The Role of the NF-kappaB Transcriptome and Proteome as Biomarkers in Human Head and Neck Squamous Cell Carcinomas

Abstract

NF-kappaB is a family of signal activated transcription factors comprised of hetero- or homo-dimers from 5 different subunits, NF-kappaB1, NF-kappaB2, RELA, cREL and RELB. NF-kappaBs normally are transiently activated in response to infection or injury, but in cancers are aberrantly activated, regulating a transcriptome of hundreds of genes and corresponding proteome that promote pathogenesis and therapeutic resistance. In head and neck squamous cell carcinomas, an important role of NF-kappaB in regulation of the altered transcriptome and proteome has been established, providing a catalog of activating and target genes and proteins that may be useful as biomarkers of alterations in this pathway for this and other cancers. An emerging appreciation that NF-kappaB and other signal pathways form an altered regulatory network highlights the need to use biomarkers and combine targeted agents for personalized therapy of cancer.

Figure 1

NF-κB activation in cancer. NF-κB activation in cancer development is linked with chronic exposure to bacteria, certain viral products, chemical promoters, and carcinogens and reactive oxygen species (ROS), which cause repeated DNA damage. Induction also occurs in response to cytotoxic chemotherapy and ionizing radiation. A. Canonical NF-κB Pathway activation occurs in response these bacterial, viral, chemical or physical stimuli; by aberrant cytokine, integrin and growth factor ligand/receptor activation (e.g., TNFR, IL-1R, Integrin α6β4, EGFR, Her2/Neu and other serum factors); expression of activating fusion proteins following translocations (BCR-ABL, MALT1); or aberrant activation by intermediate kinases (e.g., PI3K, CK2, AKT). Intermediate kinases convey signals to the Inhibitor-κB complex formed by IKKα, β and γ, and IKKβ phosphorylates Inhibitor-κB, marking it for ubiquitiantion by E3 ligase βTrCP (SCF) and proteasome degradation. P105/RELA or cREL is processed to NF-κB1 (p50)/RELA or cREL heterodimers, which translocate to the nucleus, and bind promoters of genes regulating proliferation, apoptosis, migration inflammation, angiogenesis and innate immunity. B. Alternative or Non-canonical Pathway. The alternative pathway may be activated by other TNF family members via NIK, and involves IKKα/ IKKα homodimers, which activate NF-κB2/p100 for processing into p52/RELB heterodimers. The REL-B/p52 heterodimer then translocates into the nucleus to bind the promoter of genes whose products are important for the malignant phenotype in some cancers and B cell development and adaptive immunity. Abbreviations: NF-κB, nuclear factor κB; IKK, Inhibitor-κB kinase; NIK, NF-κB inducing kinase; IL-1R, Interleukin 1 Receptor; EGFR, Epidermal Growth Factor Receptor; TNFR, Tumor Necrosis Factor Receptor; PI3-K, phosphatidylinositol 3-kinase.

Figure 2

Co-activation of IL-1R/TNFR-NF-κB and EGFR/MAPK/STAT3/AP-1 pathways and proliferative, survival and angiogenesis factor genes in HNSCC may be inhibited by proteasome and EGFR inhibitors. Through transcription factor binding sites in the regulatory promoter region of genes, the MAPK-AP1, STAT3 and NF-κB pathways have been shown to independently promote activation of proliferative and prosurvival genes cyclin D1, IAP1 and BCL-XL, as well as angiogenesis factors IL-6, IL-8, VEGF and HGF, which contribute to the malignant phenotype and therapeutic resistance of HNSCC. Thus, targeting EGFR mediated activation of MAPK and STAT3 by EGFR tyrosine kinase or antibody inhibitors, and proteasome dependent activation of NF-κB by bortezomib, has potential for combinatorial activity against co-activated pathways important in HNSCC.

Figure 3

Hierarchical clustering analysis of differentially expressed genes in UM-SCC cells (reproduced from Yan et al, Genome Biol, 2007). 1011 differentially expressed genes were extracted from 24K cDNA microarray database, based on a two-fold and above difference among human normal keratinocytes (HKC), UM-SCC cells with different p53 status (t-test score at P < 0.05, two-tailed). The hierarchical clustering tree was generated using Java Treeview. Four HKCs were grouped on the left, and 5 UM-SCC cell lines with wild-type p53-deficient (wt p53-def) expression pattern were grouped together in the middle, and 5 UM-SCC cell lines with mutant (mt) p53 grouped to the right, respectively. Over-expressed genes are indicated by red and under-expressed genes by green; and the expression level is proportional to the brightness of the color (see color bar). A, entire hierarchical clustering tree included 3 up-regulated clusters A, B and C (including sub-clusters C1-3), and 3 down-regulated D, E and F; B, cluster A consisted of 34 genes; and C, cluster B consisted of 37 genes.

Figure 4

Frequency of putative transcription factor binding sites (TFBS) in proximal regions of promoters (reproduced from Yan et al, Genome Biol, 2007). The promoter sequences were extracted from the over-expressed clusters A, B and C1-3 genes in UM-SCC cells using Genomatix Suite 3.4.1 (www.genomatix.de). The average length of these promoters was adjusted to approximately 600, including ~500 bp upstream and ~100 bp downstream from TSS. The promoter sequences from vertebrates represented 159,505 promoters, including 55,207 from human, 69,108 from mouse and 35,190 from rat in Genomatix promoter database (GPD). The P value of TFBS frequency in a given cluster was calculated by MatInspector of Genomatix Suite 3.4.1. * indicates the significantly increased frequencies of putative binding motifs on promoter regions of clustered genes when compared with the vertebrate promoters with a randomly drawn samples of the same size (P< 0.05). ^† indicates a significantly lower frequency of the AP-1 binding motif when compared with the vertebrate promoters.