Apoptosis

, Volume 14, Issue 4, pp 348–363

NFκB signaling in carcinogenesis and as a potential molecular target for cancer therapy

Authors

    • Department of Community, Occupational and Family Medicine, Yong Loo Lin School of Medicine, NUS Graduate School for Integrative Sciences and EngineeringNational University of Singapore
    • Institute for Molecular and cell Biology
Cell Death and Disease

DOI: 10.1007/s10495-009-0315-0

Cite this article as:
Shen, H. & Tergaonkar, V. Apoptosis (2009) 14: 348. doi:10.1007/s10495-009-0315-0
  • 699 Views

Abstract

It has become increasingly clear that deregulation of the NFκB signaling cascade is a common underlying feature of many human ailments including cancers. The past two decades of intensive research on NFκB has identified the basic mechanisms that govern the functioning of this pathway but uncovering the details of why this pathway works differently in different cellular contexts or how it interacts with other signaling pathways remains a challenge. A thorough understanding of these processes is needed to design better and more efficient therapeutic approaches to treat complex diseases like cancer. In this review, we summarize the literature documenting the involvement of NFκB in cancer, and then focus on the approaches that are being undertaken to develop NFκB inhibitors towards treatment of human cancers.

Keywords

NFκBp65KinaseInhibitorsSignalingCancerIKK

General introduction: NF-κB family proteins and the signaling pathways

Since its discovery over 21 years ago, nuclear factor-κB (NFκB) has been one of the most extensively investigated transcription factors for its immense importance in vertebrate physiology and for its apparent deregulation in human diseases such as cancers [1, 2]. The mammalian NFκB family consists of five members: NFκB1 (p50/p105), NFκB2 (p52/p100), c-Rel, RelA (p65) and RelB. All NFκB proteins are characterized by the presence of a highly conserved 300 amino acid Rel homology domain (RHD) which is responsible for DNA binding, dimerization, and interaction with inhibitory factors known as IκB proteins. There are seven IκB proteins—IκBα, IκBβ, IκBγ, IκBε, BCL-3 and the two precursor proteins p100 and p105. The functional activity of IκB proteins is largely driven by their abundance, which in turn is cell and tissue type specific. All IκB proteins are characterized by the presence of five to seven ankyrin repeats that are responsible for their interaction with the RHD of the NF-κB proteins. Although it was believed that IκB proteins inhibit NFκB activity mainly by masking the nuclear localization signal on NFκB proteins and hence restricting their nuclear accumulation, this hypothesis provides only partial explanation for the mechanism of IκB action for the following reasons: (a) it has been shown that NFκB bound to IκB proteins shuttle dynamically in living cells and (b) a majority of NFκB is still cytoplasmic in cells lacking all three major IκB proteins [3, 4]. A view that IκB proteins largely regulate NFκB by inhibiting its DNA binding ability which in turn regulates its nuclear accumulation has also been proposed [3]. Hence proteolytic removal of IκB is the rate limiting step in activation of NFκB. Phosphorylation of IκB proteins on specific serine residues controls their proteolysis. A 600-900KD kinase complex consisting of two major IκB kinases, IKKα (IKK1), IKKβ (IKK2) and a nonenzymatic regulatory subunit, IKKγ (or NEMO, NFκB essential modulator) has been described to carry out the signal-induced phosphorylation of IκB proteins that targets them for degradation. In addition to the IKK1/2 and NEMO, which are the core components of the IKK complex, other proteins have been described to be in the complex although it is unclear at the moment if these proteins are transiently associated with the complex or if the components of the IKK complex vary in a cell type specific manner.

In the mammalian cells, there are several distinct NFκB activation pathways [1]. Among them, the classical or canonical pathway has been the most-well studied. This pathway is activated by pro-inflammatory cytokines and stimuli via engagement of specific receptors such as tumor necrosis factor (TNF) receptor 1 and 2 (TNFR1 and TNFR2), interleukin-1 (IL-1) receptor, T-cell receptor (TCR), and Toll-like receptor (TLR), leading to activation of IKKβ, which is necessary and sufficient for phosphorylation of IκBα at ser 32 and ser 36. IκB phosphorylation then leads to polyubiquitination by βTrCP at specific lysine residues and subsequently degradation by the proteasome. The released NFκB proteins then accumulate in the nucleus, due to their binding to NFκB binding sites across the coding and non coding regions of the chromosomes. It is thought that the canonical pathway mainly controls the transcription of genes which are essential for innate immunity and for the inhibition of apoptosis under most conditions [5].

On the other hand, the non-canonical pathway of NFκB activation is mainly triggered by certain members of the TNF cytokine family such as lymphotoxin β, B-cell activating factor belonging to the TNF family (BAFF), or CD40 ligand [6]. Different from the above-mentioned canonical pathway, it is the selective activation of IKKα by the NFκB-inducing kinase (NIK) that leads to phosphorylation of NFκB2 (p52/p100). Subsequent processing of p100 gives rise to p52, which then results in the formation of the p52-RelB heterodimer that enters the nucleus and regulates the expression of genes that are known to be important for adaptive immune response and development of secondary lymphoid organs.

At present, more than 150 NFκB target genes have been identified [2], a direct indication of the vast spectrum of NFκB biological functions. Some recent reviews have given excellent summary on the signaling mechanisms and biological functions of NFκB [1, 710]. Here we attempt to summarize some of the recent developments on the involvement of NFκB in cancer, by focusing on the role of NFκB in the carcinogenic process and the potential application of NFκB inhibitors in cancer therapy.

