Patrizia Cesaro and Elisabetta Raiteri contributed equally to this work.
Expression of protein kinase C β1 confers resistance to TNFα- and paclitaxel-induced apoptosis in HT-29 colon carcinoma cells
Article first published online: 11 APR 2001
Copyright © 2001 Wiley-Liss, Inc.
International Journal of Cancer
Volume 93, Issue 2, pages 179–184, 15 July 2001
How to Cite
Cesaro, P., Raiteri, E., Démoz, M., Castino, R., Baccino, F. M., Bonelli, G. and Isidoro, C. (2001), Expression of protein kinase C β1 confers resistance to TNFα- and paclitaxel-induced apoptosis in HT-29 colon carcinoma cells. Int. J. Cancer, 93: 179–184. doi: 10.1002/ijc.1314
- Issue published online: 9 JUN 2001
- Article first published online: 11 APR 2001
- Manuscript Accepted: 9 FEB 2001
- Manuscript Revised: 9 JAN 2001
- Manuscript Received: 28 JUL 2000
- Ministero dell'Università e della Ricerca Scientifica e Tecnologica (Roma, cofin. 1997 and 1998)
- Associazione Italiana per la Ricerca sul Cancro (AIRC, Milano)
- Consiglio Nazionale delle Ricerche (CNR, target project on Biotechnology. Grant Number: 9900386PF.49
- Università del Piemonte Orientale “A. Avogadro”
- Fondazione Cavalieri Ottolenghi (Torino)
- protein kinase C;
- colon carcinoma
The expression of different protein kinase C (PKC) isoenzymes has been shown to vary with proliferation rates, differentiation or apoptosis in normal colon crypts. In addition, the activity of some PKC isoenzymes appears to be reduced in colorectal cancer. The aim of the present work was to determine whether modulation of PKC expression would affect the susceptibility of a p53-defective colon carcinoma cell line to different apoptotic treatments. HT-29 cells exhibited sensitivity to paclitaxel (Taxol) and tumor necrosis factor α (TNFα) in a dose- and time-dependent manner but were relatively resistant to etoposide. Inhibition of PKC activity augmented the susceptibility of HT-29 cells to apoptosis, and phorbol ester induction of PKC reduced such susceptibility. Transfected HT-29PKC cells, hyper-expressing the β1 isoform of PKC, were less sensitive to TNFα and paclitaxel than the normal counterpart. The present data 1) indicate that the expression of PKC influences the susceptibility of HT-29 colon cancer cells to apoptotic drugs apparently regardless of their mechanism of action, and 2) suggest paclitaxel as a potential candidate for the treatment of colon cancer, possibly in association with inhibitors of PKC (α and β) at doses not cytotoxic per se. © 2001 Wiley-Liss, Inc.
Colorectal cancer is one of the most common solid tumors world-wide. Due to its high metastatic potential and the frequent onset of resistance to chemotherapy, it is one of the four major causes of death by neoplasia in westernized countries. Loss of function of p53 occurs in more than 75% of human colorectal cancers. Since p53 regulates a complex array of cellular responses to DNA damage, including cell cycle arrest and apoptosis, its loss of function is also expected to affect the sensitivity of tumor cells to DNA-damaging antiblastic drugs.1,2
The expression of different protein kinase C (PKC) isoenzymes varies with proliferation rates, differentiation or apoptosis in normal colon crypts,3–6 and the activity of some PKC isoenzymes is reduced in colorectal cancer.7–9 Whether, in addition to loss of function of p53, altered functioning of PKCs contributes to the low sensitivity of colon carcinomas to apoptosis-based chemotherapy remains to be established. In the present work we addressed this issue by examining the effect of various apoptotic treatments on HT-29 colon cancer cells that lack functional p5310 and either do or do not express abnormal levels of the isoform β1 of PKC. We found that paclitaxel (Taxol) and TNFα, but not etoposide, efficiently induced apoptosis of HT-29 cells in a dose- and time-dependent manner.
The involvement of PKC in the apoptotic pathways triggered by paclitaxel or TNFα was assessed by using a myristoylated pseudo-substrate as inhibitor at doses not affecting cell viability. In HT-29 cells, inhibition of PKC (α and β isoforms) activity augmented by twofold the cytotoxicity of both drugs. Induction of PKC by the phorbol ester phorbol myristate acetate (PMA) decreased the susceptibility of HT-29 cells to both TNFα and paclitaxel treatments. Resistance to apoptotic treatments was consistently observed in transfected HT-29 cells hyper-expressing PKCβ1. The present data implicate PKC (α and β) as a component of the mechanism responsible for the resistance of colon carcinoma cells to apoptotic drugs.
