If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
). Initially, oncologists focused on alterations in rates of proliferation and cell cycle kinetics, but more recently an emphasis on apoptosis has dominated the fight against cancer (
). As approximately 1,000,000 individuals in the U.S.A. develop skin cancer each year, it is important to elucidate the molecular mechanisms that govern cell survival and cell death in the epidermis (
). Moreover, given that most skin cancers occur on sun-exposed skin, the pro-apoptotic and antiapoptotic response of keratinocytes (KC) to UV light is of particular relevance to the development of skin cancer (
). Whereas both squamous cell carcinoma (SCC) and basal cell carcinoma (BCC) arise from epidermal KC, it is becoming increasingly apparent that the natural history of their development, their underlying molecular pathogenesis, and potential involvement of antiapoptotic pathways are significantly different. Nonetheless, as pointed out later in the text, significant progress is being made in our understanding of the pathophysiology of these relatively common epithelial-cell-derived neoplasms. In this review we will explore four topics: first, a review of the life and death signaling pathways operative in normal human skin that prevents premature apoptosis of KC with an emphasis on nuclear factor κB (NFκB) survival signals; second, the molecular pathways that are engaged and regulate apoptosis after normal KC are exposed to ultraviolet (UV) light; third, the apoptotic resistant mechanisms that premalignant and malignant KC utilize to avoid cell death; fourth, therapeutic strategies that can render malignant cells more susceptible to apoptosis with an emphasis on a death pathway mediated by the death ligand TRAIL.
Life and Death Signaling in Normal Human Skin and Normal KC
Ironically, the maintenance of cutaneous homeostasis requires creation of the stratum corneum, which in turn depends on the properly regulated formation and function of dead KC in the outermost epidermal layers (Figure 1;
). Despite this incredibly thin layer of dead cells, these corneocytes are extremely efficient at subserving a barrier function: preventing dehydration and invasion by toxic environmental agents (
). Due to the unique anatomical and structural requirements of the epidermis, not only must it be a self-renewing tissue, but also the ongoing proliferation must be balanced with KC differentiation and ultimately cell death. Most importantly, the death of corneocytes must be delayed until early and late (i.e., terminal) differentiation has occurred (
). We have recently coined the phrase “planned cell death pathway” to emphasize the highly coordinated spatial and temporal reactions in the epidermis that ultimately give rise to corneocytes (
Figure 1Histologic view of normal human skin highlighting the spatial coordination of KC proliferation, differentiation, and death contributing to creation of stratum corneum in the so-called “planned cell death pathway”. Ironically, the vitality of skin is dependent on a highly orderly process by which proliferating KC in lower layers of epidermis (nuclei positively stained for a cell cycle marker – Ki67; white arrows) become growth arrested, undergo terminal differentiation in upper layers (stratum granulosum), followed by cell death to produce corneocytes in the outermost layers. Note that even though the stratum corneum is composed of dead KC, it remains metabolically alive, and is responsible for the barrier properties of skin.
Figure 2 portrays the potential contributions of various biochemical pathways involved in regulating cellular reactions in each discrete epidermal compartment, ultimately leading to cell death and barrier formation (i.e., corneogenesis). Beginning in the basal cell layer, the proliferating compartment involving stem cells/transiently amplifying KC, the cell cycle regulatory molecules are highlighted as being important. Next, the movement of KC into the suprabasal layer in which early differentiation occurs is accompanied by changes in the cell cycle resulting in growth arrest, activation of NFκB, with appearance of mediators that can prevent premature apoptosis by various stimuli including activation of p53. Examples of antiapoptotic mediators include cell survival gene products such as Bcl-xL, various decoy receptors, and cell cycle inhibitors (i.e., p21). In addition, we recently identified an important role for Notch signaling in the differentiation and cornification process in human epidermis that involved both NFκB and PPAR signaling (
). Once late stage and terminal differentiation programs are initiated, it is postulated that specific death effector pathways are engaged involving death receptors, death ligands, and various caspases. Experimental evidence to support these proposed pathways is provided below. Before delving into the specifics of the apoptotic machinery, a brief review of potential life and death mediators will be provided.
Figure 2Schematic presentation of cellular, biochemical, and apoptotic contributors to the “planned cell death pathway”. Note the interplay of mediators that regulate cell cycle activity, early/late differentiation, and cell survival.
