Abstract
Challenge of macrophages with DNA containing an internal CpG motif results in IL-12 p40 secretion. In the presence of IFN-γ, CpG DNA induces more p40 secretion than does LPS. In the RAW 264 macrophage cell line, both CpG DNA and LPS activate a p40 promoter-reporter construct, and the promoter response to either agent is augmented 2- to 5-fold by IFN-γ. While either LPS or CpG DNA induces p40 promoter activity, only CpG DNA induces an increase in p40 mRNA or protein secretion. Even though IFN-γ augmented LPS-driven p40 promoter activity in RAW 264 cells, the combination of IFN-γ and LPS induced less p40 mRNA or protein secretion than the combination of IFN-γ and CpG DNA. The ability of IFN-γ to augment LPS or CpG DNA-induced p40 promoter activation was observed with truncation mutants of the IL-12 promoter containing as few as 250 bp 5′ of the TATA box. Although LPS alone is a poor inducer of p40 transcription, both LPS and CpG DNA induce similar nuclear translocation of NF-κB. This binding is not augmented by costimulation with IFN-γ. Thus, CpG DNA induces p40 transcription by a mechanism that includes NF-κB translocation; however, CpG DNA appears to induce other factor(s) necessary for p40 transcription. These results illustrate fundamental differences between CpG DNA and LPS with respect to activation of IL-12 p40 secretion.
Interleukin-12 is a heterodimeric, proinflammatory cytokine that induces the production of IFN-γ, which, in turn, drives the production of a number of inflammatory cytokines. Additionally, IL-12-induced IFN-γ can direct activated T lymphocytes to differentiate into type 1 helper cells (reviewed in Refs. 1 and 2). Adequate production of IL-12 is essential for the maintenance of normal host defense mechanisms (especially against intracellular pathogens); however, excessive production of IL-12 has been associated with deleterious inflammation (3, 4, 5, 6, 7). Although the IL-12 p35 chain is produced by a number of cell types, secretion of the biologically active p70 heterodimeric is controlled at the level of p40 chain transcription (8, 9). Secretion of p40 or p70 is limited to cells of the macrophage/monocyte lineage and occurs only after activation of these cells. A number of stimuli, including microbial products and anti-CD40, induce IL-12 p70 production (10, 11, 12). A major element in the control of p40 transcription is a NF-κB half-site located at bp −132 to −122 5′ of the TATA box. Disruption of this site abolishes the activity of p40 reporter constructs (13). The T cell/NK cell cytokine IFN-γ (which is induced by IL-12) is a powerful costimulator of LPS-induced p40 transcription (8, 13). The critical role of IFN-γ is underscored by the finding that IFN consensus sequence-binding protein (ICSBP)3-deficient mice are incapable of producing p40 following challenge with virus, intracellular parasites, or LPS (14, 15). The mechanism by which IFN-γ augments p40 transcription is unclear, since the mouse IL-12 promoter does not contain a recognized IFN-γ-activated sequence or an IFN-sensitive response element.
Microbial DNA or synthetic oligonucleotides containing internal CpG motifs function as potent inducers of the inflammatory response (reviewed in Refs. 16 and 17). Recent studies have identified the cells’ response to CpG DNA and the cytokines that are induced; however, less is understood about the molecular mechanisms by which CpG DNA activates cells (18, 19, 20, 21). In earlier studies, we found that CpG DNA is a potent inducer of macrophage IL-12 secretion, and we found that IL-10 inhibits CpG-induced p40 secretion (21, 22). Although a receptor for CpG has not been identified, cells responding to CpG DNA initiate synthesis of reactive oxygen species and translocate NF-κB to the nucleus (23, 24). Here, we show that while both LPS and CpG DNA induce transcription of IL-12 promoter-reporter constructs, and both stimuli are associated with protein binding to the NF-κB half-site, CpG DNA alone (unlike LPS) induces an increase in p40 mRNA in a macrophage cell line. Additionally, the combination of IFN-γ and CpG DNA results in greater p40 mRNA levels and secreted protein than is seen after challenge with CpG DNA alone or with LPS/LPS + IFN-γ. These observations illustrate both similarities and differences between CpG DNA and LPS-induced activation of IL-12 p40 transcription.