MATERIAL AND METHODS
Cells and chemicals
The HT-29 human colon cancer cell line was obtained from the American Type Culture Collection (ATCC, Rockville, MD). HT-29PKC (clone 7) and HT-29C1 cells11 derive from HT-29 cells transfected with the cDNA encoding PKCβ1 or with the vector lacking the PKC DNA insert, respectively, and were kindly provided by Dr. Weinstein (Columbia University, New York, NY). Tissue culture medium and antibiotics were purchased from Sigma (St. Louis, MO), and FCS was from Gibco (Gaithersburg, MD). Human recombinant TNFα was obtained from REB (Abingdon Oxon, UK), and paclitaxel was purchased from Bristol-Myers Squibb (Sermoneta, LT, Italy). The myristoylated PKC-α and -β pseudo-substrate (inhibitor of PKC) was purchased from BACHEM (Bubendorf, Switzerland). Etoposide (VP-16), PMA, 4-6-diamidino-2-phenyl-indol-dihydrochloride (DAPI), propidium iodide and other chemicals were from Sigma.
Western blotting analysis of PKCβ1 expression
Analysis of PKCβ1 expression was performed by standard Western blotting. Briefly, 50 μg of cell homogenates were fragmented by electrophoresis on a 12.5% polyacrylamide gel and electroblotted onto nitrocellulose. PKCβ1 was detected by chemiluminescence reaction on a filter decorated with a specific polyclonal rabbit antiserum (Santa Cruz Biotechnology, Santa Cruz, CA) followed by horseradish peroxidase-conjugated secondary antibodies (Bio-Rad, Richmond, CA). A gel was run in parallel and stained with Coomassie Blue to verify equal loading of samples. The amount of β-actin per sample loaded was determined by re-probing the same filter with polyclonal anti-β-actin antibodies (Sigma). Bands on films were quantitated by densitometry (Bio-Rad GelDoc 1,000; software Quantity One 4.03). Pre-stained standard molecular weight proteins were from Bio-Rad.
Cell culture and treatments
HT-29, HT-29C1 and HT-29PKC cells were grown under standard culture conditions in DMEM with antibiotics and 10% heat-inactivated FCS as previously reported.12 G418 (Sigma) was included as the selecting antibiotic in the culture medium of transfected cells. For the experiments, HT-29 cells were plated in 25 cm2 flasks at 5 × 104 cells/cm2 initial density and left to adhere for 48 hr prior to treatment. By this time cell density had reached a value of approximately 12 × 104 cells/cm.2 In serum-fed HT-29 cultures, cell density was approximately 25 × 104 cells/cm2 after 24 hr and 40 × 104 cells/cm2 after 48 hr. In this respect, HT-29C1 behaved like the parental HT-29 cell line. Since the growth rate of HT-29PKC cells is lower than that of HT-29, HT-29PKC cells were plated at 8 × 104 cells/cm2 the initial density so that both cell lines had a comparable cell density by the time the cytotoxic treatments were applied. Treatments lasted for 24 or 48 hr. In the latter case the medium was renewed after the first 24 hr in order to avoid side effects from nutrient consumption and accumulation of toxic metabolites; the drug was then re-added.
Evaluation of cytotoxic effects
At the times indicated, control and treated cultures were trypsinized, and the cells were counted. The percentage of adherent viable cells was determined by counting the trypan blue-excluding cells with a hemocytometer. For a rough evaluation of morphology and cell density and for detection of apoptotic or necrotic cells, cultures were observed at 12 hr intervals under a phase-contrast microscope and photographed. The appearance of a hypodiploid (sub-G1) cell population, corresponding to apoptotic cells, was monitored by flow cytometry on whole cell populations (i.e., monolayer plus cells recovered from the medium) fixed and stained with propidium iodide as described previously.13 Chromatin alterations were revealed in fixed cell monolayers by DNA fluorescent staining with DAPI (1 μg/ml methanol for 30 min), observed under a UV microscope and photographed.
