The Role of P38 MAPK and PKC in BLP Induced TNF-α Release, Apoptosis, and NFnB Activation in THP-1 Monocyte Cells

Background. P38 mitogen activated protein kinase (p38 MAPK) is a critical mediator of the inflammatory response, which makes it a suitable candidate as a novel therapeutic strategy for inflammatory condi- tions. In this study, we set out to examine the precise role of both protein kinase C (PKC) and P38 MAPK signaling kinases in bacterial lipoprotein (BLP) in- duced nuclear factor-kappa B (NFnB) activation and tumor necrosis factor-alpha (TNFα) release in THP-1 monocytic cell line.

Materials and methods. THP-1 cells were incubated with BLP(0 –1000 ng/mL), phorbol myristate acetate (PMA; 0 –100 µg/mL) or a combination of both for 6 and 24 h, with or without pretreatment with SB202190, a specific inhibitor of p38 MAPK and bisindolylmaleim- ide I, a specific inhibitor of PKC (0 –200 µM). Cell su- pernatants were analyzed for TNF-α release and apo- ptosis. NFnB activity was analyzed by electromobility supershift assay.

Results. BLP induced TNF-α release was signifi- cantly reduced by pretreatment with SB202190 at all concentrations (428.7 ± 5.9 versus 51 ± 0.8 pg/mL, P < 0.05). Pretreatment with bis I significantly inhibited TNF-α release at higher concentrations (200 µM) (429.7 ± 5.9 versus 194.9 ± 42.68 pg/mL, P < 0.05) but this was much less effective than SB202190. PMA induced TNF-α release was not inhibited at 6 h by either SB202190 or bis I, but was significantly so at 24 h (148.5 ± 9.8 versus 24 ± 1.7 and 25.1 ± 4.4 pg/mL, P < 0.05). BLP or lipopolysaccharide (LPS) did not result in apoptosis in THP-1 cells (P > 0.05) with PMA inducing apoptosis in a time- and dose-dependent manner. In combina- tion with BLP (1000 ngmL) but not LPS (1000 ng/mL), low dose PMA resulted in a significant increase in apoptosis, 6% ± 0.5% (Control) versus 9.2% ± 0.3% (P < 1 0.05) and 7% ± 2.2% (Control) versus 7.7% ± 0.3% (P > 0.05), respectively. This synergistic effect was inhib- ited by bisindolylmaleimide 100 nM, 8.9% ± 0.9% (Con- trol) versus 9.8% ± 0.2% (P > 0.05). PMA and BLP in- duced rapid nuclear translocation of NFnB, which was inhibited by pretreatment with both SB-202190 and bis I, and SB202190 but not bis I, respectively.

Conclusions. P38 is a critical mediator of BLP in- duced TNF-α release and NFnB activation, whereas PKC is only partially responsible for its response. P38 and PKC are both critical mediators of PMA induced TNF-α release and NFnB activation.

Key Words: P38 MAP kinase; protein kinase C; TNF-α; NFnB; apoptosis; LPS; BLP.

INTRODUCTION

Tumor necrosis factor-alpha (TNF-α), a critical me- diator of acute inflammation, is produced primarily by activated monocytes and macrophages [1, 2]. It has a number of important pathophysiological responses, which ultimately result in leukocyte and endothelial cell activation, with further progression of the acute inflammation via chemokines and other mediators. TNF-α can also stimulate hepatocytes to produce amy- loid A protein and fibrinogen, which contribute to the acute phase response [3]. Both lipopolysaccharide (LPS) and bacterial lipoprotein (BLP) can stimulate monocytes to produce TNF-α, via the cell surface re- ceptors, toll-like receptors 2 and 4, respectively [4– 6]. They can also activate nuclear factor-kappa B (NFnB). NFnB is a ubiquitous translation factor, which is inti- mately involved in the acute phase responses [7–9]. It is maintained in an inactive form in the cytosol by its association with its inhibitory protein InB. Multiple stimuli including LPS and BLP can phosphorylate InB and result in the liberation of the active NFnB, which translocates to the nucleus and where it is ultimately involved in the translation of a diverse range of genes involved in acute inflammation [10, 11].

