Stimulation of the human monocytic cell line Mono Mac 6 with the synthetic lipopeptide (S)-(2,3-bis(palmitoyloxy)-(2RS)-propyl)-N-palmitoyl-(R)-Cys-(S)-Ser(S)-Lys4-OH, trihydrochloride (Pam3Cys) at 10 μg/ml induces a rapid expression of the TNF gene in a TLR2-dependent fashion. Preculture of the cells with Pam3Cys at 1 μg/ml leads to a reduced response after subsequent stimulation with Pam3Cys at 10 μg/ml, indicating that the cells have become tolerant to Pam3Cys. The CD14 and TLR2 expression is not decreased on the surface of the tolerant cells, but rather up-regulated. Analysis of the NF-κB binding in Pam3Cys-tolerant cells shows a failure to mobilize NF-κB-p50p65 heterodimers, while NF-κB-p50p50 homodimers remain unchanged. Pam3Cys-tolerant cells showed neither IκBα-Ser32 phosphorylation nor IκBα degradation but MyD88 protein was unaltered. However, IRAK-1 protein was absent in Pam3Cys-induced tolerance, while IRAK-1 mRNA was still detectable at 30% compared with untreated cells. In contrast, in LPS-tolerized cells, p50p50 homodimers were induced, IRAK-1 protein level was only partially decreased, and p50p65 mobilization remained intact. It is concluded that in Mono Mac 6 monocytic cells, inhibition of IRAK-1 expression at the mRNA and protein levels is the main TLR-2-dependent mechanism responsible for Pam3Cys-induced tolerance, but not for TLR-4-dependent LPS-induced tolerance.

Inflammation may be triggered when specific receptors, known as TLRs, recognize specific, nonself patterns of molecules derived from microbes. Following recognition of (S)-(2,3-bis(palmitoyloxy)-(2RS)-propyl)-N-palmitoyl-(R)-Cys-(S)-Ser(S)-Lys4-OH, trihydrochloride (Pam3Cys)3 by TLR2/TLR1/CD14 and LPS by TLR4/MD-2/CD14, the TLRs bind to the intracellular adaptor protein, MyD88. MyD88 recruits IL-1R-associated kinase-1 (IRAK-1) to this receptor-signaling complex. IRAK-1 is then phosphorylated and activated. Activated IRAK-1 leaves the TLR/MyD88 complex, associates with TNFR-associated factor 6, and activates downstream signaling pathways such as MAPK and (via TGF-β-activated kinase 1 and NF-κB-inducing kinase) the transcription factor NF-κB. TGF-β-activated kinase 1 and NF-κB-inducing kinase are thought to phosphorylate IκB kinase α and β, which in turn leads to the phosphorylation of specific serine residues of IκB (Ser32 and Ser36 of IκBα, or Ser19 and Ser23 of IκBβ). This is followed by ubiquitination of IκB, its degradation, and release of active NF-κB complex (NF-κB-p50p65 or NF-κB-p50p50 dimers). NF-κB is then translocated to the nucleus and activates transcription of the major proinflammatory cytokine TNF and a whole range of other proinflammatory factors (1, 2).

Although a single ligation of TLRs induces responses such as TNF production, repeated ligation will lead to a loss of response, i.e., the cells become tolerant. Tolerance to self and also to nonself is a general phenomenon preventing or diminishing inflammatory responses of the immune system, and involves either deletion of responder cells, down-regulation of the respective receptors, blockade of signal transduction, or induction of suppressive cytokines. In monocytes/macrophages, tolerance to bacterial LPS does not appear to involve deletion nor receptor down-regulation because CD14 receptor expression is up-regulated, as reported in several studies (3, 4, 5). Blockade of signal transduction appears to be the major mechanism of LPS-induced tolerance in that the interruption of the MAPK and NF-κB pathways has been described (6, 7, 8, 9). With respect to the NF-κB pathway, we have described a unique mechanism that involves induction of NF-κB p50p50 homodimers (3, 10). The p50 protein has no trans activation domain, and therefore the p50p50 homodimers cannot trans activate, but they can bind to DNA in the promoter regions of various genes, including TNF. By binding to the cognate DNA sequences, p50p50 can prevent the classical NF-κB (p50p65) heterodimer from binding and trans activating, thus blocking the expression of genes such as TNF (11).

Although Wysocka et al. (12) were still able to induce LPS tolerance in p50 knockout mice, the NF-κB (p50p50) homodimer mechanism of LPS tolerance has been confirmed by many others (13, 14, 15, 16). Also, Bohuslav et al. (17) demonstrated LPS tolerance to be absent in NF-κB p50−/− mice. Finally, Adib-Conquy et al. (18) recently demonstrated that up-regulation of p50p50 homodimers may occur in blood monocytes of patients with severe sepsis.

LPS acts via CD14 and the associated TLR4 (19, 20, 21). In the present study, we have investigated which mechanisms operate in tolerance induced via TLR2 by the synthetic lipopeptide Pam3Cys. Using this pathway of monocyte activation, an entirely different mechanism appears to operate, because mobilization of NF-κB-p50p65 heterodimers is completely blocked. The failure to mobilize classical NF-κB-p50p65 is shown to be due to a blockade at the level of IRAK-1 protein in Pam3Cys-tolerant Mono Mac 6 cells. We also confirm that LPS tolerance operates via induction of NF-κB-p50p50 homodimers and that highly purified (lipoprotein-free) LPS only partially inhibits IRAK-1 protein expression in LPS-tolerant cells.

Hence, in monocytes, different mechanisms of tolerance appear to coexist. This indicates that it is important for the host to provide more than one mechanism to ensure down-regulation of TNF and to prevent the detrimental effects of excessive amounts of proinflammatory cytokines during bacterial infection.

Mono Mac 6 cells (22) were cultured in 24-well plates (Costar, Cambridge, MA) (3 × 105/well) in RPMI 1640 medium (Biochrom, Berlin, Germany) supplemented with 2 mM l-glutamine (043-05030 H; Invitrogen Life Technologies, Gaithersburg, MD), 200 U/ml penicillin, 200 μg/ml streptomycin (043-05140 H; Invitrogen Life Technologies), 1–2× nonessential amino acids (043-01140 H; Invitrogen Life Technologies), and OPI supplement (contains oxalacetic acid, sodium pyruvate, and insulin; O-5003; Sigma-Aldrich, Munich, Germany). The culture medium, as well as all other buffers, were passed through a Gambro U-2000 ultrafiltration column (Gambro Medizintechnik, Planegg-Martinsried, Germany) to deplete inadvertent LPS contamination, and this was followed by addition of low-LPS 10% FCS (Biochrom). All medium were checked for LPS contamination using Endosafe Limulus Gel Clot assay (AE 010; Charles River Laboratories, Sulzfeld, Germany; sensitivity, 0.06 EU/ml). The cells were controlled for absence of Mycoplasma sp. on a weekly basis (for detailed method, see www.monocytes.de/) (23) and using Mycoplasma PCR ELISA (1663925; Roche, Mannheim, Germany). For tolerance induction, Mono Mac 6 cells were precultured in complete culture medium in 24-well plates for 42 h with either 1 μg/ml Pam3Cys (L2000; EMC Microcollections, Tübingen, Germany, www.microcollections.de/) or 20 ng/ml LPS from Salmonella minnesota (L6261; protein content <3%; Sigma-Aldrich). For Western blotting experiments, the highly purified LPS from Salmonella abortus equi (kind gift from C. Galanos (Freiburg, Germany) (24)) and from S. minnesota (L6261) were used. After preculture, Mono Mac 6 cells were washed and stimulated for 10 min (cytoplasmic protein extracts), 1 h (nuclear extracts and RT-PCR), 4 h (transfection experiments), or 5 h (intracellular cytokine staining) with 10 μg/ml Pam3Cys, or with 1 μg/ml LPS.

For detection of cell surface Ags, Mono Mac 6 cells (2 × 105/sample) were stained for 30 min on ice with 10 μg/ml of either anti-CD14 FITC mAbs (clone MY-4; 6603511; Coulter, Krefeld, Germany) or anti-TLR2 Alexa 488 mAbs (clone TL2.1) (25) vs IgG2b FITC (6603034; Coulter) or IgG2a Alexa 488 isotype controls, respectively. For intracellular TNF staining, Mono Mac 6 cells were stimulated (or not) for 5 h with appropriate doses of Pam3Cys in the presence of 10 μg/ml brefeldin A (B-7651; Sigma-Aldrich), then collected, washed in LPS-free PBS, fixed (5 × 105 cells/sample) for 30 min on ice in 200 μl of Cytofix/Cytoperm (2090KZ; BD Pharmingen, Heidelberg, Germany), washed twice, and incubated on ice for an additional 30 min in 50 μl of Perm/Wash buffer (2091KZ; BD Pharmingen) together with 10 μg/ml anti-TNF PE mAbs (clone MAb11; 18645A; BD Pharmingen) or mouse IgG1 PE (20605A; BD Pharmingen). Samples were then washed and resuspended in 200 μl of PBS/0.1% FCS, and cells were acquired immediately and then analyzed using a FACScan flow cytometer equipped with CellQuest software (version 3.1; BD Biosciences, Mountain View, CA). The specificity of TNF staining was determined by incubating the anti-TNF PE mAbs for 10 min at room temperature with a 10-fold molar excess of human rTNF (kindly provided by BASF-Knoll, Ludwigshafen, Germany), followed by addition to the Mono Mac 6 cells, fixation, and permeabilization, as described above. Fluorescence intensity is expressed as specific median (in channels) and determined by subtracting the median fluorescence of the isotype control from the median fluorescence of the anti-TNF mAbs. Percentage of positive cells was determined by subtraction of the negative control histogram from the positive staining histogram.