Biological functions of NFκB closely associated with cancer

NFκB and inflammation

Inflammation, in particular chronic inflammation, has been closely associated with cancer [11]. Several well-established examples include hepatitis and hepatocellular carcinoma, ulcerative colitis and colorectal cancer, as well as gastritis and gastric cancers. Emerging evidence in the last decade has suggested the critical role of NFκB in linking inflammation and cancer. NFκB regulates major inflammatory factors, including TNFα, IL-6, IL-1, IL-8, many of which are potent activators for NFκB. It is thus believed that NFκB and inflammation constitute a positive feedback loop in the milieu of inflammatory sites to induce cellular and DNA damage, promote cell proliferation and transformation, and eventually the initiation, promotion and progression of cancer [1215]. The involvement of NFκB in skin cancer and hepatocarcinogenic process will be discussed in details in the subsequent sections of this review, as examples to support the notion that NFκB acts as the molecular link bridging inflammation and cancer. As a result, NFκB becomes an ideal molecular target in controlling cancer development. In fact, many of the anti-inflammatory drugs such as nonsteroidal anti-inflammatory drugs (NSAIDs) are found to work through inhibiting NFκB in order to execute their cancer therapeutic and chemo-preventive effects [16].

On the other hand, it should be noted that the role of NFκB in inflammation is highly context and cell or tissue type dependent. One intriguing observation is that activation as well as blockage of NFκB signaling results in inflammatory skin diseases [17]. Transgenic mice expressing a nondegradable dominant negative form of IκBα or keratinocyte-specific knockouts of IKKβ and IKKγ/NEMO were found to develop severe inflammatory response in skin shortly after birth and progress with age [1820]. Similar observations were also found in other systems, as conditional ablation of IKKβ or IKKγ in intestinal epithelial cells leads to inflammation and colitis [21, 22], and deletion of IKKγ in liver parenchymal cells results in steatohepatitis and hepatocellular carcinoma [23]. More recently, IKKβ was found to have opposing functions during acute and chronic intestinal inflammation in mice with IKKβ deletion in intestinal epithelial cells and exposed to dextran sulfate, namely promotion of acute inflammation but attenuation of chronic inflammation in the intestinal tract [24]. As most of the above studies are based on transgenic animal models in which the NFκB pathway was completely blocked in one specific type of cells, presently there is no evidence suggesting that other approaches for NFκB inhibition such as small molecule NFκB inhibitors are capable of reaching similar degree of inhibition or specificity. These findings also raise caution for therapeutic usage of NFκB inhibitors in control of inflammation and cancer.

Promotion of cell proliferation

Uncontrolled cell proliferation and defective cell death machinery are two important hall-marks of cancer. NFκB is known to control the expression of some key cell cycle regulatory genes, including cyclin D1, D2, D3, cyclin E1, c-myc, and cyclin-dependent kinases (CDK 2, 4, 6). A number of excellent reviews have discussed these topics extensively [1, 2528].

It has been well established that the phosphoinositide 3-kinase (PI3K)-AKT-mammalian target of rapamycin (mTOR) pathway is one of the key elements in promoting cell proliferation, cell growth and suppression of apoptosis, and plays a critical role in tumorigenesis [2931]. Recently there is emerging evidence showing that the pro-survival and pro-growth functions of NFκB are related to its functional interaction with the PI3K-AKT-mTOR signaling pathway. PI3K is a lipid kinase that phosphorylates phosphoinositides (PI), which then activate the survival kinase AKT. The activated AKT acts on the tuberous sclerosis complex (TSC; consisting of TSC1/TSC2) and Rheb, leading to mTOR activation. The activated mTOR then controls de novo protein synthesis via its downstream molecular targets including p70S6 kinase and 4EBP1 [30, 32].

The first indication of intricate cross-talks between these two pathways came from the studies showing that the PI3K-AKT-mTOR pathway is able to positively regulate NFκB activation. It was demonstrated that in cells treated with TNFα, AKT is capable of directly phosphorylating IKKα to activate NF-κB [33]. Although this finding was initially challenged [34], subsequent studies have supported the notion that AKT functions through IKK to promote the phosphorylation and transactivation of p65 protein [24, 35, 36]. It is also clear that in cells treated with cytokines (TNFα and IL-1) and growth factors (insulin), AKT engages mainly IKKα, (preferentially to IKKβ), in promoting NFκB activation [24, 36]. A very recent report further examined the molecular mechanisms integrating these two pathways by discovering a key role for the mTOR-associated protein Raptor [37]. Correspondingly, the mTOR inhibitor rapamycin is shown to suppress NF-κB activation through a mechanism that may involve dissociation of Raptor from mTOR [37].

The second hint of a cross-talk between these two pathways comes from intriguing findings that mTOR activation leads to suppression of NF-κB activation but AKT acts downstream of TSC1-TSC2 to promote NFκB activation [38]. In MEF cells with deletion of TSC1 or TSC2, the elevated mTOR activation suppresses AKT and hence IKK and NF-κB activation in response to TNFα and regulates cell survival [38]. The exact mechanisms for the mTOR-mediated suppression of AKT and NF-κB remain to be further established.