LPS induced NFnB activation and TNF-α production in human monocytes is independent of protein kinase C (PKC) but dependent on protein tyrosine kinases [12]. However, the importance of the various protein kinases in intracellular signaling following stimulation with BLP is unclear. P38 mitogen activated protein kinase (p38 MAPK) has been implicated as a critical mediator of the release of proinflammatory cytokines and positively regulates the expression of a variety of genes involved in the acute phase response, such as TNF-α, IL-6, and other inducible enzymes [13, 14]. Many different stimuli can activate p38 MAPK, such as LPS, cytokines, BLP, and other stresses [13, 15]. P38 MAPK as a critical mediator of the inflammatory response makes it a suitable candidate as a novel ther- apeutic strategy for inflammatory conditions. P38 MAPK inhibition can improve survival following cecal ligation and puncture as a consequence of reducing circulating cytokine levels [16]. P38 MAPK inhibition has also been shown to have broad anti-inflammatory effects in an experimental model of human endotox- emia [17].

PKC is a family of ubiquitous phospholipid depen- dent enzymes involved in signal transduction path- ways, associated with a variety of cellular responses [18–20]. Unlike LPS induced NFnB activation and TNF-α release, which are independent of PKC, the role of PKC in BLP, induced NFnB activation and TNF-α release is unclear. As a result, in the present study we set out to examine the precise role of both PKC and P38 MAPK in BLP induced NFnB activation and TNF-α release in human monocytes. To achieve this we plan to use a specific inhibitor of PKC and P38 MAPK, bisin- dolylmaleimide I (bis I) and SB-202190, respectively. BLP, besides inducing NFnB activation and TNF-α release in THP-1 cells, has also been reported to induce apoptosis in human monocyte cell lines. However, this apoptosis is greatly augmented by the addition of phor- bol myristate acetate (PMA), a potent protein kinase C agonist. The precise intracellular mechanism of this is unclear as NFnB activation has demonstrable anti- apoptotic effects. As a result, in this study we also plan to investigate the precise role of BLP and PKC stimu- lation in human monocyte apoptosis.

MATERIALS AND METHODS

Reagents

RPMI 1640, fetal calf serum, penicillin, streptomycin sulfate, and glutamine were purchased from Life Technologies, Inc. (Paisley, Scotland). LPS from Escherichia coli serotype O55B5, propidine io- dine (PI), dimethyl sulfoxide, PMA, phenylmethylsulfonyl fluoride (PMSF), Nonidet P-40, dithiothreitol (DTT), HEPES, MgCl2, KCL, NaCl, sodium citrate, Tris, Triton X-100, and ethylenediamine tetraacetic acid (EDTA) were purchased from Sigma Aldrich (St. Louis, MO). BLP, a synthetic bacterial lipopeptide (Pam3-Cys-Ser-Lys4- OH) derived from the immunologically active NH2 terminus of bac- terial lipoproteins, was purchased from EMC Microcollections GmbH (Tuebingen, Germany), which was LPS-free as confirmed by the Limulus amebocyte lysate assay (Charles River Endosafe, Charles- ton, SC). SB-202190 and bis I were purchased from Biomol (Ply- mouth Meeting, PA).

Cell Culture and Cytokine Analysis

The human monocyte cell line, THP-1 (ATCC, Manassas, VA) was maintained in culture in RPMI 1640 medium supplemented with 10% heat inactivated fetal calf serum, penicillin (100 units/mL), streptomycin sulfate (100 µg/mL), and glutamine (2.0 Mm) at 37°C, in a humidified 5% CO2 atmosphere. THP-1 cells (2 × 105 cells/well) were incubated, in 24 well plates (Falcon, Lincoln Park, NJ), with control (vehicle-phosphate-buffered saline [PBS]), 100 or 1000 ng/mL of BLP for 6 and 24 h (n = 3 and duplicated). This was repeated with cells pretreated for 30 min with 50, 100, and 200 µM of SB-202190 or bis I prior to stimulation with 1000 ng/mL of BLP. Cells were again pretreated as above and then stimulated with PMA 50 µg/mL or vehicle (0.01% dimethyl sulfoxide). After incubation, the suspension of cells were collected and immediately centrifuged at 4°C for 10 min at 1500 rpm. The supernatant was collected and immediately further centrifuged at 4°C for 8 min at 10,000 rpm. Supernatant was col- lected and stored at —80°C for determination of TNF-α concentra- tions using commercially available enzyme-linked immunosorbent assay kits (R and D Systems, Minneapolis, MN) according to the manufacturer’s instructions.

