Published online before print October 4, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Liver Center, Divisions of
* Gastroenterology and
Infectious Diseases, Department of Medicine, University of Massachusetts Medical School, Worcester
1 Correspondence: Department of Medicine, University of Massachusetts Medical School, LRB 215, 364 Plantation Street, Worcester, MA 01605-2324. E-mail: gyongyi.szabo{at}umassmed.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-), interleukin (IL)-6, interferon-
(IFN-), and IL-12 (p40/p70), and increased RNA expression of proinflammatory cytokines (IL-12p40, IL-1, IL-6, IL-1ß, IL-18, IFN-) in the liver after a LPS challenge. Furthermore, P. acnes priming sensitized TLR2-deficient (TLR2–/–) but not MyD88–/– mice to LPS-induced injury, evidenced by hepatocyte necrosis, increased levels of serum TNF-, IFN-, IL-6, and liver proinflammatory cytokine mRNA expression. IFN-, a cytokine sensitizing to endotoxin, was induced by P. acnes in splenocytes of TLR2–/– and TLR9–/– but not MyD88–/– mice. These results suggest that although P. acnes triggers TLR2-mediated cell activation, TLR2-independent but MyD88-dependent mechanisms mediate in vivo sensitization by P. acnes for LPS-induced liver injury.
Key Words: TLR4 • TLR9 • LPS • IFN-
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(IFN-) production by mononuclear cells, and T helper cell type 1 lymphocytes are recruited to the liver via increased expression of various chemokines (macrophage-inflammatory protein-1, -2, -3, IFN-inducible protein 10 [2
3
4
]). All of these changes are associated with a decrease in the CD4+natural killer T cell population [5
]. It has also been shown that P. acnes increases CD14 levels in the serum and in Kupffer cells [6
, 7
]. LPS treatment in P. acnes-sensitized mice results in substantially increased proinflammatory cytokine production and massive liver injury [1
2
3
4
]. However, it is not clear how P. acnes is recognized to trigger the mechanism that leads to priming for LPS-induced injury in the liver. In addition, in vivo recognition of P. acnes is important in various human conditions, as P. acnes is believed to be involved in granuloma formation in primary biliary cirrhosis, granulomatous colitis, chest infections in chronic granulomatous disease, endocarditis after valve transplantation, central nervous system infections, and acne [8
9
10
11
12
13
14
]. Toll-like receptors (TLRs) are signaling receptors of the innate immune system, which recognize pathogen-associated molecular patterns. TLR2 recognizes a broad range of pathogens including Gram-positive bacterial components, peptidoglycan (PGN), lipoteichoic acid (LTA), and macrophage-activating lipopeptide from Mycoplasma fermentans (MALP) [15 , 16 ]. Unlike TLR2, TLR4 has a restricted pathogen-recognition profile, and it acts as a signaling receptor for LPS, a Gram-negative bacterial component [15 , 16 ]. TLR2 and TLR4 share common downstream elements in the induction of proinflammatory responses [15 16 17 ]. However, repeated stimulation with LPS has been shown to result in attenuated inflammatory activation by the second TLR4- or TLR2-induced stimulus, a phenomenon called homotolerance and heterotolerance, respectively [18 , 19 ]. TLR2 ligands have also been suggested to cause homo- or heterotolerance [20 ].
