Initial exposure of monocytes/macrophages to LPS induces hyporesponsiveness to a second challenge with LPS, a phenomenon termed LPS tolerance. Molecular mechanisms responsible for endotoxin tolerance are not well defined. We and others have shown that IL-1R–associated kinase (IRAK)-M and SHIP-1 proteins, negative regulators of TLR4 signaling, increase in tolerized cells. TGF-β1, an anti-inflammatory cytokine, is upregulated following LPS stimulation, mediating its effect through SMAD family proteins. Using a monocytic cell line, THP1, we show that LPS activates endogenous SMAD4, inducing its migration into the nucleus and increasing its expression. Secondary challenge with high dose LPS following initial low-dose LPS exposure does not increase IRAK-M or SHIP1 protein expression in small hairpin (sh)SMAD4 THP-1 cells compared with control shLUC THP1 cells. TNF-α concentrations in culture supernatants after second LPS challenge are higher in shSMAD4 THP-1 cells than shLUC THP1 cells, indicating failure to induce maximal tolerance in absence of SMAD4 signaling. Identical results are seen in primary murine macrophages and mouse embryonic fibroblasts, demonstrating the biological significance of our findings. TGF-β1 treatment does not increase IRAK-M or SHIP1 protein expression in shSMAD4 THP-1 cells, whereas it does so in shLUC THP1 cells, indicating that TGF-β1 regulates IRAK-M and SHIP1 expression through a SMAD4-dependent pathway. Knockdown of endogenous SHIP1 by shSHIP1 RNA decreases native and inducible IRAK-M protein expression and prevents development of endotoxin tolerance in THP1 cells. We conclude that in THP-1 cells and primary murine cells, SMAD4 signaling is required for maximal induction of endotoxin tolerance via modulation of SHIP1 and IRAK-M.

The endotoxin LPS is a major component of the outer cell wall of Gram-negative bacteria and is a potent inducer of inflammation. It stimulates monocytes and macrophages/histiocytes and induces them to release proinflammatory cytokines and other mediators. This results in augmentation of the host immune defense system and helps eliminate bacterial infection (1, 2). Following infection with Gram-negative bacteria, LPS, which forms a complex with LPS-binding protein, binds to the TLR4 receptor complex consisting of the adaptor molecules MyD88 and Toll-IL-1R domain-containing adaptor protein (TIRAP), which then recruits cytoplasmic proteins IL-1R–associated kinase (IRAK)1, IRAK4, and TNFR-associated factor 6. Phosphorylation and degradation of IRAK1 induces ubiquitination of TNFR-associated factor 6 and releases the adaptor protein complex into the cytoplasm, resulting in activation of downstream kinases such as TGF-β–activated kinase 1 (TAK1), inhibitor of κB kinase (IKK), and JNK (3, 4). This leads to activation of NF-κB, resulting in downstream induction of TNF-α, IL-6, and NO. TAK1, which was originally identified as a member of MAPK kinase kinase family mediating TGF-β/bone morphogenetic protein (BMP) signaling, is involved in LPS-induced activation of NF-κB and JNK (5, 6). In the TGF-β signaling pathway, TAK1 deletion leads to impaired NF-κB and JNK activation without impacting SMAD2 activation or TGF-β–induced gene expression (7).

SMAD4 is the common SMAD (co-SMAD) and is the common mediator of signal transduction by TGF-β/BMP superfamily. Signaling by TGF-β family members occurs through type I and type II receptors. The activated type I receptor kinase propagates the signal inside the cell through phosphorylation of receptor-regulated SMADs (R-SMADs: SMAD2/SMAD3 for TGF-β, SMAD1/SMAD5/SMAD8 for BMP). Activated receptor-regulated SMADs form heteromeric complexes with SMAD4. These complexes migrate into the nucleus, where they interact with transcription factors, activators, and corepressors to induce targeted gene expression (8). TGF-β1 is now well established as a potent anti-inflammatory cytokine, which plays a pivotal role in maintaining balanced host responses in immune and nonimmune inflammatory conditions (9, 10). TGF-β1 null mice as well as mice lacking the TGF-β transcription factor SMAD3 show increased TLR4 mRNA expression and increased expression of inflammatory cytokines and NO following LPS stimulation (11).

Endotoxin tolerance is a phenomenon in which previous exposure to a low level of LPS induces a transient period of hyporesponsiveness to subsequent challenge with high-dose LPS. Endotoxin tolerance has been also called deactivation, adaptation, desensitization, anergy, refractoriness, or reprogramming (12,14). The underlying molecular mechanisms for endotoxin tolerance are not clearly resolved. IRAK-M (15), SHIP1 (16), and suppressor of cytokine signaling 1 (SOCS1) (17) participate in negative regulation of LPS response. The expression of all three proteins increases following LPS restimulation in LPS tolerant cells. SHIP1 inhibits LPS- induced activation of MAPKs and cytokine production primarily by its phosphatase activity and in a PI3K-independent mechanism (18). TGF-β expression is also increased by stimulation with LPS. TGF-β is an anti-inflammatory cytokine and plays a critical role in LPS induced tolerance to repeat LPS stimulation. Interestingly, SHIP1 protein increase is mediated by the autocrine activity of LPS-induced production of TGF-β (16). In this study, we examine the regulatory role of SMAD4 in modifying cell signaling molecules SHIP1 and IRAKM involved in induction of endotoxin tolerance.

