Gavin P. Collett, Frederick C. Campbell, Overexpression of p65/RelA potentiates curcumin-induced apoptosis in HCT116 human colon cancer cells, Carcinogenesis, Volume 27, Issue 6, June 2006, Pages 1285–1291, https://doi.org/10.1093/carcin/bgi368
Download citation file:
© 2019 Oxford University PressClose
Permissions Icon Permissions
Curcumin, the yellow pigment in the spice turmeric, has potent chemopreventive activities that involve diverse molecular pathways. It is widely believed that curcumin pro-apoptotic properties are mediated by downregulation of NF kappa B (NFκB). The p65/RelA subunit of NFκB may influence cell death, in part by activation of NFκB anti-apoptotic target genes including X-linked inhibitor of apoptosis ( XIAP ), A20 , bcl-xL and inhibition of sustained activation of c-Jun N-terminal kinase (JNK). We have shown previously that curcumin inhibits NFκB, activates JNK and promotes apoptosis in HCT116 colorectal cancer cells. Here, we show that forced overexpression of p65 does not affect curcumin-induced JNK activation. Indeed, overexpression of p65 enhanced curcumin-mediated apoptosis as assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT) assay and poly(ADP-ribose) polymerase (PARP) cleavage. This potentiating effect of p65 upon curcumin-mediated apoptosis was reversed by transfection of cells with an IκB super-repressor (ΔNIκB). Curcumin treatment inhibited expression of NFκB anti-apoptotic target genes in mock-transfected and in p65-overexpressing HCT116 cells, although expression levels remained higher in the latter. Taken together, these results show that curcumin-mediated activation of JNK or induction of apoptosis does not require inhibition of p65. Furthermore, curcumin/p65 synergy in promotion of apoptosis cannot be attributed to active repression of NFκB anti-apoptotic genes.
Curcumin (diferuloylmethane) is a naturally occurring polyphenolic pigment, isolated from the rhizomes of the plant Curcuma longa (Linn.), and is commonly used as a colouring and flavouring agent in food products. Curcumin has antioxidant ( 1 ) or anti-inflammatory properties ( 2 ) and inhibits tumourigenesis in various tissues in animal models ( 3 – 6 ). These anti-tumourigenic effects are associated with induction of apoptosis ( 3 , 7 – 9 ). Curcumin has diverse molecular effects ( 10 – 14 ) including inhibition of nuclear factor kappa B (NFκB) in colon cancer and other cells ( 15 – 17 ).
The NFκB family of transcription factors consists of five Rel-domain-containing proteins: p65/RelA, RelB, c-rel, p50/NFκB1 and p52/NFκB2. In most cells the predominant form of active NFκB consists of a p65/p50 heterodimer although other homo/heterodimers also form ( 18 ). NFκB has been shown to promote cell survival through induced expression of anti-apoptotic genes such as bcl-xL ( 19 ) and X-linked inhibitor of apoptosis ( XIAP ) ( 20 ). In physiological conditions, NFκB is sequestered in inactive form in the cytoplasm by inhibitory IκB proteins. However, constitutive activation of NFκB has been observed in several cancer cell types and may contribute to apoptosis resistance ( 21 , 22 ). The p65 subunit of NFκB may promote cell survival by induction of growth arrest and DNA damage-inducing gene beta ( GADD45β ) ( 23 ) or XIAP ( 24 ) leading to inhibition of c-jun N-terminal kinase (JNK) activity. When p65 is inhibited, cells show sustained activation of JNK in response to a stimulus such as tumour necrosis factor α (TNFα) and undergo apoptosis. Inhibition of NFκB may also upregulate GADD45α , which in turn activates JNK ( 25 ).
