Inhibiting TRAF2-mediated activation of NF-kappaB facilitates induction of AP-1

Abstract

The compound 5-(4-methoxyarylimino)-2-N-(3,4-dichlorophenyl)-3-oxo-1,2,4-thiadiazolidine (P(3)-25) is known to possess anti-bacterial, anti-fungal, and anti-tubercular activities. In this report, we provide evidence that P(3)-25 inhibits NF-kappaB, known to induce inflammatory and tumorigenic responses. It activates AP-1, another transcription factor. It inhibits TRAF2-mediated NF-kappaB activation but not TRAF6-mediated NF-kappaB DNA binding by preventing its association with TANK (TRAF for NF-kappaB). It facilitates binding of MEKK1 with TRAF2 and thereby activates JNK and AP-1. We provide evidence, for the first time, that suggests that the interaction of P(3)-25 with TRAF2 leads to inhibition of the NF-kappaB pathway and activation of AP-1 pathway. These results suggest novel approaches to design of P(3)-25 as an anti-cancer/inflammatory drug for therapy through regulation of the TRAF2 pathway.

FIGURE 1.

Effect of P₃-25 on activation of NF-κB, AP-1, IKK, and JNK in Jurkat and HuT-78 cells. Jurkat and HuT-78 cells were treated with 100 nm P₃-25 for different times, and nuclear extracts were prepared. NF-κB (A1) and AP-1 (B1) DNA binding was measured by a gel shift assay. Cells were co-transfected with plasmids for NF-κB promoter DNA that had been linked to SEAP (NF-κB-SEAP) or AP-1 promoter DNA that was linked to the luciferase reporter gene (AP-1-luciferase) and GFP. After washing, cells were cultured for 12 h and then treated with P₃-25 (100 nm) for different times. The GFP-positive cells were counted, and transfection efficiency was calculated. Culture supernatant was assayed for SEAP activity. The results are represented as -fold activation over the nontransfected control (A2). After treatment, the cell pellet was extracted and assayed for luciferase activity (B2). Cells were treated with 100 nm P₃-25 for different times. The whole cell extracts (300 μg of protein) were immunoprecipitated with anti-IKKα and IKKβ (1 μg each) antibodies, and kinase was assayed using GST-IκBα as a substrate (C). The IKKα was assayed from 100 μg of protein extract by Western blot. The amount of phospho-JNK was measured from 100 μg of whole cell extracts by Western blot using anti-phospho-JNK antibody (D, top). The whole cell extracts (300 μg of protein) were immunoprecipitated with 1 μg of anti-JNK antibody, and JNK activity was assayed using GST-Jun as substrate (D, bottom). HuT-78 cells, transfected with the NF-κB-SEAP construct were treated with 100 pm P₃-25 or 5-aryl TZD for different times, and activity of SEAP was measured from culture supernatant (E).

FIGURE 2.

Effect of P₃-25 on TNF- or IL-8-induced NF-κB activation and PKAα activity. Jurkat cells were treated with 100 nm P₃-25 for different times and then stimulated with 100 pm TNF or 100 ng/ml IL-8 for 2 h. NF-κB DNA binding was measured from nuclear extracts (A). The whole cell extracts (300 μg of protein), prepared from P₃-25-treated but TNF- or IL-8-stimulated cells for 1 h, were immunoprecipitated with anti-PKAα (1 μg) antibody. PKAα activity was assayed using GST-p65 as substrate (C). The amount of PKAα was detected in the cell extract by Western blot. Jurkat cells were transfected with NF-κB-SEAP and GFP constructs cultured for 12 h and treated with P₃-25 for different times. Cells were then stimulated with TNF (100 pm) or IL-8 (100 ng/ml) for 12 h. Culture supernatants were used to detect SEAP activity, and activity is indicated as -fold activation relative to vector-transfected cells (B).

FIGURE 3.