THP1 and 293T cells were obtained from American Type Culture Collection (Manassas, VA). THP1 cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 10 mM HEPES, 1 mM sodium pyruvate, 4.5 g/l glucose, penicillin (100 U/ml), and streptomycin (100 μg/ml) (Invitrogen, Carlsbad, CA). 293T cells were cultured in low glucose DMEM supplemented with 10% FBS and antibiotics (Invitrogen). Mouse bone marrow derived macrophages and mouse embryonic fibroblast (MEF) cells were gifts from Dr. X. Zhong (Department of Pediatrics, Duke Medical Center, Duke University, Durham, NC). Bone marrow cells from femurs and tibias in the C57BL/6 mice were plated into petri dishes containing RPMI 1640 medium supplemented with 10% FBS, 15% L929 cell-conditioned medium, and antibiotics. After 2–3 d of culture at 37°C in CO2 incubator, nonadherent cells were collected and cultured in fresh medium for another 3–5 d. More than 95% of cells were CD11b+ using flow cytometry analysis. These cells were used in experiments to evaluate the role of Smad4 in endotoxin tolerance. MEF cells are grown in DMEM supplemented with 10% FBS and antibiotics.

For MEF cells, mouse embryos 12.5–14.5 d postcoitum were obtained and after removing limbs, brain, and internal organs, they were minced with a sterile razor blade. The tissue was placed in a 15-ml screw-cap tube containing 10 ml 0.05% trypsin/0.02% EDTA in PBS, and the tube was rotated end-over-end at 37°C for 10 min. A 5-ml aliquot from the incubated tube was removed, and another 5 ml fresh trypsin/EDTA was added to the original tube for 10 min. The procedure was repeated at least five times. The cell pellet was resuspended in 50 ml DMEM with 10% FBS, and the cells plated on 10-cm sterile plastic tissue culture dishes. The next day, the medium was changed, and the cells maintained in culture until confluent. The cells were split 1:10 and allowed to grow to confluency. MEF cells were maintained in DMEM supplemented with 10% heat inactivated FBS, 0.1 mmol/l nonessential amino acids, 0.1 μmol/l 2-ME, and antibiotics at 37°C in an atmosphere of 95% air and 5% CO2.

Rabbit anti-human IRAK-M and IRAK-1 Abs were purchased from ProSci (Poway, CA). Mouse anti-human SHIP1 or SMAD4 Abs and SMAD4 siRNA were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Donkey anti-rabbit F(ab′)2 and anti-mouse F(ab′)2 Abs conjugated with HRP were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA). The sources of other Abs are as follows: rabbit anti–phospho-AKT and SMAD2 Abs; mouse anti-AKT and SMAD2 Abs (Cell Signaling Technology, Danvers, MA), mouse anti–β-actin Ab (Sigma, Milwaukee, WI). LPS was obtained from Sigma (catalog number L6529). TNF-α ELISA kits were purchased from R&D Systems (Minneapolis, MN). Western Lightning Chemiluminescence Reagent was from PerkinElmer (Boston, MA).

Cells were lysed in cell lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 1% Triton X-100, and 0.1% SDS [pH 7.5]) containing proteinase inhibitor mixture (Roche, Basel, Switzerland) on ice for 30 min. The lysates were clarified at 4°C for 30 min. at 14,000 rpm/min. Protein concentrations were determined using Bio-Rad Protein Assay at OD595nm (Bio-Rad, Hercules, CA). Equivalent amounts of protein for each sample were subjected to SDS-PAGE and transferred to the polyvinylidene difluoride membrane (Bio-Rad). After blocking with 5% nonfat dry milk/PBS, the membranes were incubated with a primary Ab for 2 h, washed with TTBS (100 mM Tris [pH 7.5], NaCl 0.9%, and Tween 0.1%) three times, reacted with a secondary Ab for 45 min, and washed with Tween 20 and Tris-buffered saline three times. Protein bands were visualized by ECL (PerkinElmer Life Sciences, Waltham, MA). For nuclear protein extraction, we used the procedure described by Xu et al. (19). For densitometric analysis, the ImageJ software from NIH was employed.

The complementary oligonucleotides to generate the small hairpin (sh)RNA against human SMAD4, IRAK-M, and SHIP1 were designed using BLOCK-iT RNAi Designer from Invitrogen. Oligo sequences used were as follows: shSMAD4-RNA, 5′-GCATAGTTTGATGTGCCATAG-3′; shIRAK-RNA, 5′-GCAGTATATCAAGTGCAAACA-3′; and shSHIP1-RNA, 5′-GCCACATCTGTACTGACAACG-3′.

Double-stranded oligos were subcloned into pLKOpuro1 (20). pLKOpuro1-shLUC was used as a control. pLKOpuro1-shLUC and pLKOpuro1 plasmids were a gift from Drs. J. Yang and R.A. Weinberg (Whitehead Institute for Biomedical Research, Cambridge, MA). pLKOpuro1-shLUC and pLKOpuro1 targeted shRNA were cotransfected with pCMV-VSVG and pHR’8.2ΔR (Weinberg) into 293T cells. After 48 h, medium was collected and centrifuged for 10 min at 2000 rpm/min. Supernatants were transferred into targeted cell lines, and these infected cell lines were selected with 1 μg/ml puromycin for 2 wk.