Fifteen minutes before Pam3Cys stimulation, purified mAbs anti-CD14 (clone MY-4; 6602622; Coulter) and anti-TLR2 (clones TL2.1 and TL2.3), or relevant purified isotype controls IgG2b (6603001; Coulter) and IgG2a (clone UPC 10; M-9144; Sigma-Aldrich) were added (single or in combinations) to the cells (each of mAbs and isotype controls at a final concentration of 10 μg/ml). Mono Mac 6 cells were then stimulated for 5 h with 1 μg/ml Pam3Cys in the presence of brefeldin A. Thereafter, the cells were stained with either PE-conjugated isotype control, or PE-conjugated anti-TNF mAbs, as described above.

Quantitative PCR was performed using the LightCycler system (Roche), according to the manufacturer’s instructions, using the primer pairs, as noted below. In brief, mRNA was isolated and reverse transcribed as for conventional RT-PCR using oligo(dT) as primer and Moloney murine leukemia virus reverse transcriptase (all Applied Biosystems, Heidelberg, Germany). A total of 3 μl of cDNA was used for amplification in the SYBR Green format using the LightCycler-FastStart DNA Master SYBR Green I kit from Roche (2 239 264). Amplification was performed with 5 mM MgCl2 and 40 cycles with 1 s, 95°C; 10 s, 60°C; and 25 s, 72°C. Fluorescent signals generated during the informative log-linear phase are used to calculate the relative amount of template DNA.

The following primers were used: TNF, 5′-CAG AGG GAA GAG TTC CCC AG and 3′-CCT TGG TCT GGT AGG AGA CG (325 bp); IRAK-1, 5′-AAA GGA GGC CTC CTA TGA CC and 3′-ATG ATG CAG AGC TGC CAA G (441 bp); and IRAK-2, 5′-CTG AGG ATG AAC AGG AAG AGG and 3′-CCA GCA CAG GTA AGA CAT TGG (453 bp).

The mRNA expression was shown as fold difference compared with untreated control cells.

The pTNF-1064 luci-β reporter plasmid containing the human TNF 5′ region was obtained by exchanging the mouse β-globin promoter of the pβTATA luci-β reporter plasmid for the HindIII/BglII fragment from the TNF 5′ region pxP2 luciferase construct (3). The resultant pβTATA luci-β reporter plasmid contains 3′ to the luciferase gene the rabbit β-globin intron (26).

Mono Mac 6 cells were transfected with this construct (5 μg of DNA/107 cells), according to Shakhov et al. (27), using DEAE-dextran (62.5 μg/ml; E1210; Promega, Madison, WI), as described in detail previously (10). Cells were then cultured for 42 h with or without Pam3Cys at 1 μg/ml, followed by Pam3Cys stimulation (10 μg/ml) for 4 h. Luciferase activity in cell lysates was determined using a model LB9501 luminometer (Berthold Technologies, Wildbad, Germany) and the Luciferase Assay System (E1500; Promega). The protein concentration was determined by the method of Bradford, using a commercial kit (500-0006; Bio-Rad, Munich, Germany). Luciferase activity was expressed as relative light units/μg of cellular protein.

Nuclear extracts were isolated, according to Dignam et al. (28), in the presence of a protease inhibitors’ mixture and admixed with a 32P-labeled double-stranded oligonucleotide representing the −605 NF-κB motif of the human TNF promoter, as described previously (29). After 15 min of incubation at 21°C, samples were electrophoresed on nondenaturing polyacrylamide gels in 0.25 TBE buffer (22.5 mM Tris borate, 0.5 mM EDTA, pH 8.5). Gels were dried and exposed for 24–48 h at −80°C to x-ray films.

Cytoplasmatic protein extracts (20 μg/lane) were separated after 10 min, as previously described (10), on 4–12% Tris-glycine gels (EC60385; Invitrogen Life Technologies), followed by electroblotting using Xcell SureLock MiniCell & Xcell II Blot Module (E10002; Invitrogen Life Technologies). Blots were reacted with Abs specific for: IRAK-1 (sc-5287), MyD88 (sc-11356), actin (sc-8432) (all from Santa Cruz Biotechnology, Santa Cruz, CA), IκBα (9243), and IκBα-PSer32 (9241) (both from New England Biolabs, Frankfurt/Main, Germany), followed by appropriate goat anti-rabbit/mouse peroxidase-conjugated Abs (Sigma-Aldrich), development with ECL reagent (RPN2106; Amersham, Braunschweig, Germany), and exposure to Hyperfilm (RPN3103; Amersham). The specificity of stainings was corroborated, using as a positive control appropriate cytoplasmic protein extracts provided by the manufacturer. Blots were scanned and than analyzed using the analySIS program (Soft Imaging System, Münster, Germany).

Data were analyzed with the software program SchoolStat (version 1.0.7; WhiteAnt Occasional Publishing, Newton, MA) or GraphPad InStat (GraphPad, San Diego, CA). Significance analysis was performed using a paired Student’s t test, and two-sided p values <0.05 were considered significant.

When the CD14-positive monocytic cell line Mono Mac 6 is stimulated with Pam3Cys at 10 μg/ml, then TNF protein is rapidly expressed within 5 h. This can be detected within the cytoplasm of cells stimulated in the presence of the protein transport blocker, brefeldin A. As shown in Fig. 1 (right panel), intracellular staining with anti-TNF mAbs gives a clear signal in flow cytometry analysis as compared with staining with an isotype control. When the anti-TNF staining is performed in the presence of a 10-fold molar excess of rTNF, then the signal is reduced almost to the level of the isotype control, thereby demonstrating the specificity of the staining. Only a negligible signal is observed in unstimulated Mono Mac 6 cells (Fig. 1, left panel).

FIGURE 1.

Induction of TNF protein by Pam3Cys in Mono Mac 6 cells. Mono Mac 6 cells were either untreated (left panel) or stimulated with Pam3Cys at 10 μg/ml (right panel) for 5 h in the presence of brefeldin A. Cells were permeabilized and stained with either PE-conjugated isotype control, the PE-conjugated anti-TNF mAbs, or the anti-TNF PE mAbs admixed with a 10-fold molar excess of rTNF. One representative example of five is shown.

FIGURE 1.

Induction of TNF protein by Pam3Cys in Mono Mac 6 cells. Mono Mac 6 cells were either untreated (left panel) or stimulated with Pam3Cys at 10 μg/ml (right panel) for 5 h in the presence of brefeldin A. Cells were permeabilized and stained with either PE-conjugated isotype control, the PE-conjugated anti-TNF mAbs, or the anti-TNF PE mAbs admixed with a 10-fold molar excess of rTNF. One representative example of five is shown.

Close modal

In dose-response analysis (Fig. 2), Pam3Cys induces a weak, but clear response at 1 μg/ml (average median fluorescence intensity of 3.2 channels) with a maximum expression at 100 μg/ml (20.3 channels).

FIGURE 2.

Dose-response analysis for Pam3Cys-induced TNF production. Mono Mac 6 cells were cultured for 5 h with various doses of Pam3Cys in the presence of brefeldin A. Cells were permeabilized and stained with either PE-conjugated isotype control or PE-conjugated anti-TNF mAbs. Fluorescence intensity for each dose of Pam3Cys is expressed as specific median (in channels). The average of five experiments ± SD is given. ∗, p < 0.05 as compared with the unstimulated cells.

FIGURE 2.

Dose-response analysis for Pam3Cys-induced TNF production. Mono Mac 6 cells were cultured for 5 h with various doses of Pam3Cys in the presence of brefeldin A. Cells were permeabilized and stained with either PE-conjugated isotype control or PE-conjugated anti-TNF mAbs. Fluorescence intensity for each dose of Pam3Cys is expressed as specific median (in channels). The average of five experiments ± SD is given. ∗, p < 0.05 as compared with the unstimulated cells.