Effect of P₃-25 on TRAF2- and/or TRAF6-induced NF-κB, AP-1, IKK, and JNK activation. Jurkat cells were transfected with plasmids for NF-κB-SEAP or AP-1-luciferase and GFP. After washing, cells were cultured for 12 h and then treated with P₃-25 (100 nm) for different times. Nuclear extracts were prepared, and NF-κB (A1) and AP-1 (B1) DNA binding was measured by gel shift assay. Culture supernatant was assayed for SEAP activity (A2). Cell pellet was extracted and assayed for luciferase activity (B2). The results are represented as -fold activation over the nontransfected control. TRAF2- and/or TRAF6-transfected cells were treated with P₃-25 for different times. The whole cell extract (300 μg of protein) was immunoprecipitated with anti-IKKα and IKKβ (1 μg of each) antibodies, and IKK was assayed using GST-IκBα as substrate (C). The IKKα was assayed from 100 μg of protein extract by Western blot. The amounts of phospho-JNK were measured from 100 μg of whole cell extracts by Western blot (D, top). The whole cell extracts (300 μg of protein) were immunoprecipitated with anti-JNK antibody (1 μg), and JNK activity was measured using GST-Jun as substrate (D, bottom).

FIGURE 4.

Effect of P₃-25 on TRAF2, TANK, and MEKK1 interaction. HuT-78 cells were transfected with vector, TRAF2, or TRAF2-DN constructs. Cells were cultured for 12 h and then treated with P₃-25 (100 nm) for 12 h. The whole cell extracts were prepared and incubated with Sepharose CL 4B beads coupled with anti-TRAF2 (A) or anti-TANK (B) antibody for 4 h. Beads were packed in a 2-ml column, washed with 0.2 m NaCl solution, and then eluted with 0.5 m NaCl solution. The fractions were run in 12% SDS-PAGE, and TRAF2, MEKK1, and TANK were detected by Western blot. IP, immunoprecipitation; WB, Western blot.

FIGURE 5.

Effect of P₃-25 on interaction with TRAF2 and TRAF6 in silico. Shown is a surface representation of TRAF2, colored on the basis of electrostatic potential (−57.293k_bT/e to +57.293k_bT/e, where k_b, T, and e are the Boltzmann constant, temperature, and electron charge, respectively) and the bound P₃-25. A surface representation of TRAF6 is shown with the bound P₃-25 (A). P₃-25 bound into the binding site of TRAF2, represented in stick form (light blue), forms hydrogen bonds with Arg³⁹³, Asp³⁹⁹, and Phe⁴⁴⁷ and the binding site of TRAF6 and forms hydrogen bonds with Arg³⁹², Lys⁴⁶⁹, and Gly⁴⁷⁰ (B). The lowest energies of different clusters of TRAF2 and TRAF6 that bind with P₃-25 have been calculated and are indicated in the graph (C).

FIGURE 6.

Effect of P₃-25 and 5-aryl TZD on interaction with TRADD-TRAF2, TRAF2-TANK, or TRAF2-MEKK complexes. A surface representation of TRADD-TRAF2 complex and binding of P₃-25 and 5-aryl TZD at positive regions of the complex in silico is depicted (A1). Interaction of 5-aryl TZD and P₃-25 with TRADD-TRAF2 complex is indicated (A2). Shown is a surface representation of the TRAF2-TANK complex, colored on the basis of electrostatic potential (−57.293k_bT/e to +57.293k_bT/e, where k_b, T, and e are the Boltzmann constant, temperature and the electron charge, respectively) and the bound P₃-25 and 5-aryl TZD (B1). P₃-25 binds into the binding site of the TRAF2-TANK complex, represented in stick form (light blue), forms hydrogen bonds with Arg³⁹³, Asp³⁹⁹, and Phe⁴⁴⁷, and 5-aryl TZD forms hydrogen bonds with Arg³⁹³, Lys⁴⁴⁷, and Asn⁴⁰⁷ (B2). The electrostatic potential surface area of TRAF2-MEKK1 complex in binding with P₃-25 and 5-aryl TZD is represented (C1). P₃-25 interacts with the active site amino acid (magenta) Lys¹⁹² of MEKK1 and TRAF2 complex (C2, left). Interaction of 5-aryl TZD with MEKK1 active site residues (magenta) and TRAF2 complex is analyzed (C2, right). The graph shows the lowest binding energies of complexes with P₃-25 and 5-aryl TZD with different complexes (D).

FIGURE 7.

Schematic diagram of P₃-25-mediated interference with cell signaling to inhibit NF-κB but to activate AP-1.