Piperlongumine Chemosensitizes Tumor Cells through Interaction with Cysteine 179 of IκBα Kinase, Leading to Suppression of NF-κB–Regulated Gene Products

Cancer is a major public health problem in the United States and many other parts of the world. Currently, 1 in 3 women and 1 in 2 men in the United States will develop cancer in his or her lifetime (1). Although chemotherapy is the standard treatment for most kinds of cancers, resistance to chemotherapeutic drugs has become a major obstacle in treating cancer. In fact, multidrug resistance is now considered a main reason for the failure of chemotherapy (2). Alternatives that are inexpensive, efficacious, and safe compared with synthetic agents are sorely needed.

Epidemiologic, clinical, and experimental evidence suggests that medicines derived from plants play a pivotal role in treating most diseases, including cancer. As much as 80% of the cancer therapeutic agents currently used have their roots in natural products. One such product is the compound 5,6-dihydro-1-[(2E)-1-oxo-3-(3,4,5-trimethoxyphenyl)-2-propenyl]-2(1H)-pyridinone (Fig. 1A), called piperlongumine or piplartine, it is the pyridine alkaloid found in members of the Piper species, in particular in the fruit of the long pepper (Piper longum Linn.). Piperlongumine has been used widely in traditional medicine, including the Indian Ayurvedic system of medicine, traditional Chinese medicine, Tibetan medicine, and the folk medicine of Latin America. Piperlongumine has multiple pharmacologic activities: it has been described as a platelet aggregation inhibitor (3), an anxiolytic agent (4), an antidepressant (5), an antinociceptive agent (6), an anti-atherosclerotic agent (7), an antidiabetic agent (8), an anti-inflammatory agent (9), and an infectious pathogen inhibitor (10). Furthermore, piperlongumine has been reported to kill multiple types of cancer cells, inhibit metastasis, and have antitumor activities in a variety of animal models (11–22).

Although this compound was first isolated in 1967, major interest in it did not emerge until the 2011 publication in Nature by Raj and colleagues (18). This group reported that piperlongumine selectively induces cell death in a wide variety of tumor cell types and does not affect noncancerous cell types even at high doses. They also reported on its antitumor effects in murine models of melanoma, bladder cancer, breast cancer, and lung cancer while having only minimal toxic effects in these models. How piperlongumine acts as an anticancer agent is not yet clear. Several studies have demonstrated that it induces cytotoxicity through a variety of mechanisms, such as activation of caspases (14, 20, 21), activation of the ERK pathway (11), inhibition of cdc-2, cdk2, and cyclin D1 (19, 20), and downregulation of anti-apoptotic genes (Bcl-2, Raf-1, and survivin) and metastatic genes (VEGF, CD31, HIF-2, and Twist; refs. 17–20). The compound has also been shown to inhibit the enzymatic activity of GST π 1 and carbonyl reductases and to decrease glutathione levels, indicating that piperlongumine-mediated apoptosis of cancer cells is induced by a reactive oxygen species (ROS)–dependent mechanism (18). In addition, piperlongumine-dependent cytotoxicity may involve suppression of NF-κB, MYC, and LMP-1 in Burkitt lymphoma (14). Piperlongumine was shown to inhibit NF-κB activity in prostate cancer cells recently (23); however, the underlying mechanism is unclear.

NF-κB activation has been linked to the survival, proliferation, invasion, angiogenesis, and metastasis of various types of cancers. Activation of NF-κB requires the activation of IκB kinase (IKK). IKKβ, a subunit of the IKK complex, is essential for the activation of NF-κB in response to various proinflammatory stimuli. Cys-179 in the activation loop of this IKKβ plays a major role in IKK activation as mutation of this residue to alanine decreased its activity (24). Mutation of Cys179 to alanine in IKKβ also caused reduced phosphorylation of serine residues at positions 177 and 181, required for IKKβ activation. In addition, phosphorylation of tyrosine residues at positions 188 and 199 are also crucial for IKKβ activation as mutations of these residues have been shown to abolish the NF-κB activity (25).