Close modal

We then analyzed the type of receptors involved in Pam3Cys stimulation of Mono Mac 6 cells. In these studies, a set of two anti-TLR2 mAbs (clones TL2.1 and TL2.3) reduced TNF protein expression by ∼50%, as did the anti-CD14 mAb, MY-4 (Fig. 3). The combination of the anti-TLR2 mAbs with the anti-CD14 mAbs reduced Pam3Cys-induced TNF expression to background level, while the addition of isotype controls, in the same final concentration as anti-TLR2 and anti-CD14 mAbs, had no effect. These data indicate that Pam3Cys does use the CD14-TLR2 receptor complex for signal transduction.

FIGURE 3.

Blockade of Pam3Cys-induced TNF production in Mono Mac 6 cells by anti-receptor Abs. Fifteen minutes before Pam3Cys stimulation, mAbs or their combinations (each mAb at final concentration of 10 μg/ml) and relevant isotype controls were added to the cells. Mono Mac 6 cells were then stimulated for 5 h with 1 μg/ml Pam3Cys in the presence of brefeldin A. Thereafter, the cells were permeabilized and stained with either PE-conjugated isotype control or PE-conjugated anti-TNF mAbs. Fluorescence intensity is expressed as specific median and shown as percentage of Pam3Cys stimulation (second column). The average of three experiments ± SD is given. ∗, p < 0.05 as compared with the culture with isotype controls (sixth column).

FIGURE 3.

Blockade of Pam3Cys-induced TNF production in Mono Mac 6 cells by anti-receptor Abs. Fifteen minutes before Pam3Cys stimulation, mAbs or their combinations (each mAb at final concentration of 10 μg/ml) and relevant isotype controls were added to the cells. Mono Mac 6 cells were then stimulated for 5 h with 1 μg/ml Pam3Cys in the presence of brefeldin A. Thereafter, the cells were permeabilized and stained with either PE-conjugated isotype control or PE-conjugated anti-TNF mAbs. Fluorescence intensity is expressed as specific median and shown as percentage of Pam3Cys stimulation (second column). The average of three experiments ± SD is given. ∗, p < 0.05 as compared with the culture with isotype controls (sixth column).

Close modal

We have next asked the question as to whether Pam3Cys can induce a state of tolerance in Mono Mac 6 cells. For this purpose, the cells were first precultured for 42 h in the presence of 1 μg/ml Pam3Cys, and after washing were stimulated (in the presence of brefeldin A) for 5 h with 10 μg/ml Pam3Cys. As expected, Mono Mac 6 cells that were precultured without Pam3Cys (naive cells) gave a strong intracellular TNF signal after Pam3Cys stimulation (Fig. 4, second panel). By contrast, when the cells were precultured in the presence of the low Pam3Cys dose, followed by Pam3Cys (10 μg/ml) stimulation for 5 h, then there was no intracellular TNF expression observed (Fig. 4, fourth panel). This demonstrates that the cells have become tolerant to Pam3Cys stimulation. In an average of three experiments, the Pam3Cys-induced TNF expression was 8.5 ± 0.9 channels in the naive cells, and it was reduced to 0.1 ± 0.7 channels in tolerant cells (p < 0.05).

FIGURE 4.

Induction of tolerance to Pam3Cys at the TNF protein level. Mono Mac 6 cells were precultured for 42 h in the presence of 1 μg/ml Pam3Cys or in medium alone (P3C Pre), and thereafter stimulated or not for 5 h with 10 μg/ml Pam3Cys (in the presence of brefeldin A) (P3C Stim). Then cells were permeabilized and stained with either PE-conjugated isotype control or PE-conjugated anti-TNF mAbs. In naive cells stimulated with Pam3Cys TNF expression was 36.3% with a fluorescence intensity of 9.53 channels (specific median). In tolerant cells stimulated with Pam3Cys, the respective values were 0.12% and 0.69 channels. One representative experiment of three is shown. The averages were 39.5 ± 2.9% and 1.1 ± 1.4% for naive stimulated and tolerant stimulated cells, respectively.

FIGURE 4.

Induction of tolerance to Pam3Cys at the TNF protein level. Mono Mac 6 cells were precultured for 42 h in the presence of 1 μg/ml Pam3Cys or in medium alone (P3C Pre), and thereafter stimulated or not for 5 h with 10 μg/ml Pam3Cys (in the presence of brefeldin A) (P3C Stim). Then cells were permeabilized and stained with either PE-conjugated isotype control or PE-conjugated anti-TNF mAbs. In naive cells stimulated with Pam3Cys TNF expression was 36.3% with a fluorescence intensity of 9.53 channels (specific median). In tolerant cells stimulated with Pam3Cys, the respective values were 0.12% and 0.69 channels. One representative experiment of three is shown. The averages were 39.5 ± 2.9% and 1.1 ± 1.4% for naive stimulated and tolerant stimulated cells, respectively.

Close modal

We then analyzed whether the down-regulation of TNF protein in Pam3Cys tolerance is regulated at the transcriptional level. For this purpose, we performed RT-PCR using the LightCycler technology for real-time monitoring of cDNA amplification. In Fig. 5,A, the SYBR Green fluorescence, as a measure of dsDNA on the y-axis, is plotted vs cycle number on the x-axis. When analyzed at a fluorescence intensity of 3 U, unstimulated Mono Mac 6 cells reached this level at 25.8 cycles, i.e., they express appreciable amounts of TNF-mRNA. After stimulation with 10 μg/ml Pam3Cys, the same level of fluorescence intensity was reached at 20.1 cycles, reflecting an increase of TNF-mRNA by the factor of 52. Preculture of Mono Mac 6 cells with 1 μg/ml Pam3Cys reduced the constitutive TNF-mRNA (3 U at 27.3 cycles), and the stimulation of the precultured cells with 10 μg/ml Pam3Cys increased the expression only by the factor of 4 (3 U at 25.2 cycles). The level reached by the stimulation of the tolerant cells only minimally exceeded the level seen in unstimulated naive cells (Fig. 5,A). The levels for the constitutively expressed α-enolase gene were comparable for all four samples (Fig. 5,B). In an average of three experiments, Pam3Cys-stimulated TNF-mRNA levels were induced ∼45-fold in naive cells and 5-fold in Pam3Cys-tolerant cells, i.e., induction was 9-fold lower (Fig. 5 C).

FIGURE 5.

Induction of tolerance to Pam3Cys at the TNF-mRNA level. Real-time monitoring of TNF-cDNA amplification in tolerant Mono Mac 6 cells by RT-PCR using the LightCycler technology was applied. A and B, The SYBR Green fluorescence was analyzed at a fluorescence intensity level of 3 U as a measure of dsDNA on the y-axis vs cycle number on the x-axis. Mono Mac 6 cells were precultured for 42 h in the presence of 1 μg/ml Pam3Cys or in medium alone (Pre), and thereafter stimulated or not for 1 h with 10 μg/ml Pam3Cys (Stim). Isolated RNA was reverse transcribed and amplified for α-enolase (B) and for TNF (A). One representative experiment of three is shown. C, Average of TNF-mRNA expression ± SD (in relative units) from three independent experiments is shown. ∗, p < 0.05 as compared with the stimulated naive cells (second column).

FIGURE 5.

Induction of tolerance to Pam3Cys at the TNF-mRNA level. Real-time monitoring of TNF-cDNA amplification in tolerant Mono Mac 6 cells by RT-PCR using the LightCycler technology was applied. A and B, The SYBR Green fluorescence was analyzed at a fluorescence intensity level of 3 U as a measure of dsDNA on the y-axis vs cycle number on the x-axis. Mono Mac 6 cells were precultured for 42 h in the presence of 1 μg/ml Pam3Cys or in medium alone (Pre), and thereafter stimulated or not for 1 h with 10 μg/ml Pam3Cys (Stim). Isolated RNA was reverse transcribed and amplified for α-enolase (B) and for TNF (A). One representative experiment of three is shown. C, Average of TNF-mRNA expression ± SD (in relative units) from three independent experiments is shown. ∗, p < 0.05 as compared with the stimulated naive cells (second column).

Close modal

Tolerance was not restricted to TNF, because when looking at IL-8 mRNA by LightCycler technology expression in Pam3Cys-tolerant cells was 10-fold lower (n = 4).

To demonstrate that down-regulation of TNF-mRNA in Pam3Cys tolerance is due to reduced trans activation of the TNF gene, we transfected Mono Mac 6 cells with the pTNF-1064 luci-β reporter plasmid containing the human TNF promoter sequence up-stream of the luciferase reporter gene. Transfected cells were then precultured and stimulated with Pam3Cys. In these studies, in naive cells trans activation was 16-fold, while the Pam3Cys-tolerant cells showed only a 2-fold trans activation (Fig. 6). These data indicate that Pam3Cys tolerance is regulated at the transcriptional level.

FIGURE 6.