The third aspect a cross-talk between these two pathways is based on observations that the NFκB signaling pathway is also capable of promoting the function of mTOR. A recent study by Lee et al. [39] provided convincing evidence that IKKβ upregulates mTOR activity through direct interaction, phosphorylation (at Ser487 and Ser511) and inactivation of TSC1. Furthermore, IKKβ-mediated TSC1 phosphorylation and inactivation is associated with enhanced angiogenesis and poor prognosis in some breast cancer patients [39, 40]. Taken together, it is likely that NFκB and the mTOR pathways constitute a positive feedback loop and understanding such a reciprocal relationship between these pathways would be important for development of effective anti-cancer agents with dual effects, especially in those tumors with TSC deficiency and high mTOR activation.

Suppression of cell death, both apoptosis and necrosis

Owning to extensive studies in the past two decades, the anti-apoptotic function of NFκB has been well illustrated as the key mechanism which provides the positive engagement of NFκB in tumorigenesis. The anti-apoptotic function of NFκB is mainly achieved through the transcriptional regulation of an array of anti-apoptotic proteins that are capable of inhibiting (1) the death receptor-mediated apoptosis pathway, and (2) mitochondria-dependent apoptosis pathway. The first group of the anti-apoptotic factors mainly includes inhibitor of apoptosis proteins (IAPs) 1 and 2, X-linked IAP (XIAP), cellular Fas-associated death domain-like IL-1β-converting enzyme (FLICE) inhibitory protein (cFLIP), TNF receptor associated factor 1 (TRAF1) and TRAF2 [4143]. The second group mainly refers to Bcl-2 family members, including Bcl-2, Bcl-xL, A1 (also known as Bfl-1) and A20 [28, 44, 45]. In addition, NFκB may also execute its anti-apoptotic function via its regulatory effects on JNK, the PI3K-AKT-mTOR pathway and the tumor suppressor p53, and the details of such interactions will be discussed in the other sections of this review.

Apoptosis and necrosis are two distinct forms of cell death. Although necrosis has traditionally been described as accidental cell death occurring only in cases of severe pathological damage, recent studies have revealed that necrotic cell death could also be programmed and occur during normal cell physiology and development [4648]. One good example is that cell death ligands (such as FasL and TNFα) are able to induce necrotic cell death independent of the caspase cascade, a process involving enhanced production of reactive oxygen species (ROS) and oxidative stress [47, 4951].

Downstream of ROS, c-Jun-N-terminal kinase (JNK) has been identified as the key effector molecule in ROS-mediated necrotic cell death, either in cells exposed to higher level of exogenous ROS [52], or in cells treated with TNFα [50, 53]. Moreover, activated JNK can potentiate TNFα-stimulated necrosis by increasing the production of ROS [51, 53]. Thus, it is likely that ROS and JNK constitute a vicious cycle which culminates in mediating necrotic cell death. Currently the exact mechanism for JNK-mediated necrosis is still not clear. One clue is that JNK may contribute to the sustained activation of poly (ADP-ribose) polymerase-1 (PARP-1) via direct protein–protein interaction and phosphorylation [54].

At present there is emerging evidence suggesting that NFκB is also capable of protecting from non-apoptotic cell death or necrosis, in addition to its potent inhibitory effect on typical apoptotic cell death. The anti-necrotic function of NFκB is achieved via suppression of ROS and subsequent downregulation of JNK activation [5558]. Earlier studies indicated that some anti-apoptotic proteins such as Gadd45b and XIAP are responsible for the inhibitory effect of NFκB on JNK activation [59, 60]. It is now well recognized that NFκB downregulates JNK activation via transcriptional control of some key antioxidant proteins such as MnSOD and FHC that are capable of eliminating ROS and then preventing JNK activation [61, 62]. Therefore, the cross-talk between NFκB and JNK is a critical factor in determining the susceptibility to cell death, in particularly to necrosis via maintaining the redox balance in the cell [55, 56, 58, 63]. The biological relevance of the anti-necrosis function of NFκB in carcinogenesis and cancer development is a important remaining question to be further examined. Recently it has been proposed that necrosis is closely associated with inflammation and autophagy, both of which are connected to cancer development [64, 65]. Therefore, understanding of role of NFκB and JNK in necrosis provides new opportunities for development of more effective anti-cancer agents.

NFκB and autophagy

Autophagy refers to an evolutionarily conserved process in which intracellular proteins and organelles are sequestered in autophagosomes and subsequently degraded by lysosomal enzymes for the purpose of recycling cellular components to sustain metabolism during nutrient deprivation and to prevent accumulation of damaged proteins and organelles [6668]. Recently autophagy has entered the research spotlight owing to the increasing understanding of its regulatory mechanisms and its connection with many physiological and pathological processes and human diseases such as cancers and neurodegenerative disorders [67, 69].

NFκB is generally recognized as a major pro-survival and anti-apoptosis pathway, whereas the role of autophagy in cell death is still a matter of dispute [70, 71]. Therefore it would of great interest to study the functional interaction between the NFκB signaling pathway and autophagy. The work from Codogno’s group provided the initial evidence showing that in several cancer cell lines activation of NFκB in response to TNFα led to suppression of autophagy [72]. Consistently, the suppression of NFκB leads to an enhancement of autophagy in response to starvation in leukemia cells [73]. More importantly, both studies found that autophagy is capable of promoting apoptotic cell death and the inhibitory effect of NFκB on autophagy may constitute part of its anti-apoptotic effect [74, 75]. At present, the exact mechanisms responsible for the inhibitory effect of NFκB on autophagy remains to be further examined.