Apoptosis of HT-29 cells in response to TNFα or paclitaxel is dose and time dependent
Allelic deletion or inactivating mutations of the p53 tumor suppressor gene are well documented in colorectal carcinomas. Disruption of p53 function has profound cellular consequences including reduced sensitivity to apoptotic treatments.1, 2, 14 We evaluated HT-29 colorectal cancer cells, lacking functional p53,10 for their sensitivity to apoptotic drugs that act at different cellular levels and are of potential interest for the treatment of this cancer. For this purpose we incubated HT-29 cells with the antiblastic drugs etoposide (VP-16) or paclitaxel or with the cytokine TNFα. Cells were exposed to various concentrations of the cytotoxic drugs for periods ranging from 4 to 48 hr. To ensure optimal conditions for cell growth, the medium was renewed after 24 hr of culture, and, where indicated, the cytotoxic substances were re-added. At the end of the incubation, surviving adherent cells were counted. The occurrence of cell death by apoptosis was ascertained by cytofluorometric analysis of propidium iodide-labeled DNA and DAPI decoration of nuclei.
Some significant results of such treatments, expressed as percentage of cell survival in treated cultures compared with controls, are summarized in Table I. Paclitaxel only exerted a cytotoxic effect within the first 16 hr of treatment. Etoposide, at a concentration shown to be effective on many cell types,13, 15, 16 displayed a low cytotoxicity on HT-29 cells treated for as long as 48 hr. A higher rate of apoptosis was obtained by treating HT-29 cells for 48 hr with 100 ng/ml TNFα, a rather high concentration compared with those sufficient to induce apoptosis in other tumor cell types17–19 (Table I). Cytofluorometric analysis of cell cycle distribution (data not shown) indicated that TNFα initially induced growth arrest (first 24 hr of treatment) and later also induced cell death, which became apparent by 48 hr. Paclitaxel was a quite powerful apoptogenic agent for HT-29 cells since it was effective at a concentration as low as 20 nM (data not shown). In cultures treated with 5 μM paclitaxel, about 30% of cells underwent apoptosis within 4 hr, and more than 65% of cells underwent apoptosis by 24 hr (data not shown). With 50 nM of paclitaxel, cell survival accounted to 57% of control by 24 hr and this percentage decreased to about 12% by 48 hr (Table I). With 500 nM of paclitaxel, cell viability was even more drastically reduced at any time.
|time (h)||VP-16 (μM)||TNFα (ng/ml)||Paclitaxel (nM)|
|16||100||92||100||100||88 ± 9||77 ± 9|
|24||100||78 ± 9||85 ± 7||88 ± 4||57 ± 4||25 ± 4|
|48||90 ± 9||70 ± 8||78 ± 5||54 ± 6||12 ± 4||4 ± 2|
In a separate set of experiments the treatment was prolonged to 48 hr in unchanged medium and without re-adding the substances. In these experimental conditions cytotoxicity by TNFα and by etoposide increased, although in the case of the latter, necrosis was also observed. Because of its low apoptogenic efficacy on HT-29 cells, etoposide was not used further.
We examined TNFα- and paclitaxel-treated cultures for morphological detection of apoptosis by DAPI staining. When exposed to 100 ng/ml TNFα, HT-29 became larger in size and showed cytoplasmic extension and intracytoplasmic vacuoles within the first 24 hr; chromatin condensation and formation of DNA-containing apoptotic bodies became evident by 48 hr of treatment (Fig, 1). Twenty-four hours after exposure to 50 nM paclitaxel, HT-29 cells became smaller and rounded, displaying typical apoptotic features, such as chromatin condensation and fragmentation of nuclei into small spherical particles (Fig. 1).
In the following experiments, we explored the role of PKC in the cellular response of HT-29 cells to TNFα and paclitaxel. Because of their different efficacies, we used TNFα and paclitaxel under experimental conditions that caused a comparable cytotoxic effect on the cells (50 nM paclitaxel for 24 hr and 100 ng/ml TNFα for 48 hr).
Modulation of classical protein kinase C activity alters the sensitivity to cytotoxic agents
PKC serves as second messenger in pathways signaled by survival, growth and pro-differentiative factors.20, 21 Inhibition of PKC activity by staurosporine or its derivatives results in cell cycle block and eventually apoptosis in HT-29 cells.22, 23 We wondered whether modulation of PKC activity would impact the sensitivity of HT-29 colon carcinoma cells to TNFα or paclitaxel. In a first set of experiments, we inhibited PKC activity by using a cell-permeable myristoylated pseudo-substrate that selectively inhibits the isoforms α and β of PKC.24 We assessed the optimal concentration of this drug (hereafter referred to as inhibitor of PKC [IPKC]) that did not affect the growth of the cultures for as long as 72 hr. As shown in Figure 2, inhibition of PKC activity increased by nearly twofold the sensitivity of HT-29 cells to both the apoptotic drugs used. This confirms and extends previous data23 suggesting that in p53-deficient colon carcinoma cells PKC (specifically α and β isoforms) is also a central molecule in pathways that promote cell survival and counteract apoptosis.