To maintain a proper homeostatic thickness and function of the epidermis, KC proliferation and survival must be counterbalanced by cell death (Figure 3). A partial listing of molecules that promote KC survival is depicted on the left side of Figure 3, whereas those factors that decrease KC survival are shown on the right side. As can be appreciated from the rather large and diverse set of molecules presented, the overall regulation of KC survival and death is likely to be complex, redundant, and highly coordinated to ensure that on the one hand there is not premature apoptosis of KC prior to terminal differentiation, but also on the other hand not such enhanced longevity of KC that precludes formation of the stratum corneum (
). Because cancer may be viewed as a disease of deregulated survival it is important to delineate which, if any, of the antiapoptotic factors present in normal KC are also utilized by malignant cells (
Figure 3Many molecular mediators contribute to the life and death balance of epidermal KC. The epidermis is not simply a barrier between external and internal environments, but is the site for numerous signaling pathways that have life and death consequences. Those factors that may enhance KC survival are depicted on the left-hand side, and those factors that decrease KC survival are depicted on the right-hand side. The precise epidermal layers that contain individual survival and death mediators should not be inferred by the location within the triangles, e.g., p63 is primarily expressed in basal layer KC.
There is an intense search under way by several laboratories to elucidate the biochemical mediators that regulate apoptosis in normal and diseased epidermis as depicted in Figure 2 and Figure 3 (
). Some components of the biochemical machinery that may contribute to regulating apoptosis in normal human epidermis are portrayed in Figure 4. By fractionating normal human epidermis into distinct bands using a Percoll gradient (
), whole cell extracts can be produced and examined by Western blot. The buoyancy of each cell layer can be confirmed not only by cytologic examination of fractions 1–6, but by the maturational state of KC verified by differential expression of loricrin, keratin-1, and β4 integrin (Figure 4). Perhaps the most significant initial impression of this profile for components of the apoptotic machinery in normal human skin is the nonrandom spatial distribution of individual molecules. For example, whereas the TRAIL death receptor 4 (DR-4) is present in all cell fractions, the DR-5 becomes more abundant in the upper cell layer fractions. By contrast, the tumor necrosis factor receptor 1 (TNF-R1) is more conspicuous in the basal layer and mid-epidermal layers, but decreases in the upper cell layers. The decoy receptor-2 for TRAIL is detectable in all cell layers, but is enhanced near the surface where TRAIL levels tend to be the highest (Nickoloff, unpublished observation). The intact or pro-form of caspase 14 is only transiently detected in fractions 2 and 3, which may support it as initially being synthesized in the mid-epidermal layers, and becomes activated (i.e., cleaved) in fraction 1. Caspase 8 appears to be synthesized in the basal layer and may not be activated until fraction 3, whereas Bcl-xL levels are relatively constant throughout all cell layers.
Figure 4Apoptotic machinery in normal human skin. By fractionating normal human epidermis according to maturational state into six fractions, the nonrandom spatial distribution of various molecular components of the apoptotic machinery can be appreciated. There are many natural defense mechanisms that counterbalance the mitogenic activity inherent in a self-renewing tissue such as the human epidermis. Potentially important regulators of KC life and death levels include Bcl-xL; caspases 8, 14; TNF-R1; TRAIL – decoy and death receptors. β-actin levels are portrayed to demonstrate equivalent protein loading for each cell fraction.
As can be seen by reviewing the aforementioned data, molecular mediators of KC cell death in human epidermis may include various death ligands, death receptors, and caspases. Even though it is currently unclear if differentiation and cornification (including apoptosis) share biochemical mediators, the overall molecular process must be coordinated in such a manner that terminal differentiation is completed prior to the onset of apoptosis (
). As regards death ligands/receptors and intracellular signaling of apoptosis in KC, members of the TNF family – including Fas ligand (Fas L) and TRAIL, as well as caspase 14 – have become leading contenders mediating the demise of KC in the epidermal compartment of normal skin (
). Besides these molecular mediators of cell death, KC also possess intracellular signaling pathways such as NFκB that can promote KC survival.