On the other hand, autophagy has also been found to negatively regulate NFκB activation. Firstly, autophagy is able to suppress the NFκB signaling pathway by selectively degrading IKK and NIK proteins [76, 77]. Inversely, it was reported that in cells with deletion of TSC, the elevated level of mTOR (and thus reduced autophagy) is associated with reduced NFκB activation [38], thus providing the indirect evidence that activation of autophagy (by suppression of the TSC-mTOR signaling pathway) would enhance NFκB activation. It is thus believed that the nature of interaction between NFκB and autophagy machinery is likely to be cell type and stimulus specific.

Although it seems that both NFκB and autophagy have similar pro-survival function in cell death, their involvement in cancer development has been found to be at odds. While NFκB is widely considered as a pro-tumorigenesis pathway [13, 28], autophagy, in contrast, emerges as a potential anti-cancer mechanism [78, 79]. Therefore, it would be interesting and important to elucidate the cross-talks between NFκB and autophagy in the context of tumor development.

NFκB and p53

Many cellular signals such as DNA damage culminate in the activation of both the p53 and the NFκB pathways. It has been documented that NFκB can negatively regulated p53 function (hence could contribute to tumorigenesis) by upregulating the levels of p53 E3 ubiquitin ligase, Mdm2, thereby negatively regulating p53 stability [80, 81]. Another mechanism by which NFκB can suppress p53 function is by upregulating anti-apoptotic target genes that can antagonize the pro-apoptotic functions of p53 [7]. Finally the p65 subunit of NFκB subunit has been show to compete with limiting pools of p300 and CBP coactivators, which are required for transcriptional activities of both p53 and NFκB [82]. However, the rate-limiting step regulating the competition between p53 and NFκB was only deciphered recently. In unstimulated cells, CBP (and not p300) is preferentially bound to p53, but some NFκB activating stimuli, lead to phosphorylation of CBP at serines 1382 and 1386, which leads to its dissociation from p53 and this now recruits CBP to the p65 subunit of NFκB [83]. The yin and yan nature of p53-NFκB interactions is likely to be further uncovered using better reagents and more biologically relevant systems. Interestingly the kinase that phosphorylates CBP such that it now is recruited to p65 was none other than the IκB kinase IKKα [83].

It is important to note that IKKα is the major kinase that drives the non-canonical NFκB activation and the non-canonical pathway was also found to regulate p53 function through an alternative mechanism [84]. This report documented that the NFκB subunit p52 could activate or repress p53 target promoters in a context specific manner. It will be interesting to investigate if inhibition of p53 activity is a mechanism by which the non-canonical pathway could contribute to tumorigenesis. It has also been recently reported that some drugs can simultaneously activate p53 and inhibit NFκB [85, 86]. Thus, designing bitargeted (or dual targeted?) drugs that are capable of inhibiting NFκB and simultaneously activating p53 could be one efficient way in cancer therapy [86, 87].

Involvement of NF-κB in carcinogenesis and cancer development

NF-κB and angiogenesis

Angiogenesis refers to a process of new blood vessel formation and is often closely associated with cancer development and metastasis. The role of NFκB in angiogenesis appears to be more complicated than originally thought. On one hand, it has been well established that some of the key angiogenesis factors such as vascular endothelial growth factor (VEGF), interlukine-6 (IL-6), monocyte chemotactic protein-1 (MCP-1), and metalloproteinases (MMPs) are known target genes of NFκB [88, 89]. Moreover, many of the NFκB inhibitors are also known to suppress angiogenesis [9093]. On the other hand, there is evidence showing an inhibitory role for NFκB in tumor angiogenesis. Kisseleva et al. [94] reported a striking increase in tumor vascularization in mice with a mutated IκBα, indicating an inhibitory role of NFκB in angiogenesis in vivo. Moreover, the effects of some known angiostatic agents such as 16-kDa N-terminal fragment of human prolactin (16K hPRL), angiostatin, and Neovastat are closely associated with activation of NFκB [9597]. Since both suppression of NFκB and inhibition of angiogenesis are important approaches in cancer therapy, such findings thus raise a cautionary note about the pharmaceutical agents that block NFκB in cancer therapy [96, 98].

In addition to NFκB, the mTOR pathway is another important regulatory mechanism that influences angiogenesis. A study by Hung and coworkers has provided evidence linking components of the NFκB pathway with mTOR-mediated angiogenesis independent of the transcriptional activation of NFκB. In their study, IKKβ was found to upregulate mTOR activity through direct phosphorylation of TSC1 at ser487 and ser511 [39]. In fact such observations provide additional link between inflammation and cancer development since inflammatory cytokines such as TNFα are capable of promoting angiogenesis via the IKKβ-mTOR pathway.

NF-κB and cancer metastasis

Cancer metastasis is a complex cascade of biological events, including adhesion, migration, and invasion, which finally allows tumor cells to escape from primary site and invade and proliferate at ectopic environments. It has been well documented that many of the above-mentioned processes are under the effluence of NFκB via its transcriptional control of an array of target genes, including vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), MMPs, chemokine receptor CXCR4, and serine protease urokinase-type plasminogen activator (uPA) [99102]. One good example is the observation that the IKKβ/IκBα/NF-κB pathway is required for the induction and maintenance of epithelial-mesenchymal transition (EMT) and blockage of NFκB suppress EMT and then abrogates the metastatic potential of mammary epithelial cells in a mouse model [99, 103].