A total of 1 × 107 THP1 cells from a 75-cm2 flask were split into two 175-cm2 flasks with 50 ml fresh medium and cultured for 2 additional days. On the afternoon of the second day, LPS at a final concentration of 10 ng/ml was added into one of two flasks. On the morning of the third day, LPS at a final concentration of 100 ng/ml was added into each of two flasks. In this study, we specify some terminologies: 100 ng/ml LPS stimulation stands for one-time stimulation without low dose of LPS priming; 10–100 ng/ml stands for 100 ng/ml treatment following 10 ng/ml LPS priming overnight. Cells were harvested at a specified period, and supernatants were stored for ELISA analysis. TNF-α concentrations in cultured medium were determined by using human TNF-α/TNFSF1A kit (R&D Systems). The OD of each well was read by using a microplate reader at 450 nm with 540-nm correction (Molecular Devices, Sunnyvale, CA). Each sample was executed in triplicates. Each experiment was repeated at least three times.

THP1 xells were grown on chamber slides (BD Discovery Labware, Bedford, MA) in growth medium supplemented with PMA (1 μg/ml) for 24 h at 37°C. The adherent cells were cultured in serum-free medium overnight, followed by adding LPS for 3–6 h. After three washes with PBS, cells were fixed in 4% paraformaldehyde/PBS for 15 min, and permeabilized with 0.2% Triton X-100/PBS for 20 min and preblocked in 5% BSA/PBS for 1 h. The slides were then incubated with mouse anti-human SMAD4 Ab (diluted in 1/100 in blocking solution) for 1 h, washed three times with PBS, and incubated with FITC-conjugated anti-mouse Ab (Jackson ImmunoResearch Laboratories) for 1 h. After additional washes, the slides were incubated with DAPI for 3 min, and mounted with antifade solution. Cells were examined under the Zeiss LSM510 inverted confocal microscopy (Light Microscopy Core Facility, Duke University).

In this study, we used THP1 cells as a model to define the role of SMAD4 in the development of endotoxin tolerance. THP1 cells were treated with low-dose LPS (10 ng/ml) overnight and subsequently stimulated with 100 ng/ml LPS for 0, 3, 6, and 24 h. This activation protocol is referred to as “10–100 ng/ml.” Control cells were treated with 100 ng/ml LPS for similar time points but without prior stimulation with low dose LPS. This activation protocol is referred to as “100 ng/ml.” Assessment of TNF-α secretion by ELISA showed that TNF-α did not increase over time in low-dose LPS-primed cells compared with unprimed cells, indicating induction of tolerance by pretreatment with low-dose LPS (Fig. 1A). Cytosolic SMAD4 protein level decreased at 3 and 6 h and increased at 24 h after 100 or 10–100 ng/ml LPS stimulation. At similar time points, nuclear SMAD4 protein level, like p65, increased with the above treatments (Fig. 1B). To further confirm SMAD4 activation, we evaluated the levels of phospho-SMAD2, which forms heteromeric activation complex with SMAD4 during TGF-β signaling. We found that phospho-SMAD2 increased following LPS treatment (Fig. 1C). Confocal microscopy confirmed nuclear localization of SMAD4 after LPS stimulation (Fig. 2). Endotoxin tolerance in THP1 cells can be induced by pretreatment with low-dose LPS and SMAD4 is activated following LPS stimulation. When cells pretreated with low-dose LPS (10 ng/ml tolerance model) were restimulated with higher dose of LPS, there was no increase in nuclear SMAD4, but rather SMAD4 was already in the nucleus. Similarly, p65 is already in the nucleus when cells are pretreated with 10 ng/ml LPS. These findings suggest an important role of SMAD4 in tolerance. In the tolerance model (10 ng pretreatment), SMAD4 and p65 are already in the nucleus, and activation of SMAD4 and p65 is sustained following 100 ng/ml restimulation.

Both PI3K and NF-κB pathways play important roles in LPS-induced endotoxin tolerance development (18). To explore the role SMAD4 plays in cell signaling following LPS stimulation or restimulation, we knocked down endogenous SMAD4 by using shSMAD4 RNA technology. Western blot analysis showed that SMAD4 shRNA successfully knocked down endogenous SMAD4 compared with control shLUC (Fig. 3A). SHIP-1 and IRAK-M levels were also decreased in shSMAD4 THP-1 cells. To investigate p-AKT and p-IκBα phosphorylation activities, both shLUC and shSMAD4 THP1 cells were treated with 100 or 10–100 ng/ml LPS for 0, 10, 30 and 60 min. Both p-AKT and p-IκBα levels were sustained or slightly decreased in shSMAD4 THP1 following second exposure to LPS, whereas they decreased considerably with second LPS treatment in shLUC THP1 cells (Fig. 3B). LPS-induced degradation of total IκB-α protein was enhanced in shSMAD4 THP1 compared with shLUC THP1 cells following second exposure to LPS. These results demonstrate that both p-AKT and p-IκB-α continue to be activated in shSMAD4 THP1 cells following repeated exposure to LPS.