We have shown previously that curcumin induces sustained activation of JNK, which promotes apoptosis in human HCT116 colon cancer cells ( 26 ). As HCT116 cells exhibit constitutive NFκB activity, which in turn may be inhibited by curcumin ( 26 ), we hypothesized that sustained activation of JNK and induction of apoptosis by curcumin may occur in these cells as a result of inhibition of NFκB transcriptional activity. In order to test this hypothesis we sought to override the inhibitory effect of curcumin on constitutive NFκB activity by overexpressing p65 in HCT116 cells. While p65 may impede apoptosis, growing evidence suggests that it may also potentiate cell death depending upon the nature of the apoptotic stimulus ( 27 – 29 ). Indeed, p65 can be both an activator and repressor of its target genes depending upon the manner in which it is induced ( 30 ). Here, we show that overexpression of p65 has no effect on curcumin-induced JNK activation but potentiates curcumin-induced apoptosis. Furthermore, enhancement of curcumin-induced apoptosis by p65 overexpression cannot be attributed to active repression of NFκB anti-apoptotic target genes, XIAP , bcl-xL or A20 .
Materials and methods
Curcumin was 80% pure (98% curcuminoid content) and was obtained from Sigma-Aldrich (Poole, Dorset, UK). All antibodies were obtained from New England Biolabs (Hitchin, Hertfordshire, UK), except for anti-β-actin, which was purchased from Sigma-Aldrich. All other reagents were widely available commercially.
The NFκB-responsive luciferase reporter construct, 3EnhConALuc ( 31 ), was kindly provided by Prof. Ron Hay, University of St Andrews. The p65 expression vector, pCMV-p65 ( 32 ), was a kind gift from Dr Warner Greene, University of California, San Francisco. The pCMV-ΔNIκBα construct, an N-terminal deletion mutant of IκBα ( 33 ), was kindly provided by Dr Dean Ballard, Vanderbilt University, Nashville.
Human colon cancer cell line HCT116 was obtained from the European Collection of Cell Cultures (ECACC; Salisbury, UK) and grown as monolayers at 37°C in a humidified atmosphere with 5% CO 2 in Dulbecco's modified Eagle's medium (DMEM) containing 10% (v/v) heat inactivated foetal calf serum, glutamine (2 mM), penicillin (50 U/ml) and streptomycin (50 µg/ml). Curcumin was dissolved in dimethyl sulfoxide (DMSO) and used in the previously determined optimal concentration of 35 µM ( 26 ). Controls were treated with 0.1% DMSO alone. All treatments were carried out on cells at 60–80% confluence.
Transient transfections and luciferase assay
Transfection was performed in 24-well plates using Lipofectin reagent (Invitrogen, Paisley, UK) according to the manufacturer's instructions. Briefly, cells were grown in 24-well plates and transfected with the appropriate vector (250 ng) the following day. After 5 h the transfection mix was removed and replaced with complete medium. Cell treatments were then carried out 24 h post-transfection as indicated. Luciferase activity was determined using the luciferase assay system with reporter lysis buffer from Promega, Southampton, UK. Results, which were expressed as relative luciferase activity, were corrected for differences in transfection efficiency by co-transfection with a pCMV-βGal construct followed by β-galactosidase assay (Promega).
Cell monolayers were washed with phosphate-buffered saline (PBS) and lysed directly into SDS–PAGE loading buffer. Soluble protein (30 µg) was resolved by SDS–PAGE and transferred to nitrocellulose membrane. For all experiments, equal protein loading was confirmed by staining the nitrocellulose with Ponceau S or by probing with an antibody to β-actin. Membranes were probed with the appropriate primary antibodies. Reactions were visualized with a suitable secondary antibody conjugated to horseradish peroxidase (HRP) (Dako, Ely, UK) using an enhanced chemiluminescence system (Santa Cruz, California, US).
Cell viability assay
Cells were seeded at 1 × 10 4 cells/well in 96-well plates and transfected as described above. Cells were treated with curcumin (35 µM) for 24 h and then 5 mg/ml MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide; 20 µl] was added and incubated for 1 h. Cells were then incubated overnight with 20% SDS and the formation of coloured formazan dye was assessed colorimetrically at 550 nm. Results are expressed as percentage loss of cell viability compared with control.