Similar to the IKK-dependent but NFκB transcription-independent control of angiogenesis, there is also a specific link between IKK and metastasis. Luo et al. [104] found that activation and nuclear localization of IKKα in prostatic epithelial tumor cells represses the expression of the metastasis suppressor gene maspin, thus leading to metastasis of malignant prostatic epithelial cells. Conversely, genetic inhibition of IKKα kinase activity was found to promote maspin expression and reduce metastatic potential of the cancer cells [104].

Based on the discussion above, it is obvious that the NFκB pathway is a legitimate target for development of effective agents and approaches in treatment or prevention of metastasis in cancers. Both pharmacologic inhibitors or genetic approaches have been demonstrated to restrain the adhesion, migration and invasion processes and ultimately suppress or prevent cancer cell metastasis in vivo and in vitro [93, 105, 106]. In addition to chemical inhibitors, a recent study revealed that microRNA-146a and microRNA-146b (miR-146a/b) are capable of negatively regulating NFκB activity and metastatic potential of the highly metastatic human breast cancer cell line MDA-MB-231, thus suggesting that modulating miR-146a/b levels has therapeutic potential to suppress breast cancer metastases [107].

NF-κB in UV-induced skin cancer

Ultraviolet (UV) exposure is the most important etiological factor for the non-melanoma skin cancer, in particular squamous cell carcinoma (SCC) [108]. Both UVB (wave length ranging from 280 to 320 nm) and UVA (320–400 nm) have been demonstrated to be potent carcinogens capable of inducing initiation, promotion and progression of skin cancer in various mouse models [109, 110]. On the other hand, the critical role of NFκB in skin physiology and pathology has been well studied. It is recognized that NFκB has unique two-faced functions in skin: both activation and inhibition of NF-κB causes inflammation and promotes cell proliferation [17].

At present the involvement of NFκB in UV-induced skin carcinogenesis has been controversial. At one end, NFκB may contribute to UV-induced carcinogenesis and one of the supporting evidence is based on the observations that UV exposure readily activates NFκB signaling pathway in keratinocytes [110, 111], although the signaling mechanism controlling UV-induced NFκB activation has not been fully elucidated. It is known that UV irradiation-induced NFκB activation is via a distinct pathway from that of other stimuli such as TNFα. First, UV-induced NFκB could be IκBα-independent [110, 112]. Second, UV exposure leads to ubiquitin-dependent degradation of IκBα, but this process does not depend on phosphorylation of the N-terminal serines or activation of IKK [113, 114]. At present, the upstream kinases responsible for UV-induced IκBα phosphorylation remain largely elusive, except one serine/threonine kinase casein kinase II that has bee implicated in UV-and other DNA damage induced IκBα phosphorylation and degradation [115].

The second line of evidence supporting the tumor promoting function of NFκB in UV-induced skin carcinogenesis is from studies that suppression of NFκB signaling is capable of suppressing UV-induced skin cancer. For instance, parthenolide is a well established NFκB inhibitor via specific suppression of the IKK complex [106] and earlier work in our laboratory has found that parthenolide can effectively block UVB-induced in skin carcinogenesis in hairless mice [116]. Similarly, topical treatment of other known NFκB inhibitors such as aspirin, sodium salicylate, or glycolic acid delayed UVB-induced tumor formation in SKH-1 mice [117, 118]. However, caution should be excised in the extrapolation of these findings since most of these chemical inhibitors are not specific to NFκB and UV is capable of eliciting many other biological responses such as activation of AP-1. It remains to be further determined as to whether NFκB serves a real and the sole molecular target for the chemopreventive activity of these compounds.

On the other hand, there is growing evidence suggesting that NFκB may possess tumor suppressive function in skin cancer. For instance, selective inhibition of Rel/NFκB signaling in murine skin, by targeted overexpression of a super-repressor form of IκBα, results in an increased basal frequency of apoptotic cells and the spontaneous development of SCC [119]. Consistently, IKKα was recently identified as a tumor suppressor in SCC [120]. Based on the evidence from IKKα−/− mouse, IKKα plays a vital function in controlling epidermal development and differentiation, independent of its kinase activity and its ability to control NFκB signaling [12, 121, 122]. Moreover, it was found that reduced IKKα expression provides a selective growth advantage to promote the formation of mouse SCC induced by chemical carcinogens [123], a process likely involving activation of the epidermal growth factor receptor (EGFR)-driven pathway [124]. It remains to be tested whether suppression of the IKKα function would affect UV-induced skin carcinogenesis. Nevertheless, such findings prompt us to rethink the strategy of using NFκB inhibitors for control of skin cancer related to environmental carcinogens, in particularly UV.

NFκB in chemical-induced heptocarcinogenesis

Hepatocellular carcinoma (HCC) is one of the most prevalent cancers worldwide, especially in males in Asia and Africa. Infection with hepatitis viruses and exposure to environmental hepatocarcinogens such as aflatoxins are probably the two most important etiological factors for HCC [125]. However, the molecular mechanisms linking the risk factors with the carcinogenic processes are largely unknown. Recently, a number of elegant studies have provided experimental evidence implicating the role of NF-κB in hepatocarcinogenesis.