Therefore, we postulated that piperlongumine modulates this NF-κB signaling pathway. In this study, we investigated in detail the effect of piperlongumine on NF-κB pathway regulation. We found that the compound suppressed NF-κB activation pathways induced by inflammatory stimuli, growth factors, carcinogens, and tumor promoters through the direct inhibition of cysteine 179 (Cys¹⁷⁹) of IKKβ, which led to the inhibition of IκBα, the suppression of NF-κB–regulated gene products, and the enhancement of apoptosis in tumor cells.

Piperlongumine was obtained from Indofine Chemical Company. Bacteria-derived human recombinant TNFα was provided by Genentech. Penicillin, streptomycin, RPMI-1640 medium, DMEM, FBS, and the kit for the LIVE/DEAD assay were obtained from Invitrogen. Phorbol 12-myristate 13-acetate (PMA), lipopolysaccharide (LPS), okadaic acid (OA), H₂O₂, MTT, Hoechst 33342, and antibodies against FLAG and β-actin were obtained from Sigma. Antibodies against p65, p50, cyclin D1, MMP-9, COX-2, PARP, IAP-1, IAP-2, Bcl-2, Bcl-xL, survivin, c-Myc, ICAM-1, caspase-3, caspase-8, and caspase-9 were purchased from Santa Cruz Biotechnology, as was the Annexin V staining kit. Phospho-specific anti-IκBα (Ser^32/36) and anti-p65 (Ser⁵³⁶) antibodies were obtained from Cell Signaling. Antibodies against CXCR4, phospho-IKKβ (Tyr¹⁸⁸) and IKKα/β (Ser^180/181) antibody was obtained from Abcam. Antibody against survivin and an ELISA system kit for human IL6 were purchased from R&D Systems. Anti-VEGF antibody was purchased from NeoMarkers. Antibodies against IκBα, IKKα, and IKKβ were obtained from Imgenex. Velcade (PS-341) was obtained from Millennium Pharmaceuticals.

Human multiple myeloma U266 cells, human embryonic kidney A293 cells, and human breast MCF-7 cells were obtained in 2000, and human T-cell leukemia Jurkat cells were obtained in 1997 from ATCC. Human lung adenocarcinoma H1299 cells and human squamous cells carcinoma SCC4 cells were obtained in 2005 from Dr. Reuben Lotan, and human chronic myeloid leukemia KBM-5 cells were obtained from Dr. Michael Andreeff of The University of Texas MD Anderson Cancer Center (Houston, TX). KBM-5 cells were cultured in Iscove's modified Dulbecco's medium with 15% FBS; Jurkat, H1299, MCF-7, and U266 cells were cultured in RPMI-1640 medium with 10% FBS; and A293 cells were cultured in DMEM supplemented with 10% FBS. SCC-4 cells were cultured in DMEM containing 10% FBS, nonessential amino acids, pyruvate, glutamine, and vitamins. All culture media were supplemented with 100 μg/mL streptomycin and 100 units/mL penicillin. The above-mentioned cell lines have not been recently tested for authentication in our laboratory.

The effect of piperlongumine on the cytotoxic potential of TNFα was determined using the MTT dye uptake method as described previously (26).

To measure the effect of piperlongumine on apoptosis induced by TNFα, we used the LIVE/DEAD assay to determine plasma membrane integrity (red fluorescent ethidium homodimer-1) and intracellular esterase activity (green fluorescent calcein-AM). This assay was performed as described previously (27).

The Annexin V assay is used to detect early apoptosis. This assay takes advantage of the rapid translocation and accumulation of the membrane phospholipid phosphatidylserine from the cytoplasmic interface of the cell to the extracellular surface. This loss of membrane asymmetry can be detected using the binding properties of Annexin V. Briefly, 1 × 10⁶ cells were pretreated with 10 μmol/L piperlongumine for 4 hours, treated with 1 nmol/L TNFα for 24 hours, and subjected to Annexin V staining. Cells were washed in PBS and resuspended in 100 μL of binding buffer containing FITC-conjugated Annexin V and then analyzed by flow cytometry (FACS Calibur; BD Biosciences) after the addition of propidium iodide.