Semi-quantitative reverse transcriptase–polymerase chain reaction (RT–PCR)
Total RNA was prepared using Trizol reagent (Gibco, Paisley, Scotland). RT–PCR was carried using a One-Step RT–PCR kit (Qiagen, Crawley, UK) in a final volume of 12.5 µl according to the manufacturer's instructions. Primer sequences were as follows: bcl-xL forward, 5′-CCCAGAAAGATACAGCTGG; bcl-xL reverse, 5′-GCGATCCGACTCACCAATAC; XIAP forward, 5′-CAGATAGGCTTAACAAATGGAGCT; XIAP reverse, 5-ATAGTGTCCCGCCACTTGCATT; A20 forward, 5′-ACTCTTTGGGTTATTACTGTC; A20 reverse, 5′-ACAGGTTATTATTATTTAGCC. Reactions were normalized by evaluating the level of amplification of the glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) transcript using the following primers: GAPDH forward, 5′-AGTTCAACGGATTTGGTCGTA, reverse, 5′-AAATGAGCCCCAGCCTTCT. For semi-quantitative analysis, each sample was amplified within the exponential phase of the PCR ( bcl-xL and XIAP , 26 cycles; A20 , 28 cycles; GAPDH , 21 cycles). Amplified DNA was electrophoresed on 1.5% agarose gels and visualized by ethidium bromide staining.
Data are presented as mean ± SD. Comparisons between treatment groups were made with the Student's t -test. Values of P < 0.05 were considered statistically significant.
Overexpression of p65 increases NFκB transcriptional activity in HCT116 cells
HCT116 cells express constitutive NFκB transcriptional activity, which is inhibited by curcumin ( 26 ). In order to confirm that overexpression of p65 resulted in an increase in NFκB activity, we co-transfected cells with pCMV-p65 expression construct and an NFκB-dependent luciferase reporter plasmid ( 31 ). p65 overexpression was confirmed by western blotting ( Figure 1A ). Overexpression of p65 resulted in a significant increase in constitutive NFκB transcriptional activity as measured after 6 and 18 h, respectively ( Figure 1B and C ). Curcumin (35 µM) treatment had no significant effect on NFκB transcriptional activity after 6 h in mock- or p65-transfected cells ( Figure 1B ) but resulted in a significant inhibition after 18 h ( Figure 1C ). NFκB transcriptional activity was significantly higher in p65-transfected cells compared with mock-transfected cells after 6 and 18 h, respectively, even after curcumin treatment ( Figure 1B and C ).
Activation of JNK signalling by curcumin is unaffected by overexpression of p65
To determine the effects of overexpression of p65 on JNK activation, we transiently transfected HCT116 cells with p65 or empty vector and assessed whole-cell lysates from curcumin-treated cells by western blotting analysis using an antibody that specifically recognizes the phosphorylated form of JNK. As shown in Figure 2 , curcumin treatment resulted in sustained phosphorylation of both p46 and p54 isoforms of JNK. Western blotting with an antibody that recognizes JNK regardless of its phosphorylation status revealed that overall levels of JNK were unaffected by curcumin. The activation of JNK by curcumin was confirmed by western blotting using an antibody specific to the phosphorylated form of the JNK substrate c-jun. Overexpression of p65 had no significant effect on sustained curcumin-induced activation of JNK or c-jun ( Figure 2 ).
p65 overexpression enhances curcumin-induced apoptosis
We have previously shown that curcumin induces apoptosis in HCT116 cells ( 26 ). In order to assess the effects of overexpression of p65 on curcumin-induced apoptosis, we transfected cells with p65 or empty vector and then treated with 35 µM curcumin for 24 h. Cell death was assessed by MTT assay and Poly(ADP-ribose) polymerase (PARP) cleavage assay. Figure 3A shows that p65 overexpression resulted in a significant reduction in cell viability after curcumin treatment compared with cells transfected with empty vector. Similarly, p65-overexpressing cells showed increased PARP cleavage in comparison with mock-transfected controls after curcumin treatment ( Figure 3B ), indicating an increase in apoptosis. Basal levels of apoptosis in untreated cells were unaffected by p65 overexpression. Treatment with vehicle (DMSO) alone did not induce apoptosis in either mock- or p65-transfected cells ( Figure 3B ).
