Published online before print August 4, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||



,1
* Division of Rheumatology and Immunology, Department of Medicine, Duke University, Durham, North Carolina;
Laboratory of Biomedical Science, North Shore University Hospital-New York University School of Medicine, Manhasset; and
Medical Research Services, Durham Veterans Affairs Medical Center, North Carolina
1Correspondence: 151G Durham VA Medical Center, 508 Fulton Street, Durham, NC 27705. E-mail: piset001{at}mc.duke.edu
|
|
|---|
and interleukin-12, it failed to cause HMGB1 release into the media. Although unable to induce HMGB1 release by itself, the CpG ODN nevertheless potentiated the action of LPS. With RAW 264.7 cells, lipoteichoic acid and polyinosinic-polycytidylic acid, like LPS, stimulated HMGB1 release as well as cytokine production. These results indicate that the effects of CpG DNA on macrophages differ from other ligands of Toll-like receptors and may lead to a distinct pattern of immune cell activation in the context of infection or its use as an immunomodulatory agent.
Key Words: oligonucleotide lipopolysaccharide Toll-like receptor
|
|
|---|
Although CpG DNA and LPS induce responses in common, they nevertheless differ in their signaling pathways and range of activities. Thus, CpG DNA acts through TLR9, and LPS acts through TLR4 [5 , 6 ]. It is important that TLR9 activation occurs at an intracellular site, with access to DNA requiring internalization and passage through an endosomal compartment [7 ]. In contrast, TLR4 is a surface receptor. Another difference between activation by CpG DNA and LPS concerns the relative induction, directly or indirectly, of various cytokines and other mediators, including glucocorticoids [8 ]. The basis for these differences is not known, as LPS and CpG DNA, despite use of different TLRs, nevertheless involve similar downstream signaling molecules such as myeloid differentiaton factor 88 [9 ].
Among molecules whose expression is modulated by LPS, high-mobility group (HMG)B1 serves as a late mediator of its action. HMGB1 is a nonhistone nuclear molecule with dual function. Inside the cell, HMGB1 plays a role in chromatin structure and transcriptional activity [10 11 12 ]. Outside the cell, HMGB1 serves as a cytokine and induces an array of inflammatory responses such as those of LPS [13 ]. The release of HMGB1 from the cell during activation involves hyperacetylation, translocation from the nucleus, and secretion from vesicles [14 ]. A central role of HMGB1 in inflammation comes from studies in animal models showing that inhibition of HMGB1 by antibody treatment can attenuate sepsis and collagen-induced arthritis [15 , 16 ].
To explore further the activity of CpG DNA in immunity, we have investigated whether CpG DNA can function like LPS and can induce HMGB1 release from macrophages. For this purpose, we have compared the effects of a synthetic CpG ODN and LPS using the murine macrophage cell lines RAW 264.7 and J774A.1 as models. HMGB1 release from cells was assessed by Western blotting of supernatants as well as confocal microscopy. In studies presented herein, we show that activation of macrophages by CpG DNA, unlike that of LPS, does not induce the translocation and release of HMGB1. Nevertheless, CpG DNA can act synergistically with LPS to induce the cellular release of this mediator. Together, these results indicate important differences in the consequences of TLR activation and distinguish CpG DNA from LPS in its potential for stimulating innate immunity and inducing inflammation.
|
|
|---|
Western blotting
Supernatants described above were collected after 24 h of stimulation and concentrated by Centricon YM-10 (Millipore, Billerica, MA). The volume of the concentrated supernatants was adjusted to
70 µl, and samples were resolved on 412% NuPAGE® Tris-Bis sodium dodecyl sulfate-polyacrylamide gel (Invitrogen). Protein was transferred to polyvinylidene difluoride membranes (Invitrogen), blocked with 5% dry milk in Tris-buffered saline-Tween, and blotted with rabbit anti-HMGB1 polyclonal antibody. The membrane was then incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit immunoglobulin G (IgG) followed by Super Signal® West Femto substrate (Pierce, Rockford, IL). Images were captured by exposing the membrane to a charged-coupled deivce camera (FluorChem8900, Alpha Innotech, San Leandro, CA).