In this study, we investigated the role of TLR2 in P. acnes-mediated sensitization for LPS-induced liver injury. We found that P. acnes triggers cell activation via TLR2; however, TLR2-independent mechanisms caused sensitization for LPS-induced liver injury by P. acnes. Furthermore, we found that myeloid differentiation primary-response protein 88 (MyD88) plays a key role in P. acnes-triggered cell activation.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In vitro studies
Human embryonic kidney (HEK) cells stably transfected with human TLR2 or TLR4/myeloid differentiation protein-2 (MD-2) or RAW 264.7 murine macrophages were stimulated with P. acnes, PGN, or phenol-purified LPS (pLPS) as indicated and plated in 96-well plates in Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY) with 10% fetal bovine serum (FBS) [22
]. After overnight stimulation, cytokine production was measured in the cell-free supernatants by enzyme-linked immunosorbent assay (ELISA). To assess the role of CD14 in P. acnes-induced cell stimulation, HEK/TLR2 or HEK/TLR4/MD-2 were transiently transfected with human CD14 (hCD14) using GeneJuice (Novagen, Madison, WI), allowed to express CD14 for 48 h, and then stimulated as indicated in figure legends. Cell-free supernatants were analyzed for IL-8 by ELISA. Chinese hamster ovary (CHO) cells stably transfected with human TLR2 and/or hCD14 were described previously [23
]. The cells were plated in six-well plates (106 cells/ml) in F-12 nutrient mixture (HAM, Gibco) with 10% FBS and stimulated for 1 h as indicated. Nuclear fraction was extracted as described previously [21
] and nuclear factor (NF)-
B activation was measured by electromobility shift assay (EMSA) using equal amounts of nuclear protein (5 µg) and a 32P-labeled NF-B consensus (Promega, Madison, WI) oligonucleotide sequence as we described [21
]. Mouse peritoneal macrophages were collected 3 days after i.p. injection of 2 ml 4% thioglycolate (Difco, Detroit, MI), cultured in RPMI 1640 medium supplemented with 10% fetal calf serum (1.25x105/ml), and stimulated overnight as indicated in figure legends. Cytokine production was determined in the culture supernatants by ELISA. Freshly removed spleens were homogenized in cold RPMI 1640, spleen suspensions were filtered through 70 µm nylon cell strainers (BD Falcon, Bedford, MA), and 8.3 g/l ammonium chloride in 0.01 M Tris-HCl buffer, pH 7.5 (Sigma Chemical Co.) was added to lyse the erythrocytes. Splenocytes (107/ml) were resuspended in 10% FBS/RPMI 1640 and stimulated as indicated in the figure legends. Cytokine production was analyzed in the cell-free culture supernatants by ELISA.
Measurement of tumor necrosis factor (TNF-), IL-12p40/p70, IL-6, IFN-, and IL-8 levels
TNF-, IL-12, IL-6, and IFN- levels from the serum and cell supernatants and human IL-8 levels from HEK cell supernatants were determined by ELISA, according to the manufacturers’ instructions (Endogen, Woburn, MA; BD PharMingen, San Diego, CA; and eBioscience, San Diego, CA).
RNase protection assay (RPA)
RPAs were carried out using the RiboQuant multiprobe assay system (BD PharMingen), according to the manufacturer’s instructions and as we described [24
]. Briefly, 32P-labeled RNA probes were transcribed with T7 polymerase using the multiprobe template set mCK-2b. RNA (10 µg) was hybridized with 3.5 x 105 counts per minute/µl probe overnight at 56°C. Samples were then digested with RNase followed by proteinase K treatment, phenol:chloroform extraction, and ethanol precipitation and resolved on 5% acrylamide-bisacrylamide (19:1) urea gels. Dried gels were visualized using the Fuji FLA5000 phospho-imager system and analysis software (Image Guage, Emlsford, NY).
Histopathological analysis
Formalin-fixed liver samples were embedded in paraffin and stained with hematoxylin-eosine (H&E) to assess liver inflammation, granuloma formation, and necrosis.
