| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Letters to the Editor |
–Related Apoptosis-Inducing Ligand–Induced Apoptosis
Department of Cancer Biology, Genomics Institute of the Novartis Research Foundation, San Diego, California
Department of Protein Sciences, Genomics Institute of the Novartis Research Foundation, San Diego, California
To the Editor:
Considerable excitement has surrounded the observation that the cytokine tumor necrosis factor-
–related apoptosis-inducing ligand (TRAIL) induces apoptosis in many tumor cell types but has little or no effect on nontransformed cells. Two recent articles attempt to elucidate the genetic basis for this tumor specificity. Nesterov et al. (1) and ourselves (2) employ similar approaches using a transformation model based on one originally established by Hahn et al. (3). In this system, normal human fibroblasts (BJ) or embryonic kidney cells (HEK) are transformed from normalcy to malignancy by introducing a series of defined oncogenic elements, thereby allowing specific characteristics of transformed cells (e.g., TRAIL sensitivity) to be attributed to specific genetic alterations. However, as Nesterov et al. test only a subset of the oncogenes we tested, the two groups reach differing conclusions. Because these results have potentially important implications for the clinical application of TRAIL-related therapeutics, we wish to clarify and reconcile the apparently contradictory results from the two groups.
In their article, Nesterov et al. conclude that expression of an activated RAS allele (HRASV12) is sufficient to confer TRAIL sensitivity to otherwise nontransformed human cells. In contrast, we propose that RAS transformation is not sufficient but that overexpression of the MYC oncoprotein is both necessary and sufficient (2). Although these results may seem to be contradictory, they are in fact not. Like Nesterov et al., we also observed that HRASV12-expressing cells are reproducibly more susceptible to apoptosis than isogenic non-HRASV12 cells using either TRAIL or DR5-A (an agonistic antibody against the relevant TRAIL receptor, DR5) as an apoptotic stimulus. However, side-by-side comparison in primary BJ fibroblasts shows the effects of HRASV12 to be very modest in comparison with those of MYC (Fig. 1A), an oncogene that was not tested by Nesterov et al. Similar results were obtained using immortalized BJ and HEK cells (ref. 2; data not shown; see also Fig. 1B). Differences in the absolute amount of cell death observed in HRASV12-expressing cells by Nesterov et al. and ourselves can be attributed to the different durations of treatment employed by each group (72 hours for Nesterov et al. versus 16-24 hours for our experiments). Differences were not due to the differing apoptotic stimuli used, as we observed similar effects with TRAIL ligand as with the DR5-A antibody (data not shown).
|
If RAS acts through MYC, then are both groups correct? We believe so, with any apparent discrepancies explainable by the differing assay conditions employed by each group. However, as determining which genes contribute to TRAIL (or any other drug) sensitivity is of importance ultimately for its therapeutic implications, we would like to raise a concern about the findings of Nesterov et al. We have shown that treatment of mice bearing genetically defined human tumor xenografts with therapeutic doses of DR5-A does not result in a significant response in HRASV12-expressing tumors, unless they also ectopically express MYC (2). The claim that RAS sensitizes cells to TRAIL-induced apoptosis, although technically correct, should therefore be qualified, as this sensitization seems insufficient to be of therapeutic relevance, at least in this experimental treatment setting.
Perhaps more importantly, if RAS, acting through MEK, is indeed a significant contributor to TRAIL sensitivity in tumor cells, this would imply that drugs targeting upstream regulators of RAS (e.g., receptor tyrosine kinases or farnesyl transferases), or the RAS downstream effectors RAF and MEK, would be poor choices to apply in combination with TRAIL-based therapies, because inhibition of RAS activity would negate the apoptotic effects of TRAIL. However, if RAS is not a significant determinant of TRAIL sensitivity, then such combinations might be reasonable to try. We propose the latter situation to be the case. In tumor cells, MYC stability is frequently dissociated from RAS signaling, either by mutation of the relevant phosphorylated region on the MYC protein (4), by loss of the MYC ubiquitin ligase FBW7 (5–7), or by as-yet unidentified mechanisms. Thus, the TRAIL-sensitive colon carcinoma cell line HCT15, which carries an activated allele of the frequently mutated RAS family member (KRAS), does not show decreased MYC accumulation in response to MEK inhibitors and is not protected from TRAIL- or DR5-A–induced apoptosis (Fig. 1D). In fact, because RAS has multiple additional downstream effectors, including some, such as phosphatidylinositol 3-kinase, that transduce antiapoptotic signals, inhibiting RAS activity may in fact sensitize cells to DR5-A. We have indeed observed such a sensitizing effect using the phosphatidylinositol 3-kinase inhibitor LY294002 (Fig. 1D). Thus, contrary to what Nesterov et al. observed using the T24 cell line (1), we propose that in many TRAIL-sensitive tumor cells inhibitors of RAS function may enhance rather than diminish TRAIL treatment efficacy.
In summary, although we agree with the validity of the results of Nesterov et al., we have concerns about their potential therapeutic implications. We believe the ability of RAS to sensitize normal human cells to TRAIL apoptotic signaling is insufficiently potent to be clinically relevant. Nonetheless, their work is of interest in that it identifies an experimental system in which the mechanisms underlying TRAIL sensitivity can be elucidated. Here, we propose such a mechanism, reconciling their observed results with our own: that RAS stimulates MYC stability and activity and thereby sensitizes cells to TRAIL-induced apoptosis.