To investigate the role SMAD4 plays in endotoxin tolerance, both shLUC and shSMAD4 THP1 cells were treated with 100 or 10–100 ng/ml LPS for 0, 1, 3, 6, and 24 h. TNF-α concentrations, measured by ELISA, were significantly higher in the shSMAD4 THP1 cells than in the shLUC THP1 cells at the same time points (Fig. 4A). The abrogation of SMAD4 expression by shRNA in THP1 cells resulted in higher levels of TNF-α release following 100 and 10–100 ng/ml LPS stimulation compared with control, which indicates a partial abrogation of tolerance.

We investigated whether SMAD4, the co-SMAD, which mediates signal transduction by TGF-β/BMP superfamily (8) also regulates SHIP1 and IRAK-M in tolerized cells. Endogenous SMAD4 was knocked down with small hairpin (sh)SMAD4 RNA. We found that both SHIP1 and IRAK-M protein expressions were reduced in SMAD4 knockdown THP1 cells compared with control shLUC THP1 cells (Fig. 3A). Further studies showed that both SHIP1 and IRAK-M were not induced upon second exposure to LPS (Fig. 4B), whereas they were strongly induced in shLUC following similar stimulation with LPS. These data demonstrate that SMAD4 plays an important role in endotoxin tolerance development through regulation of SHIP1 and IRAK-M.

As shown above, abrogating SMAD4 signaling with shSMAD4 in THP1 cells prevented the development of LPS-induced endotoxin tolerance. To test whether TGF-β–induced upregulation of SHIP1 and IRAK-M is dependent on SMAD4, shSMAD4, and shLUC THP1 cells were treated with TGF-β1 (1 ng/ml) for 0, 3, 6, and 24 h. Western blot analysis showed decreased induction of both SHIP1 and IRAK-M protein expressions after TGF-β treatment in shSMAD4 THP1 cells compared with shLUC THP1 cells (Fig. 5). The above results indicate that TGF-β–induced upregulation of SHIP-1 and IRAK-M is dependent on SMAD4.

To confirm the induction of SMAD4, SHIP1 and IRAK-M by LPS or TGF-β in primary cells, mouse bone marrow-derived macrophages were treated with 10 ng/ml LPS for 0, 6, and 24 h. For TGF-β stimulation assay, the mouse bone marrow-derived macrophages were starved in serum-free medium overnight and subsequently stimulated with 10 ng/ml TGF-β for 0, 6, and 24 h. Western blot analysis was used to measure the above indicated protein expressions. The results showed that SMAD4, IRAK-M, and SHIP1 protein expressions increase upon LPS or TGF-β stimulation in primary mouse macrophages (Fig. 6A).

Furthermore, the SMAD4 gene of the mouse bone marrow-derived macrophages and mouse embryonic fibroblasts was knocked down with SMAD4 small interfering RNA (siRNA), and control siRNA was used as control. The cells were treated with low-dose LPS (10 ng/ml) overnight and subsequently stimulated with 100 ng/ml LPS for 0, 6, and 24 h. The cells were also treated with 100 ng/ml LPS for 0, 6, and 24 h as control. The indicated protein levels were analyzed by Western blot (Fig. 6B). Induced IRAK-M and SHIP1 protein expression levels were increased following priming with low dose LPS and subsequent stimulation with high dose LPS in control siRNA murine bone macrophages but unchanged upon LPS stimulation in the SMAD4 siRNA murine bone marrow macrophages. Similar results were obtained in control siRNA MEF cells and SMAD4 siRNA MEF cells (data not shown).

To study the role SMAD4 plays in endotoxin tolerance in primary cells, subconfluent MEF cells were transfected with siSMAD4 or control siRNA. The transfected MEF cells were respectively treated with 100 and 10–100 ng/ml LPS for 0, 6, and 24 h. LPS-induced TNF-α release from primary cells was measured by ELISA. Western blot analysis showed that endogenous SMAD4 was knocked down in MEF cells (data not shown). TNF-α secretion was significantly higher in the SMAD4 siRNA MEF cells than in the control siRNA MEF cells in both nontolerized (100 ng/ml) and tolerized (10–100 ng/ml) MEF cells (Fig. 6C). This result further demonstrates that SMAD4 is a negative regulator of proinflammatory cytokines, not only in THP-1 cells but also in primary murine cells.

LPS-induced upregulation of SHIP1 is essential for endotoxin tolerance (16). SHIP1 is a phosphatase that hydrolyzes phosphatidylinositol-3,4,5-triphosphate to phosphatidylinositol-3,4-biphosphate. To verify whether this occurs in THP1 cells, we knocked down the endogenous SHIP1 by using shSHIP1 RNA. Western blot analysis showed that shSHIP1 RNA successfully knocked down SHIP1 protein expression in THP1 cells (Fig. 7A). p-AKT level is upregulated in shSHIP1 cells following LPS stimulation (Fig. 7B). Consistent with others (16), TNF-α secretion significantly increases in shSHIP1 cells after restimulation with LPS than in shLuc cells at similar time points (Fig. 7C). These results confirm that SMAD4 regulated SHIP1 plays an essential role in endotoxin tolerance development.