The BD BioCoat tumor invasion system (BD Biosciences), which includes a light-tight polyethylene terephthalate membrane with 8-μm diameter pores and a thin layer of reconstituted Matrigel basement membrane matrix, was used to assess cell invasion. H1299 cells (2.5 × 10⁴) were suspended in serum-free medium and seeded into the upper wells. After incubation overnight, cells were treated with different concentrations of piperlongumine for 4 hours and then stimulated with 1 nmol/L TNFα for 24 hours in the presence of 1% FBS. The invasive cells were fixed and stained with Diff-Quik stain (Siemens Healthcare Diagnostics) and counted in 5 random microscopic fields (Nikon).

An ELISA kit (R&D Systems) was used to detect human IL6. U266 cells were treated with different concentrations of piperlongumine for 24 hours, and cell-free supernatants were collected for detection of IL6 by following the manufacturer's protocol.

To assess NF-κB activation, we performed electrophoretic mobility shift assay (EMSA) as described previously (28). Briefly, nuclear extracts prepared from TNFα-treated cells (2 × 10⁶/mL) were incubated with ³²P-end–labeled 45-mer double-stranded NF-κB oligonucleotide (15 μg of protein with 16 fmol DNA) from the human immunodeficiency virus long terminal repeat (5′-TTGTTACAA GGGACTTTC CGCTG GGGACTTTC CAGGGAGGCGTGG-3′) for 30 minutes at 37°C, and the DNA–protein complex that formed was separated from free oligonucleotide on 6.6% native PAGE. A double-stranded mutated oligonucleotide (5′-TTGTTACAA CTCACTTTC CGCTG CTCACTTTC CAGGGAGGCGTGG-3′) was used to examine the specificity of binding of NF-κB to the DNA. The dried gels were visualized with a Storm 820 phosphorimager (Molecular Dynamics), and radioactive bands were quantitated using ImageQuant software (Molecular Dynamics).

Cytoplasmic, nuclear, or whole-cell extracts were prepared, and Western blot analysis was performed. Briefly, 30 μg of protein was resolved with SDS-PAGE. After electrophoresis, the proteins were electrotransferred to nitrocellulose membranes, probed with specific antibodies, and detected by enhanced chemiluminescence reagent (GE Healthcare Life Sciences).

To determine the effect of piperlongumine on TNFα-induced IKK activation, we performed a kinase assay as described previously (29).

The effect of piperlongumine on the induction of NF-κB–dependent reporter gene transcription by TNFα, TNFR1, TRADD, TRAF2, NIK, TAK1/TAB1-β, IKKβ, and p65 was analyzed using a SEAP assay, as described previously (30).

The aim of this study was to investigate the effect of piperlongumine on the NF-κB activation induced by various carcinogens and inflammatory stimuli in cancer cells. Most of the studies were performed using the human myeloid cell line KBM-5 because these cells express TNF receptors, and the inflammatory pathway in these cells is well understood. Other cell types were used to test the specificity of the effect of piperlongumine.

We first investigated the dose and duration of piperlongumine exposure required to suppress TNFα-induced NF-κB activation in KBM-5 cells. EMSA results showed that pretreatment of cells with piperlongumine inhibited TNFα-induced NF-κB activation in a dose- (Fig. 1B) and time-dependent manner (Fig. 1C).

To confirm that the band visualized in TNFα-treated cells was indeed NF-κB, we incubated the nuclear extracts from TNFα-pretreated KBM-5 cells with anti-p50 or -p65 antibodies. The bands shifted to higher molecular masses when incubated with the antibodies, suggesting that the TNFα-activated NF-κB activation complex consisted of both p50 and p65. Addition of preimmune serum and mutated oligonucleotide had no effect on DNA binding, whereas the addition of excess unlabeled NF-κB (cold oligonucleotide; 100-fold excess) caused a decrease in the intensity of the band (Fig. 1D).

The endotoxin LPS, OA, PMA, H₂O₂, and cigarette smoke condensate (CSC) are known to activate NF-κB by different mechanisms. We found that all these agents activated NF-κB in KBM-5 cells and that piperlongumine suppressed this activation (Fig. 1E). These results suggested that piperlongumine acts at a step in the NF-κB activation pathway that is common to all 5 of these agents.