To investigate this issue further, we examined the sensitivity to TNFα or paclitaxel of HT-29 cells (pre)-treated with PMA under conditions that stimulated intracellular PKC activity. Experimental conditions (PMA concentration, absence of serum and time of incubation) were assessed based on data from the literature25 and on our own experience (unpublished data). In the present experiment, serum was omitted to avoid activation of PKC by other exogenous stimuli. In cultures treated for as long as 72 hr with 50 nM PMA, cell growth was slowed, but apoptotic death was not apparent. As shown in Figure 3a, the percentage of cell survival after 48 hr of incubation with TNFα approached 57% in PMA-untreated cells, 75% in cultures pre-treated for 24 hr with PMA and 87% in cultures in which PMA was present for 24 hr prior to and 48 hr throughout the incubation with the cytokine. Parallel experiments were performed in which paclitaxel was employed in the last 24 hr of incubation. Survival under paclitaxel treatment increased to about 80% in cultures exposed to PMA only for the first 24 hr and reached nearly 100% in cultures exposed to PMA for the whole experimental period (Fig. 3b). Therefore, stimulation of PKC by PMA, which mainly acts on α and β isoforms, appears to render HT-29 cells more resistant to TNFα and to paclitaxel action.
Hyper-expression of PKCβ1 modifies the response of HT-29 colorectal cancer cells to apoptotic treatments
We wished to clarify further the role of classical PKCs in HT-29 cells and therefore focused our attention on the β isoform, taking advantage of the fact that HT-29 cells transfected with the cDNA encoding PKCβ1 and hyper-expressing this PKC isoform11 were available in our laboratory (kindly provided by Dr. Weinstein). We examined the sensitivity of the transfected HT-29PKC cells to TNFα or paclitaxel. HT-29C1 cells (the parental clone transfected with the empty vector) were shown to be indistinguishable from the parental untransfected HT-29 cells with respect to morphology, growth characteristic (duplication time, serum requirements, saturation density), PKC levels and response to PMA (data not shown and Fig. 4; see also ref. 26). Transfection of PKCβ1 resulted in a nearly 15-fold increase in PKC activity.11
We analyzed by Western blotting the expression of the PKCβ1 protein in HT-29 cells and in the clones HT-29C1 and HT-29PKC. As shown in Figure 4, HT-29 and HT-29C1 cells expressed equal amounts of the protein, whereas HT-29PKC expressed a much higher level of the protein, as expected. Given the similarity of HT-29 and HT-29C1 with respect to growth characteristics and PKC expression, for the sake of homogeneity in the following experiment we utilized the parental HT-29 cell line as the counterpart for HT-29PKC. When exposed to apoptotic treatments, HT-29PKC cells were more resistant than the wild-type counterpart to both paclitaxel and TNFα by almost 1.4-fold. In particular, the percentage of HT-29PKC cells that survived after a 24 hr paclitaxel treatment was about 85% (whereas in HT-29 culture a similar treatment led to a 58% survival): after 48 hr, TNFα treatment survival was about 72% (whereas in HT-29 culture a similar treatment led to a 55% survival; Table II). Thus, it can be concluded that a high intracellular level of PKCβ1 confers resistance to apoptosis in HT-29 colon carcinoma cells. Taken together, the results of the experiments reported in Figure 3 and Table II would imply that stimulation of conventional PKC protects HT-29 cells more from paclitaxel than from TNFα.
|HT-29||100||58 ± 4||55 ± 4|
|HT-29PKC||100||85 ± 5||72 ± 4|
In the present work we assayed the sensitivity of HT-29 cells to different apoptogenic drugs, namely, etoposide (VP-16), TNFα and paclitaxel, which act through different cellular pathways; we wondered whether, together with p53 loss of function, altered PKC expression would affect the response to these apoptotic treatments. Two main conclusions can be drawn from the present study: 1) TNFα and paclitaxel, but not etoposide, are cytotoxic for HT-29 cells, presumably through p53-independent pathways; and 2) α and β PKCs are involved in the apoptotic pathways activated by TNFα and paclitaxel.