Materials and Methods
Mice and cell cultures
BALB/c mice were purchased from Harlan Sprague-Dawley (Indianapolis, IN) and were maintained in a conventional mouse facility with cage-top filters. The RAW 264.7 mouse macrophage cell line was purchased from the American Type Culture Collection (Manassas, VA), and WEHI 231 cells were the gift of Dr. Gail Bishop (University of Iowa, Iowa City, IA). Cell lines were maintained in RPMI 1640 supplemented with penicillin, streptomycin, l-glutamine, and 10% FBS. Primary cultures of bone marrow-derived macrophages were prepared by flushing mouse femurs and iliac wings with cold media. Cells were cultured at an initial concentration nucleated cells/ml in complete RPMI supplemented with 15% conditioned mouse L cell medium, which served as a source of M-CSF and GM-CSF (25). After 4 days of culture, plates were washed twice with warm medium, and adherent cells were removed with a cell lifter. Cell surface staining revealed the recovered cells were >95% Mac1 bright.
In vitro cultures
Bone marrow-derived macrophage, or RAW 264 cells (5 × 105/ml) were cultured in 0.2 ml in 96-well plates (Costar, Corning, NY) of complete RPMI in the presence of various stimuli. Supernatant fluids were harvested at 24 h, and IL-12 concentration was determined by ELISA. Cultures were stimulated with LPS (Escherichia coli 0127:B8; Sigma, St. Louis, MO), the CpG-containing, phosphorothioate backbone oligonucleotide designated 1826 (5′-TTCATGACGTTCCTGACGTT-3′) or a control phosphorothioate backbone oligonucleotide (designated 1982) in which the internal CpG motif is reversed (5′-TCCAGGACTTCTCTCAGGTT-3′). These two oligonucleotides have been shown in previous studies (20, 22
IL-12 p40 ELISA
Measurement of IL-12 p40 promoter activity
A segment of the IL-12 promoter (−703 to +54) coupled to a luciferase-encoding plasmid was the gift of Dr. Kenneth Murphy (Washington University, St. Louis, MO). This construct was used as a PCR template for the generation of the following truncations: −697, −543, −403, −250, and −100 to +53 (all numbers refer to position relative to the TATA box). The PCR products were 5′ and 3′ tailed with BamHI and HindIII restriction sites to allow directional cloning into the multiple cloning site of the luciferase reporter plasmid Luc Link. After confirmation of orientation and sequence, plasmid DNA was isolated from transformed E. coli and was purified by CsCl2 centrifugation. To measure promoter activity, we used transient transfection of RAW 264 cells or WEHI 231 cells (electroporation at 260 MV/960 mFD). Cells were cotransfected with 20 μg of the p40-luciferase constructs and 5 mg of a plasmid containing β-galactosidase (β-gal) under control of a constitutive CMV promoter. In WEHI 231 cells, we also transfected with a reporter construct in which luciferase was driven by the NF-κB-containing promoter from the MHC class II invariant chain (26). After electroporation, 106 cells were cultured overnight in a volume of 5 ml in 6-well tissue culture plates (Costar). Cells were then treated with the indicated stimulus and were harvested 24 h later into 100 ml of harvest buffer (100 mM KPO4 (pH 7.8), 1.0 mM DTT, 1% Triton X-100). Lysates were mixed with 100 ml of assay buffer (200 mM KPO4 (ph 7.8), 10 mM ATP, 20 mM MgCl2), followed by 100 ml of 1 mM luciferin. Luciferase activity was measured as light units using a Moonlight 2010 luminometer (Analytical Luminescence Laboratory, San Diego, CA). β-gal expression was assayed using an assay kit purchased from Tropix (Bedford, MA). In all groups, the relative expression of luciferase light units by transfected cells represents: light units experimental/light units unstimulated. The transfected but unstimulated cells showed luciferase activity, which was consistently <4× the machine background of 100 U. For each experimental group, luciferase expression was corrected for the simultaneous expression of β-gal by multiplying the fold increase in luciferase by the fraction: β-gal stimulated/β-gal control. This calculation corrects for variation in cell number or ability to synthesize protein.
p40 mRNA measurement
Total RNA was isolated from unstimulated or stimulated (4 h) RAW 264 cells using RNA stat (Tel-Test “B” Inc., Friendswood, TX), according to the manufacturer’s recommendation. Ten micrograms of RNA were separated on a denaturing 1.5% agarose gel (MOPS running buffer with 3% formaldehyde). Following electrophoretic separation, RNA was transferred and UV-linked to nitrocellulose paper, then sequentially blotted with 32P-labeled probes for p40 and GAPDH. The p40 cDNA was prepared by PCR amplification of a 610-bp section (bases 22–632) of the p40 cDNA. Following isolation of total RNA, cDNA was synthesized as described previously (27). The PCR amplification was accomplished with 100 pM of sense (5′-GCAGCAAAGCAAGATGTGTCC-3′) and antisense (5′-CAGTTGGGCAGGTGACATCC-3′) primers. Each primer included a BamHI site tailed to the 5′ to permit excision of the segment from the multiple cloning site of the pGEM T vector system (Promega, Madison, WI), which was used to propagate the PCR product. The identity of the p40 cDNA was confirmed by sequence. The GAPDH control cDNA was a gift of Dr. Arthur Krieg (University of Iowa).