We next determined whether piperlongumine-mediated inhibition in TNFα-induced NF-κB activation is cell-type–specific. We observed an inhibitory effect in not only KBM-5 cells but also multiple myeloma (U266), T-cell leukemia (Jurkat), head and neck squamous cell carcinoma (SCC4), breast carcinoma (MCF-7), lung adenocarcinoma (H1299), and kidney (A293) cells (Fig. 1F). Thus, piperlongumine appears to be able to suppress NF-κB activation in a variety of human tumor cells.

Some NF-κB inhibitors, such as plumbagin (31) and bharangin (32), directly modify NF-κB to suppress its binding to DNA. To determine whether piperlongumine suppresses NF-κB activation through a similar mechanism, we incubated the nuclear extract from TNFα-treated KBM-5 cells with piperlongumine and found that the compound did not modify the DNA-binding ability of NF-κB protein (Fig. 2A). These results suggested that inhibition of NF-κB activation by piperlongumine does not suppress DNA binding directly. This compound may, however, use the inhibitory mechanism of nimbolide (33) and γ-tocotrienol (34), which inhibit NF-κB activation indirectly.

NF-κB activation requires the phosphorylation, polyubiquitination, and subsequent degradation of its inhibitory subunit, IκBα. To determine whether the inhibition of TNFα-induced NF-κB activation we observed in KBM-5 cells was due to the inhibition of IκBα degradation, KBM-5 cells were pretreated with piperlongumine and then exposed to TNFα for various time periods. We then analyzed the cells for nuclear NF-κB by EMSA and for IκBα degradation in the cytoplasm fraction by Western blotting. TNFα activated NF-κB in the control cells in a time-dependent manner but not in piperlongumine-pretreated cells (Fig. 2B). Moreover, TNFα induced IκBα degradation in control cells as early as 5 minutes (Fig. 2C). IκBα was completely degraded after 10 to 15 minutes and was resynthesized after 30 minutes. In piperlongumine-pretreated cells, however, this degradation was completely reversed.

Because IκBα degradation requires IκBα phosphorylation (35), we explored whether the inhibition of TNFα-induced IκBα degradation was due to the inhibition of IκBα phosphorylation. We blocked IκBα degradation by using the proteasome inhibitor N-acetyl-leucyl-leucylnorleucinal (ALLN). Western blotting results showed that co-treatment with TNFα and ALLN induced IκBα phosphorylation and that pretreatment with piperlongumine strongly suppressed this phosphorylation (Fig. 2D). These results indicated that piperlongumine suppresses TNFα-induced IκBα degradation by inhibiting IκBα phosphorylation.

Phosphorylation of IκBα is regulated by the upstream kinase IKK complex. Because piperlongumine inhibits the phosphorylation and degradation of TNFα-induced IκBα, we determined the effect of this compound on IKK activation. Our results revealed that TNFα activated IKK in a time-dependent manner and piperlongumine suppressed this activation (Fig. 2E). Expression of IKKα or IKKβ proteins was not notably affected by TNFα or piperlongumine.

To determine whether piperlongumine inhibits IKK activity directly or indirectly, we treated the kinase assay mixture prepared from TNFα-pretreated KBM-5 cells with various concentrations of piperlongumine. Results from the kinase assay showed that it inhibited IKK activity (Fig. 2F), suggesting that the compound suppresses TNFα-induced IKK activation directly.

Because IKKβ contains various cysteine residues, we investigated whether piperlongumine suppresses IKK activation by modifying one or more of these cysteine residues. The reducing agent dithiothreitol (DTT) was used to determine whether the modulation of IKK activity by piperlongumine was through the modification of critical cysteine residues. We found that the addition of DTT to the kinase reaction mixture reversed the piperlongumine-mediated inhibition of TNFα-induced IKK activity (Fig. 2G). The finding suggested that a cysteine residue is involved in this pathway.

