzo Diagnostics, Farmingdale, NY, USA). Briefly, the denatured EC DNA, hexanucleotide primers, deoxynucleotide mixture containing fluorescein-12-dUTP and Klenow DNA polymerase were mixed in the final volume of 50 μl distilled water, and incubated at 37°C in a water bath for 16 h. The fluorescein-labeled EC DNA was purified as following. The reaction mixture (200 μl) was mixed with 40 μl 4 M LiCl, followed by 800 μl prechilled pure ethanol. After incubation at −70°C for 30 min, DNA was spun down at 10 000 r.p.m. in a tabletop centrifuge. The DNA pellet was washed with pure ethanol, dried at room temperature and resuspended in TE buffer.

Generation and methylation of plasmid pGT

Vector pUK21, which contains a ColE1 replication region, a kanamycin resistance gene and a polylinker, was provided by Dr Martin Schleef at Qiagen (Qiagen, Hilden, Germany).35 Plasmid pcDNA3 was purchased from Invitrogen (Carlsbad, CA, USA). The CMV promoter (from pcDNA3 position 209 to 863) was amplified by PCR and inserted into the pUK21 polylinker between XbaI and NotI sites. BGH polyA (from pcDNA3 position 1018 to 1249) was amplified by PCR and and ligated between XhoI and StuI sites of pUK21. Site-directed mutagenesis was subsequently performed on the vector by overlapping extension PCR as described by Ge and Rudolph.36 Overall, 75 point mutations were generated to result in vector pGT. The sequence of pGT is available upon request. Plasmid pGT was methylated as described previously.5 Briefly, plasmid pGT, together with SssI or MspI methylase (2 U/μg DNA) and S-adenosylmethionine (New England Biolab) were incubated at 37°C in a water bath for 4 h. S-adenosylmethionine was added again and the mixture was incubated for another 4 h. The completeness of methylation was confirmed by digestion with HpaII and MspI. The SssI-methylated DNA was completely cut by MspI but not by HpaII. Plasmid pGT methylated by MspI methylase could not be cut by MspI. The methylated DNA was purified by phenol–chloroform extraction.

EMSA

Murine spleen cells were cultured for 6 h with either 4 μg/ml EC, 4 μg/ml calf thymus DNA, 1 μg/ml LPS, 4 μg/ml EC + 4 μg/ml calf thymus DNA or 1 μg/ml LPS + 4 μg/ml calf thymus DNA. Cells were then resuspended in a hypotonic buffer for 15 min on ice followed by addition of NP-40 detergent (0.1% final concentration) and centrifugation for 5 min at 1000 g. The nuclear pellet was resuspended in the extraction buffer and centrifuged again for 10 min at 14000 g. Protein concentration in the supernatants was measured by Bradford's protein assay (Bio-Rad, Hercules, CA, USA) and samples were stored at −70°C before use. For EMSA, 32P-end-labeled specific dsDNA-NF-κB probe (5′-AGTTGAGGGGACTTTCCC AGG-3′, sc-2505; Santa Cruz Biotechnologies, Santa Cruz, CA, USA) or AP-1 probe (5′-CGCTTGATGACTCAGCCG GAA-3′, sc-2501 (Santa Cruz Biotechnologies) were preincubated with 1 μg of nuclear extracts for 30 min at room temperature in a final volume of 10 μl of the binding buffer (5% glycerol, 10 mM Tris-HCl pH 7.5, 50 mM KCl, 1 mM EDTA, 0.25% NP-40, 1 mM DTT, 100 μg/ml poly (dI-dC)). Mixtures were loaded on 6% native polyacrylamide gels and electrophoresed in 0.5 × TBE buffer at 150 V for 50 min. Gels were dried and exposed to X-ray films for 6–12 h at −70°C with intensifying screens (autoradiography). Specificity of the NF-κB and AP-1 bands was confirmed by competition experiments with 100-fold excess of the cold dsDNA oligonucleotides.

ELISA for IFNγ and TNFα

IFNγ and TNFα detection was done according to the Cytokine ELISA protocol from PharMingen (San Diego, CA, USA). Briefly, flat-bottom Immulon 1B plates (Dynatech Laboratories, Chantilly, VA, USA) were coated with 2 μg/ml rabbit anti-mouse monoclonal Ab overnight at 4°C. The plates were washed with T-PBS (0.05% Tween in PBS) and blocked with 10% fetal calf serum in PBS. Mouse sera or supernatants from cell cultures were added to the plates and incubated at 4°C overnight. After washing with T-PBS, 1 μg/ml of biotinylated anti-mouse IFNγ or TNFα Ab was added and incubated at room temperature for 1 h. The plates were washed again and 2.5 μg/ml of avidin peroxidase (Sigma Chemical) was added and incubated for 30 min. After washing with T-PBS, 100 μl of tetramethylbenzidine substrate (Sigma) was added and the plates were read at 650 nm on a microplate reader (Cambridge Technology, Watertown, MA, USA) after development for 30 min at room temperature.