Effect of Pam3Cys tolerance induction on trans activation directed by the human TNF promoter. Mono Mac 6 cells were DEAE-dextran transfected with 5 μg of pTNF-1064 luci-β reporter plasmid. Cells were then precultured with Pam3Cys at 1 μg/ml for 42 h, followed by washing (Pre) and stimulation with Pam3Cys at 10 μg/ml for an additional 4 h (Stim). Luciferase activity was then determined in cell extracts. The average of three experiments ± SD is given. ∗, p < 0.05 as compared with the stimulated naive cells (second column).

FIGURE 6.

Effect of Pam3Cys tolerance induction on trans activation directed by the human TNF promoter. Mono Mac 6 cells were DEAE-dextran transfected with 5 μg of pTNF-1064 luci-β reporter plasmid. Cells were then precultured with Pam3Cys at 1 μg/ml for 42 h, followed by washing (Pre) and stimulation with Pam3Cys at 10 μg/ml for an additional 4 h (Stim). Luciferase activity was then determined in cell extracts. The average of three experiments ± SD is given. ∗, p < 0.05 as compared with the stimulated naive cells (second column).

Close modal

We therefore asked the question: which mechanism upstream of transcription might be responsible for the trans activation failure in Pam3Cys tolerance? We first analyzed expression of the two cell surface receptors involved in Pam3Cys-monocyte activation. In this study, after 42-h preculture of Mono Mac 6 cells with 1 μg/ml Pam3Cys, CD14 expression increased dramatically. Furthermore, TLR2 cell surface expression showed an increase from an average of 12 channels to an average of 17 channels (Fig. 7). In parallel, in the LightCycler experiments, we found the CD14-mRNA expression to be increased by the factor of 17 and the TLR2-mRNA expression to be unchanged (data not shown).

FIGURE 7.

Effect of Pam3Cys tolerance induction on expression of cell surface receptors on Mono Mac 6 cells. Mono Mac 6 cells were cultured for 42 h with or without 1 μg/ml Pam3Cys, followed by cytofluorimetric analysis after staining with anti-CD14 mAbs (MY-4-FITC), or anti-TLR2 mAbs (TL2.1-Alexa 488). Fluorescence intensity is expressed as specific median in channels (ch.) and is given in the upper right corner of the respective histograms. In an average of five experiments, CD14 expression increased from 23 ± 8 ch. in naive cells to 426 ± 231 ch. in tolerant cells (p < 0.05), and TLR2 expression increased from 12 ± 5 ch. to 17 ± 4 ch. (p < 0.05), respectively. One representative experiment of five is shown.

FIGURE 7.

Effect of Pam3Cys tolerance induction on expression of cell surface receptors on Mono Mac 6 cells. Mono Mac 6 cells were cultured for 42 h with or without 1 μg/ml Pam3Cys, followed by cytofluorimetric analysis after staining with anti-CD14 mAbs (MY-4-FITC), or anti-TLR2 mAbs (TL2.1-Alexa 488). Fluorescence intensity is expressed as specific median in channels (ch.) and is given in the upper right corner of the respective histograms. In an average of five experiments, CD14 expression increased from 23 ± 8 ch. in naive cells to 426 ± 231 ch. in tolerant cells (p < 0.05), and TLR2 expression increased from 12 ± 5 ch. to 17 ± 4 ch. (p < 0.05), respectively. One representative experiment of five is shown.

Close modal

NF-κB is a crucial transcription factor for the regulation of TNF gene expression. LPS tolerance has been shown to act via NF-κB, either by reduced mobilization or by shifting to p50p50 homodimers. We therefore performed gel-shift analysis using the −605 NF-κB motif from the human TNF promoter (GGGGCTGTCCC). In these studies, we confirmed the intact mobilization of p50p65 heterodimers and the strong induction of p50p50 homodimers in the LPS-tolerant Mono Mac 6 cells (Fig. 8, lane 7). In the case of Pam3Cys tolerance, an entirely different pattern emerged in that there was only a slight increase in p50p50 and a failure to mobilize p50p65 (Fig. 8, lane 4).

FIGURE 8.

Gel-shift analysis of NF-κB proteins in Pam3Cys tolerance Mono Mac 6 cells. Mono Mac 6 cells were precultured with either LPS (S. minnesota, L6261) or Pam3Cys for 42 h, as indicated, and after washing they were stimulated with either LPS or Pam3Cys for 1 h, and nuclear extracts were prepared. They were tested for binding to the −605 NF-κB motif of the human TNF promoter. One of three experiments is shown.

FIGURE 8.

Gel-shift analysis of NF-κB proteins in Pam3Cys tolerance Mono Mac 6 cells. Mono Mac 6 cells were precultured with either LPS (S. minnesota, L6261) or Pam3Cys for 42 h, as indicated, and after washing they were stimulated with either LPS or Pam3Cys for 1 h, and nuclear extracts were prepared. They were tested for binding to the −605 NF-κB motif of the human TNF promoter. One of three experiments is shown.

Close modal

This indicates that the signaling cascade leading from the cell surface receptor to the mobilization of the classical NF-κB p50p65 heterodimer is blocked in Pam3Cys tolerance.

Of the elements of the NF-κB signaling cascade, the MyD88, IRAK-1, IκBα, and IκBα-PSer 32 proteins were studied by Western blotting. As shown in Fig. 9 A, the MyD88 protein expression did not change with Pam3Cys stimulation and during tolerance induction. By contrast, IκBα was degraded upon Pam3Cys stimulation (lane 2), but it remained unchanged in stimulated tolerant cells (lane 4). Serine-phosphorylated IκBα did show an inverse pattern, in that it was only detectable in Pam3Cys-stimulated naive cells (lane 2) and there was no phosphorylation in Pam3Cys-stimulated tolerant cells (lanes 4). These data indicate that elements upstream of IκBα phosphorylation and degradation are blocked in Pam3Cys tolerance.

FIGURE 9.

Western blots for MyD88, IκBα, and IκBα-Ser32 in Pam3Cys and LPS tolerance in Mono Mac 6 cells. Cell extracts for MyD88 and IκBα-Ser32 blots were prepared from tolerized Mono Mac 6 cells after stimulation with 10 μg/ml Pam3Cys or 1 μg/ml LPS (S.m. = S. minnesota, L6261). A, Pam3Cys tolerance; B, LPS tolerance. One of three experiments is shown.

FIGURE 9.

Western blots for MyD88, IκBα, and IκBα-Ser32 in Pam3Cys and LPS tolerance in Mono Mac 6 cells. Cell extracts for MyD88 and IκBα-Ser32 blots were prepared from tolerized Mono Mac 6 cells after stimulation with 10 μg/ml Pam3Cys or 1 μg/ml LPS (S.m. = S. minnesota, L6261). A, Pam3Cys tolerance; B, LPS tolerance. One of three experiments is shown.

Close modal

By contrast, when looking at LPS tolerance in the Mono Mac 6 cells, then IκBα degradation was evident after LPS stimulation both in naive and in tolerant cells (Fig. 9 B).

We then studied IRAK-1 during Pam3Cys tolerance. As shown in Fig. 10, upper panel, IRAK-1 protein was strongly expressed in naive cells. In Pam3Cys-tolerant cells, however, no signal for IRAK-1 was detectable. When analyzing blots by densitometry (Fig. 10, left-hand graph in upper panel), then Pam3Cys-tolerant cells in an average of six experiments gave only a background signal (<5% of the level in naive cells). In contrast, LPS from S. minnesota only partially depleted IRAK-1 protein after preculture (Fig. 10, middle panel). Because this commercial LPS may still contain some lipoprotein, we also looked at a highly purified form of LPS from S. abortus equi (kindly provided by C. Galanos). Also, with this type of LPS, IRAK-1 protein was only moderately reduced in tolerance (Fig. 10, lower panel). Densitometry showed that in an average of three experiments the level of IRAK-1 in tolerant cells was 36% for LPS S. minnesota and 39% for S. abortus equi (+/+ lane in comparison with the naive cells). Thus, in LPS-tolerant cells the IRAK-1 adaptor protein, albeit at lower levels, is still available for signaling.

FIGURE 10.

Western blot for IRAK-1 in Pam3Cys tolerance in Mono Mac 6 cells. Cell extracts were prepared at 10 min of stimulation with 10 μg/ml Pam3Cys or 1 μg/ml LPS in appropriate tolerized cells. LPS was from S. minnesota (L-6261; Sigma-Aldrich) or S. abortus equi (kindly provided by C. Galanos). Blots were reacted with anti-IRAK-1 or anti-actin for control. Blots were analyzed by densitometry and normalized for actin content. The average results ± SD are given in the left-hand panels. Reduction of IRAK-1 protein in stimulated tolerant cells was down to <5% for Pam3Cys, 36.1 ± 19.6% for LPS S. minnesota, and 38.9 ± 11.2% for LPS S. abortus equi. ∗, p < 0.05 as compared with the unstimulated naive cells.