The cysteine residue Cys¹⁷⁹, which is positioned within the activation loop of the IKK catalytic subunits, is critical for IKKβ activity. To determine whether Cys¹⁷⁹ is involved in piperlongumine-induced IKK inhibition, we transfected A293 cells with wild-type FLAG-IKKβ or mutated FLAG-IKKβ (alanine substituting for Cys¹⁷⁹). Piperlongumine inhibited wild-type IKKβ but had no apparent effect on mutated IKKβ (Fig. 2H). These results suggested that piperlongumine inhibits IKKβ activity by directly modifying the Cys¹⁷⁹ residue.

Furthermore, we determined whether piperlongumine modulates phosphorylation of other residues of IKKβ such as serine-181 and tyrosine-188, which are also responsible for its full activity. Results showed that TNF-induced phosphorylation of IKKβ at both serine-181 and tyrosine-188 residues and piperlongumine inhibited the phosphorylation of both the residues (Fig. 2I).

The phosphorylation and degradation of IκBα are essential to the NF-κB activation pathway, whereas the stabilization of IκBα is important in preventing the nuclear translocation of p65. The effect of piperlongumine on TNF-induced nuclear translocation of p65 was examined by Western blot analysis. For this experiment, KBM-5 cells were treated with piperlongumine and then exposed to TNFα for different time periods. The nuclear extracts were prepared and examined for p65 by Western blot analysis. We observed that TNFα induced the nuclear translocation of p65 and that piperlongumine suppressed the TNFα-induced p65 nuclear translocation (Fig. 3A).

Whether piperlongumine affects TNFα-induced p65 phosphorylation at Ser⁵³⁶ was also examined. The results showed that TNFα induced the phosphorylation of p65 and that piperlongumine suppressed this phosphorylation (Fig. 3A).

Although we observed that piperlongumine inhibits NF-κB activation, DNA binding alone does not always result in NF-κB–dependent gene transcription. For this reason, we assessed whether piperlongumine can affect TNFα-induced reporter gene transcription. A 12-fold increase in SEAP activity was observed after simulation with TNFα compared with the vector control, and the induction was nearly abolished by dominant-negative IκBα (Fig. 3B), indicating specificity. When the A293 cells were pretreated with piperlongumine, TNFα-regulated NF-κB–dependent SEAP expression was inhibited in a dose-dependent manner (Fig. 3B). These results suggested that piperlongumine inhibits the NF-κB–dependent reporter gene expression induced by TNFα.

TNFα-induced NF-κB activation is mediated through the sequential interaction of TNF receptor 1 (TNFR1) with TNFR-associated death domain (TRADD), TNFR-associated factor 2 (TRAF2), NF-κB–inducing kinase (NIK), TGFβ-activated protein kinase 1 (TAK1)/TAK1-binding protein-1 (TAB1), and IKKβ, which results in phosphorylation of IκBα and leads to degradation of IκBα and p65 nuclear translocation. We determined the site of action for piperlongumine in TNFα signaling in A293 cells. This agent significantly suppressed NF-κB–dependent reporter gene expression induced by TNFR1, TRADD, TRAF2, NIK, TAK1/TAB1, and IKKβ plasmids (Fig. 3C), indicating that piperlongumine acts at a step upstream of p65.

NF-κB regulates the expression of the antiapoptotic proteins Bcl-2, Bcl-xL, cellular inhibitor of apoptosis proteins 1 and 2 (c-IAP-1, c-IAP-2), and survivin, which are known to be induced by TNFα and to play an important role in cell survival. We used Western blotting to determine whether piperlongumine inhibits the expression of these gene products in KBM-5 cells. Our results showed that the compound downregulated the expression of all 5 of these TNFα-induced proteins (Fig. 4A).

We also used Western blotting to investigate whether piperlongumine modulates the expression of the proliferative proteins c-myc, cyclin D1, and COX-2. We observed that TNFα induced the expression of these proliferative proteins and that piperlongumine inhibited it (Fig. 4B).

Finally, we examined whether piperlongumine affects NF-κB–regulated gene products involved in tumor cell invasion and metastasis. Western blotting revealed that TNFα induced the expression of intercellular adhesion molecule-1 (ICAM-1), matrix metallopeptidase 9 (MMP-9), CXC chemokine receptor 4 (CXCR-4), and VEGF and that piperlongumine downregulated the expression of all these proteins (Fig. 4C).