IkB reverses the potentiation of curcumin-induced apoptosis after p65 overexpression
In order to confirm that activation of NFκB transcriptional activity was involved in the increase in curcumin-induced apoptosis after overexpression of p65, we transfected cells with a super-repressor form of IκBα (ΔNIκBα) that has been shown to efficiently block the activation of NFκB ( 33 ). As shown in Figure 4A , co-expression of p65 and ΔNIκBα resulted in a potent reduction in NFκB transcriptional activity, in both untreated and curcumin-treated cells, compared with cells transfected with p65 only. MTT and PARP cleavage assays showed that inhibition of NFκB transcriptional activity in cells overexpressing p65 was associated with a decrease in curcumin-induced cell death ( Figure 4B and C ). Cells co-transfected with pCMV4 and ΔNIκBα showed reduced NFκB transcriptional activity compared with transfection with pCMV4 alone, but this did not result in a change in basal levels of apoptosis in untreated cells. Curcumin-induced apoptosis was also unaffected by transfection with pCMV4 and ΔNIκBα, compared with PCMV4 alone. Taken together, these results show that increased NFκB transcriptional activity resulting from overexpression of p65 results in a potentiation of curcumin-induced apoptosis.
Curcumin downregulation of NFκB-responsive anti-apoptotic genes is not enhanced by overexpression of p65
Previous studies have shown that pro-apoptotic effects of p65 may be mediated by active repression of anti-apoptotic gene expression ( 30 ). To assess the expression of such genes in curcumin-treated cells overexpressing p65 we carried out semi-quantitative RT–PCR of XIAP , bcl-xL and A20 genes, which have been shown to be actively repressed by p65 in pro-apoptotic contexts, on RNA prepared from cells transfected with the p65 expression vector or empty vector and treated with curcumin. Figure 5 shows that expression of XIAP and bcl-xL was similar in mock- and p65-transfected cells, whereas A20 was significantly upregulated by p65 overexpression. Curcumin treatment resulted in a reduction in expression of XIAP and bcl-xL to a similar degree in both mock- and p65-transfected cells. A20 expression was reduced by curcumin in mock-transfected cells but was more highly expressed in p65-overexpressing cells at all time points.
Pro-apoptotic effects of curcumin are central to its chemopreventive efficacy. Multiple intracellular signalling pathways including NF-kappaB, JNK and caspases interact in the regulation of programmed cell death ( 29 , 30 ). NFκB is an ubiquitous transcriptional control factor, involved in diverse cellular processes. To dissect any NFκB/JNK signalling interactions implicated in curcumin-mediated cell death, the current study overexpressed p65, the major transactivating subunit of NFκB, in HCT116 cells and assessed effects on curcumin-induced JNK activation, NFκB transcriptional activity and apoptosis.
In this study, overexpression of p65 in HCT116 cells resulted in increased p65 protein expression, enhanced NFκB transcriptional activation and upregulation of the NFκB responsive gene, A20 . Curcumin treatment, at 35 µM, which optimally induces apoptosis in HCT116 cells ( 26 ), inhibited NFκB transcriptional activity and expression of NFκB target genes in both mock transfectants and p65-overexpressing cells. Curcumin suppression of NFκB was incomplete in p65 transfectants since activity remained at least 3-fold higher than in mock-transfected cells. Curcumin induced phosphorylation of JNK and c-jun as shown previously ( 26 ) and was unaffected by p65 overexpression. While the precise molecular mechanisms involved in curcumin-mediated activation of JNK remain unclear, upstream signalling pathways that could potentially be implicated include MKK4 and MKK7 kinases ( 34 ), apoptosis-signal-regulating kinase ( 35 ), PKC-δ (protein-regulating kinase-δ) ( 36 ), mixed lineage kinases ( 37 ) and reactive oxygen species ( 38 ). Taken together, these results show that curcumin-induced activation of JNK occurs independently of its inhibition of NFκB transcriptional activity, in HCT116 cells.