FIGURE 10.

Western blot for IRAK-1 in Pam3Cys tolerance in Mono Mac 6 cells. Cell extracts were prepared at 10 min of stimulation with 10 μg/ml Pam3Cys or 1 μg/ml LPS in appropriate tolerized cells. LPS was from S. minnesota (L-6261; Sigma-Aldrich) or S. abortus equi (kindly provided by C. Galanos). Blots were reacted with anti-IRAK-1 or anti-actin for control. Blots were analyzed by densitometry and normalized for actin content. The average results ± SD are given in the left-hand panels. Reduction of IRAK-1 protein in stimulated tolerant cells was down to <5% for Pam3Cys, 36.1 ± 19.6% for LPS S. minnesota, and 38.9 ± 11.2% for LPS S. abortus equi. ∗, p < 0.05 as compared with the unstimulated naive cells.

Close modal

By contrast, in Pam3Cys tolerance, IRAK-1 protein is completely ablated such that no signaling through this pathway can occur.

Next, we analyzed the time course for IRAK-1 mRNA and protein expression in Mono Mac 6 cells precultured for up to 48 h in the presence of 1 μg/ml Pam3Cys. As shown in Fig. 11 A, the mRNA levels showed some variation over time, but up to the 24-h time point the transcript level was essentially unchanged. Only at 48 h the mRNA decreased to 33.8 ± 8.6% as compared with cells at time 0.

FIGURE 11.

Time course of IRAK-1 expression in Pam3Cys-stimulated Mono Mac 6 cells. Mono Mac 6 cells were stimulated for indicated (0.5–48 h) periods of time with 1 μg/ml Pam3Cys. A, Real-time monitoring of cDNA amplification using the LightCycler technology was applied. Isolated RNA was reverse transcribed and amplified for IRAK-1 and α-enolase as a control. Results were normalized for α-enolase expression. Reduction of IRAK-1 mRNA expression in cells precultured for 48 h was down to 33.8 ± 8.6% in comparison with 0-h, untreated cells. Average of IRAK-1 mRNA expression, in comparison with unstimulated Mono Mac 6 cells, ± SD (in relative units) from five independent experiments is shown; ∗, p < 0.01 as compared with naive cells. B, One representative time course of a Western blot for IRAK-1 protein expression in Pam3Cys-precultured Mono Mac 6 cells is shown. C, Western blots were analyzed by densitometry and normalized for actin content, and the average results from four independent experiments ± SD (in gray relative units), in comparison with 0-h cells, are given. Reduction of IRAK-1 protein content in stimulated cells was down to 48.8 ± 26.5% after 4 h, 37 ± 19.9% after 6 h, 2.5 ± 1.3% after 24 h, and 2.6 ± 1.6% after 48 h of Pam3Cys stimulation; ∗, p < 0.05 as compared with naive cells; ∗∗, p < 0.01 vs naive cells.

FIGURE 11.

Time course of IRAK-1 expression in Pam3Cys-stimulated Mono Mac 6 cells. Mono Mac 6 cells were stimulated for indicated (0.5–48 h) periods of time with 1 μg/ml Pam3Cys. A, Real-time monitoring of cDNA amplification using the LightCycler technology was applied. Isolated RNA was reverse transcribed and amplified for IRAK-1 and α-enolase as a control. Results were normalized for α-enolase expression. Reduction of IRAK-1 mRNA expression in cells precultured for 48 h was down to 33.8 ± 8.6% in comparison with 0-h, untreated cells. Average of IRAK-1 mRNA expression, in comparison with unstimulated Mono Mac 6 cells, ± SD (in relative units) from five independent experiments is shown; ∗, p < 0.01 as compared with naive cells. B, One representative time course of a Western blot for IRAK-1 protein expression in Pam3Cys-precultured Mono Mac 6 cells is shown. C, Western blots were analyzed by densitometry and normalized for actin content, and the average results from four independent experiments ± SD (in gray relative units), in comparison with 0-h cells, are given. Reduction of IRAK-1 protein content in stimulated cells was down to 48.8 ± 26.5% after 4 h, 37 ± 19.9% after 6 h, 2.5 ± 1.3% after 24 h, and 2.6 ± 1.6% after 48 h of Pam3Cys stimulation; ∗, p < 0.05 as compared with naive cells; ∗∗, p < 0.01 vs naive cells.

Close modal

Parallel analysis of the IRAK-1 protein is shown in a representative Western blot in Fig. 11,B, which demonstrates a decrease of the protein beginning at 4 h posttreatment with 1 μg/ml Pam3Cys, and the protein is strongly depleted at 24 and 48 h. At the same point in time, actin bands are unchanged. Quantitation of the IRAK-1 protein by densitometry as compared with the actin expression revealed for an average of four experiments that IRAK-1 was reduced to 50% at 4 h and to 40% at 6 h. At 24 and 48 h, the level was <3% of the level seen in untreated cells at 0 h (Fig. 11 C). These data show that the mRNA prevalence is stable and shows only a moderate decrease after 2 days of preculture with 1 μg/ml Pam3Cys. At the same time, IRAK-1 protein gradually decreases over the first hours to reach background levels only after 1 day. These data indicate that IRAK-1 can be the main point of regulation during tolerance induction after Pam3Cys stimulation.

In the present study, we analyzed whether tolerance induction by Pam3Cys can be achieved in human monocytes and what mechanisms might be operative. We showed that Pam3Cys is an efficient inducer of TNF in the human monocytic Mono Mac 6 cell line and that this induction can be blocked completely by a combination of anti-TLR2 and anti-CD14 Abs. This is consistent with earlier studies that reported Pam3Cys to act via TLR2 (30, 31, 32).

When preincubating Mono Mac 6 cells with a low dose of Pam3Cys for 2 days, followed by stimulation with a high dose of Pam3Cys, intracellular TNF protein accumulation could not be induced anymore. The same pattern was evident at the TNF-mRNA level. The action of Pam3Cys cannot be ascribed to contaminant LPS, because in previous studies we have shown that addition of polymyxin B to Pam3Cys does not affect its TNF-inducing capacity (33). Furthermore, in experiments comparing LPS and Pam3Cys in this study, we show that LPS induces p50 homodimers, while Pam3Cys has little such activity, thus demonstrating that Pam3Cys cannot contain appreciable amounts of LPS.

The high precision and sensitivity of the LightCycler analysis allowed us to also demonstrate a decrease of the basal level of TNF-mRNA in tolerant cells. Stimulation of these tolerant cells by Pam3Cys showed some induction of the TNF-mRNA. This induction was, however, substantially lower than that seen after stimulation of naive cells, and it barely exceeded the basal level found in naive Mono Mac 6 cells (Fig. 5 A). These data clearly show that Pam3Cys tolerance occurs at the transcript level. Similar to what has been observed with LPS tolerance (10), trans activation directed by the TNF promoter is also clearly reduced in Pam3Cys-tolerant cells. This indicates that Pam3Cys tolerance acts by decreasing transcription of the gene rather than by increasing mRNA decay.

We have therefore analyzed crucial upstream elements that are involved in regulation of TNF gene transcription. The earliest step at which tolerance might occur is at the level of the cell surface receptor. In this study, Wang et al. (34) have shown by Western blots in lipoprotein-tolerant cells a down-regulation of TLR2 in cellular lysates. In the present study, CD14 and TLR2 showed strong and moderate cell surface expression in tolerant cells, respectively. This indicates that Pam3Cys tolerance does not act by down-regulating these cell surface receptors, but we have not studied expression of TLR1, which is the partner for TLR2 in Pam3Cys signaling (35). In a study by Nomura et al. (36), murine macrophages tolerant to LPS failed to stain with an Ab that recognizes the complex of TLR4 and MD-2, which suggests that these molecules are down-regulated or that they do not interact anymore. Currently, we cannot exclude the possibility that, in Pam3Cys tolerance, there is a disruption of the interaction of the TLR1, TLR2, and CD14 receptor components.

Downstream of the receptor complex, mechanisms of tolerance may interfere with signaling cascades that lead to mobilization and activation of the transcription factor NF-κB. Alternatively, the NF-κB p50p50 homodimer mechanism may operate. To distinguish between these possibilities, we have performed gel-shift analysis for determination of NF-κB-binding proteins. These studies confirmed that in LPS tolerance, mobilization of p50p65 was intact and p50p50 homodimers were induced, as reported earlier (3, 10). In contrast, Pam3Cys-tolerant Mono Mac 6 cells analyzed in the same experiments showed no change of p50p50, but mobilization of p50p65 was blocked (see Fig. 8). Hence, it appears that tolerance induction by the TLR4 ligand LPS and the TLR2 ligand Pam3Cys operates by different mechanisms, i.e., LPS mainly via the p50p50 homodimer mechanism and Pam3Cys via interruption of the NF-κB signaling cascade.