The survival of normal KC is clearly dependent on NFκB (Figure 2) as disrupting this signaling pathway either pharmacologically using the proteosome inhibitor MG132 (
), is accompanied by enhanced susceptibility to apoptosis in vitro. In vivo, disrupting NFκB signaling in transgenic murine epidermis leads to an increased frequency of apoptotic KC (
). As regards susceptibility to apoptosis, it appears that NFκB signaling can also regulate levels of decoy receptors that normally protect KC against TRAIL-induced apoptosis (
). When NFκB is disrupted using the IκBαDN retrovirus, decoy receptor levels are significantly reduced, and exposure of these cells to TRAIL, which normally are resistant to apoptosis, renders them susceptible to apoptosis. Besides these newly discovered cell survival pathways mediated by NFκB, earlier studies uncovered many other antiapoptotic mechanisms at the disposal of cells that activate the pathway NFκB (
During NFκB activation, several subunits such as p50 and p65 form heterodimers and recruit coactivators such as p300 to enhance transcription of many antiapoptotic genes (
). Of note, in many cell types NFκB activation is conducive for cell proliferation, but in KC NFκB triggers abrupt growth arrest – presumably via induction of the cyclin-dependent kinase inhibitor, p21 (
). Thus, it should not be surprising that for premalignant or malignant cells to escape cell cycle check points on their route to becoming immortalized, cells such as HaCaT cells possess altered NFκB signaling pathways (
). As will be discussed later, immortalized cells that have dampened down or eliminated the NFκB response may be particularly susceptible to apoptosis because the loss of cell cycle restraint exacts a toll on the cell – namely a diminution or loss of various cell survival proteins. As can be appreciated from this brief review, there are many molecular tools at the disposal of normal KC to resist apoptosis, and considerable work remains to better understand how these diverse mediators are integrated and coordinated to properly balance life and death events in the epidermis. It should also be mentioned that in some cell systems activation of NFκB is required for the p53-mediated apoptotic response (
). Previously, the functional role of these apoptotic KC was obscure, but recent evidence suggests that sunburn cells are not just markers for severe sun damage, but this apoptotic process is important for preventing skin cancer (
Caspase activation and disruption of mitochondrial membrane potential during UV radiation-induced apoptosis of human keratinocytes requires activation of protein kinase C.
The UV (ribotoxic) stress response of human keratinocytes involves the unexpected uncoupling of the Ras-extracellular signal-regulated kinase signaling cascade from the activated epidermal growth factor receptor.
As illustrated in Figure 5, the response of KC to UV light probably activates multiple death pathways within the irradiated cell including: the ability of UV light to multimerize or activate clustering of death receptors (i.e., DR) in the plasma membrane including Fas and other TNF family members (
). Several transgenic mouse models revealed important roles for Fas/FasL, as well as Bcl-2/Bcl-x, and p53 in mediating sensitivity to UV-induced apoptosis of KC (
Human keratin-1 bcl-2 transgenic mice aberrantly express keratin 6, exhibit reduced sensitivity to keratinocyte cell death induction, and are susceptible to skin tumor formation.
). In addition, UV light also activates the MAP kinase families including extracellular regulated kinase (ERK), p38, and stress activated protein kinase/c-jun N-terminal kinase (JNK) in KC that impacts their survival (
Differential phosphorylation of mitogen-activated protein kinase families by epidermal growth factor and ultraviolet B irradiation in SV40-transformed human keratinocytes.
). On the flip side various growth factors and cytokines including epidermal growth factor, interferon-γ (IFN-γ), and interleukin-1 can protect KC against UV-induced apoptosis (
Keratinocyte apoptosis induced by ultraviolet B radiation and CD95 ligation – differential protection through epidermal growth factor receptor activation and Bcl-xL expression.
). Whereas much has been learned regarding the decision of whether a KC will undergo apoptosis after UV irradiation, many molecular mechanistic insights remain to be determined for this critical process in photo and multistage-epidermal carcinogenesis (
Figure 5How to kill a KC. There are at least three different pathways of importance in the skin that can contribute to the death of a KC when exposed to UV light. UV light may oligomerize death receptors in the plasma membrane (left-hand side); or UV light may trigger release of proapoptotic mediators from the mitochondria (center pathway); or UV light my cause DNA damage and increase p53 activity leading to induction of Apaf-1 (right-hand side). Following the proximal molecular events various caspase cascades are set into motion as depicted with cross-talk between pathways. Ultimately, final effector caspases are activated, leading to DNA fragmentation and apoptosis.