Enzyme-linked immunosorbent assay (ELISA)
To determine cytokine levels in cultured RAW 264.7 cells, supernatants were collected at 24 h after stimulation. Cytokine [tumor necrosis factor
(TNF-
) and interleukin (IL)-12] concentrations in culture media were determined by ELISA. Capture antibodies, anti-TNF-
(R&D Systems, Minneapolis, MN) or anti-IL-12 p40/p70 (BD PharMingen, San Diego, CA), were coated overnight at 4°C on Immunlon® 96-well plates. After three washes, samples were added in duplicate and incubated at room temperature for 2 h. Biotinylated anti-TNF-
(R&D Systems) or anti-IL-12 p40/p70 detection antibodies (BD PharMingen) were added and incubated for an additional 2 h at room temperature. Avidin-conjugated HRP was added with 30 min incubation at room temperature followed by 0.015% 3,3',5,5'-tetramethylbenzidine, 0.01% H2O2in 0.1 M citrate buffer, pH 4.0, substrate until color development. Three phosphate-buffered saline (PBS) washes were performed between steps, and plates were read at optical density of 650 using an automated microtiter plate reader.
Confocal imaging
After 24 h stimulation, cells were washed twice with ice-cold PBS, scraped off the plates, and replated in eight-well chamber slides in RPMI 1640 supplemented with 10% FBS. Cells were allowed to adhere for 23 h, washed twice with ice-cold PBS. Cells are then fixed and permeablized using Cytofix/Permeable kit (BD PharMingen), followed by incubation with rabbit-anti-HMGB1 antibody and Alexa Fluor 488 goat anti-rabbit IgG (Molecular Probes, Eugene, OR). Images were captured by a confocal laser-scanning microscope (Zeiss LSM 510, Carl Zeiss, Oberkochen, Germany).
Endotoxin assays
A commercial chromogenic limulus amebocyte lysate (LAL) test (BioWhittaker, Walkersville, MD) was used to determine endotoxin levels in the TLR ligands. To further determine the effects of endotoxin on HMGB1 release, polymixin B was added in the cultures stimulated with different TLR ligands. Supernatants were collected at 2024 h after stimulation and were subjected to Western blotting.
|
|
|---|
and IL-12 levels were also measured. As shown in Figure 1C
, the levels of cytokines induced by ODN1826 were similar to those of LPS. This stimulatory effect was sequence-specific, as the control ODN lacking a CpG motif was inactive. Similar results were obtained with the J774A.1 cell line, although background levels of extracellular HMGB1 were higher than those observed with the RAW 264.7 cells (Fig. 1A)
.
![]() View larger version (17K): [in a new window] |
Figure 1. The effect of CpG DNA and LPS on HMGB1 release. RAW 264.7 and J774.A1 cells were stimulated with medium, LPS (0.5 µg/ml), SAK1 (1.5 µM), and 1826 (1.5 µM) for 24 h. Culture medium was collected at 24 h and subjected to Western blotting (A) and cytokine ELISA assay (C; RAW 264.7 cells only). The effects of increasing doses of LPS and 1826 on RAW 264.7 cell release of HMGB1 release into culture medium were also assayed by Western blotting (B).
|
![]() View larger version (17K): [in a new window] |
Figure 2. Synergistic effects of CpG DNA and LPS on HMGB1 release. RAW 264.7 cells were stimulated with a suboptimal dose of LPS (0.025 µg/ml), SAK1, or 1826 (1.5 µM) or a combination of DNA and LPS for 24 h. Supernatants were collected and assayed for HMGB1 (A) and cytokine (B). To determine a dose range for CpG DNA-induced synergy in HMGB1 release, 1826, at a concentration ranging from 0.015 to 1.5 µM, was added to RAW 264.7 cell culture medium containing a low-dose LPS (0.025 µg/ml). Twenty-four hours after the stimulation, culture medium was subject to Western blotting (C).
|
![]() View larger version (62K): [in a new window] |
Figure 3. HMGB1 translocation in RAW 264.7 cells by confocal microscopy. Stimulated RAW 264.7 cells were plated on chamber slides and allowed to attach. Cells were fixed and permeabilized, followed by rabbit anti-HMGB1 and anti-rabbit IgG conjugated with Alexa Fluor 488.
|
production in association with HMGB1 release and differ from CpG DNA in this property (Fig. 4B) . Confirming the Western blot, confocal microcopy demonstrated HMGB1 translocation (Fig. 4C)
, although translocation was only observed in some cells in the cultures.