Recent studies indicate that the IRAK-1 protein may be a crucial switch in the induction of tolerance in monocytes/macrophages (9, 37, 38, 39). It has been shown that IRAK-1 protein is degraded in LPS-tolerant THP-1 cells, and this can be prevented by pretreatment with GM-CSF and IFN-γ (40). Therefore, the level of IRAK-1 protein expression appears to be a main factor influencing activation of IRAK-1-dependent genes.

In our study, Pam3Cys-induced, TLR2-dependent tolerance is associated with a complete depletion of IRAK-1 protein expression in the cytoplasm of Mono Mac 6 cells. Time course analyses have shown that preculture with Pam3Cys at 1 μg/ml will gradually deplete IRAK-1 protein over several hours with a significant decrease first seen at 4 h posttreatment. This slow depletion is consistent with studies by Adib-Conquy and Cavaillon (40), who reported on a clear decrease of IRAK-1 protein only after 6 h of treatment when using a low dose of LPS. Higher doses of a stimulus may result in more rapid depletion of the protein, as seen by Li et al. (9) for LPS in THP-1 cells.

Of note, depletion of IRAK-1 protein did not accompany a depletion of the respective mRNA (Fig. 11, A and C). Adib-Conquy and Cavaillon (40) also noted no decrease of IRAK-1 mRNA when the protein expression decreased. These data suggest that IRAK-1 protein is either less efficiently synthesized or more rapidly degraded in cells tolerized with either Pam3Cys or LPS. Reduced synthesis may be due to translational silencing, as shown for T cell-restricted intracellular Ag-1, which acts on the TNF gene (41). No such silencer has been described for IRAK-1 protein synthesis, as yet. In contrast, given the previous findings of IRAK-1 phosphorylation and ubiquitination (9, 40), which target protein degradation, we assume that degradation is the relevant mechanism. Pulse-chase analysis is, however, required to distinguish between these possibilities.

Although we see a strong depletion of IRAK-1 with induction of Pam3Cys tolerance, this protein is only moderately reduced during LPS tolerance (Fig. 10) such that NF-κB mobilization can still occur upon stimulation with LPS. This is in contrast to other studies (9, 40, 42), which suggested IRAK-1 degradation to be the relevant mechanism in LPS tolerance. In studies not shown, we have seen no change of IRAK-1 protein in THP-1 cells that were pretreated with LPS from Escherichia coli (Sigma-Aldrich; L4391). Our data in Mono Mac 6 and THP-1 suggest that a complete ablation of IRAK-1 by LPS may depend on the type and purity of LPS. We propose that TLR2-dependent IRAK-1 depletion is the specific mechanism operating during Pam3Cys-triggered tolerance induction. This depletion is demonstrated in this study in Mono Mac 6 cells, and also in our studies on THP-1 and primary monocytes. In the latter studies, preculture with Pam3Cys led to an 82% decrease of IRAK-1 protein for THP-1 and an 83% decrease for monocytes (data not shown).

Dobrovolskaia et al. (43) demonstrated a reduced IRAK-1 kinase activity in Pam3Cys tolerance in murine peritoneal macrophages, but IRAK-1 protein levels were reported to be unchanged. This difference to our findings is probably due to differences in species (mouse vs human) and/or cell type (peritoneal cells vs monocytic cells).

Changes in IRAK-1 protein expression and activity after LPS/TLR4-dependent tolerance induction observed in other studies may be partially dependent on the use of commercially available LPS contaminated with other bacterial cell wall compounds, such as lipoproteins (44, 45). For Porphyromonas gingivalis natural lipid A, it was recently shown that contamination by only trace amount of lipopeptide may induce cell activation via TLR2 (46). These lipoproteins may target TLR2 and thereby induce IRAK-1 degradation. In contrast, differences in cell types and treatment protocols may explain the more pronounced IRAK-1 protein depletion seen with LPS in other studies.

The association of TLR2 with other TLRs, i.e., with TLR1 and TLR6 (35, 47, 48), may be of importance for the mechanism of tolerance induction. For example, pretreatment of cells with TLR2/6-dependent mycoplasmal lipopeptide-2 only slightly decreases IRAK-1 protein expression and IRAK-1 phosphorylation (49), while lipopeptide Pam3Cys, which appears to act via TLR2/1, completely depletes IRAK1, as shown in the present study. This depletion can be explained in part by a decrease in IRAK-1-mRNA expression, but with 30% of the mRNA remaining, we assume that additional active protein degradation may substantially contribute to the complete disappearance of IRAK-1 protein in tolerant cells.

Similar to the action of Pam3Cys, triggering via TLR7 with imidazoquinoline R-848 substantially reduces IRAK-1 protein in peritoneal macrophages, although some protein remains (49). Taken together, tolerance induced by triggering different TLRs (TLR2, TLR7, and to some extent TLR4) may target IRAK-1 degradation. IRAK-1 may also be crucial to tolerance without being degraded, e.g., release of IRAK-1 from the TLR5 complex was shown to be blocked in flagellin-induced tolerance (38). Of note, we have recently shown that tolerance induction in monocytes by incubation with tumor cells will also decrease IRAK-1 protein (50), suggesting that the IRAK-1 mechanism may not be restricted to tolerance induced by microbial products.

Upstream of IRAK-1 the MyD88 adaptor may determine tolerance. Medvedev et al. (8) have shown that after LPS tolerance induction there is a failure of MyD88 to be recruited to TLR4. There is the possibility that following LPS stimulation, MyD88 could be alternatively spliced such that IRAK-1 phosphorylation and/or autophosphorylation are inhibited (51, 52). In our experiments, we did not observe a change in MyD88 protein expression, but the splice variant may not readily show in Western blots.

In Pam3Cys tolerance, with the complete disappearance of IRAK-1 protein, it is clear that no signaling can occur through this kinase. It is, however, unclear whether the depletion of IRAK-1 in Pam3Cys tolerance is sufficient to explain the degree of blockade of TNF gene expression. It appears that LPS signaling can still occur in IRAK-deficient mouse cells (53). There may be a role for IRAK-4 and IRAK-M in tolerance after repeated TLR stimulation, as IRAK-4 may be deficient or functionally inactive (52, 54, 55, 56) or IRAK-M may be overexpressed (57, 58). In fact, IRAK-M has been shown to be up-regulated in the RAW264.7 mouse macrophage cell line made tolerant by high doses of peptidoglycan (59).

In our studies, we also observed an up-regulation by 20% of IRAK-2 mRNA in tolerant Mono Mac 6 cells (data not shown). Recently, Ruckdeschel et al. (60) described that Yersinia enterocolitica-induced apoptosis involves a proapoptotic signal delivered through MyD88 and IRAK-2, which potentially targets the Fas-associated death domain protein/caspase-8 apoptotic pathway, whereas IRAK-1 and TNFR-associated factor 6 counteract the bacteria-induced cytotoxic response by signaling macrophage survival. Hence, it is possible that in tolerant monocytes, an imbalance of IRAK-2 over IRAK-1 may lead to apoptosis. Therefore, apoptosis mediated by changes of the IRAK proteins could be another mechanism of tolerance induction in monocytes/macrophages.

In addition to the classical mechanism of tolerance that involves NF-κB p50p50 homodimers and the mechanisms targeting IRAK-1, recent studies have shown a role for suppressor of cytokine signaling 1 in LPS tolerance (61, 62). This may suggest a role for the JAK-STAT pathway in TNF gene expression, but suppressor of cytokine signaling 1 might also act by binding to IRAK-1 (61).

Taken together, our findings demonstrate that in the same type of cell two different mechanisms of tolerance can coexist: 1) induction of p50 homodimers for LPS/TLR4 signals, and 2) depletion of IRAK-1 protein for Pam3Cys/TLR2 signals. This indicates that it may be of importance to the host to ensure down-regulation of TNF and to prevent the detrimental effects of excessive amounts of this cytokine after repeated stimulation, i.e., during a persistent bacterial infection that will involve engagement of more than one type of TLR.

We acknowledge B. W. Kiernan’s excellent support in data management.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by Deutsche Forschungsgemeinschaft Grant Zi 288/1, by Travel Grant POL 98/058 awarded by Internationales Büro des Bundesministerium fur Bildung, Wissenschaft, Forschung und Technologie, and by Polish State Committee for Scientific Research Grant 6 P05A 095 21. M.S. was supported by Alexander von Humboldt-Stiftung.

3

Abbreviations used in this paper: Pam3Cys, (S)-(2,3-bis(palmitoyloxy)-(2RS)-propyl)-N-palmitoyl-(R)-Cys-(S)-Ser(S)-Lys4-OH, trihydrochloride; IRAK, IL-1R-associated kinase.