IL6 is a proinflammatory cytokine that is regulated by NF-κB and augments the proliferation of a wide variety of tumor cells. We examined whether piperlongumine affects the IL6 level produced by human multiple myeloma U266 cells. Our ELISA results showed that piperlongumine suppressed IL6 production in a dose-dependent manner (Fig. 4D).

Using KBM-5 cells, we next examined whether the downregulation by piperlongumine of the tumor cell survival proteins leads to potentiation of the apoptosis induced by TNFα or chemotherapeutic agents. Using the MTT assay, we found that piperlongumine did, in fact, potentiate the cytotoxicity induced by TNFα, 5-fluorouracil (5-FU), or bortezomib in a dose-dependent manner (Fig. 5A). The apoptotic effect of piperlongumine was also examined by phosphatidylserine externalization, a marker of early apoptosis, using Annexin V staining and flow cytometry. The number of Annexin V–positive cells was increased significantly when KBM-5 cells were pretreated with piperlongumine before TNFα (Fig. 5B).

Whether piperlongumine can enhance the TNFα-induced activation of caspases-3, -8, and -9 was also examined. Our Western blotting results showed that the piperlongumine sensitized the cells to caspase activation (Fig. 5C).

We also used Western blotting to examine whether piperlongumine can enhance TNFα-induced PARP cleavage in KBM-5 cells. We found that TNFα alone had little effect on PARP cleavage, but when the cells were pretreated with piperlongumine, a remarkable increase in PARP cleavage was observed (Fig. 5D). These results also demonstrate a synergistic interaction between the 2 agents. Taken together, these results suggested that piperlongumine potentiates the apoptotic effects of TNFα and chemotherapeutic agents.

Ubiquitous intracellular esterase activity and an intact plasma membrane are characteristics of live cells. We used the LIVE/DEAD assay to discriminate between live and dead KBM-5 cells by simultaneously staining with calcein-AM and ethidium homodimer-1. This assay indicated that piperlongumine alone induces loss of membrane integrity (an indicator of apoptosis) and that treatment with TNFα enhances the piperlongumine-induced apoptosis (Fig. 6A).

Because piperlongumine suppressed the TNFα-induced expression of proteins such as CXCR-4 or MMP-9, which are linked to tumor cell invasion, we next examined whether piperlongumine suppresses TNFα-induced tumor cell invasion. For this experiment, we used H1299 cells. Our results showed that TNFα increased tumor cell invasion by 1.4-fold and that pretreatment with piperlongumine decreased the level of TNFα-induced tumor cell invasion to below baseline (Fig. 6B).

We also determined whether piperlongumine inhibits the expression of MMP-9 and CXCR4 involved in invasion and metastasis, respectively. We found that TNFα induced the expression of MMP-9 and CXCR4, and the pretreatment with piperlongumine significantly suppressed the expression of these proteins in H1299 cells (Fig. 6C). These results suggest that inhibition in MMP-9 and CXCR4 protein expression may contribute to the inhibitory effects of piperlongumine on invasion activity of H1299 cells.

Although radiotherapy and chemotherapy are used to treat cancer, both activate the NF-κB pathway, which leads to radioresistance and chemoresistance, respectively. Thus, agents that are safe and efficacious in downmodulating the NF-κB pathway are urgently needed for radio- and chemosensitization. One such agent is piperlongumine, which is derived from the fruit of the “long pepper” plant and is consumed traditionally to address a range of inflammatory diseases. However, the mechanism of action of this compound is not fully understood. Two recent studies from Harvard researchers published in Nature (18) and PNAS (15) described the activity of piperlongumine against a wide variety of human cancers in cell culture and animal models, but how this compound mediates its anticancer and other activities is poorly understood. Our current study provides insight into the mechanism of action of piperlongumine.

The specific objective of our study was to determine the effect of piperlongumine on the NF-κB activation pathway and on NF-κB–regulated gene products that are associated with inflammation, tumor cell proliferation, invasion, and angiogenesis. We demonstrated that piperlongumine suppressed NF-κB activation that was induced by various carcinogens and inflammatory agents through the inhibition of IKK activation, which led to the suppression of IκBα phosphorylation and degradation and the suppression of p65 nuclear translocation and phosphorylation. Piperlongumine also promoted the apoptosis induced by TNFα and various chemotherapeutic agents, and it suppressed cancer cell invasion.