Apoptosis Analysis

THP-1 cells (5 × 105 cells/well) were incubated in 6-well plates (Falcon) with control (vehicle-PBS), 10, 100, or 1000 ng/mL of BLP or LPS for 6 and 24 h (n = 3 and duplicated). Cells were then incubated with either LPS or BLP (0, 100, 1000 ng/mL) with high (50 and 100 µg/mL) or low (6.25 and 12.5 µg/mL) dose of PMA. This was repeated with cells pretreated with 50, 100, and 200 µM of bis I prior to stimulation. After the incubation period had finished the suspension of cells were collected in nonadherent FACS tubes (Falcon), and immediately centrifuged at 4°C for 10 min at 1500 rpm. The cell pellet were gently resuspended in 0.5 mL of hypotonic fluorochrome solution (50 µg/mL PI, 3.4 mM sodium citrate, 1 mM Tris, 0.1 mM EDTA, 0.1% Triton X-100), and incubated in the dark before they were analyzed on a FACScan flow cytometer (Becton Dickinson, Lincoln Park, NJ). The forward scatter and side scatter of THP-1 cells were simultaneously measured. The PI fluorescence of individ- ual nuclei with an acquisition of fluorescence channel 2 was plotted against forward scatter, and the data were registered on a logarith- mic scale. A minimum number of 10,000 events were collected and analyzed using the CellQuest software (Becton Dickinson). Apoptotic THP-1 cell nuclei were distinguished by their hypodiploid DNA con- tent from the diploid DNA content of normal THP-1 cell nuclei. THP-1 cell debris was excluded from analysis by raising the forward scatter. All measurements were performed under the same instru- ment setting.

Isolation of Nuclear Proteins and NFnB Electrophoretic Mobility Gel-Shift Assay

Following treatment as described above, cells 6 × 106 per exper- iment were washed in cold PBS twice. After centrifugation at 1500 rpm for 15 min, the pellet was resuspended in 1 mL of cold lysis buffer A (10 mM HEPES-NaOH pH 7.9, 1.5 mM MgCl2, 10 Mm KCl, 0.5 mM DTT, 0.5 mM PMSF) and centrifuged at 14,000 rpm for 10 min. The supernatant was discarded and the pellet resuspended in 20 µL of cold lysis buffer A and 0.1% (vol/vol) Nonidet P-40 and incubated on ice for 10 min with gentle intermittent vortex. The sample was centrifuged at 14,000 rpm for 10 min, supernatant discarded and nuclear pellet resuspended in 15 µL of cold buffer B (20 mM HEPES, 420 mM NaCl, 1.5 mM MgCl2, 0.2 Mm EDTA, 25% (wt/vol) glycerol, 0.5 mM DTT, 0.5 mM PMSF, pH 7.9) on ice for 15 min with gentle intermittent vortex. Following centrifugation at 14,000rpm for 10 min the supernatant (nuclear extract) was col- lected, diluted with 75 µL of cold buffer C (20 mM HEPES, 50 mM KCl, 0.2 Mm EDTA pH 8.0, 20% (wt/vol) glycerol, 0.5 mM DTT, 0.5 mM PMSF), and stored at —80°C. Protein concentrations were de- termined using a micro BCA protein assay reagent kit (Pierce, Rock- ford, IL).

NFnB consensus oligonucleotide (double stranded 5=-AGTTGA- GGGGACTTTCCCAGG C-3=, (Promega, Madison, WI) was 5= end labeled with γP-32 ATP with T4 polynucleotide kinase (Promega) by incubation for1h at 37°C. Unincorporated oligonucleotide was removed using Microspin G-25 centrifugation columns (Amesham Pharmacia Biotechnology) 6.5 ug of nuclear protein was incubated with labeled nucleotide (150,000 –200,000 cpm) in binding buffer for 30 min at room temperature. Oligonucleotide labeled protein and unlabeled probe were separated by electrophoresis on a 3% native polyacryl- amide gel in tris-glycine. Gels were dried and exposed to an inten- sifying screen and developed using a Storm BSO Phospho-imaging System (Molecular Dynamics, Sunnyvale, CA). Luminosity values were obtained by Quantikine image analysis software supplied with the phospho-imaging hardware.