1
Medzhitov, R..
2001
. Toll-like receptors and innate immunity.
Nat. Rev. Immunol.
1
:
135
.
2
Zhang, G., S. Ghosh.
2001
. Toll-like receptor-mediated NF-κB activation: a phylogenetically conserved paradigm in innate immunity.
J. Clin. Invest.
107
:
13
.
3
Ziegler-Heitbrock, H. W. L., A. Wedel, W. Schraut, M. Ströbel, P. Wendelgass, T. Sternsdorf, P. Bäuerle, J. G. Haas, G. Riethmüller.
1994
. Tolerance to lipopolysaccharide involves mobilization of nuclear factor κB with predominance of p50 homodimers.
J. Biol. Chem.
269
:
17001
.
4
Labeta, M. O., J.-J. Durieux, G. Spagnoli, N. Fernandez, J. Wijdenes, R. Herrmann.
1993
. CD14 and tolerance to lipopolysaccharide: biochemical and functional analysis.
Immunology
80
:
415
.
5
Mathison, J., E. Wolfson, S. Steinemann, P. Tobias, R. Ulevitch.
1993
. Lipopolysaccharide (LPS) recognition in macrophages.
J. Clin. Invest.
92
:
2053
.
6
Durando, M. M., K. E. Meier, J. A. Cook.
1998
. Endotoxin activation of mitogen-activated protein kinase in THP-1 cells; diminished activation following endotoxin desensitization.
J. Leukocyte Biol.
64
:
259
.
7
Kraatz, J., L. Clair, J. L. Rodriguez, M. A. West.
1999
. In vitro macrophage endotoxin tolerance: defective in vitro macrophage MAP kinase signal transduction after LPS pretreatment is not present in macrophages from C3H/HeJ endotoxin resistant mice.
Shock
11
:
58
.
8
Medvedev, A. E., A. Lentschat, L. M. Wahl, D. T. Golenbock, S. N. Vogel.
2002
. Dysregulation of LPS-induced Toll-like receptor 4-MyD88 complex formation and IL-1 receptor-associated kinase 1 activation in endotoxin-tolerant cells.
J. Immunol.
169
:
5209
.
9
Li, L., S. Cousart, J. Hu, C. E. McCall.
2000
. Characterization of interleukin-1 receptor-associated kinase in normal and endotoxin-tolerant cells.
J. Biol. Chem.
275
:
23340
.
10
Kastenbauer, S., H. W. Ziegler-Heitbrock.
1999
. NF-κB1 (p50) is up-regulated in lipopolysaccharide tolerance and can block tumor necrosis factor gene expression.
Infect. Immun.
67
:
1553
.
11
Ziegler-Heitbrock, L..
2001
. The p50-homodimer mechanism in tolerance to LPS.
J. Endotoxin Res.
7
:
219
.
12
Wysocka, M., S. Robertson, H. Riemann, J. Caamano, C. Hunter, A. Mackiewicz, L. J. Montaner, G. Trinchieri, C. L. Karp.
2001
. IL-12 suppression during experimental endotoxin tolerance: dendritic cell loss and macrophage hyporesponsiveness.
J. Immunol.
166
:
7504
.
13
Goldring, C. E., S. Reveneau, D. Pinard, J. F. Jeannin.
1998
. Hyporesponsiveness to lipopolysaccharide alters the composition of NF-κB binding to the regulatory regions of inducible nitric oxide synthase gene.
Eur. J. Immunol.
28
:
2960
.
14
Liu, H., P. Sidiropoulos, G. Song, L. J. Pagliari, M. J. Birrer, B. Stein, J. Anrather, R. M. Pope.
2000
. TNF-α gene expression in macrophages: regulation by NF-κB is independent of c-Jun or C/EBPβ.
J. Immunol.
164
:
4277
.
15
Fujihara, M., S. Wakamoto, T. Ito, M. Muroi, T. Suzuki, H. Ikeda, K. Ikebuchi.
2000
. Lipopolysaccharide-triggered desensitization of TNF-α mRNA expression involves lack of phosphorylation of I-κB-α in a murine macrophage-like cell line, P388D1.
J. Leukocyte Biol.
68
:
267
.
16
Jin, L., D. P. Raymond, T. D. Crabtree, S. J. Pelletier, C. K. Rudy, T. L. Pruett, R. G. Sawyer.
2002
. Preexposure of murine macrophages to CpG-containing oligonucleotides results in nuclear factor κB p50 homodimer-associated hyporesponsiveness.
Surgery
132
:
245
.
17
Bohuslav, J., V. V. Kravchenko, G. C. Parry, J. H. Erlich, S. Gerondakis, N. Macklman, R. C. Ulevitch.
1998
. Regulation of an essential innate immune response by the p50 subunit of NF-κB.
J. Clin. Invest.
102
:
1645
.
18
Adib-Conquy, M., C. Adrie, P. Moine, K. Asehnoune, C. Fitting, M. R. Pinsky, J.-F. Dhainaut, J.-M. Cavaillon.
2000
. NF-κB expression in mononuclear cells of patients with sepsis resembles that observed in lipopolysaccharide tolerance.
Am. J. Respir. Crit. Care Med.
162
:
1877
.
19
Kopp, E. B., R. Medzhitov.
1999
. The Toll-receptor family and control of innate immunity.
Curr. Opin. Immunol.
11
:
13
.
20
Beutler, B..
2000
. Endotoxin, Toll-like receptor 4, and the afferent limb of innate immunity.
Curr. Opin. Microbiol.
3
:
23
.
21
Means, T. K., D. T. Golenbock, M. J. Fenton.
2000
. Structure and function of Toll-like receptor proteins.
Life Sci.
68
:
241
.
22
Ziegler-Heitbrock, H. W. L., E. Thiel, A. Fütterer, V. Herzog, A. Wirtz, G. Riethmüller.
1988
. Establishment of a human cell line (Mono Mac 6) with characteristics of mature monocytes.
Int. J. Cancer
41
:
456
.
23
Ziegler-Heitbrock, H. W. L., R. Burger.
1987
. Rapid removal of mycoplasma from cell lines mediated by a direct effect of complement.
Exp. Cell Res.
173
:
388
.
24
Galanos, C., O. Luderitz, O. Westphal.
1979
. Preparation and properties of a standardized lipopolysaccharide from Salmonella abortus equi (Novo-Pyrexal).
Zentralbl. Bakteriol. Orig. A
243
:
226
.
25
Lien, E., T. J. Sellati, A. Yoshimura, T. H. Flo, G. Rawadi, R. W. Finberg, J. D. Carroll, T. Espevik, R. R. Ingalls, J. D. Radolf, D. T. Golenbock.
1999
. Toll-like receptor 2 functions as a pattern recognition receptor for diverse bacterial products.
J. Biol. Chem.
274
:
33419
.
26
Kastenbauer, S., A. Wedel, M. Frankenberger, T. Wirth, H. W. L. Ziegler-Heitbrock.
1996
. Analysis of promoter activity by polymerase chain reaction amplification of reporter gene mRNA.
Anal. Biochem.
233
:
137
.
27
Shakhov, A. N., M. A. Collart, P. Vassalli, S. A. Nedospasov, C. V. Jongeneel.
1990
. κB-type enhancers are involved in lipopolysaccharide-mediated transcriptional activation of the tumor necrosis factor-α gene in primary macrophages.
J. Exp. Med.
171
:
35
.
28
Dignam, J. D., R. M. Lebovitz, R. G. Roeder.
1983
. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei.
Nucleic Acids Res.
11
:
1475
.
29
Ziegler-Heitbrock, H. W. L., T. Sternsdorf, J. Liese, B. Belohradsky, C. Weber, A. Wedel, R. Schreck, P. Bäuerle, M. Ströbel.
1993
. Pyrrolidine dithiocarbamate inhibits NF-κB mobilization and TNF production in human monocytes.
J. Immunol.
151
:
6986
.
30
Thoma-Uszynski, S., S. M. Kiertscher, M. T. Ochoa, D. A. Bouis, M. V. Norgard, K. Miyake, P. J. Godowski, M. D. Roth, R. L. Modlin.
2000
. Activation of Toll-like receptor 2 on human dendritic cells triggers induction of IL-12, but not IL-10.
J. Immunol.
165
:
3804
.
31
Hirschfeld, M., C. J. Kirschning, R. Schwandner, H. Wesche, J. H. Weis, R. M. Wooten, J. J. Weis.
1999
. Cutting edge: inflammatory signaling by Borrelia burgdorferi lipoproteins is mediated by Toll-like receptor 2.
J. Immunol.
163
:
2382
.
32
Vasselon, T., W. A. Hanlon, S. D. Wright, P. A. Detmers.
2002
. Toll-like receptor 2 (TLR2) mediates activation of stress-activated MAP kinase p38.