The notion that NFκB functions as a tumor promoter in HCC was derived from a transgenic animal model in which inhibition of the NFκB signaling pathway by a hepatocyte-specific inducible expression of IκB-supper-repressor resulted in tumor promotion [14]. Such observation was consistent with earlier findings that NFκB was often activated in human HCC associated with hepatitis [125], as well as the tumor promoting function of IKKβ in colitis-associated cancer in mice [21]. Such observations thus suggest that suppression of NFκB could serve as the practical strategy for controlling the carcinogenic process for some common cancers including HCC.

Interestingly, opposing findings were obtained from other transgenic models in which the NFκB signaling pathway was disrupted. An elegant study by Maeda et al. [126] provided convincing evidence that hepatocyte-specific deletion of IKKβ in mice significantly promoted diethylnitrosamine (DEN)-induced HCC. More importantly, this study revealed the cell type-specific effect of NFκB on the carcinogenic process as the specific deletion of IKKβ in Kuppfer cells, the macrophage-type of cells in liver, had opposite effect, resulted in reduced susceptibility to DEN-induced hepatocarcinogenesis [126]. This is probably the first study showing cell-type specific effect of NFκB on carcinogenesis in which specific suppression of NFκB was achieved in different cell types in the same animal model. At present, it is believed that IKKβ-deletion in hepatocytes results in augmented cell death due to accumulation of ROS, prolonged JNK activation in hepatocytes with IKKβ deletion, and subsequently leads to enhanced compensatory cell proliferation and tumor promotion [53, 127]. Understanding the cell-type specific effect of NFκB is obviously important in developing effective agents for cancer prevention by targeting NFκB.

A recent report using another transgenic animal model extended the above findings by demonstrating that specific deletion of NEMO/IKKγ in hepatocytes was capable of inducing spontaneous development of HCC in mice [23], indicating a physiological function of NFκB in protecting cancer development in liver. Both IKKβ and IKKγ are the key components of the IKK complex responsible for the canonical activation of NFκB [128]. Interestingly, the enhanced risk of spontaneous HCC was not observed in mice with IKKβ deletion, suggesting that NEMO/IKKγ may play a more critical role as a tumor suppressor in liver than IKKβ and IKKβ deletion in hepatocytes failed to completely abolish the NFκB activation [23]. It would be of interest to test whether deficiency in IKKβ in the above transgenic mice would affect chemical-induced HCC as reported earlier [126].

Taken together, it is clear that a basal and signal-induced activation of NFκB is required for the protection of the liver from spontaneous or carcinogen-induced tumorigenesis, at least in experimental settings. Although the anti-tumor function of NFκB based on the above studies is somehow contradictory to the common belief about the tumor promoting function of NFκB, such findings carry obvious implications in developing NFκB inhibitors as anti-tumor agents. More studies are needed to clarify whether suppression of NFκB in specific cell type may actually enhance cancer risk.

NFκB in virus-induced leukemia/lymphoma

The study with human T-cell leukemia virus type I (HTLV-I) and adult T-cell leukemia/lymphoma (ATL) probably provides the most convincing evidence linking NFκB with cancer development [129131]. HTLV-I is a type of retrovirus that has been etiologically associated with acute T-cell malignancy in humans. The first evidence supporting the role of NFκB in HTLV-I-mediated carcinogenesis was from observations that the transcriptional activator Tax from HTLV-I is a potent inducer for NFκB activation [132, 133]. Tax has been well studied as the key viral protein from HTLV-I in the pathogenesis of ATL [134]. Tax is able to activate NFκB via both pathways. In the canonical pathway, Tax activates NFκB by stimulating the activity of the IKK via direct protein–protein interaction, leading to phosphorylation and degradation of IκB [135]. In addition, Tax also activates the noncanonical pathway via p100 processing, a process mainly involving IKKα [136]. At present, it has been well established that Tax-induced activation of NFκB plays a critical role in the transformation and clonal proliferation of HTLV-1-infected cells. For instance, recombinant HTLV-1 virus carries the Tax M22 mutant that can activate cyclic adenosine monophosphate response element binding protein (CREB) but not NFκB is unable to immortalize T cells. In contrast, the M47 mutant, which can activate NFκB but not CREB, is able to induce T-cell immortalization [137]. Additional mechanism may also involve the tumor suppressor p53 as Tax-induced activation of NFκB results in the inactivation of p53 [138, 139].

Moreover, recent reports also suggested the presence of constitutive NFκB activation in ATL cells independent of HTLV-1 infection and Tax protein [140], which further highlights the importance of NFκB in the pathogenesis of ATL.

Based on the above discussion, it is obvious that NFκB would serve as a legitimate target for development of therapeutic agents against ALT. At the present time, there are several trials using the IKK inhibitor Bay11-7082 [141] and the proteasome inhibitor PS-341 [142] for treatment of ATL. Recently, a new NFκB inhibitor, dehydroxy-methylepoxy-quinomicin (DHMEQ), has been found to inhibit both conical and non-conical NFκB pathways induced by TAX-as well as the constitutive NFκB activation in primary ATL cells, without affecting control peripheral blood mononuclear cells, thus making DHMEQ an ideal candidate for development of new class of anti-ATL therapeutic agents [143, 144].