Etoposide at a concentration that was cytotoxic for many cell types in vitro had a low cytotoxicity.13, 15, 16 This fact might be related to the mechanism of action of this drug that causes DNA damage through inhibition of topoisomerase II.27 It is likely that the absence of functional p53 negatively influenced the onset of apoptosis in etoposide-treated HT-29 cells. This interpretation is consistent with the report by Lowe et al.,1 who demonstrated that p53-deficient fibroblasts developed resistance to apoptosis induced by several DNA damaging agents, etoposide included. HT-29 cells have been previously reported to be rather resistant to TNFα, which in this cell type mainly exerts cytostatic effects.28 In general, TNFα elicits apoptotic effects when administered along with metabolic inhibitors such as cycloheximide or actinomycin D. Here we show that TNFα alone is cytotoxic for HT-29 cells in a dose- and time-dependent manner. Cytotoxicity by this cytokine was enhanced in HT-29 cells treated for as long as 48 hr in unchanged medium, a condition that negatively impacts on cellular metabolism.
TNFα binds to its membrane receptor and triggers the activation of a cascade of kinases and proteases that culminate into apoptosis.29 The extent of cytostatic or cytotoxic effects mediated by TNFα depends on the amount of specific receptors that are engaged as well as on the fine tuning of the intracellular signals that are recruited upon their activation. Therefore, the cytokine concentration, cell density and level of membrane expression of the receptor as well as the integrity of the signaling pathways involved are all factors that influence the outcome of the treatment. In this respect, it is worth noting that pre-incubation with interferon γ sensitized HT-29 cells to TNFα cytotoxicity.28, 30 One can speculate that interferon γ increases not only the synthesis31 but also the number of TNF receptors at the plasma membrane level. We found that co-treatment with an inhibitor of PKC (α and β isoforms) at doses not cytotoxic per se for as long as 48 hr sensitized HT-29 cells to TNFα cytotoxicity. Conversely, PMA-induced stimulation of conventional PKC activity protected HT-29 cells from this cytokine. We further assessed the role of the β isoform of PKC in this process. To this end we employed a transfected clone of HT-29 cells, which was shown to express about sevenfold higher levels of the PKCβ1 protein than the parental cell line (Fig. 4). We found that transfected HT-29PKC were more resistant to TNFα (and to paclitaxel) than their untransfected parental counterpart. On the whole, the present data consistently demonstrate that PKC is clearly involved in TNFα cytotoxicity in HT-29 cells.
High rates of apoptosis were obtained by treating HT-29 cells with paclitaxel, a chemotherapeutic drug not commonly used for the treatment of colorectal cancer. Inhibition of PKC activity enhanced paclitaxel-induced cytotoxicity, and paclitaxel was consistently more effective on HT-29 cells than on HT-29PKC. Moreover, paclitaxel was virtually not effective in cultures (pre-)treated with PMA under conditions that stimulated α and β PKC isoforms (Fig. 3). Therefore, PKC appears to be clearly involved in the molecular mechanisms initiated by paclitaxel and leading to apoptosis of colorectal cancer cells. Clinically, paclitaxel is an active agent in the treatment of many tumor types including leukemia and breast, ovary and prostate cancers.32, 33, 34 Based on data reported here, paclitaxel should be considered a potential candidate for the chemotherapeutic treatment of colon carcinomas and possibly also in association with inhibitors of PKC at doses not cytotoxic per se.
PKCβ isoforms have been shown to be expressed at low levels in colorectal cancer, and this has been linked to the loss of the enterocytic differentiated phenotype.7, 9 Consistently with this view, hyper-expression of PKCβ1 was associated in HT-29 cells with (partial) recovery of cell proliferation control and the ability to acquire an enterocytic-like differentiated phenotype under adequate experimental conditions.11, 26 Here we have shown that active PKCβ1 (and probably other PMA-sensitive PKC isoforms) confers resistance in HT-29 cells exposed to various apoptogenic stimulus, apparently regardless of their type and mechanism of action. Therefore, altered expression of PKC should be considered among the mechanisms responsible for the resistance to cytotoxic drugs exhibited by colorectal cancer cells.
Thanks are due to Dr. Weinstein (Columbia University, New York, NY) for providing us with the HT-29PKC and HT-29C1 clones. R.C. was supported by a fellowship from the Fondazione Cavalieri Ottolenghi (Torino, Italy). M.D. is a postdoctoral fellow at the Università del Piemonte Orientale “A. Avogadro”.