![]() View larger version (21K): [in a new window] |
Figure 4. The effects of TLR2 and TLR3 stimulation on HMGB1 release. RAW 264.7 cells were stimulated with LTA and poly (I:C) at varying concentrations for 24 h. Culture medium was collected and analyzed by Western blotting (A) and cytokine ELISA assays (B). Stimulated RAW 264.7 cells were plated on chamber slides, fixed, permeablized, and stained for HMGB1, and translocation was visualized by confocal microscopy (C).
|
![]() View larger version (8K): [in a new window] |
Figure 5. The effects of polymixin B on HMGB1 release stimulated by LPS, LTA, or poly (I:C). RAW 264.7 cells were stimulated with different TLR ligands as above in the presence of polymixin B. Supernatants were collected and analyzed by Western blotting.
|
|
|
|---|
and IL- 12. As such, these findings suggest selective stimulation of inflammatory mediator release by different PAMPs, a finding that is relevant to the activation of innate immunity in bacterial infection as well as settings in which PAMPs may be used as adjuvants or immune modulators.
As shown by studies in vitro and in vivo, HMGB1 serves as a late mediator of the action of LPS and displays a wide array of immunostimulatory actions, which includes the induction of TNF-
, IL-1, and nitric oxide (NO) [21
22
23
]. These actions become manifest when HMGB1 exits the cell into the external milieu, where it displays its alternative function as a cytokine. This action reflects the activity of a structural domain of HMGB1the B box [24
]. It is interesting that another structural domainthe A boxcan inhibit responses of HMGB1 [16
]. Although the signaling pathways elicited by HMGB1 are not fully defined, there is evidence that the triggering occurs via the receptor for advanced glycation end-products [25
] as well as TLR2 and TLR4 [26
]. The importance of this pathway for immune activation is evident in studies demonstrating that HMGB1 levels are elevated in patients with sepsis [13
] and that anti- HMGB1 antibodies can block sepsis and inflammation in animal models [16
].
In these studies, we have used a synthetic oligonucleotide as a model for CpG DNA. As shown in many studies, this ODN is a potent immune stimulator, which can activate cells via TLR9 and elicit a wide array of inflammatory responses that include B cell activation and stimulation of proinflammatory mediators including cytokines and NO [3 , 27 , 28 ]. Nevertheless, this ODN failed to induce release of HMGB1 and, moreover, failed to cause the translocation of this molecule from the nucleus. Other TLR9 ligands, including a natural DNA as well as a second-generation immunomer, had similar behavior as the ODN, although these sources of CpG DNA induced cytokine production (data not shown). These findings suggest that TLR9 ligands, in general, are incapable of inducing HMGB1 release from macrophages.
Although CpG DNA alone did not cause HMGB1 release, it nevertheless potentiated the action of LPS. Thus, in the presence of CpG DNA, the doses of LPS required for HMGB1 release were reduced dramatically. Synergism between CpG DNA and LPS is also manifest in other responses such as the production of TNF-
and IL-12. Together, these findings indicate that CpG DNA and LPS, although leading to similar signaling events such as the activation of mitogen-activated protein kinases and nuclear factor-
B, nevertheless differ in the ultimate effects of stimulation. Furthermore, CpG DNA differs from ligands of TLR2 and TLR3, which, like LPS, can induce cytokine expression and HMGB1 release. In the context of infection, although CpG alone may not trigger HMGB1 release, it nevertheless may act in concert with other PAMPs to augment and prolong this response.
The differences between stimulation by CpG DNA and LPS in the HMGB1 release are notable, given their categorization as PAMPs and postulated roles in innate immunity. Previous studies have also demonstrated that CpG DNA and LPS are not equivalent in their action. Thus, LPS induces significant release of glucocorticoids under conditions in which CpG DNA is much less effective in this response, despite similar induction of cytokines [8 , 29 ]. Furthermore, in this setting in vivo, LPS, but not CpG DNA, can induce extensive lymphocyte apoptosis, which can culminate into the release of nucleosomes into the blood [8 ]. Such cell loss may contribute to the immune suppression that occurs during sepsis.
Previous studies about the induction of shock by CpG DNA involved mice treated with galactosamine [30
]. This sugar inhibits transcription in the liver and renders mice extraordinarily sensitive to the effects of TNF-
. In mice treated with galactosamine, LPS can stimulate shock at doses 1000 less than those required for an untreated animal. Although CpG DNA can induce shock in galactosamine-treated mice, the effect may result from an enhanced effect of the TNF induced. In the intact mouse, however, although CpG DNA can stimulate TNF production, the other conditions necessary for shock may not be present. As such, stimulation by CpG DNA may be well-tolerated. In this conceptualization, CpG DNA may be "safer" than other "danger" molecules.