Statistical analysis
Statistical values were determined using the paired Student’s t-test for cytokine evaluations in mice and primary mouse cells or nonpaired Student’s t-test (two-tailed distribution) for cell lines. A P< 0.05 was selected before the study as the level of significance.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B (P<0.02) in CHO/TLR2/CD14 but not in CHO/CD14 cells, suggesting that P. acnes was a TLR2 ligand. The selective TLR2 ligand (PGN) activated NF-B in CHO/TLR2/CD14 and not in CHO/CD14 cells, and the TLR4 ligand, pLPS, activated both cell lines, consistent with expression of endogenous TLR4 in CHO cells [23
]. To investigate the specificity of TLR2 in P. acnes-mediated cell activation, next, we studied peritoneal macrophages from TLR2–/– mice. Consistent with the requirement for TLR2 expression for cell activation by P. acnes, we found no activation of peritoneal macrophages by P. acnes in TLR2–/– mice (Fig. 1B)
. The TLR2 ligands PGN, zymosan, and P. acnes activated cells only from wild-type and not from TLR2–/– mice, and pLPS, a TLR4 ligand, induced IL-6 in macrophages from TLR2–/– and wild-type mice. Polymixin B (PB) prevented IL-6 induction by pLPS but not by P. acnes or any of the TLR2 ligands (PGN, LTA) in wild-type macrophages (Fig. 1B)
. Furthermore, P. acnes stimulated HEK cells stably transfected with human TLR2 but failed to induce cell-activation in HEK/TLR4/MD-2 cells, suggesting that it was a TLR2 ligand and that it was endotoxin-free (Fig. 1C)
. The adaptor molecule, MyD88, is recruited to TLRs upon stimulation [15
16
17
]. Although TLR4 signals through MyD88-dependent and -independent pathways, TLR2-induced cell activation is strictly MyD88-dependent [25
]. Thus, we tested P. acnes-induced cell activation in murine peritoneal macrophages deficient of MyD88 expression. Data in Figure 2A
demonstrate the lack of TNF- induction in MyD88–/– peritoneal macrophages upon stimulation with P. acnes or a TLR2 ligand compared with MyD88+/+ cells. Consistent with the presence of MyD88-independent signaling, TLR4 stimulation with LPS induced a minimal increase in TNF- production in MyD88–/– peritoneal macrophages. CD14 is a coreceptor, which can augment TLR2 signaling [26
]. Thus, we wished to evaluate the role of CD14 in TLR2-mediated cell activation by P. acnes. After transient transfection of a hCD14 construct into HEK293/TLR2 cells, TLR2-mediated IL-8 induction was augmented significantly in response to the TLR2 ligand, PGN. Likewise, coexpression of CD14 in HEK/TLR2 cells augmented IL-8 induction by the lower dose of P. acnes (Fig. 2B)
. Augmentation of CD14 in TLR2-mediated cell activation by P. acnes was also evident in murine peritoneal macrophages, where lower TNF- was induced in cells of CD14–/– animals compared with the wild-type controls (CD14+/+; Fig. 2C ).
 View larger version (33K): [in a new window] |
Figure 1. P. acnes-induced cell activation requires TLR2 expression. (A) CHO cells stably transfected with human TLR2/hCD14 or hCD14 alone (106/ml) were challenged with P. acnes, phenol-pLPS, PGN, and 5 µg/ml Polymyxin B (PB), where indicated, for 1 h to assess NF-B binding by EMSA. Representative data from four independent experiments are shown. (B) Wild-type and TLR2–/– peritoneal macrophages were challenged with P. acnes, pLPS, PGN, zymosan, and 5 µg/ml PB, as indicated overnight, and IL-6 production was measured by ELISA. Mean ± SD of four independent experiments is shown. (C) HEK cells expressing TLR2 (HEK/TLR2) or TLR4 + MD-2 (HEK/TLR4+MD2) were stimulated with P. acnes, PGN, or pLPS as indicated, and IL-8 production was analyzed 16 h later by ELISA. Mean ± SD from three experiments is shown.
|
 View larger version (20K): [in a new window] |
Figure 2. TLR2-mediated cell activated by P. acnes is MyD88-dependent and augmented by CD14 coexpression. (A) Peritoneal macrophages from wild-type and MyD88–/– mice were stimulated as indicated for 16 h, and TNF- production was determined by ELISA. (B) HEK 293/TLR2 cells (0.3x106/ml) were transfected with a hCD14 or control construct and stimulated overnight with PGN, P. acnes, human IL-1, or pLPS, as indicated. IL-8 production was measured in the cell-free supernatants by ELISA (n=4). (C) Wild-type (WT) and CD14–/– mouse peritoneal macrophages (1.25x105/ml) were stimulated overnight with P. acnes, pLPS, and PGN as indicated. TNF- production was measured by ELISA (n=4).