References
–related apoptosis-inducing ligand-induced apoptosis. Cancer Res 2004;64:3922–7.Bothell, Washington
Hollings Cancer Center, Medical University of South Carolina Medical Center, Charleston, South Carolina
In Response:
As discussed by Wang et al. (1), we employed a genetically defined transformation system to investigate how conversion of human cells from normal to tumorigenic renders them susceptible to tumor necrosis factor--related apoptosis-inducing ligand (TRAIL)-induced apoptosis. Human embryonic kidney cells and foreskin fibroblasts were first immortalized by the combination of the early region of SV40 and telomerase and then transformed with either oncogenic allele of RAS (H-ras-V12), two effector loop mutants of RAS that do not activate extracellular signal-regulated kinase (ERK) pathway (H-ras-V12C40 and H-ras-V12G37), or gain-of-function mutant of MEK1 (MEK1-Q56P). Our results showed that RAS-mediated activation of ERK pathway sensitized cells to TRAIL-induced apoptosis by up-regulating one of TRAIL receptors (DR5) and by increasing recruitment of caspase 8 to TRAIL DISC.
Using essentially the same experimental system, Wang et al. showed that up-regulation of another important oncoprotein (MYC) also rendered immortalized human cells susceptible to TRAIL-induced apoptosis (2). The results of Wang et al. are supported by the recent data from El-Deiry group, demonstrating that overexpression of MYC sensitized cells to TRAIL-induced apoptosis, whereas its down-regulation by small interfering RNA lead to TRAIL resistance (3). Importantly, similarly to the effect of RAS (1), overexpression of MYC resulted in up-regulation of DR5 (2) and potentiated TRAIL-induced processing of caspase 8 (2, 3).
Wang et al. propose that both RAS and MYC lie within the same proapoptotic signaling pathway. As activation of the RAS/RAF/ERK pathway is known to enhance the accumulation of MYC (4), it is conceivable that RAS and ERK-dependent sensitization of human cells to TRAIL may occur through up-regulation of MYC level. In this correspondence, Wang et al. present a set of data strongly suggesting that this may indeed be the case.
Although we agree with the results of Wang et al., we take issue with some conclusions drawn by this group. First, at this point, it is hard to determine whether MYC is better apoptosis sensitizer than RAS, as it depends on relative overexpression levels of these proteins. As RAS lies upstream of MYC within the same proapoptotic signaling pathway, both proteins were sufficient to trigger apoptosis. Second, MYC is deregulated in about one third of tumors (5), and as reported by the El-Deiry group (3), not all TRAIL-sensitive cells contained above average MYC levels and not all cells could be efficiently rescued from TRAIL-induced apoptosis by small interfering RNA-mediated knockdown of MYC. Alternatively, some other mechanisms may also sensitize cancer cells to TRAIL-induced apoptosis. Third, we believe that the therapeutic relevance of RAS/RAF/mitogen-activated protein kinase/ERK kinase/ERK pathway in sensitization of cancer cells to TRAIL-induced apoptosis may be cancer or cell type specific. In our experiments (1), both farnesyl transferase inhibitor SCH 66336 and mitogen-activated protein kinase/ERK kinase inhibitor PD 98059 efficiently suppressed proapoptotic effects of TRAIL on bladder cell line T24, which carries an activated allele of H-ras and possesses constitutively activated ERK (6). Wang et al. have shown that although the mitogen-activated protein kinase/ERK kinase inhibitor U0126 did not protect HCT15 cells from TRAIL-induced apoptosis, the same compound efficiently blocked TRAIL-induced apoptosis in MYC-expressing immortalized HEK cells. In the latter case, the antiapoptotic effect of the mitogen-activated protein kinase/ERK kinase inhibitor was accompanied by down-regulation of MYC. Therefore, it is logical to assume that for tumor cells where accumulation of MYC is mediated by ERK the combination of TRAIL with drugs targeting RAS/RAF/mitogen-activated protein kinase/ERK kinase/ERK pathway may be a poor therapeutic choice.
In summary, we believe that the Wang et al. data reconcile the results of these two articles and clearly show the power of genetically defined transformation systems in elucidating the basis of TRAIL selectivity.
References
-related apoptosis-inducing ligand-induced apoptosis. Cancer Res 2004;64:3922–7.
This article has been cited by other articles:
|
|
 |
|
  A. Panner, J. L. Nakamura, A. T. Parsa, P. Rodriguez-Viciana, M. S. Berger, D. Stokoe, and R. O. Pieper mTOR-Independent Translational Control of the Extrinsic Cell Death Pathway by RalA Mol. Cell. Biol., October 15, 2006; 26(20): 7345 - 7357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Rottmann, Y. Wang, M. Nasoff, Q. L. Deveraux, and K. C. Quon From the Cover: A TRAIL receptor-dependent synthetic lethal relationship between MYC activation and GSK3{beta}/FBW7 loss of function PNAS, October 18, 2005; 102(42): 15195 - 15200. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