Expression of both SHIP1 and IRAK-M expressions increases in tolerized cells. To explore whether SHIP1 regulates IRAK-M ex-pression in monocytes, we blotted cell lysates from both shSHIP1 and shLUC THP1 cells with IRAK-M Ab. Western blot analysis shows that IRAK-M expression is reduced in shSHIP1 cells (Fig. 8A). IRAK-M expression was not induced following LPS stimulation in shSHIP1 cells compared with control cells (Fig. 8A). TGF-β stimulation in THP1 cells increased both SHIP1 and IRAK-M expression. To further determine whether the induced IRAK-M expression is dependent on shSHIP1 expression, we treated both shSHIP1 and shLUC cells with TGF-β. Induced IRAK-M expression is more marked in shLUC cells following treatment with TGF-β than in shSHIP1 cells (Fig. 8B). These data indicate that SHIP1 regulates IRAK-M expression in endotoxin tolerance development and that this is enhanced via a TGF-β–dependant mechanism.

Endotoxin tolerance has also been termed hyporesponsiveness, deactivation or desensitization. Tolerized cells produce less proinflammatory cytokines and NO in response to a second dose of LPS. However, LPS tolerance is not a global downregulation of signaling proteins and mediators. LPS-tolerant animals and cells can still respond to further LPS challenge and express anti-inflammatory proteins, including IL-10, IRAK-M, SHIP1, SOCS1, and TGF-β (9, 1517, 21, 22). TGF-β1 inhibits LPS-induced NF-κB activation and TNF-α release in mouse RAW264.7 cells and microglial cells (23, 24). In vitro tolerance of human monocytes can be partially mimicked by IL-10 and TGF-β, and the use of anti–IL-10 and anti–TGF-β Abs during the step of tolerization can prevent the phenomenon of endotoxin tolerance (25). Clearly TGF-β is an anti-inflammatory cytokine. However, LPS activates TAK1, which can be activated by TGF-β. SMAD4 is the co-SMAD mediating signal transduction by TGF-β/BMP superfamily. We aimed to determine whether upregulation of SHIP1 and IRAK-M are dependent on SMAD4.

In this study, we demonstrate that upon LPS stimulation, SMAD4 is translocated from the cytosol into the nucleus within 3 h. Also, there is an upregulation of SMAD4 expression upon stimulation with 100 or 10–100 ng/ml LPS within a period of 24 h. The abrogation of SMAD4 expression resulted in higher level of TNF-α release following 100 or 10–100 ng/ml LPS stimulation compared with control cells, which indicates a partial failure of induction of endotoxin tolerance and highlights the critical role of SMAD4 signaling in this phenomenon. LPS-induced increase in SHIP1 is mediated by autocrine-activity of TGF-β (16). Our studies show that both SHIP1 and IRAK-M expression are reduced in the quiescent shSMAD4 cells and showed decreased induction following 100 or 10–100 ng/ml LPS stimulation in shSMAD4 THP1 compared with shLUC THP1 cells. SHIP1 is a negative mediator of AKT activities. The second exposure to LPS leads to the reduced phosphorylation of AKT and IκBα in control shLUC cells, but not in shSMAD4 THP1 cells (Fig. 3B), because of reduced SHIP1 in shSMAD4 cells. Total IκBα degradation is much faster in shSMAD4 THP1 cells than in shLUC THP-1 cells. In vitro TGF-β upregulation of SHIP1 and IRAK-M takes place in shLUC but not in shSMAD4 THP1 cells following TGF-β treatment (Fig.5). Induction of IRAK-M by TGF-β is a novel finding as is the fact that it is partially through a SMAD4-dependent pathway. Thus, SMAD4 negatively regulates LPS signaling through upregulation of both SHIP1 and IRAK-M expression.

Taken together, AKT is activated in THP1 cells (nontolerized) upon the first exposure to LPS and AKT is inactivated in LPS restimulated cells (tolerized). AKT activation is retained in both non-tolerized and tolerized shSMAD4 cells, along with higher TNF-α production. It has been reported that AKT promotes NF-κB activation and inhibition of PI3K decreases LPS-induced transcriptional activity of NF-κB (2628). This is in contrast to other reported data (29, 30). In their studies, AKT dampens NF-κB activation and subsequent production of proinflammatory cytokines. It is unclear how AKT can mediate these distinctly opposing effects on NF-κB activation. Perhaps different cell types and LPS origins or doses may contribute to these differences. Clearly, AKT activation, p38 phosphorylation, and NF-κB activation are diminished in tolerized cells. As a result, tolerized cells produce less proinflammatory cytokines and NO in response to a second dose of LPS. In the meantime, anti-inflammatory proteins, including SHIP1, IRAK-M, and SOCS1, are produced by tolerized cells. It is known that bone marrow-derived macrophages and mast cells show increased TGF-β expression in response to LPS, which increases SHIP1, IRAK-M, and SOCS1 expression (1517). SHIP1 negatively regulates PI3K and therefore inactivates AKT.

Consistent with data by Sly et al. (16), we observe that THP1 cells with knockdown of SHIP1 expression do not develop endotoxin tolerance following a second exposure to LPS. shSHIP1 cells have reduced IRAK-M expression. IRAK-M expression is much less induced following 100 or 10–100 ng/ml LPS in shSHIP1 cells than in shLUC THP1 cells. In vitro TGF-β upregulation of IRAK-M takes place in shLUC but not in shSHIP THP1 cells following TGF-β treatment (Fig. 8B). These findings indicate that SMAD4 regulates SHIP1 expression, which in turn controls IRAK-M expression.