Flow cytometry

Balb/c spleen cells (5 × 105) were incubated with fluorescein-labeled EC DNA at 4 μg/ml in the presence or absence of CT DNA at the same concentration in 37°C incubator. Two hours after incubation, the cells were washed with PBS three times and resuspended in PBS and analyzed on a FACScan (Becton Dickinson, Mountain View, CA, USA).

Luciferase assay

Plasmid pCMV-luc (2 μg) with or without CT DNA (2 μg) was dissolved in 50 μl normal saline and was injected i.m. into Balb/c mouse TA. Mouse TA muscles were excised 3 days after i.m. injection. Luciferase activity was measured using the Promega luciferase assay system (Promega, Madison, WI, USA) as described previously.26 Data were plotted as group means + s.e.m. of relative light units (RLU)/s/mg protein, with protein contents determined by the BioRad microassay procedure (BioRad, Mississauga, ON, Canada).

References

  1. 1

    Messina JP, Gilkeson GS, Pisetsky DS . Stimulation of in vitro murine lymphocyte proliferation by bacterial DNA J Immunol 1991 147: 1759–1764

  2. 2

    Tokunaga T et al. Antitumor activity of deoxyribonucleic acid fraction from Mycobacterium bovis BCG. I. Isolation, physicochemical characterization, and antitumor activity J Natl Cancer Inst 1984 72: 955–962

  3. 3

    Shimada S et al. Antitumor activity of the DNA fraction from Mycobacterium bovis BCG. II. Effects on various syngeneic mouse tumors J Natl Cancer Inst 1985 74: 681–688

  4. 4

    Krieg AM et al. Sequence motifs in adenoviral DNA block immune activation by stimulatory CpG motifs Proc Natl Acad Sci USA 1998 95: 12631–12636

  5. 5

    Krieg AM et al. CpG motifs in bacterial DNA trigger direct B-cell activation Nature 1995 374: 546–549

  6. 6

    Sato Y et al. Immunostimulatory DNA sequences necessary for effective intradermal gene immunization Science 1996 273: 352–354

  7. 7

    Klinman DM, Yamshchikov G, Ishigatsubo Y . Contribution of CpG motifs to the immunogenicity of DNA vaccines J Immunol 1997 158: 3635–3639

  8. 8

    Zabner J et al. Comparison of DNA–lipid complexes and DNA alone for gene transfer to cystic fibrosis airway epithelia in vivo J Clin Invest 1997 100: 1529–1537

  9. 9

    Freimark BD et al. Cationic lipids enhance cytokine and cell influx levels in the lung following administration of plasmid: cationic lipid complexes J Immunol 1998 160: 4580–4586

  10. 10

    Yew NS et al. Contribution of plasmid DNA to inflammation in the lung after administration of cationic lipid:pDNA complexes Hum Gene Ther 1999 10: 223–234

  11. 11

    Schwartz DA et al. CpG motifs in bacterial DNA cause inflammation in the lower respiratory tract J Clin Invest 1997 100: 68–73

  12. 12

    Deng GM et al. Intra-articularly localized bacterial DNA containing CpG motifs induces arthritis Nat Med 1999 5: 702–705

  13. 13

    Cowdery JS, Chace JH, Yi AK, Krieg AM . Bacterial DNA induces NK cells to produce IFN-gamma in vivo and increases the toxicity of lipopolysaccharides J Immunol 1996 156: 4570–4575

  14. 14

    Sparwasser T et al. Bacterial DNA causes septic shock Nature 1997 386: 336–337

  15. 15

    Gao JJ et al. Cutting edge: bacterial DNA and LPS act in synergy in inducing nitric oxide production in RAW 264.7 macrophages J Immunol 1999 163: 4095–4099

  16. 16

    Li S et al. Effect of immune response on gene transfer to the lung via systemic administration of cationic lipidic vectors Am J Physiol 1999 276: L796–L804

  17. 17

    Tan Y, Li S, Pitt BR, Huang L . The inhibitory role of CpG immunostimulatory motifs in cationic lipid vector-mediated transgene expression in vivo (see comments) Hum Gene Ther 1999 10: 2153–2161