To dissect out various potential apoptotic pathways involved in UV-light-induced apoptosis, several inhibitors have been employed. To determine if plasma membrane-based death receptors participate in UV-induced apoptosis, KC were infected with a retrovirus designed to overexpress a dominant negative form of FADD (i.e., FADD-DN). As the death receptor complex in the plasma membrane is coupled to proximal caspases (i.e., caspase 8) by adaptor molecules such as FADD, introducing a dominant negative version of FADD is a method to identify the apoptotic contribution of the aforementioned death receptor pathways following various stimuli. After confirming overexpression of FADD-DN, and blocking TRAIL-mediated apoptosis, the apoptotic response to UV light was examined and found not to be influenced by the overexpression of FADD-DN (
). Also, addition of anti-Fas antibody (100 ng per ml) did not inhibit UV-induced apoptosis (Denning et al, 2002). Hence, it does not appear that plasma membrane-based death receptors play a major role in the UV-light-induced apoptotic pathway for KC.
Moving to mitochondrial-based components of the apoptotic machinery, however, several results indicate a significant participatory role for those mediators in the UV response of KC. If KC are infected with retroviral constructs designed to overexpress either Bcl-2 or Bcl-x (which associate with mitochondria), then the subsequent apoptotic response to UV light is significantly decreased (
). Moreover, addition of the protein kinase C (PKC) inhibitor GF 109203x(GF), which can block mitochondrial-associated PKC-δ, also reduces the susceptibility of KC and HaCaT cells to UV-induced apoptosis (Denning et al, 2001;
). Finally, employing an inhibitor of free oxygen radicals that can damage mitochondria, such as pyrrolidine dithiocarbamate (PDTC), also reduces the susceptibility of KC and HaCaT cells to UV-induced apoptosis (Figure 6A, B, respectively). The precise mechanism by which PDTC inhibits apoptosis remains to be determined, but is not likely to involve p53 because it can equally protect normal KC as well as HaCaT cells (which have mutated both p53 alleles), and because there was no increase in p53 functional activity using a luciferase-based reporter assay or increase in total p53 protein levels (data not shown). One mechanism by which PDTC inhibits apoptosis in KC is by reducing the release of cytochrome C from mitochondria (Figure 6C). Interestingly, whereas PDTC did not increase NFκB activity, it did increase the antiapoptotic cell-cycle-dependent kinase inhibitor p21 as determined at the mRNA and protein levels (data not shown).
Figure 6PDTC protects KC and HaCaT cells against UV-induced apoptosis. (A) KC cells; (B) HaCaT cells. Whereas PDTC (10, 50, 100 μM) by itself had little effect on the viability of KC (left panel) or HaCaT cells (right panel), addition of PDTC for 2 h prior to UV light (25 mJ per cm2) reduced the number of apoptotic cells in a dose-dependent fashion. Apoptotic cells were measured after 18 h of UV light following incubation with propidium iodide and cell sorting with sub-G0 DNA content containing cells defined as apoptotic as previously described (
). (C) Western blot analysis demonstrating release of cytochrome C by mitochondria isolated from KC before (lane 1) and 18 h following UV light (25 mJ per cm2) in the absence (lane 2) and presence (lane 3) of PDTC (2 h pretreatment; 50 μM). Note the UV-light-induced release of cytochrome C is blocked by PDTC. Equal protein loading confirmed by β-actin levels.
). The precise death effector molecules involved in this pathway remain to be determined, but evidence using the human papillomavirus (HPV) derived E6 protein indicates a role for the pro-apoptotic protein Bak (
). Moreover, as certain strains of HPV are implicated in the development of skin cancer, the ability of E6 to block Bak-mediated apoptosis has been postulated as a mechanism by which premalignant and malignant HPV-infected cells can enhance their survival, thereby promoting carcinogenesis in skin (
Returning to Figure 5, it should be emphasized that there are multiple pathways by which KC can be triggered to undergo apoptosis. It is probable that such a wide assortment of pathways exists to ensure that a KC dies when given an apoptotic signal, even if one pathway is inactivated. Obviously these multiple apoptotic pathways have implications for skin cancer development, as activation of any one of these distinct pathways could eliminate a potentially cancerous cell. Although more details will be provided in the next section, a brief mention of the p53 pathway is relevant. On the one hand, whereas p53 is frequently mutated in sun-exposed normal human skin, these cells apparently have little to no precancerous potential (Ponten et al, 1995). Thus, eliminating such KC via a different apoptotic mechanism that is p53-independent is valuable and perhaps best highlighted by the fact that Li–Fraumeni patients (individuals with germ-line p53 mutations) do not develop an excess of skin cancers. Such clinical observations emphasize the importance of further studies on both preneoplastic and malignant cells in which distinct apoptotic pathways are dissected and viewed from a therapeutic targeting perspective (
). In the next section, this line of inquiry will be further examined and discussed with a focus on skin cancers rather than normal epidermis.