NFκB as the molecular targets for cancer therapy

NFκB inhibitors as cancer therapeutics

Based on the evidence describing the role of NFκB in various clinical and experimental settings, it is evident that NFκB plays a pivotal role in suppression of apoptosis, promotion of cell proliferation and inflammation, and is closely associated with cancer development. Thus, the NFκB signaling pathway becomes a legitimate target for cancer therapy [45]. Tremendous effort has been invested by the academic community, and the biotech and pharmaceutical industry in development of NFκB inhibitors as cancer therapeutics. Based on their chemical nature, the NFκB inhibitors that are currently in preclinical or clinical trials include the following groups: (1) small molecule compounds from natural/dietary sources, such as curcumin, resvetriol, genistein, and parthenolide, (2) synthetic small molecule compounds, such as Bortezomib (formely PS-341), MG-132, flavopiridol, Bay 11-7082, and NSAIDs, (3) cell permeable peptides such as SN-50, and (4) gene therapy such as overexpression of IκBα-supper repressor and NFκB decoy oligodeoxynucleotide (ODN) [2, 8, 145148]. These inhibitors are capable of acting on various steps of the NFκB signaling pathway, including (1) direct inhibition of IKK (by Bay 11-7082), (2) blockage of the proteasome activity to inhibit IκBα degradation (by Bortezomib), (3) prevention of nuclear translocation of the NFκB protein (by SN-50), and (4) suppression of NFκB binding to DNA (by NSAIDs). So far, one successful story is Bortezomib (formerly PS-341), a specific proteasome inhibitor capable of suppression NFκB activation, has obtained FDA approval for treatment of multiple myeloma [149]. It is expected to see more new anti-cancer drugs targeting NFκB to be available in clinical oncology in the coming years.

NFκB inhibitors as sensitizers in cancer therapy

Although NFκB inhibitors alone hold great potential as cancer therapeutics, in recent years, various NFκB inhibitors have emerged as potent chemo-sensitizers to the established chemotherapy, biotherapy and radiation-therapy. The rationale for such combined therapy is based on (1) NFκB is constitutively activated in human cancers, (2) many cancer chemotherapeutics are able to activate NFκB, and (3) NFκB is closely related to drug resistance [147, 150153]. Therefore, suppression of NFκB signaling pathway would offer the benefit of overcoming the resistance, reducing the dosage of the anti-cancer drugs and alleviating the side effects [153].

Table 1 summarizes some of the recent reports in which small molecule NF-κB inhibitors were used with some of the common chemotherapeutic agents, mostly direct DNA damage agents. In most of these studies, the NFκB inhibitors are able to promote cell death (apoptosis) both in cultured cancer cells in vitro, as well as in xenografted tumors in nude mice in vivo. As chemo-resistance is one of the major issues in cancer chemotherapy, it is believed that these NFκB inhibitors have the potential to be used as the adjuvant agents and further studies, in particular clinical trials for those combinations with promising results from in vivo animal models, would be required to develop the combined therapy in cancer. Recently a number of clinical trials (phase I and II) with the proteasome inhibitor Bortezomib as the adjuvant agent in chemotherapy have been reported, mostly in advanced or refractory solid tumors and multiple myeloma [154157].

TNFα and TRAIL are members of the TNF superfamily that are capable of inducing cell death via death receptors-mediated apoptosis pathway in tumor cells, holding potential as cancer bio-therapeutic agents [158161]. One of the major issues hindering their clinical application is the adaptive resistance and activation of the NFκB pathway is one of underlying molecular mechanisms responsible for the resistance [158, 159]. Therefore, suppression of NFκB becomes a logical and important approach in overcoming the resistance to TNF and TRAIL. Table 2 summarized some of the recent studies in which various small molecule NFκB inhibitors were found to promote apoptosis and inhibit tumor growth in both in vitro and in vivo systems in combination with TNF or TRAIL. Such findings provide important information for further evaluation with clinical trials to establish the regime of combined therapy in human cancers.

Radiotherapy is another well established therapeutic modality against cancer in clinical oncology. Ionizing radiation is able to kill cancer cells mainly via induction of DNA damage and mitochondria-dependent cell death pathway [162, 163]. Similar to chemotherapeutic agents and cell death ligands (TNFα and TRAIL) as discussed above, radiation is also a potent inducer of NFκB which is known to be one of the underlying mechanisms responsible for resistance to radiotherapy in cancer cells [164, 165]. Table 3 lists some of the recent reports using small molecule NFκB inhibitors in sensitizing cancer cells to γ-irradiation. In most of these studies, suppression of the NFκB signaling pathway was found to promote apoptosis in vitro and to enhance the efficacy of γ-irradiation in animal models in vivo. Most of these combinations certainly deserve further evaluation in clinical trials.
Table 1

NFκB inhibitors as sensitizing agents on cancer chemotherapeutics

NFκB inhibitors

Chemo-therapeutics

Cancer type

Effect/action

References

Genistein

Docetaxel

Cisplatin

Doxorubicin

Prostate

Breast

Lung

Pancreas

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[171173]

Curcumin

Paclitaxel/taxol

Gemcitabine

Breast

Cervical

Bladder

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[174176]

Resveratrol

Bortezomib thalidomide

Multiple myeloma

(1) Sensitize to apoptosis in vitro

[177]

Parthenolide

Paclitaxel

Docetaxel

NS398

Arsenic trioxide

Breast

Liver

Prostate

Leukemia

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[178182]

DHMEQ

Taxanes

Cisplatin

Thyroid

Liver

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[183, 184]