|
, IL-12, IL-6, and IFN- compared with LPS-stimulated mice without P. acnes priming (Fig. 3A
). In contrast, in animals challenged with selective TLR2 ligands, PGN, LTA, or PGN + LTA (5 µg/g b.w., respectively), there was no sensitization for LPS-induced cytokine induction. Consistent with the elevated serum cytokine levels, there was profound proinflammatory cytokine induction in the liver in the P. acnes-primed but not in the PGN-, LTA-, or PGN + LTA-primed mice at the mRNA levels of IL-1, IL-1ß, IL-12 (p40), IL-18, and IL-6 compared with LPS-stimulated, nonprimed control animals (Fig. 3B
and 3C)
. Hyperactivation of the inflammatory responses in the P. acnes-primed, LPS-stimulated livers was also evident by the necrotic changes on histopathology analysis. P. acnes alone induced granuloma formation (Fig. 4
). LPS stimulation without P. acnes priming induced only minimal inflammatory cell recruitment (Fig. 4)
. There was only minimal inflammatory cell recruitment and no necrosis after a LPS challenge in the lesions of mice primed with TLR2 ligands. These results suggested that unlike P. acnes, selective TLR2 activation was insufficient to sensitize for LPS-induced liver injury and proinflammatory activation.
 View larger version (52K): [in a new window] |
Figure 3. Selective TLR2 ligands cannot prime for LPS-induced liver injury. C57BL/6 mice were primed with heat-killed P. acnes (P.a; 1000 µg, i.p.) PGN (5 µg/g b.w., i.p.), LTA (5 µg/g b.w., i.p.), PGN plus LTA (5 µg/g b.w., each, i.p.), or saline and 1 week later, challenged with LPS (0.5 µg/g b.w., i.p.). (A) Serum TNF-, IL-12 (p40/p70; both at 1.5 h post-LPS stimulation), IL-6, and IFN- (both at 4 h post-LPS stimulation) were assessed by ELISA. The average of three experiments is shown. (B) RNA levels of inflammatory cytokines were analyzed by RPA in the liver 1.5 h post-LPS stimulation. IL-1Ra, IL-1 receptor antagonist; MIF, migration inhibitory factor. (C) Densitometry analysis of RPA in B indicates the average of relative density units normalized to L32 from three experiments. *, P < 0.05. For better visualization, IL-12, IL-18, IFN- and IL-6 bands are proportionally, digitally enhanced from the same blot.
|
 View larger version (124K): [in a new window] |
Figure 4. Histopathology of the liver. C57BL/6 mice were primed with heat-killed P. acnes (1000 µg, i.p.), PGN (5 µg/g b.w., i.p.), LTA (5 µg/g b.w., i.p.), PGN plus LTA (5 µg/g b.w., i.p., each), or saline and then 1 week later, challenged with LPS (0.5 µg/g b.w., i.p.). Histopathological changes in the liver were determined by H&E staining (original magnification, 10x) after stimulations as indicated. Open arrows indicate inflammatory cell recruitment, and solid arrows show granuloma formation. One representative of three experiments is shown.
|
induction by the second stimulation with P. acnes, suggesting TLR-2 homotolerance. Furthermore, initial stimulation with P. acnes resulted in heterotolerance to the TLR4 ligand, pLPS (Fig. 5A) . Similar to the pattern found in RAW264.7 macrophages, investigation of peritoneal macrophages from wild-type mice after P. acnes priming showed tolerance to subsequent LPS (TLR4) or PGN stimulation (Fig. 5B)
. Consistent with the absence of TLR2, no heterotolerance to LPS or homotolerance to PGN was seen in TLR2– peritoneal macrophages. These results suggested that in vitro in macrophages, P. acnes was recognized as a TLR2 ligand and resulted in TLR2 homo- and TLR4 heterotolerance.