IRAK-M expression is upregulated in both nontolerized and tolerized cells (15, 31). In the nontolerized cells, IRAK-M expression, at least in part, depends on the activation of Tpl2/ERK and PI3K/AKT1 signaling pathways (32). In tolerized cells, both ERK1/2 and AKT activation are dramatically reduced, but IRAK-M expression is retained and elevated. Other pathways involving SHIP1, SOCS1, or other anti-inflammatory proteins might play important roles. Knockdown of SMAD4 by shRNA reduces both SHIP1 and IRAK-M expression. Interestingly, knockdown of SHIP1 by shRNA results in significant reduction of IRAK-M expression induced by TGF-β and LPS. How SHIP1 regulates IRAK-M expression needs to be explored. The levels of IRAK-M expression in PI3K inhibitor pretreated cells rose more rapidly and reached higher levels after 6 h compared with levels in cells that were not pretreated (31). SHIP1 induction takes place in both nontolerized and tolerized cells. Overexpressed SHIP1 inactivates AKT and increases IRAK-M protein level. In contrast, activated AKT contributes to IRAK-M reduction. Hence, reduced SHIP1 by shRNA results in diminished IRAK-M expression, increasing proinflammatory cytokine secretion.

SMAD4 is probably induced both by TLR4 signaling as well as by autocrine production of TGF-β. We show that stimulation of THP-I cells with LPS (Fig. 4B) or with TGF-β (Fig. 5) results in induction of SMAD4. We also shows that primary murine cells, stimulated with either LPS or TGF-β, show induction of SMAD4 (Fig. 6A). To study the role SMAD4 plays in endotoxin tolerance in primary cells, LPS-induced TNF-α release from primary cells was measured by ELISA. TNF-α secretion was significantly higher in the SMAD4 siRNA MEF cells than in the control siRNA MEF cells in both non-tolerized (100 ng/ml) and tolerized (10–100 ng/ml) cells (Fig. 6C). This result further demonstrates that SMAD4 is a negative regulator of proinflammatory cytokines, not only in THP-1 cells but also in primary mouse cells.

On the basis of the above studies, we demonstrate that TGF-β, through SMAD4, negatively regulates LPS signaling during endotoxin tolerance development. In contrast, there are reports of elevated levels of circulating TGF-β in patients with sepsis syndrome (33). Hepatic overexpression of TGF-β1 promotes LPS-induced inflammatory cytokine secretion by liver cells and endotoxin-induced shock (34). Furthermore, there have been reports of increased mortality, blunted production of NO, and increased production of TNF-α in endotoxemic TGF-β1 transgenic mice (35). These results indicate that TGF-β1 promotes LPS-signaling and releases various cytokines that can lead to septic shock in patients. TAK1, which mediates TGF-β/BMP signaling, is phosphorylated concomitant with its activation in LPS-stimulated macrophages and its activity is necessary for the activation of NF-κB (3, 36). The effects of TGF-β on macrophages can be either stimulatory or inhibitory, depending on the other cytokines present and the state of differentiation or tissue origin of the cells (14). Therefore, the clinical administration of exogenous TGF-β1 to induce endogenous production of TGF-β1 may require careful in vivo studies to evaluate the utility of this cytokine in the clinical management of septic shock.

In conclusion, we provide the first evidence that SMAD4 regulates SHIP1 and IRAK-M expression during LPS-induced endotoxin tolerance development. In addition to inhibiting LPS-induced PI3K activation, SHIP1 regulates IRAK-M expression, which controls TNF-α release. These results suggest that manipulation of SMAD4 expression or properly targeting TGF-β pathway might provide novel therapeutic strategies to modulate response to repeated in vivo exposure to LPS in patients with sepsis.

Disclosures The authors have no financial conflicts of interest.

This work was supported by National Institutes of Health/National Institute on Aging Grant R21 AG 024208-02 (to S.A.L.-D.).

Abbreviations used in this paper:

BMP

bone morphogenetic protein

co-SMAD

common SMAD

IKK

inhibitor of kB kinase

IRAK

IL-1R-associated kinase

MEF

mouse embryonic fibroblast

sh

small hairpin

shLUC

control shRNA

shSHIP1

shSHIP1 RNA

shSMAD4

shSMAD4 RNA

siRNA

small interfering RNA

SOCS1

suppressor of cytokine signaling 1

TAK1

TGF-β–activated kinase 1

TIRAP

Toll-IL-1R domain-containing adaptor protein.