Statistical Analysis

Cytokine and apoptosis data were compared using one-way analysis of variance with Tukey post-hoc analysis. Inter group differences were compared using Student’s t-test. All values were expressed as mean (±SEM). P < 0.05 was considered to be sta- tistically significant.

RESULTS

BLP induced a rapid release of TNF-α from THP-1 cells, 428.7 ± 5.9 pg/mL at 6 h, which was still present 24 h after stimulation, however at a reduced level (285.1 ± 5.3 pg/mL), while unstimulated monocytes had extremely low levels of spontaneous TNFα release through out the experiments (Fig. 1). SB-202190 pre- treatment prior to BLP stimulation resulted in a sig- nificant reduction in TNFα release compared with no pretreatment (P < 0.0001) (Fig. 2A). Pretreatment with 200 µM of SB-202190 resulted in a significantly greater reduction compared with 50 or 100 µM of SB- 202190, however, this reduction was small from mean values of 55 ± 1.5 and 51.8 ± 0.8 pg/mL to 32.9 ± 0.9 pg/mL at 6 h for 50, 100, and 200 µM SB202190. These values were still all significantly greater than unstimu- lated cells, accounting for 12.8%, 12%, and 7.7% of overall TNF-α production. Pretreatment with greater concentrations of inhibitor did not produce any greater reduction in TNFα release. A similar pattern was seen at 24 h, except pretreatment with both 100 and 200 µM of SB-202190 resulted in a significantly greater reduc- tion compared to 50 µM of SB-202190 (Fig. 2B).

FIG. 1. BLP induced TNF-α release at 6 and 24 h, following stimulation with 1000 ng/mL of BLP. THP-1 cells, 2 × 105 were treated with control and 1000 ng/mL of BLP for 6 and 24 h.

FIG. 2. The effect of pretreatment with 50, 100, and 200 µM of SB-202190, on BLP induced TNF-α release at (A) 6 and (B) 24 h. THP-1 cells were pretreated with 0.50, 100, and 200 µM of SB202190 and then stimulated with 1000 ng/mL of BLP for 6 and 24 h. Pre- treatment with SB202190 resulted in a significant reduction in TNF-α release compared to controls, *P < 0.05. Pretreatment with 200 µM of SB 202190 resulted in a significantly greater reduction than 50 or 100 µM at 6 h (†P < 0.05) and pretreatment with 200 and 100 µM of SB202190 resulted in significantly greater reduction than 50 µM at 24 h (‡P < 0.05).

Bis I pretreatment prior to BLP stimulation did not result in a significant reduction in TNF-α release com- pared with no pretreatment (P > 0.05) with concentra- tions of 50 or 100 µM, but did with a concentration of 200 µM of bis I (Fig. 3A). This dose of bis I produced a reduction in TNF-α level from 429.7 ± 5.9 pg/mL to 194.9 ± 42.6 pg/mL, a 54.7% reduction in overall TNF-α production. Pretreatment with higher concen- trations of inhibitor did not produce any greater reduc- tion in TNF-α release and again a similar pattern was seen at 24 h (Fig. 3B).

FIG. 3. The effect of pretreatment with 50, 100, and 200 µM of bis-I, on BLP induced TNF-α release at (A) 6 and (B) 24 h. THP-1 cells were pretreated with 0, 50, 100, and 200 µM of bis I and then stimulated with 1000 ng/mL of BLP for 6 and 24 h. Pretreatment with 200 µM of bis I, but not 50 or 100 µM, resulted in a significant reduction in TNF-α release compared to controls at both 6 and 24 h, *P < 0.05.

PMA stimulation induced TNFα release in THP-1 cells at 6 h 12.5 ± 0.6 pg/mL and to a much greater extent at 24 h 148.5 ± 9.8 pg/mL, both of which were significantly greater than unstimulated cells (P < 0.001) (Fig. 4). At 6 h, pretreatment with either SB- 202190 (200 µM) or bis I (200 µM) did not result in any significant reduction in PMA induced TNF-α release (P > 0.05).