J. Leukocyte Biol.
71
:
503
.
33
Belge, K.-U., F. Dayyani, A. Horelt, M. Siedlar, M. Frankenberger, B. Frankenberger, T. Espevik, L. Ziegler-Heitbrock.
2002
. The proinflammatory CD14+CD16+DR++ monocytes are a major source of TNF.
J. Immunol.
168
:
3536
.
34
Wang, J. H., M. Doyle, B. J. Manning, Q. Di Wu, S. Blankson, H. P. Redmond.
2002
. Induction of bacterial lipoprotein tolerance is associated with suppression of Toll-like receptor 2 expression.
J. Biol. Chem.
277
:
36068
.
35
Takeuchi, O., S. Sato, T. Horiuchi, K. Hoshino, K. Takeda, Z. Dong, R. L. Modlin, S. Akira.
2002
. Cutting edge: role of Toll-like receptor 1 in mediating immune response to microbial lipoproteins.
J. Immunol.
169
:
10
.
36
Nomura, F., S. Akashi, Y. Sakao, S. Sato, T. Kawai, M. Matsumoto, K. Nakanishi, M. Kimoto, K. Miyake, K. Takeda, S. Akira.
2000
. Cutting edge: endotoxin tolerance in mouse peritoneal macrophages correlates with down-regulation of surface Toll-like receptor 4 expression.
J. Immunol.
164
:
3476
.
37
Hu, J., R. Jacinto, C. McCall, L. Li.
2002
. Regulation of IL-1 receptor-associated kinases by lipopolysaccharide.
J. Immunol.
168
:
3910
.
38
Mizel, S. B., J. A. Snipes.
2002
. Gram-negative flagellin-induced self-tolerance is associated with a block in interleukin-1 receptor-associated kinase release from Toll-like receptor 5.
J. Biol. Chem.
277
:
22414
.
39
Yeo, S. J., J. G. Yoon, S. C. Hong, A. K. Yi.
2003
. CpG DNA induces self and cross-hyporesponsiveness of RAW264.7 cells in response to CpG DNA and lipopolysaccharide: alterations in IL-1 receptor-associated kinase expression.
J. Immunol.
170
:
1052
.
40
Adib-Conquy, M., J. M. Cavaillon.
2002
. γ 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
.
41
Piecyk, M., S. Wax, A. R. Beck, N. Kedersha, M. Gupta, B. Maritim, S. Chen, C. Gueydan, V. Kruys, M. Streuli, P. Anderson.
2000
. TIA-1 is a translational silencer that selectively regulates the expression of TNFα.
EMBO J.
19
:
4154
.
42
Jacinto, R., T. Hartung, C. McCall, L. Li.
2002
. Lipopolysaccharide- and lipoteichoic acid-induced tolerance and cross-tolerance: distinct alterations in IL-1 receptor-associated kinase.
J. Immunol.
168
:
6136
.
43
Dobrovolskaia, M. A., A. E. Medvedev, K. E. Thomas, N. Cuesta, V. Toshchakov, T. Ren, M. J. Cody, S. M. Michalek, N. R. Rice, S. N. Vogel.
2003
. Induction of in vitro reprogramming by Toll-like receptor (TLR)2 and TLR4 agonists in murine macrophages: effects of TLR “homotolerance” versus “heterotolerance” on NF-κB signaling pathway components.
J. Immunol.
170
:
508
.
44
Hirschfeld, M., Y. Ma, J. H. Weis, S. N. Vogel, J. J. Weis.
2000
. Cutting edge: repurification of lipopolysaccharide eliminates signaling through both human and murine Toll-like receptor 2.
J. Immunol.
165
:
618
.
45
Lee, H. K., J. Lee, P. S. Tobias.
2002
. Two lipoproteins extracted from Escherichia coli K-12 LCD25 lipopolysaccharide are the major components responsible for Toll-like receptor 2-mediated signaling.
J. Immunol.
168
:
4012
.
46
Ogawa, T., Y. Asai, M. Hashimoto, O. Takeuchi, T. Kurita, Y. Yoshikai, K. Miyake, S. Akira.
2002
. Cell activation by Porphyromonas gingivalis lipid A molecule through Toll-like receptor 4- and myeloid differentiation factor 88-dependent signaling pathway.
Int. Immunol.
14
:
1325
.
47
Ozinsky, A., D. M. Underhill, J. D. Fontenot, A. M. Hajjar, K. D. Smith, C. B. Wilson, L. Schroeder, A. Aderem.
2000
. The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between Toll-like receptors.
Proc. Natl. Acad. Sci. USA
97
:
13766
.
48
Takeuchi, O., T. Kawai, P. F. Muhlradt, M. Morr, J. D. Radolf, A. Zychlinsky, K. Takeda, S. Akira.
2001
. Discrimination of bacterial lipoproteins by Toll-like receptor 6.
Int. Immunol.
13
:
933
.
49
Sato, S., O. Takeuchi, T. Fujita, H. Tomizawa, K. Takeda, S. Akira.
2002
. A variety of microbial components induce tolerance to lipopolysaccharide by differentially affecting MyD88-dependent and -independent pathways.
Int. Immunol.
14
:
783
.
50
Mytar, B., M. Woloszyn, R. Szatanek, M. Baj-Krzyworzeka, M. Siedlar, I. Ruggiero, J. Wieckiewicz, M. Zembala.
2003
. Tumor cell-induced deactivation of human monocytes.
J. Leukocyte Biol.
74
:
1094
.
51
Janssens, S., K. Burns, J. Tschopp, R. Beyaert.
2002
. Regulation of interleukin-1- and lipopolysaccharide-induced NF-κB activation by alternative splicing of MyD88.
Curr. Biol.
12
:
467
.
52
Burns, K., S. Janssens, B. Brissoni, N. Olivos, R. Beyaert, J. Tschopp.
2003
. Inhibition of interleukin 1 receptor/Toll-like receptor signaling through the alternatively spliced, short form of MyD88 is due to its failure to recruit IRAK-4.
J. Exp. Med.
197
:
263
.
53
Thomas, J. A., J. L. Allen, M. Tsen, T. Dubnicoff, J. Danao, X. C. Liao, Z. Cao, S. A. Wasserman.
1999
. Impaired cytokine signaling in mice lacking the IL-1 receptor-associated kinase.
J. Immunol.
163
:
978
.
54
Li, S., A. Strelow, E. J. Fontana, H. Wesche.
2002
. IRAK-4: a novel member of the IRAK family with the properties of an IRAK-kinase.
Proc. Natl. Acad. Sci. USA
99
:
5567
.
55
Suzuki, N., S. Suzuki, W. C. Yeh.
2002
. IRAK-4 as the central TIR signaling mediator in innate immunity.
Trends Immunol.
23
:
503
.
56
Suzuki, N., S. Suzuki, G. S. Duncan, D. G. Millar, T. Wada, C. Mirtsos, H. Takada, A. Wakeham, A. Itie, S. Li, et al
2002
. Severe impairment of interleukin-1 and Toll-like receptor signalling in mice lacking IRAK-4.
Nature
416
:
750
.
57
Wesche, H., X. Gao, X. Li, C. J. Kirschning, G. R. Stark, Z. Cao.
1999
. IRAK-M is a novel member of the Pelle/interleukin-1 receptor-associated kinase (IRAK) family.
J. Biol. Chem.
274
:
19403
.
58
Kobayashi, K., L. D. Hernandez, J. E. Galan, C. A. Janeway, Jr, R. Medzhitov, R. A. Flavell.
2002
. IRAK-M is a negative regulator of Toll-like receptor signaling.
Cell
110
:
191
.
59
Nakayama, K., S. Okugawa, S. Yanagimoto, T. Kitazawa, K. Tsukada, M. Kawada, S. Kimura, K. Hirai, Y. Takagaki, Y. Ota.
2004
. Involvement of IRAK-M in peptidoglycan-induced tolerance in macrophages.
J. Biol. Chem.
279
:
6629
.
60
Ruckdeschel, K., O. Mannel, P. Schrottner.
2002
. Divergence of apoptosis-inducing and preventing signals in bacteria-faced macrophages through myeloid differentiation factor 88 and IL-1 receptor-associated kinase members.
J. Immunol.
168
:
4601
.
61
Nakagawa, R., T. Naka, H. Tsutsui, M. Fujimoto, A. Kimura, T. Abe, E. Seki, S. Sato, O. Takeuchi, K. Takeda, et al
2002
. SOCS-1 participates in negative regulation of LPS responses.
Immunity
17
:
677
.
62
Kinjyo, I., T. Hanada, K. Inagaki-Ohara, H. Mori, D. Aki, M. Ohishi, H. Yoshida, M. Kubo, A. Yoshimura.
2002
. SOCS1/JAB is a negative regulator of LPS-induced macrophage activation.
Immunity
17
:
583
.