 View larger version (13K): [in a new window] |
Figure 5. P. acnes induces TLR2 homo- and TLR4 heterotolerance. (A) RAW264.7 murine macrophages were stimulated with pLPS (0.01 µg/ml), P. acnes (25 µg/ml), or PGN (10 µg/ml). Twenty-four hours later, macrophages were rechallenged with LPS (1 µg/ml), PGN (10 µg/ml), or P. acnes (10 µg/ml). TNF- levels were measured in the cell supernatants 24 h after the second stimulation. (B) Peritoneal macrophages from wild-type and TLR2–/– mice were stimulated with LPS (0.01 µg/ml), P. acnes (10 µg/ml), or PGN (1 µg/ml). Twenty-four hours later, cells were rechallenged with LPS (1 µg/ml) or PGN (10 µg/ml), and TNF- secretion was measured in the supernatants 24 h after the second stimulation by ELISA.
|
, IFN-, and IL-6 levels were significantly greater in P. acnes-primed TLR2–/– as well as in TLR2+/+ mice compared with the corresponding LPS-challenged, nonprimed mice (TNF-: 39.05±12.6 vs. 1.43±0.46 ng/ml, P<0.01; IFN-: 7.2±3.5 ng/ml vs. nondetectable, P<0.001; IL-6: 122±11 vs. 92.5±26 ng/ml, P<0.04; Fig. 6A
). The magnitude of the LPS-induced inflammatory response was comparable between the TLR2–/– and wild-type (TLR2+/+) mice, consistent with the presence of TLR4 in both animal strains (Figs. 3A
and 6A) . In contrast to wild-type mice, we found no significant increase in serum IL-12 levels in the P. acnes primed, LPS-stimulated TLR2 compared (Figs. 3A
and 6A)
. Analysis of liver RNA levels revealed significantly higher IL-6 (P<0.02), IL-18 (P<0.01), IL-1ß (P<0.005), IL-12p40 (P<0.04), IFN- (P<0.02), and MIF (P<0.01) RNA levels induced by LPS in P. acnes-primed TLR2–/– animals, similar to TLR2+/+ mice (Fig. 6B and 6C)
. These data indicate that the Gram-positive P. acnes is capable of sensitizing the liver for LPS-induced injury in the absence of TLR2 expression. Consistent with this, histopathology analysis revealed extended inflammatory cell recruitment in the liver of the P. acnes-primed TLR2–/– mice after LPS stimulation and relatively milder inflammatory cell immigration after a single LPS stimulation (Fig. 6D)
. These histopathology features were similar to those seen in the wild-type mice (Fig. 4) .
 View larger version (54K): [in a new window] |
Figure 6. P. acnes results in sensitization to LPS in TLR2–/– mice, which were challenged with heat-killed P. acnes (1000 µg, i.p.) or saline and then 1 week later, with LPS (0.5 µg/g b.w., i.p.). (A) Serum TNF-, IL-12 (p40/p70), IL-6, and IFN- levels were determined by ELISA. Average of three independent experiments is shown. (B) Liver RNA levels of the indicated cytokines were determined by RPA. (C) Density units of an average of three experiments normalized to L32, respectively. (D) Histopathological changes in the liver H&E staining (original magnification, 10x) after stimulations indicated. Open arrows indicate inflammatory cell recruitment, and solid arrows show granuloma formation. Representative data from three experiments are shown.