Another potential difference between CpG ODN and other TLR ligands on the induction of HMGB1 release concerns the effects of these agents on apoptosis. Thus, in these culture systems, RAW 264.7 cell activation by LPS, LTA, and dsRNA is accompanied by an increase in apoptosis as assessed by activation of caspase 3, although CpG DNA does not cause this response (W. Jiang and D. S. Pisetsky, preliminary observations). Although HMGB1 release has been considered a property of necrotic but not apoptotic cells, this issue has been investigated extensively only in HeLa cells [31 ]. It is possible therefore that certain cell types (e.g., macrophages), undergoing apoptosis or secondary necrosis, release HMGB1, accounting for the differences among TLR ligands in this response. This issue is currently under investigation.
CpG ODN has been investigated as an adjuvant as well as immunomodulator to enhance host defense in the setting of infection [32 , 33 ]. As suggested by the current studies as well as those previously published, CpG DNA may lead to a unique pattern of immune activation characterized by induction of cytokines in the absence of HMGB1 release as well as other immune effects, including lymphocyte apoptosis [8 ] and glucocorticoid induction [8 , 29 ]. Although this pattern may be beneficial for an adjuvant, its role in the stimulation of innate immunity is speculative.
While this manuscript was being prepared, Dumitriu et al. [34
] reported that CpG ODN induced human plasmacytoid DCs to release HMGB1, a result in contrast to our findings with macrophages. Among differences between our studies and theirs are the cell type, species, maturation stage, and use of cell lines as opposed to cells grown from peripheral blood. Furthermore, upon CpG ODN stimulation, plasmacytoid dendritic cells produce large amounts of interferon-
, and macrophages produce IL-12 and NO [35
]. These mediators may determine whether HMGB1 is released. In this regard, although activation via different TLRs may induce similar signaling pathways, it is also possible that depending on cell type, CpG DNA may trigger a unique secondary signal, as yet to be identified, which prevents the release of HMGB1. Future studies will define the pathways by which triggering different TLRs leads to HMGB1 release and the interaction of these pathways occurring during infection.
Received April 20, 2005; revised June 29, 2005; accepted July 5, 2005.
|
|
|---|
induces high mobility group box 1 protein release partly through a TNF-dependent mechanism J. Immunol. 170,3890-3897
B by Toll-like receptor 3 Nature 413,732-738[CrossRef][Medline]
Proc. Natl. Acad. Sci. USA 93,2879-2883
-mediated shock Eur. J. Immunol. 27,1671-1679[Medline]This article has been cited by other articles:
![]() |
H. Kitamura, M. Ito, T. Yuasa, C. Kikuguchi, A. Hijikata, M. Takayama, Y. Kimura, R. Yokoyama, T. Kaji, and O. Ohara Genome-wide identification and characterization of transcripts translationally regulated by bacterial lipopolysaccharide in macrophage-like J774.1 cells Physiol Genomics, October 8, 2008; 33(1): 121 - 132. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Zicari, C. Centonze, M. Realacci, B. Buchetti, A. Pietropolli, and C. Ticconi Estradiol 17-{beta} and Progesterone Modulate Inducible Nitric Oxide Synthase and High Mobility Group Box 1 Expression in Human Endometrium Reproductive Sciences, July 1, 2008; 15(6): 559 - 566. [Abstract] [PDF] |
||||
![]() |
S. Ivanov, A.-M. Dragoi, X. Wang, C. Dallacosta, J. Louten, G. Musco, G. Sitia, G. S. Yap, Y. Wan, C. A. Biron, et al. A novel role for HMGB1 in TLR9-mediated inflammatory responses to CpG-DNA Blood, September 15, 2007; 110(6): 1970 - 1981. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Jiang, C. W. Bell, and D. S. Pisetsky The Relationship between Apoptosis and High-Mobility Group Protein 1 Release from Murine Macrophages Stimulated with Lipopolysaccharide or Polyinosinic-Polycytidylic Acid J. Immunol., May 15, 2007; 178(10): 6495 - 6503. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Tarang, A. Sodhi, and P. Chauhan Differential expression of Toll-like receptors in murine peritoneal macrophages in vitro on treatment with cisplatin Int. Immunol., May 1, 2007; 19(5): 635 - 643. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Tang, Y. Shi, R. Kang, T. Li, W. Xiao, H. Wang, and X. Xiao Hydrogen peroxide stimulates macrophages and monocytes to actively release HMGB1 J. Leukoc. Biol., March 1, 2007; 81(3): 741 - 747. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Jiang and D. S. Pisetsky The Role of IFN-{alpha} and Nitric Oxide in the Release of HMGB1 by RAW 264.7 Cells Stimulated with Polyinosinic-Polycytidylic Acid or Lipopolysaccharide. J. Immunol., September 1, 2006; 177(5): 3337 - 3343. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||