  18. 18

    Yew NS et al. Reduced inflammatory response to plasmid DNA vectors by elimination and inhibition of immunostimulatory CpG motifs Mol Ther 2000 1: 255–262

  19. 19

    Wloch MK, Pasquini S, Ertl HC, Pisetsky DS . The influence of DNA sequence on the immunostimulatory properties of plasmid DNA vectors Hum Gene Ther 1998 9: 1439–1447

  20. 20

    Pisetsky DS, Reich CF . Inhibition of murine macrophage IL-12 production by natural and synthetic DNA Clin Immunol 2000 96: 198–204

  21. 21

    Kimura Y et al. Binding of oligoguanylate to scavenger receptors is required for oligonucleotides to augment NK cell activity and induce IFN J Biochem (Tokyo) 1994 116: 991–994

  22. 22

    Yi AK et al. CpG motifs in bacterial DNA activate leukocytes through the pH-dependent generation of reactive oxygen species J Immunol 1998 160: 4755–4761

  23. 23

    Yi AK, Krieg AM . Rapid induction of mitogen-activated protein kinases by immune stimulatory CpG DNA J Immunol 1998 161: 4493–4497

  24. 24

    Sparwasser T et al. Macrophages sense pathogens via DNA motifs: induction of tumor necrosis factor-alpha-mediated shock Eur J Immunol 1997 27: 1671–1679

  25. 25

    Stacey KJ, Sweet MJ, Hume DA . Macrophages ingest and are activated by bacterial DNA J Immunol 1996 157: 2116–2122

  26. 26

    Raz E et al. Preferential induction of a Th1 immune response and inhibition of specific IgE antibody formation by plasmid DNA immunization Proc Natl Acad Sci USA 1996 93: 5141–5145

  27. 27

    Gribaudo G et al. Interferons inhibit onset of murine cytomegalovirus immediate–early gene transcription Virology 1993 197: 303–311

  28. 28

    Ritter T et al. Stimulatory and inhibitory action of cytokines on the regulation of hCMV-IE promoter activity in human endothelial cells Cytokine 2000 12: 1163–1170

  29. 29

    Qin L et al. Promoter attenuation in gene therapy: interferon-gamma and tumor necrosis factor-alpha inhibit transgene expression Hum Gene Ther 1997 8: 2019–2029

  30. 30

    Harms JS, Splitter GA . Interferon-gamma inhibits transgene expression driven by SV40 or CMV promoters but augments expression driven by the mammalian MHC I promoter Hum Gene Ther 1995 6: 1291–1297

  31. 31

    Hug M et al. Transcriptional repression by methylation: cooperativity between a CpG cluster in the promoter and remote CpG-rich regions FEBS Lett 1996 379: 251–254

  32. 32

    Parker SE et al. Cancer gene therapy using plasmid DNA: safety evaluation in rodents and non-human primates (see comments) Hum Gene Ther 1995 6: 575–590

  33. 33

    Zhao H, Cheng SH, Yew NS . Requirements for effective inhibition of immunostimulatory CpG motifs by neutralizing motifs Antisense Nucleic Acid Drug Dev 2000 10: 381–389

  34. 34

    Davis HL, Millan CL, Watkins SC . Immune-mediated destruction of transfected muscle fibers after direct gene transfer with antigen-expressing plasmid DNA Gene Therapy 1997 4: 181–188

  35. 35

    Vieira J, Messing J . New pUC-derived cloning vectors with different selectable markers and DNA replication origins Gene 1991 100: 189–194

  36. 36

    Ge L, Rudolph P . Simultaneous introduction of multiple mutations using overlap extension PCR Biotechniques 1997 22: 28–30

Download references

Acknowledgements

We thank Martin Schleef (Qiagen, Hilden Germany) for providing the pUK21 vector. AM Krieg is supported by a Career Development Award from the Department of Veterans Affairs and the National Institutes of Health (DK25295, DK54759, and CA66570), DARPA, and the Cystic Fibrosis Foundation. Additional support was provided by the Coley Pharmaceutical Group, Wellesley, MA.

Author information

Correspondence to AM Krieg.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Chen, Y., Lenert, P., Weeratna, R. et al. Identification of methylated CpG motifs as inhibitors of the immune stimulatory CpG motifs. Gene Ther 8, 1024–1032 (2001). https://doi.org/10.1038/sj.gt.3301482

Download citation

Keywords

  • gene therapy
  • CpG
  • endotoxin
  • immune stimulation
  • bacterial DNA
  • interferon γ

Further reading