|
in splenocytes in a TLR2-independent, MyD88-dependent manner in the pathobiology of P. acnes-induced priming for LPS-induced liver injury [4
, 27
]. Furthermore, IFN- has been suggested to prevent the TLR-mediated tolerance phenomenon [24
, 28
]. Thus, we investigated the capacity of P. acnes and other TLR2 ligands to induce IFN-. Data in Figure 7
demonstrate that LPS induced IFN- in a dose-dependent manner in wild-type and TLR2–/– mice. Consistent with its property as a selective TLR2 ligand, PGN stimulated IFN- production in splenocytes only in the wild-type mice. In contrast, P. acnes stimulation resulted in IFN- induction in the wild-type and TLR2–/– splenocytes, suggesting that IFN- induction by P. acnes was TLR2-independent. TLR-mediated signal transduction requires recruitment adaptor molecules, such as MyD88, a common adaptor for TLR2, TLR4, and TLR9. Our results indicated that P. acnes activates cells independent of TLR2 and did not require TLR4; thus, we hypothesized that MyD88 may play a key role in this process. As TLR9 is a receptor for DNA and is MyD88-dependent, we used cells from TLR9–/– mice to determine whether P. acnes may require TLR9 for cell activation. As shown in Figure 8A
, P. acnes failed to induce IFN- in MyD88–/– cells, but it resulted in IFN- induction in wild-type and in TLR9–/– splenocytes. MyD88–/– splenocytes were resistant to TLR2 (PGN), TLR4 (LPS), and TLR9 (CpG DNA) ligand stimulation but produced IFN- upon stimulation with recombinant TNF-, a positive control (Fig. 8A)
. Consistent with no IFN- induction in splenocytes in vitro, in vivo administration of P. acnes to MyD88–/– mice resulted in no development of liver granulomas, which was in contrast with the granuloma formation in the wild-type mice (Fig. 8B)
. Consistent with the requirement for MyD88 in cell activation by P. acnes, there was no TNF- induced in MyD88–/– peritoneal macrophages by TLR2, -4, and -9 ligands, and P. acnes induced TNF- in TLR9–/– and wild-type peritoneal macrophages (Fig. 8C)
. These results suggested that P. acnes induces cell activation via one or multiple receptors that use MyD88.
 View larger version (25K): [in a new window] |
Figure 7. P. acnes induces IFN- in mouse splenocytes. Isolated splenocytes from wild-type (TLR2+/+) or TLR–/– mice were stimulated with LPS, PGN, or P. acnes for 24 h as indicated. IFN- production was determined in the supernatants by ELISA. The average of duplicates is shown.
|
 View larger version (48K): [in a new window] |
Figure 8. MyD88 is instrumental for IFN- induction in mouse splenocytes. (A) Splenocytes from wild-type, MyD88–/–, and TLR9–/– mice were stimulated with TNF- (25 ng/ml), P. acnes, LPS, or CpG [oligodeoxynucleotide (ODN)1826], as indicated for 16 h. Cell-free supernatants were analyzed for IFN- by ELISA. Mean values from three mice per group are shown. (B) Wild-type and TLR9–/– mice were primed with heat-killed P. acnes (1000 µg/g b.w., i.p.) for 7 days, followed by challenge with LPS (0.5 µg/g b.w., i.p.) for 1.5 h. Histopathological changes in the liver H&E staining (original magnification, 10x) after stimulations are indicated. Filled arrows indicate granuloma formation. Representative data from two experiments are shown. (C) Peritoneal macrophages from wild-type, MyD88–/–, and TLR9–/– mice were stimulated with LPS, PGN, CpG (ODN1826), or P. acnes, as indicated, for 16 h. Cell-free supernatants were analyzed for TNF- by ELISA. Mean values from three mice per group are shown.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
production in macrophages from TLR2–/– mice [24
]. TLR2-mediated, downstream signaling involves MyD88-dependent activation of IL-1R-associated kinase and NF-B activation, leading to activation of proinflammatory cytokine genes [15
16
17
]. In murine peritoneal macrophages deficient of MyD88 expression, P. acnes failed to induce activation, confirming the requirement for the TLR2-dependent signaling pathway. Further, consistent with downstream TLR2 activation, P. acnes induced NF-B activation and TNF- production. Previous studies showed that TLR2 signaling can be augmented by the presence of CD14 [26
, 32
]. We found that coexpression of CD14 with TLR2 could increase P. acnes-induced IL-8 production. Lower IL-6 induction by P. acnes in peritoneal macrophages of CD14–/– mice also supported the notion that CD14 augments TLR2-induced activation by P. acnes. Recognition of PGN and LTA requires dimerization of TLR2 and TLR6, lack of TLR2 receptor completely prevents PGN- and LTA-induced signaling, and lack of TLR6 significantly reduces the signaling of both. MALP signaling requires the interaction of TLR6 and TLR2 [16
]. In a previous study, P. acnes induced activation in macrophages from TLR1- as well as from TLR6–/– mice, suggesting that the co-use of TLR1 and TLR6 by TLR2 may not be needed for P. acnes recognition [13
].