1
Ulevitch
R. J.
,
Tobias
P. S.
.
1995
.
Receptor-dependent mechanisms of cell stimulation by bacterial endotoxin.
Annu. Rev. Immunol.
13
:
437
457
.
2
West
M. A.
,
Heagy
W.
.
2002
.
Endotoxin tolerance: a review.
Crit. Care Med.
30
(
Suppl. 1
)
S64
S73
.
3
Irie
T.
,
Muta
T.
,
Takeshige
K.
.
2000
.
TAK1 mediates an activation signal from Toll-like receptor(s) to nuclear factor-κB in lipopolysaccharide-stimulated macrophages.
FEBS Lett.
467
:
160
164
.
4
Deng
L.
,
Wang
C.
,
Spencer
E.
,
Yang
L.
,
Braun
A.
,
You
J.
,
Slaughter
C.
,
Pickart
C.
,
Chen
Z. J.
.
2000
.
Activation of the IκB kinase complex by TRAF6 requires a dimeric ubiquitin-conjugating enzyme complex and a unique polyubiquitin chain.
Cell
103
:
351
361
.
5
Yamaguchi
K.
,
Shirakabe
K.
,
Shibuya
H.
,
Irie
K.
,
Oishi
I.
,
Ueno
N.
,
Taniguchi
T.
,
Nishida
E.
,
Matsumoto
K.
.
1995
.
Identification of a member of the MAPKKK family as a potential mediator of TGF-β signal transduction.
Science
270
:
2008
2011
.
6
Wang
C.
,
Deng
L.
,
Hong
M.
,
Akkaraju
G. R.
,
Inoue
J.
,
Chen
Z. J.
.
2001
.
TAK1 is a ubiquitin-dependent kinase of MKK and IKK.
Nature
412
:
346
351
.
7
Shim
J. H.
,
Xiao
C.
,
Paschal
A. E.
,
Bailey
S. T.
,
Rao
P.
,
Hayden
M. S.
,
Lee
K. Y.
,
Bussey
C.
,
Steckel
M.
,
Tanaka
N.
, et al
.
2005
.
TAK1, but not TAB1 or TAB2, plays an essential role in multiple signaling pathways in vivo.
Genes Dev.
19
:
2668
2681
.
8
ten Dijke
P.
,
Hill
C. S.
.
2004
.
New insights into TGF-β-Smad signalling.
Trends Biochem. Sci.
29
:
265
273
.
9
Letterio
J. J.
,
Roberts
A. B.
.
1998
.
Regulation of immune responses by TGF-β.
Annu. Rev. Immunol.
16
:
137
161
.
10
Letterio
J. J.
2005
.
TGF-β signaling in T cells: roles in lymphoid and epithelial neoplasia.
Oncogene
24
:
5701
5712
.
11
McCartney-Francis
N.
,
Jin
W.
,
Wahl
S. M.
.
2004
.
Aberrant Toll receptor expression and endotoxin hypersensitivity in mice lacking a functional TGF-β1 signaling pathway.
J. Immunol.
172
:
3814
3821
.
12
Cavaillon
J. M.
,
Adrie
C.
,
Fitting
C.
,
Adib-Conquy
M.
.
2003
.
Endotoxin tolerance: is there a clinical relevance?
J. Endotoxin Res.
9
:
101
107
.
13
Cavaillon
J. M.
,
Adib-Conquy
M.
.
2006
.
Bench-to-bedside review: endotoxin tolerance as a model of leukocyte reprogramming in sepsis.
Crit. Care
10
:
233
235
.
14
Fan
K.
,
Ruan
Q.
,
Sensenbrenner
L.
,
Chen
B.
.
1992
.
Transforming growth factor-β1 bifunctionally regulates murine macrophage proliferation.
Blood
79
:
1679
1685
.
15
Kobayashi
K.
,
Hernandez
L. D.
,
Galán
J. E.
,
Janeway
C. A.
 Jr.
,
Medzhitov
R.
,
Flavell
R. A.
.
2002
.
IRAK-M is a negative regulator of Toll-like receptor signaling.
Cell
110
:
191
202
.
16
Sly
L. M.
,
Rauh
M. J.
,
Kalesnikoff
J.
,
Song
C. H.
,
Krystal
G.
.
2004
.
LPS-induced upregulation of SHIP is essential for endotoxin tolerance.
Immunity
21
:
227
239
.
17
Nakagawa
R.
,
Naka
T.
,
Tsutsui
H.
,
Fujimoto
M.
,
Kimura
A.
,
Abe
T.
,
Seki
E.
,
Sato
S.
,
Takeuchi
O.
,
Takeda
K.
, et al
.
2002
.
SOCS-1 participates in negative regulation of LPS responses.
Immunity
17
:
677
687
.
18
An
H.
,
Xu
H.
,
Zhang
M.
,
Zhou
J.
,
Feng
T.
,
Qian
C.
,
Qi
R.
,
Cao
X.
.
2005
.
Src homology 2 domain-containing inositol-5-phosphatase 1 (SHIP1) negatively regulates TLR4-mediated LPS response primarily through a phosphatase activity- and PI-3K-independent mechanism.
Blood
105
:
4685
4692
.
19
Xu
Q.
,
Schett
G.
,
Li
C.
,
Hu
Y.
,
Wick
G.
.
2000
.
Mechanical stress-induced heat shock protein 70 expression in vascular smooth muscle cells is regulated by Rac and Ras small G proteins but not mitogen-activated protein kinases.
Circ. Res.
86
:
1122
1128
.
20
Yang
J.
,
Mani
S. A.
,
Donaher
J. L.
,
Ramaswamy
S.
,
Itzykson
R. A.
,
Come
C.
,
Savagner
P.
,
Gitelman
I.
,
Richardson
A.
,
Weinberg
R. A.
.
2004
.
Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis.