EMSA
After 1 or 3 h of stimulation, nuclear extracts were prepared as described previously (28), and 5 mg of extract was reacted with a 32P-labeled double-stranded probe derived from the IL-12 promoter containing both the NF-κB half-site and the 5′ Pu.1 site (5′-GGGGAGGGAGGAACTTCTTAAAATTCCCCCAGAATGTTT-3′) or a labeled probe in which the sequence of the Pu.1 site was replaced with random nucleotides (5′ATGTTTACTAGACAAAATTCCCCCAGAATGTTT). Other experiments used a labeled 27-bp oligonucleotide containing a consensus NF-κB site from the mouse IgH promoter (5′-GTAGGGGACTTTCCGAGCTCGAGATCCTATG-3′). Binding reactions were conducted for 30 min at room temperature as described (29). In some reactions, the nuclear extracts were coincubated with both labeled oligonucleotide and 1000× excess of an unlabeled specific or nonspecific oligonucleotide competitor (containing the SP.1 recognition site). Oligonucleotide specific competitors included both 39- and 33-bp oligonucleotides and the 27-bp oligonucleotide containing the consensus NF-κB site. After reaction with nuclear extracts, oligonucleotides and oligonucleotide-protein complexes were resolved on 6% nondenaturing polyacrylamide gel and visualized using a phosphorimaging system (Molecular Dynamics, Sunnyvale, CA).
Results
IFN-γ augments CpG DNA-induced secretion of IL-12 p40
Bacterial DNA, or oligonucleotides with internal CpG motifs, are known to induce IL-12 secretion by macrophages or dendritic cells, but the mechanism by which CpG DNA-induces IL-12 secretion is unknown. Since IFN-γ augments the IL-12 response to other stimuli (such as LPS), we evaluated the influence of IFN-γ on CpG DNA and LPS-induced p40 secretion. Culture of bone marrow-derived macrophages with a CpG-containing oligonucleotide (designated 1826) resulted in a significant, dose-dependent IL-12 p40 response (Fig. 1⇓). A similar oligonucleotide in which the internal CpG is switched to GpC (designated 1982) did not induce p40 secretion. Fig. 1⇓ also shows that LPS alone induced less p40 secretion than did CpG DNA. Although IFN-γ alone did not induce p40 secretion, costimulation with IFN-γ and 1826 (but not 1982) resulted in a marked increase in the p40 response compared with the response to oligonucleotide alone. The combination of IFN-γ and CpG DNA induced a p40 response that exceeded that seen following challenge with IFN-γ and LPS.
IFN-γ is a powerful costimulator of CpG DNA-induced IL-12 p40 secretion. Cultures (0.2 ml) containing 105 bone marrow-derived adherent cells were stimulated overnight with the indicated CpG oligonucleotide 1826 (0.01–1 μg/ml) or LPS (.01–1 μg/ml); non-CpG oligonucleotide 1982 (1 μg/ml), with or without murine IFN-γ (100 U/ml). Ordinate units represent μg/ml concentration. Supernatant p40 concentration was determined by ELISA. Values represent the mean of four replicate cultures. This is one of three similar experiments. 1826, containing oligonucleotide; 1982, control (non-CpG-containing) oligonucleotide.
Because the response observed in ex vivo-obtained macrophages could be influenced by other cells, we evaluated the ability of CpG DNA to induce p40 secretion in a murine macrophage cell line (RAW 264). The use of a cell line also permits manipulation of conditions to favor or antagonize p40 production. The results presented in Fig. 2⇓ demonstrate that IFN-γ again augments CpG DNA-induced p40 secretion. Fig. 2⇓ also shows that in RAW 264 cells, LPS alone does not induce p40 secretion and that the IFN-γ induced augmentation of the LPS response is less than the synergy observed between IFN-γ and CpG DNA. As we observed in bone marrow-derived macrophages, IFN-γ alone did not induce appreciable p40 secretion nor did the combination of IFN-γ and the control oligonucleotide 1982 (data not shown).