At 24 h pretreatment with either SB-202190 and bis I resulted in a significant reduction in PMA induced TNFα release, from 148.5 pg/mL to 24 and 25.1 pg/mL for SB202190 and bis I respectively (P < 0.001), a reduction of 83.9% and 83.1% of overall PMA induced TNF-α release.BLP or LPS, at concentrations of 0 –1000 ng/mL and over periods of 6 –24 h, did not induce apoptosis in THP-1 cells (P > 0.05) (data not shown). PMA stimu- lation induced apoptosis in a dose and time dependent manner, which plateaus at a dose of 100 µg/mL at 6 h and for lower doses at 24 h (Fig. 5A). THP-1 cells were costimulated with BLP or LPS (0, 100, and 1000 ng/ mL) and PMA at high doses of 50 and 100 µg/mL for 6 and 24 h, and we found that any induced apoptosis was due to the addition of PMA alone. THP-1 cells were then costimulated with BLP or LPS (0, 100, and 1000 ng/mL) and PMA at low doses of PMA 6.25 and 12.5 µg/mL for 6 and 24 h and again found that any apo- ptosis was due to the addition of PMA alone as any observed apoptosis was inhibited by bis-I 100 µM. We did note, however, that costimulation of THP-1 cells with both 1000 ng/mL of BLP and 12.5 µg/mL of PMA did induce a small but significant increase in apoptosis, 6% ± 0.5% for control versus. 9.2% ± 0.3% for cells co stimulated with BLP and PMA (P < 0.05) (Fig. 5B). This synergistic effect was indeed due to PMA as it was inhibited by bis I (100 µM), 8.9 ± 0.9% (Control) versus 9.8 ± 0.2% (P > 0.05) (Fig. 5C).
BLP induced a rapid nuclear translocation of NFnB in THP-1 cells. This translocation was inhibited by pretreatment with SB-202190, whereas pretreatment with bis I attenuated but did not fully inhibit its trans- location. PMA stimulation resulted in translocation of NFnB and this translocation was inhibited by both SB-202190 and bis I (Fig. 6).

DISCUSSION

BLP and PMA induce NK-nB activation and TNF-α production in THP-1 cells, a human monocyte cell line. BLP appears to have a rapid onset and a gradual reduction of stimulation, whereas PMA appears to have the reverse effect, slower onset with a more sustained effect. We found that BLP induced NFnB activation and TNF-α release were attenuated with a specific p38 MAPK inhibitor, but not completely ab- rogated. Previous studies have demonstrated a role for protein tyrosine kinases in LPS induced TNF-α pro- duction in THP-1 cells. Using nonspecific protein kinase inhibitors, a similar profile of attenuation of LPS induced TNF-α production in THP-1 cells was achieved [12]. However the level of attenuation achieved here is much greater using an inhibitor specific for p38 MAPK, up to 88% attenuation of overall TNF-α production. This implies that p38 MAPK is a critical mediator of BLP induced TNF-α production. We note that since p38 MAPK inhibition did not completely abrogate BLP induced TNF-α production, other pathways must play a small role in its production.

FIG. 4. The effect of pretreatment with SB-202190 and bis-I, on PMA induced TNF-α release at 6 and 24 h. THP-1 cells were pre- treated with 200 µM of SB-20290 and bis I and then stimulated with 50 µg/mL of PMA for 6 and 24 h. PMA induced a significant release of TNF-α at both 6 and 24 h compared with controls, *P < 0.05. Pretreatment with 200 µM of SB-202190 and bis I resulted in a significant reduction in PMA induced TNF-α release at 24 h but not at 6 h, †P < 0.05.

FIG. 5. (A) Dose responses for the percentage apoptosis induced by PMA 0 –100 µg/mL at 6 and 24 h. THP-1 cells were treated with 0, 25, and 50 µg/mL of PMA for 6 and 24 h. PMA at doses of 50 and 100 µg/mL at 6 h and at doses of 25, 50, and 100 µg/mL at 24 h induced apoptosis which was significantly increased compared with control (*P < 0.05, significant increase compared with. control) (B) The effect of costimulation with BLP and PMA on apoptosis in THP-1 cells. THP-1 cells were costimulated with PMA 6.25 µg/mL and 100 and 1000 ng/mL of BLP for 6 and 24 h and apoptosis assessed (*P < 0.05 compared with all other groups). (C) The effect of bis-1 pretreatment on PMA induced apoptosis in THP-1 cells. THP-1 cells were pretreated with 200 µM of bis I and then stimulated with 6.25 and 12.5 µg/mL of PMA for 6 h.