Cell
117
:
927
939
.
21
Assoian
R. K.
,
Fleurdelys
B. E.
,
Stevenson
H. C.
,
Miller
P. J.
,
Madtes
D. K.
,
Raines
E. W.
,
Ross
R.
,
Sporn
M. B.
.
1987
.
Expression and secretion of type β transforming growth factor by activated human macrophages.
Proc. Natl. Acad. Sci. USA
84
:
6020
6024
.
22
Grotendorst
G. R.
,
Smale
G.
,
Pencev
D.
.
1989
.
Production of transforming growth factor β by human peripheral blood monocytes and neutrophils.
J. Cell. Physiol.
140
:
396
402
.
23
Naiki
Y.
,
Michelsen
K. S.
,
Zhang
W.
,
Chen
S.
,
Doherty
T. M.
,
Arditi
M.
.
2005
.
Transforming growth factor-β differentially inhibits MyD88-dependent, but not TRAM- and TRIF-dependent, lipopolysaccharide-induced TLR4 signaling.
J. Biol. Chem.
280
:
5491
5495
.
24
Le
Y.
,
Iribarren
P.
,
Gong
W.
,
Cui
Y.
,
Zhang
X.
,
Wang
J. M.
.
2004
.
TGF-β1 disrupts endotoxin signaling in microglial cells through Smad3 and MAPK pathways.
J. Immunol.
173
:
962
968
.
25
Randow
F.
,
Syrbe
U.
,
Meisel
C.
,
Krausch
D.
,
Zuckermann
H.
,
Platzer
C.
,
Volk
H. D.
.
1995
.
Mechanism of endotoxin desensitization: involvement of interleukin 10 and transforming growth factor β.
J. Exp. Med.
181
:
1887
1892
.
26
Ojaniemi
M.
,
Glumoff
V.
,
Harju
K.
,
Liljeroos
M.
,
Vuori
K.
,
Hallman
M.
.
2003
.
Phosphatidylinositol 3-kinase is involved in Toll-like receptor 4-mediated cytokine expression in mouse macrophages.
Eur. J. Immunol.
33
:
597
605
.
27
Ozes
O. N.
,
Mayo
L. D.
,
Gustin
J. A.
,
Pfeffer
S. R.
,
Pfeffer
L. M.
,
Donner
D. B.
.
1999
.
NF-κB activation by tumour necrosis factor requires the Akt serine-threonine kinase.
Nature
401
:
82
85
.
28
Gustin
J. A.
,
Korgaonkar
C. K.
,
Pincheira
R.
,
Li
Q.
,
Donner
D. B.
.
2006
.
Akt regulates basal and induced processing of NF-κB2 (p100) to p52.
J. Biol. Chem.
281
:
16473
16481
.
29
Luyendyk
J. P.
,
Schabbauer
G. A.
,
Tencati
M.
,
Holscher
T.
,
Pawlinski
R.
,
Mackman
N.
.
2008
.
Genetic analysis of the role of the PI3K-Akt pathway in lipopolysaccharide-induced cytokine and tissue factor gene expression in monocytes/macrophages.
J. Immunol.
180
:
4218
4226
.
30
Fukao
T.
,
Koyasu
S.
.
2003
.
PI3K and negative regulation of TLR signaling.
Trends Immunol.
24
:
358
363
.
31
Escoll
P.
,
del Fresno
C.
,
García
L.
,
Vallés
G.
,
Lendínez
M. J.
,
Arnalich
F.
,
López-Collazo
E.
.
2003
.
Rapid up-regulation of IRAK-M expression following a second endotoxin challenge in human monocytes and in monocytes isolated from septic patients.
Biochem. Biophys. Res. Commun.
311
:
465
472
.
32
Zacharioudaki
V.
,
Androulidaki
A.
,
Arranz
A.
,
Vrentzos
G.
,
Margioris
A. N.
,
Tsatsanis
C.
.
2009
.
Adiponectin promotes endotoxin tolerance in macrophages by inducing IRAK-M expression.
J. Immunol.
182
:
6444
6451
.
33
Marie
C.
,
Cavaillon
J. M.
,
Losser
M. R.
.
1996
.
Elevated levels of circulating transforming growth factor-β1 in patients with the sepsis syndrome.
Ann. Intern. Med.
125
:
520
521
.
34
Garcia-Lazaro
J. F.
,
Thieringer
F.
,
Lüth
S.
,
Czochra
P.
,
Meyer
E.
,
Renteria
I. B.
,
Galle
P. R.
,
Lohse
A. W.
,
Herkel
J.
,
Kanzler
S.
.
2005
.
Hepatic over-expression of TGF-β1 promotes LPS-induced inflammatory cytokine secretion by liver cells and endotoxemic shock.
Immunol. Lett.
101
:
217
222
.
35
Vodovotz
Y.
,
Kopp
J. B.
,
Takeguchi
H.
,
Shrivastav
S.
,
Coffin
D.
,
Lucia
M. S.
,
Mitchell
J. B.
,
Webber
R.
,
Letterio
J.
,
Wink
D.
,
Roberts
A. B.
.
1998
.
Increased mortality, blunted production of nitric oxide, and increased production of TNF-α in endotoxemic TGF-β1 transgenic mice.
J. Leukoc. Biol.
63
:
31
39
.
36
Jang
S. B.
,
Won
J.
,
Kim
H.
,
Kim
J.
,
Lee
K. H.
,
Han
H.
,
Rha
H. K.
,
Choi
C. R.
.
2002
.
TAK1 mediates lipopolysaccharide-induced RANTES promoter activation in BV-2 microglial cells.
Mol. Cells
14
:
35
42
.