IFN-γ exhibits synergy with CpG DNA in the induction of p40 secretion by the macrophage cell line RAW 264. Cultures containing 105 log-phase RAW 264 cells were stimulated as in Fig. 1⇑, and p40 was measured in a manner identical to that depicted in Fig. 1⇑. The results represent one of four representative experiments.
IFN-γ augments CpG DNA-induced p40 promoter activity
To address the mechanism(s) by which IFN-γ augments CpG DNA-induced IL-12 secretion, we used an IL-12 promoter-luciferase construct to test the ability of CpG DNA (in the presence or absence of IFN-γ) to activate the IL-12 p40 promoter. We transiently transfected RAW 264 cells with a plasmid containing a −703 to +54 segment of the p40 promoter linked to the luciferase reporter gene. Fig. 3⇓ illustrates that, while both CpG DNA and LPS induce reporter-driven luciferase activity, the addition of IFN-γ to either of these stimuli results in a marked increase in reporter gene activity. Interestingly, LPS is a more potent activator of this promoter segment than CpG DNA (in the absence or presence of IFN-γ). This observation contrasts sharply with our finding that LPS is a less effective inducer of p40 protein secretion in macrophages or RAW 264 cells (Figs. 1⇑ and 2⇑). Because reporter assays may not reflect actual mRNA transcription, we used Northern blot analysis to confirm that IFN-γ increased the amount of p40 mRNA in CpG DNA-treated cells (Fig. 4⇓). Here again, we found that CpG DNA alone (but not LPS alone) induced an increase in the level of p40 mRNA. In the presence of IFN-γ, CpG DNA induced greater levels of p40 mRNA than the levels seen in RAW 264 cells treated with IFN-γ and 1 μg/ml LPS. Thus, the reporter assay does accurately reflect the impact of IFN-plus CpG DNA on intracellular p40 mRNA level, but, at the same time, reveals a discordance between LPS-induced promoter activity (present) and LPS-induced p40 secretion (absent). This discordance between regulation of p40 promoter constructs and regulation of p40 mRNA transcription has also been observed by others (30).
IFN-γ exhibits synergy with both CpG DNA and LPS in activating a p40 reporter construct. RAW 264 cells were transiently transfected with a p40 promoter construct consisting of the 5′ 703-bp of the promoter coupled to luciferase. To control for transfection efficiency and viability, cells were cotransfected with a β-gal reporter under control of the CMV constitutive promoter. After 24 h, cultures were stimulated with LPS (1 μg/ml) 1826 0.3 μg/ml, with or without IFN-γ (100 U/ml), with the indicated stimuli, and reporter activity was assayed 24 h later.
IFN-γ costimulates with both LPS and CpG DNA to induce an increase in IL-12 p40 mRNA level. Cultures (20 ml) containing 2 × 107 RAW 264 cells were stimulated for 4.5 h with LPS (1 μg/ml), the CpG-containing oligonucleotide 1826 (0.3 μg/ml), in the presence or absence of IFN-γ (100 U/ml). Following electrophoretic separation and blotting to nitrocellulose, mRNA band was visualized by the use of labeled cDNA probes specific for either the coding region of p40 or GAPDH.
CpG DNA-induced p40 promoter activation requires elements 3′ of the −250 position and induces transcription factor binding to the NF-κB half-site
To localize the region(s) of the p40 promoter necessary for CpG DNA-induced promoter activity, we used PCR to prepare 5′ truncations of the −703 to +54 segment. These consisted of segments beginning 5′ at −697, −543, −406, −250, and −100 and extending 3′to +54 (Fig. 5⇓A). These truncated constructs were cloned into a luciferase-containing plasmid that was used to transiently transfect RAW 264 cells. After stimulation of the transfected cells, it was apparent that stimulation of cells transfected with truncations of −250 or longer resulted in increased luciferase activity, but the −100 construct was inactive (Fig. 5⇓B). As was seen with the −703 construct, IFN-γ markedly increased the activity of all CpG-stimulated functional promoter constructs. Since the promoter region between −250 and −100 contains the NF-κB half-site (−122 to 132) and an adjacent 5′ Pu.1 site (13), we utilized a gel shift assay to analyze the DNA binding properties of nuclear extracts from treated and control RAW 264 cells. Fig. 6⇓ shows that, after 3 h of stimulating RAW 264 cells, CpG DNA induces nuclear translocation of at least two proteins that bind to a 39-bp oligonucleotide encompassing both the NF-κB half-site and the Pu.1 site. Addition to the extracts of a 1000× excess of unlabeled 39 mer was able to compete the bands induced by CpG DNA or LPS. Additionally, the unlabeled 39 mer eliminated visualization of the lower band that was present in both resting and activated cells. Addition of a 1000× excess of unlabeled 33 mer (containing only the NF-κB half-site) or addition of an unlabeled oligonucleotide containing the B cell NF-κB IgH enhancer element, competed only the upper two bands. Thus, the lower band likely represents binding to the Pu.1 site by a constitutively produced protein.