We found that piperlongumine inhibited NF-κB activation that was induced by a wide variety of agents (TNFα, LPS, OA, PMA, H₂O₂, and CSC2) and that this suppression was not cell-type–specific. These results are in agreement with recent reports that piperlongumine can suppress LPS-induced NF-κB activation in endothelial cells (9) and PMA-induced NF-κB activation in murine macrophages (36). Our results are also consistent with another recent report that piperlongumine inhibits constitutive NF-κB expression in Burkitt lymphoma cells (14).

How piperlongumine inhibits NF-κB activation has not been reported and thus was investigated in detail. We found, for the first time, that piperlongumine suppresses TNFα-induced IKK activation, which leads to suppression of the phosphorylation and degradation of IκBα. Our study also showed that piperlongumine directly inhibits TNFα-induced IKK activity. This inhibition could be reversed with the reducing agent DTT, which suggests that inhibition of IKK activation involves the interaction of piperlongumine with -SH groups within IKK. Cys¹⁷⁹ of IKK, a redox-sensitive cysteine residue, is positioned within the activation loop of the IKK catalytic subunits (37). We found that substitution of Cys¹⁷⁹ with alanine prevents the inhibition of IKK activity by piperlongumine, thus suggesting the role of this residue in the action of this alkaloid. This is first report to suggest the role of piperlongumine with -SH groups of any protein, although irreversible protein glutathionylation as a process associated with cellular toxicity of this alkaloid has been demonstrated (15).

We also found, for the first time, that TNFα induces the phosphorylation of the p65 subunit of NF-κB at Ser⁵³⁶ and that piperlongumine completely suppresses the phosphorylation. IKK has been shown to cause the phosphorylation of p65 at Ser⁵³⁶ (38). Unlike IκBα, multiple kinases have been implicated in the phosphorylation of p65 at Ser⁵³⁶ (39), and it is very likely that suppression of IKK activity is involved in inhibiting the phosphorylation of both IκBα and p65 by piperlongumine. Interestingly, phosphorylation of RelA/p65 on Ser⁵³⁶ has been shown to be independent of IκBα (40), indicating that IKK has numerous substrates. NF-κB p65 phosphorylated at Ser⁵³⁶ was also shown to be an independent prognostic factor in Swedish patients with colorectal cancer (41), suggesting the potential of using piperlongumine to treat such patients.

Our results indicate that despite the involvement of at least 50 different proteins in the NF-κB activation pathway, each containing multiple cysteines (42), piperlongumine inhibits NF-κB by modifying Cys¹⁷⁹ in the IKKβ activation loop. Nimbolide (33), butein (43), 3-formylchromone (44), and other compounds have been reported to exhibit similar effects on Cys¹⁷⁹. In addition, we found that piperlongumine also blocked the phosphorylation of serine 181 and tyrosine 188 in IKKβ induced by TNFα. Serine kinases, such as TAK-1, have been shown to phosphorylate IKKβ at residues 181 (see 25 for reference). In addition, tyrosine residue 188 is within the activation loop of IKKβ and is in close proximity to serine181. The fact that the tyrosine kinase c-Src binds to IKKβ suggests that tyrosine phosphorylation of IKKβ is potentially important for regulation of its activity (25). Thus, these results suggest that piperlongumine may inhibit IKKβ through multiple mechanisms.

The suppression of NF-κB activation induced by TNFα, LPS, OA, PMA, H₂O₂, and CSC suggests that piperlongumine acts at a step common to all these activators. NF-κB activation by TNFα is mediated through sequential recruitment of TNFR1, TRADD, TRAF2, NIK, TAK1/TAB1, and IKK. Our transfection experiments showed that piperlongumine inhibited the NF-κB reporter gene expression induced by these molecules. Similar results have been reported with γ-tocotrienol (34), flavopiridol (27), and celastrol (45).