We found opposite effects of P. acnes treatment on LPS challenge, depending on its in vitro or in vivo administration. In macrophages in vitro, P. acnes priming attenuated TNF- induction by TLR2 and TLR4 ligands, suggesting homo- and heterotolerance. In contrast, in vivo, P. acnes resulted in sensitization to LPS. There could be several different explanations for these findings. First, it has been shown previously that IFN- is a key cytokine induced by P. acnes, which mediates sensitization to LPS [4
, 27
, 33
]. Studies with IFN--deficient mice found no sensitization by P. acnes to LPS, suggesting a key role for this cytokine [27
]. Previous results from our laboratory demonstrated that IFN- could prevent P. acnes-induced heterotolerance to LPS in macrophages [24
]. Recent studies showed that that IFN- could overcome LPS homotolerance [28
]. Second, during in vivo administration, mixed cell populations respond to P. acnes and LPS, as TLR2 and TLR4 receptors are expressed on a variety of cell types including immune cells, stellate cells, and hepatocytes [34
35
36
37
38
]. Different effects of P. acnes on the diverse cell populations during in vivo activation may explain sensitization rather than desensitization to LPS by P. acnes. Here, we found that P. acnes and not the selective TLR2 ligand, PGN, induced IFN- in splenocytes of TLR2–/– mice, supporting a unique role for P. acnes-induced IFN-. Thus, IFN-, induced by P. acnes priming in vivo as opposed to activation of a single cell population in vitro, may provide explanation for our findings. Third, our data suggest that in vivo, P. acnes may also be recognized by mechanisms other than TLR2. This notion is supported by our observation that TLR2–/– mice are sensitized by P. acnes for LPS-induced liver injury similar to the wild-type animals. The cell wall of the Gram-positive bacterium, P. acnes, consists of PGN, total carbohydrates, neutral sugars, various amino acids and amino sugars, polysaccharides, and fatty acids [39
]. The PGN component in the cellular wall of P. acnes contains a cross-linkage region of peptide chains with L,L-diaminopimelic acid and D-alanine, in which two glycine residues combine with amino and carboxyl groups of two L,L-diaminopimelic acid residues. Furthermore, the different configuration of P. acnes-derived PGN compared with S. aureus PGN may play a role in receptor recognition. In addition to TLRs, PGN can be recognized by other pattern recognition receptors, PGN recognition proteins (PGRPs). Recently, three different human PGRPs have been described [40
]. PGRP-L is expressed in the liver and PGRP-S, in the bone marrow and to a lesser extent, in neutrophils and fetal liver. Thus, recognition of the PGN component of P. acnes by PGRP is a plausible mechanism for TLR2-independent activation seen in our in vivo experiments.