Activation of the p40 promoter by CpG DNA or LPS depends on elements contained in the 250-bp 5′ of the TATA box. A, Promoter truncation mutants were constructed using PCR amplification from the −703-bp construct presented in Fig. 3⇑. Truncations were directionally cloned into the luciferase-containing plasmid Luc Link. B, Individual promoter constructs were cotransfected with CMV-β-gal into RAW 264 cells using the protocol described in Fig. 3⇑. Corrected reporter activity was determined after 24 h.
EMSA showing that stimulation of RAW 264 cells with either CpG DNA or LPS induces nuclear translocation of an NF-κB binding factor. Cells were stimulated as indicated (1826, 0.3 μg/ml; LPS, 1 μg/ml; IFN-γ, 100 U/ml) for 3 h, and 5 μg of nuclear extracts were incubated with a labeled 39-bp oligonucleotide encompassing both the NF-κB half-site and the adjacent Pu.1 site. Where indicated, the reaction mixture contained a 1000× molar excess of the following unlabeled oligonucleotide competitors: SP.1 (nonspecific competitor); 39 (identical to the labeled probe); 33 (oligonucleotide with the NF-κB half-site, but not the Pu.1 site); and NF-κB (oligonucleotide containing the NF-κB recognition element contained in the mouse κ-chain enhancer.
Because the 3-h stimulation period might not reveal differences in the transcription factor activity that was present at an earlier time, we compared RAW 264 nuclear extracts at 1 and 3 h poststimulation with 1826, LPS, IFN-γ, or IFN-γ combined with either 1826 or LPS. The results shown in Fig. 7⇓, A and B, show no appreciable difference in the EMSA pattern at 1 and 3 h. Additionally, Fig. 7⇓B shows more intense oligonucleotide binding after 3 h. As found in Fig. 6⇑, the induced bands were effectively completed by an unlabeled oligonucleotide containing an NF-κB site. We found some increase in NF-κB binding in cells treated with IFN-γ alone, but the intensity of the bands was less than observed with 1826 or LPS (with or without IFN-γ). While LPS activates NF-κB translocation and drives p40 promoter constructs as effectively as CpG DNA, LPS alone is a weak inducer of p40 mRNA transcription. Thus, CpG DNA has p40 promoter-activating properties that are distinct from those of LPS (although the activity of both are augmented by IFN-γ).
EMSA showing that stimulation of RAW 264 cells for 1 or 3 h induces a similar pattern of nuclear translocation of NF-κB. Nuclear extracts (5 μg) were incubated with a labeled 27-bp oligonucleotide containing a consensus NF-κB recognition site. Where indicated, the reaction mixture contained a 1000× molar excess of the unlabeled NF-κB oligonucleotide.
Discussion
CpG-containing DNA can serve as a potent inducer of inflammatory cytokines, and, for this reason, there is much interest in its adjuvant properties. We and others have shown that CpG-containing DNA is a potent inducer of IFN-γ production (31, 32). Subsequent work revealed that CpG DNA-induced IFN-γ production is, in fact, IL-12-dependent (19, 20). Thus, IL-12 secretion may represent the most proximal proinflammatory cytokine that is induced by CpG DNA. Our finding that IFN-γ augments CpG DNA-induced IL-12 secretion is consistent with the documented activity of IFN-γ on other IL-12-inducing stimuli, and is also consistent with the observation that ICSBP-deficient mice do not transcribe p40 after microbial challenge (14, 15). Although LPS and CpG DNA share a number of biologic properties, our findings demonstrate that CpG DNA (like LPS) induces the secretion of more IL-12 p40 and in the presence of IFN-γ; however, CpG DNA alone is a stronger inducer of p40 transcription/secretion than is LPS alone.