Previous studies have shown that curcumin inhibition of NFκB is accompanied by induction of apoptosis and it is widely considered that this mechanism is causal ( 16 , 39 , 40 ). Indeed, overexpression of p65 in L929 mouse fibrosarcoma cells induced resistance to curcumin-induced apoptosis ( 41 ). The present study shows that curcumin inhibits p65 expression, NFκB-dependent transcriptional activity and expression of NFκB-dependent anti-apoptotic genes. While these pathways have a well-defined role in apoptosis in various cell types ( 42 ), the present study indicates that they may be superfluous to curcumin-mediated apoptosis in HCT116 colorectal cancer cells. Indeed, overexpression of p65 sensitized the cells to curcumin-mediated apoptosis, as assessed by PARP cleavage and MTT assay. Differences between our work and the previous study ( 41 ) may be attributed to species or cell-type-specific differences in expression of unknown curcumin target genes or other signalling pathways implicated in cell death. Cell-type-specific effects of curcumin have been shown previously. For example, curcumin-induced apoptosis is p53-dependent in MCF-7 cells ( 43 ) but independent of p53 in melanoma cells ( 44 ).
Overexpression of p65 has been shown to have pro- or anti-tumourigenic properties in various tissues. For example, prostatic intraepithelial neoplasia shows overexpression of p65 compared with benign prostatic epithelium ( 45 ). Increased expression of p65 correlates with colorectal tumourigenesis ( 46 ). Furthermore, p65 overexpression in thyroid carcinoma cell lines increases proliferation rate and colony formation in soft agar ( 47 ). Conversely, overexpression of p65 in MCF7ADR cells reduces their tumourigenic ability in nude mice ( 48 ). In breast cancer cells, overexpression of p65 inhibits TRAIL-induced apoptosis through inhibition of caspase 8 and DR4 and DR5 expression ( 49 ), whereas in a pro-B cell line it induces G 1 arrest and apoptosis ( 50 ). Clearly, the role of p65 in the carcinogenic process is complex and may involve interactions with multiple signalling pathways in a context-specific manner. It is suggested that p65 may function as an inhibitor or activator of apoptosis, depending upon the nature of the stimulus ( 27 – 29 ). For example, ultraviolet light and daunorubicin induce p65–DNA binding but inhibit NFκB transcriptional activity and suppress NFκB anti-apoptotic target genes. Indeed, p65 can be both an activator and repressor of its target genes ( 30 ). However, pro-apoptotic effects of p65 overexpression in combination with curcumin cannot be explained on this basis since we found that (i) NFκB transcriptional activity in the presence of curcumin was increased in p65-transfected cells and (ii) downregulation of bcl-xL , XIAP and A20 mRNA expression by curcumin was not enhanced in p65-transfected cells compared with mock-transfected cells. In this study, p65 pro-apoptotic effects only occurred in synergy with curcumin treatment since cell death was unchanged in p65-overexpressing cells that were untreated by curcumin. Our results suggest that the pro-apoptotic effect of p65 in concert with curcumin could result from activation of one or more unknown pro-apoptotic target genes. NFκB has been shown to induce pro-apoptotic genes such as fas ligand ( 51 ) and fas ( 52 ) in response to arsenic trioxide and etoposide, respectively. Stabilization of p53 ( 53 ) and activation of polo-like kinase 3 ( 54 ) have also been implicated in apoptosis induction by NFκB. Conceivably, one or more of these pathways may be involved in the synergistic effect of curcumin and p65 observed in this study. This theme will be explored in further work.
In summary, the present study shows that inhibition of NFκB by curcumin is not essential for its anti-apoptotic effects in HCT116 cells and suggests that other unidentified regulators of cell death may be implicated.
We thank the following: Prof. Ron Hay, University of St Andrews, for providing the 3EnhConALuc construct; Dr Warner Greene, University of California, San Francisco, for providing the pCMV-p65 construct; and Dr Dean Ballard, Vanderbilt University, Nashville, for providing the pCMV–ΔNIκB construct. This work was supported by the Research and Development Office, Department of Health, Social Services and Public Safety Northern Ireland.
Conflict of Interest Statement : None declared.