I3C/DIM

Erlotinib

Docetaxel

Pancreas

Breast

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[185187]

Bay 11-7082

UCN-01

HDACIs

Vincristine

Doxorubicin

Paclitaxel

Cisplatin

Multiple myeloma

Leukemia

Ovarian

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[188192]

Proteasome inhibitors (Bortezomib, PS-341)

HDACIs

CPT-11

Squamous cell

Colorectal

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[193, 194]

Table 2

NFκB inhibitors as sensitizing agents on cell death ligands

NFκB inhibitors

Cytokines/cell death ligands

Cancer type

Effect/action

References

Proteasome inhibitors (Bortezomib, MG132, NPI-0052)

TRAIL

Lung

Prostate

Liver

Glioblastomas

Pancreas

Colon

Leukemia

Kidney

Sensitize to apoptosis in vitro

[195202]

Curcumin

TRAIL

TNF

Prostate

Bladder

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[175, 203205]

Resveratrol

TRAIL

Melanomas

Sensitize to apoptosis in vitro

[206]

Luteolin

TRAIL

TNF

Lung

Colon

Liver

Cervical

Sensitize to apoptosis in vitro

[207, 208]

Parthenolide

TRAIL

TNF

Breast

Multiple myeloma

Colon

Cervical

Nasopharyngeal

Sensitize to apoptosis in vitro

[209211]

I3C/DIM

TRAIL

Colon

Liver

Cervical

Sensitize to apoptosis in vitro

[212]

Table 3

NFκB inhibitors as sensitizing agents on γ-radiation

NFκB inhibitors

Cancer type

Effect/action

References

Curcumin

Neuroblastoma

Colon

Cervical

(1) Sensitize to apoptosis in vitro,

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

[213215]

Resveratrol

Melanoma

Lung

Sensitize to apoptosis in vitro

[216, 217]

Genistein

Prostate

Sensitize to apoptosis in vitro

[218]

Parthenolide

Prostate

Sensitize to apoptosis in vitro

[219]

Proteasome inhibitors (Bortezomib, Pro1, MG132)

Cervical

Ki-Ras-transformed prostate epithelial cells

Melanoma

Head-and-neck squamous cell carcinoma

Prostate

Colorectal

(1) Sensitize to apoptosis in vitro

(2) Enhance therapeutic efficacy in nude-mice xenograft in vivo

(3) Enhance radiotherapy in patients in clinical trials

[220225]

Upregulation of IAP (inhibitor of apoptosis) proteins has been noted in many cancer types that. It is believed that their increased level/activity contributes to resistance of cancer cells to apoptosis. Endogenous proteins such as Smac that inhibit IAPs do exist and have served as a model to design drugs that can inhibit IAP in clinical settings. Three independent reports in the last year documented that these smac mimetic drugs cause autoubiquitination and degradation of cIAPs [166, 167]. This removal of cIAPs leads to NFkB dependent autocrine upregulation of TNFα and hence TNFa -mediated cell death. Re-establishing the apoptotic program in tumor cells is a much dreamed about strategy for cancer therapy. The use of such Smac mimetics in combination with other forms of chemo/radiation therapy should be great new ways of enhancing the efficacy of cancer therapies.

Taken together, great efforts in the past two decades in searching for effective and selective NFκB inhibitors as anti-cancer agents has been an excellent example linking bench (basic research) to bedside (clinical application). These NFκB inhibitors, either standing alone as therapeutic agents or acting as sensitizers to chemo-, bio- and radiation-therapy, are expected to provide a new line of hope in improving the current therapeutic regimes in cancer.

Conclusions and perspectives

The recent advance in our knowledge of the NFκB signaling has been vital in propelling our efforts towards making a therapeutic remedy for ailments with deregulated NFκB activity such as cancers. However, we have also learnt in the very recent past that NFκB can do diametrically opposite things in response to the same or similar signaling in different tissues under different contexts. Apart from p53, NFκB pathway has also been documented to interact with many other signaling cascades such as Notch that are also known to be deregulated in cancers [168]. How does one then begin to make specific inhibitors of NFκB that block its effects in a given tissue while leaving its physiological functions in other tissues largely intact? Understanding and cataloguing how or why NFκB functions differently in different tissues in response to a number of well known signaling cascades will be the first step that we can take towards this. The identification of NFκB in the vertebrate organism zebrafish [169, 170] has given us another powerful tool to understand the role of NFκB in organogenesis and early development. Indeed Zebrafish is being used as a model to study cancer by investigators around the world. Using such model organisms will also facilitate the application of large scale genetic screens to study the function of NFκB inhibitors. The development of various, high-throughput, cell based assays to screen large number of compounds, siRNA, cDNA and miRNA libraries and the knowledge obtained will be instrumental to further develop leads into drugs for various diseases including cancers. Not having the crystal structure of IKK complex as a whole or of some parts of its components is surely hindering our progress to develop better small molecules that could specifically inhibit this pathway. An impressive number of clinical trials are underway currently which are testing the efficacy and specificity of rationally designed drugs that inhibit NFκB. Hopefully, the next few years will lead to the identification of the new therapies that will inhibit NFκB without inducing severe side-effects.

Acknowledgments

The work in HMS’s Lab is in part supported by research grants from Singapore Biomedical Research Council (BMRC), Singapore National Medical Research Council (NMRC), and University Research Council (URC), NUS.

Copyright information

© Springer Science+Business Media, LLC 2009