Previous studies found that PGN can synergize with LPS to induce TNF-, nitric oxide, and shock in rats [41
]. There are also additional data from animal models suggesting that in vivo recognition of pathogens may be more complex than that suggested by in vitro studies using TLR2-expressing cell lines. For example, in Listeria monocytogenes infection, mounting of innate immunity was found to be MyD88-dependent but TLR2-independent [42
]. A critical role for MyD88 was also found in early clearance of L. monocytogenes [43
]. The recently described, complete genome sequence of P. acnes suggests that newly identified dipeptide proline-threonine repetitive proteins (PTRPs), PPA1880 and PPA2127, in addition to previously known PTRPs, PPA1715, PPA2210, and PPA2270, have characteristics of surface proteins and may play a role in host-microbe interaction [44
]. Furthermore, several heat shock proteins (hsp) were described in P. acnes: DnaJ, GrpE, GroEL, DnaK, and a 18-kDa protein PPA737, the homologues of which are major stimulators of the immune system during infections with Mycobacterium tuberculosis and Mycobacterium leprae [44
, 45
]. At least two members of the hsp family can signal through TLRs: hsp60 signals through TLR2, and hsp70 can activate inflammatory pathways via TLR2 and TLR4 [46
, 47
], thus leaving open the possibility that P. acnes-derived hsp may play a role in liver sensitization. Finally, the contribution of other TLRs cannot be ruled out in the in vivo recognition of the heat-killed P. acnes used in our experiments. A possible candidate is TLR9, as it recognizes bacterial CpG DNA [25
, 48
]. Furthermore, ligand recognition by TLR9, similar to TLR2, results in MyD88-dependent downstream activation [15
, 25
]. Considering that our results demonstrate TLR2-independent P. acnes priming but MyD88-dependent cytokine induction, a possible role for TLR9 activation should be considered. We found that P. acnes resulted in IFN- induction in splenocytes and TNF- induction in peritoneal macrophages of TLR9–/– mice. This could be explained by recognition of P. acnes by TLR2 in the TLR9–/– system. However, a recent work by Kalis et al. [49
] found no P. acnes-induced priming to LPS in TLR9–/– mice, suggesting a potential role for TLR9 in P. acnes-induced priming. In summary, our results suggested that selective TLR2 activation could not substitute for the complex signals involved in priming of the liver by P. acnes for LPS injury. Although recognition of P. acnes via the TLR2 receptor results in cell activation, TLR2 is not sufficient in the process of priming with Gram-positive P. acnes for LPS-induced liver injury. Additional MyD88-dependent mechanism(s) are likely to play an important role in P. acnes-mediated sensitization to LPS-induced liver injury.
Received August 9, 2004; revised April 8, 2005; accepted July 27, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
- and Fas ligand-mediated hepatotoxic pathways in endotoxin-induced liver injury in mice J. Immunol. 159,3961-3967[Abstract]-mediated apoptosis of hepatocytes followed by inflammatory T-cell-mediated granulomatous hepatitis in mice J. Biomed. Sci. 6,349-356[Medline]-deficient mice by a concomitant reduction of TNF-, IL-12, and IL-18 production J. Immunol. 162,1049-1055 B signaling pathway components J. Immunol. 170,508-519B activation by ethanol in human monocytes Int. Immunol. 11,1781-1790-deficient mice to low-dose lipopolysaccharide-induced acute liver injury Hepatology 35,805-814[CrossRef][Medline] Interferon and granulocyte/monocyte colony-stimulating factor prevent endotoxin tolerance in human monocytes by promoting interleukin-1 receptor-associated kinase expression and its association to MyD88 and not by modulating TLR4 expression J. Biol. Chem. 277,27927-27934 induction by gram-negative bacteria based on STAT4 activation by type I IFN and IL-18 signaling J. Immunol. 169,1665-1668 B activation in mouse hepatocytes stimulated with lipopolysaccharide Infect. Immun. 70,3433-3442, nitric oxide production, shock, and multiple organ injury/dysfunction in the rat Shock 15,135-142[Medline]This article has been cited by other articles:
 |
|
 |
|
  S.-J. Jiang, C.-C. Kuo, M. W. Berry, A. W. Lee, and L. A. Campbell Identification and Characterization of Chlamydia pneumoniae-Specific Proteins That Activate Tumor Necrosis Factor Alpha Production in RAW 264.7 Murine Macrophages Infect. Immun., April 1, 2008; 76(4): 1558 - 1564. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. McCaskill, K. D. Chason, X. Hua, I. P. Neuringer, A. J. Ghio, W. K. Funkhouser, and S. L. Tilley Pulmonary Immune Responses to Propionibacterium acnes in C57BL/6 and BALB/c Mice Am. J. Respir. Cell Mol. Biol., September 1, 2006; 35(3): 347 - 356. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