Most approaches used in cancer treatment, such as chemotherapy and radiotherapy, kill cancer cells by inducing apoptosis; however, cancer cells often develop resistance to such treatment. Furthermore, many cancer therapies indirectly activate apoptosis by chemically or physically damaging DNA. This damage may have the unintended effect of generating a pool of heavily mutated cancer cells and increasing the chances of their developing resistance (46). We showed that piperlongumine potentiated cell death induced by TNFα or the chemotherapeutic agents 5-FU and bortezomib in KBM-5 cells. Our results suggest that chemotherapy or radiotherapy followed by treatment with piperlongumine could reduce the tumor burden.

The loss of caspase activation appears to be central to the prevention of most cell death events in cancer. Finding strategies to overcome caspase inhibition will be valuable for the development of novel cancer treatments (47). We investigated in detail how piperlongumine potentiates apoptosis and found that piperlongumine enhanced the cleavage of precursors of caspases-3, -8, and -9, which suggested that the compound activated cell death signaling through 2 pathways: the extrinsic, death receptor pathway (caspase-8) and the intrinsic, mitochondrial pathway (caspase-9). NF-κB–regulated cell survival proteins, such as the Bcl-2 family of proteins and IAPs, are selectively overexpressed in various types of tumors and are involved in the apoptotic pathway defect in many tumor cells (48). Overexpression of these proteins is involved in tumor cell survival (49) and might be used to reverse the apoptotic pathway defect in cancer. Bcl-2 was found to be downregulated by piperlongumine in previous studies (18, 20). Furthermore, we found that piperlongumine suppressed the expression of cell survival proteins, including Bcl-2, Bcl-xl, c-IAP-1, c-IAP-2, and survivin, suggesting that these gene products could be therapeutic targets for potentiation of apoptosis by piperlongumine.

We also found that treatment with piperlongumine downregulated the expression of NF-κB–regulated gene products involved in cell proliferation (c-Myc, cyclin D1, and COX-2). The previously reported antiproliferative effects of piperlongumine against various tumor cells (7, 14, 16) could be due to downregulation of these gene products. In agreement with our results, compounds such as bharangin (32), nimbolide (33) flavopiridol (27), and celastrol (45) have also shown antiproliferative effects in different tumor cells.

Furthermore, our results demonstrated that piperlongumine suppressed TNFα-induced expression of ICAM-1, MMP-9, CXCR-4, and VEGF, which are major mediators involved in tumor invasion and metastasis (50–52). The downregulation of these invasion-related proteins suggests that piperlongumine has a role in preventing tumor cell invasion and metastasis. That this compound suppressed TNFα-induced invasion in cancer cells further suggests its anti-invasive activity. The inflammatory cytokine IL6, which is known to promote tumorigenesis, was also downregulated by piperlongumine.

Overall, our results suggest that piperlongumine has significant potential for tumor treatment: it inhibits the NF-κB pathway by targeting IKK as well as NF-κB–associated biomarkers that are involved in cell proliferation, angiogenesis, invasion, and metastasis (Fig. 7). On the basis of these results, further studies are required in animals and in patients to explore the potential of piperlongumine as an anticancer agent.

No potential conflicts of interest were disclosed.

Conception and design: J.G. Han, S.C. Gupta, B.B. Aggarwal

Development of methodology: J.G. Han

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.G. Han

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J.G. Han, S.C. Gupta, B.B. Aggarwal

Writing, review, and/or revision of the manuscript: J.G. Han, S.C. Gupta, S. Prasad, B.B. Aggarwal

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.G. Han

Study supervision: B.B. Aggarwal

The authors thank Elizabeth L. Hess from the Department of Scientific Publications for carefully proofreading the article.

B.B. Aggarwal was supported in part by a grant from Malaysian Palm Oil Board and a grant from the Center for Targeted Therapy of The University of Texas MD Anderson Cancer Center (chartstring 404400-80-111139-21). B.B. Aggarwal is also Ransom Horne Jr. Professor of Cancer Research. J.G. Han was supported by a training scheme for excellent talents from China (2011D003034000003), Program for Outstanding Medical Academic Leader of Beijing (2009-1-03), the Basic and Clinical Cooperation Project of Capital Medical University (10JL04), and a Youth Training Abroad Grant of Beijing Chaoyang Hospital.

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.