Since IL-12 is a potent inducer of IFN-γ secretion by T and NK cells, the ability of CpG DNA to induce some IL-12 secretion can lead to the subsequent production of IFN-γ by T and NK cells (which can, in turn, augment CpG DNA-induced IL-12 production). Although CpG DNA alone induces p40 transcription/secretion by RAW 264 cells (while LPS alone does not), the two stimuli have similar activity with respect to driving a p40 promoter construct containing up to 703 bp of sequence proximal to the TATA box. Both CpG DNA and LPS induce luciferase activity in promoter constructs containing as few as 250 bp 5′ of the TATA box. A more drastic truncation, leaving only the proximal 100 bp of the p40 promoter could not be induced to drive luciferase activity in RAW 264 cells. This finding is consistent with the observation that the NF-κB half-site located at position −132 to −122 is critical for promoter function (13).
We observed that IFN-γ augments the ability of either CpG DNA or LPS to activate the promoter-reporter constructs that extended 250 bp (or farther) 5′ of the TATA box. The promoter segments that we tested do not contain sites that are known be recognized by either STAT transcription factors or ICSBP. Additionally, the ability of IFN-γ to augment promoter activity is not due to increased cell survival/proliferation posttransfection, since the activity of a CMV β-gal control reporter was similar in the presence of absence of IFN-γ. Thus, the effect of IFN-γ may indirectly regulate an as yet unidentified transcription factor; alternatively, there may be an unidentified sequence in the proximal 250 bp of the p40 promoter that binds a known IFN-γ-induced transcription factor. Although the use of promoter constructs is useful to dissect mechanisms responsible for p40 transcription, our study also illustrates limitations in the interpretation of results. In RAW 264 cells, LPS alone can drive a number of p40 reporter constructs, however LPS does not (in the absence of IFN-γ) induce either p40 secretion or an increase in p40 mRNA level. Thus, other as yet unidentified factors may operate outside the promoter region that we evaluated in our constructs.
We found that stimulation with either CpG DNA or LPS induce similar translocation of proteins that bind to the NF-κB half-site. Additionally, there is constitutive nuclear translocation of a protein that binds to the adjacent Pu.1 site, and this binding is not markedly influenced by CpG DNA, LPS, or IFN-γ. Binding to the NF-κB half-site appears to be necessary for both CpG DNA and LPS-induced p40 secretion, but binding to this site alone clearly is not sufficient to activate p40 transcription. Despite the fact that nuclear extracts from cells stimulated with LPS or CpG DNA show an identical pattern of binding to an oligonucleotide containing the NF-κB half-site, only CpG DNA is able to increase levels of p40 mRNA or induce p40 secretion in the absence of IFN-γ. Our observed discordance between the induced activity of p40 promoter constructs and the actual transcription of p40 mRNA is similar to observations by others who noted that, while IL-10 was a potent inhibitor of p40 transcription, IL-10 did not inhibit the induced activity of a p40 promoter construct (28). Our findings suggest that there may be an element(s) in the proximal p40 promoter that (together with NF-κB) is necessary for the activation of transcription. A possible explanation of our findings is that CpG DNA induces this factor that, together with NF-κB, induces p40 transcription. Alternatively, CpG DNA may inhibit a repressor, thus permitting NF-κB-induced p40 transcription. Since CpG DNA is not more effective than LPS in activating the 703-bp p40 promoter construct, it is likely that the CpG DNA-sensitive element lies outside the 703-bp promoter segment.
It is important to note that p40 transcription/secretion does not always directly correlate with the level of the secreted p70 heterodimer. Activation-induced transcription of p40 is required for p70 secretion. The regulation of p70 secretion is complex and has been shown to be inhibited by the p40 homodimer (33). Treatment of mice with CpG DNA results in a sustained increase in serum IL-12 and in resistance to challenge with Listeria monocytogenes (34). The increased serum IL-12 levels and the resistance were dependent on IFN-γ. Our studies suggest that the observed influence of IFN-γ on the in vivo response to CpG DNA may be a consequence of enhanced p40 gene transcription.
Acknowledgments
We thank Shelly Forbes for her assistance in preparation of the manuscript.
Footnotes
-
↵1 This work is supported by funding from the Department of Veterans Affairs and U.S. Public Health Service Grant AI10112.
-
↵2 Address correspondence and reprint requests to Dr. John S. Cowdery, Department of Internal Medicine, University of Iowa College of Medicine, C31-0 GH, Iowa City, IA 52242. E-mail address: john-cowdery{at}uiowa.edu
-
↵3 Abbreviations used in this paper: ICSBP, IFN consensus sequence binding-protein; β-gal, β-galactosidase.
- Received November 18, 1998.
- Accepted March 17, 1999.
- Copyright © 1999 by The American Association of Immunologists
References
In this issue