In contrast to this effect, inhibition with a specific PKC inhibitor resulted in a much less marked attenu- ation of BLP induced TNF-α production. Indeed, this inhibition was only significant at 200 µM concentra- tions of inhibitor and increasing concentrations of in- hibitor did not lead to any further reduction. This would suggest that although not as significant a medi- ator as p38 MAPK, PKC does play a role in BLP in- duced TNF-α release. This is in contrast to LPS, where some studies have shown that LPS induced TNF-α release in human monocytes is independent of PKC [12]. However, other studies have demonstrated a role for PKC in LPS induced TNF-α release in different cell types [21, 22]. In parallel to the above observed roles of p38 MAPK and PKC in BLP induced TNF-α release, these were mirrored in the observed patterns of NFnB activation.

If PKC truly is a mediator of BLP induced TNF-α release in human monocytes then a PKC agonist should induced its release. Indeed as with other stud- ies, we have shown that PMA, a phorbol ester and PKC agonist induces TNF-α release and NFnB activation in human monocytes [12]. We then went on to demon- strate that both p38 MAPK inhibition and PKC inhi- bition have no effect on PMA induced TNF-α release at 6 h following stimulation, but at 24 h following stimu- lation both had an equal effect on reducing PMA in- duced TNF-α release. This would suggest that both p38 MAPK and PKC are required and thus are critical mediators of PMA induced TNF-α release. Further- more, our results demonstrate that PMA induced TNF-α release must also, in some small part, be inde- pendent of both p38 MAPK and PKC and dependent on other pathways.

FIG. 6. Electrophoretic mobility shift assays with percentage lumi- nosity histograms demonstrating (A) effect of SB-202190 and bis I on BLP induced NFnB activation and (B) effect of SB-202190 and bis I on PMA induced NFnB activation. THP-1 cells were pretreated with 200 µM of SB-20290 and bis I and then stimulated with BLP 1000 ng/mL or 50 µg/mL of PMA.

The role of apoptosis in sepsis and its clinical se- quelae of systemic inflammatory response syndrome and multiple organ dysfunction syndromes remain con- troversial and elusive [23]. Previous studies have shown that BLPs and LPS can induce apoptosis in human monocytes, which for BLPs are toll-like recep- tor 2 dependent and augmented by CD14 and PMA [24, 25]. Other studies looking at the role of PKC in tumor proliferation and vascular endothelial growth factor release have demonstrated that PKC would appear to enhance proliferation and vascular endothelial growth factor release, and also appear to increase the invasive- ness of cancer cell lines [26, 27]. Other studies have shown that PKC is an important mediator of cell death with inhibition preventing cell death and stimulation with PMA altering the PKC isoforms resulting in al- tered responses to death signals [18, 28, 29]. Indeed, δ-PKC is an important regulator of TNF antiapoptotic signaling, but p38 MAPK is not, nor does p38 inhibi- tion augment neutrophil apoptosis in sepsis [30] [20]. In our study, neither BLP nor LPS induced apoptosis. Monocytes treated with PMA demonstrated apoptosis in a time- and dose-dependent manner, whereas other studies have shown that PMA mediated PKC activa- tion protects against TNF-α related apoptosis-inducing ligand (TRAIL)-induced toxicity [31]. Other studies have found that NFnB activation inhibits the apoptotic process, probably through the induction of expression of antiapoptotic molecules. In our study, we have shown that BLP and PMA induce TNF-α release and activate NFnB, however, PMA is a potent apoptotic stimulus where as BLP is not. Furthermore, neither BLP nor LPS protected THP-1 from PMA induced ap- optosis. Therefore, this would suggest that death sig- nals mediated via PKC do not have the protective antiapoptotic effects of NFnB activation. Interestingly, other reports have also demonstrated a proapoptotic effect following NFnB activation [32–34]. This appar- ent dual effect of NFnB appears to be dependent on a number of factors, including the stimulus, the cell type, and the particular NFnB subunit activated [35]. There- fore, the ultimate fate of the cell may be related to the individual stimulus and its interaction with specific signal transduction pathways mediating Bisindolylmaleimide I their effect and how both of these interact with the various death signaling pathways.