Molecular Determinants of Life or Death Pathways in Skin Cancer
In this section emphasis moves from normal epidermal KC to cancer cells that comprise SCC and BCC of the skin. Given space constraints, the focus will be confined to a few important cell survival pathways mediated by NFκB, Bcl-x, p53, a death pathway involving TRAIL, and the Hedgehog (Hh) signaling system. There is compelling in vitro and in vivo data pointing to a critically important role for NFκB-mediated signaling events that produce a death-defying phenotype to SCC (
Expression of a dominant-negative mutant inhibitor -κBα of nuclear factor κB in human head and neck squamous cell carcinoma inhibits survival, proinflammatory cytokine expression, and tumor growth invivo.
). Experimental protocols in which NFκB signal transduction is disrupted render the SCC cells susceptible to apoptosis. As mentioned earlier, several cell survival pathways are regulated by NFκB. It should be noted that the NFκB pathway is a bit more complicated than just influencing cell survival, as NFκB activation is also associated with growth arrest because agents that stimulate NFκB produce enhanced p21 levels (
). The net effect of disrupting NFκB signaling in the epidermis, however, is not only to impact terminal differentiation but to promote the appearance of SCC (
Immunohistochemical-based investigations of cell survival profiles in skin cancer have revealed overexpression of Bcl-2 and Bcl-x in BCCs and SCC, respectively (
). Surprisingly, many tumor cells, while displaying increased cell cycle activity using antibodies to detect either proliferating cell nuclear antigen (PCNA) or Ki67, also expressed elevated levels of the pro-apoptotic protein Bax along with the Bcl-2/Bcl-x. As mentioned earlier, abrogation of Bak expression is associated with skin cancer (
). Thus, rather than just focusing on a single cell survival protein or proliferation marker, it is important to examine the overall balance of all life and death determinants (
Human keratin-1 bcl-2 transgenic mice aberrantly express keratin 6, exhibit reduced sensitivity to keratinocyte cell death induction, and are susceptible to skin tumor formation.
); however, it should be noted that the overexposure of either Bcl-2 or Bcl-x did not by itself result in skin cancer, but only following various oncogenic stimuli. Moreover, somewhat unexpectedly when Bcl-2 was targeted for overexpression by hair follicle epithelium, there was premature loss of hair, rather than a trichotrophic response (
). Thus, there are still many unknown biologic processes that remain to be elucidated concerning the regulation of apoptosis in the epidermis and hair follicle.
p53 is the most commonly mutated gene identified in human cancers (
). UV-induced mutation in the p53 gene has been implicated as an important factor for developing skin cancer, as a reduced susceptibility to apoptosis would favor survival and clonal expansion of mutated KC (
). Another downstream effector pathway regulated by p53 involves Fas/FasL, and hence alteration in p53 may confer a death defying phenotype by deregulating the Fas/FasL apoptotic pathway as observed in several types of skin cancer including SCC and melanoma (
). Not only can skin tumor cells fail to express Fas, but they can concomitantly express FasL, and thereby kill infiltrating antitumor T cells that express Fas (
Besides tumor cells down-modulating the death receptor Fas (CD95), it was recently observed that premalignant KC in actinic keratoses and malignant cells in SCC also reduce their expression of the death receptor for TRAIL (
). Thus, tumor cells in skin utilize many strategies to avoid apoptosis.
Finally, before leaving the discussion, it is important to point out that another signaling pathway is relevant for the development of BCC – namely the Hh signal transduction pathway (
). Even though several groups have linked genetic alteration in the Hh pathway to BCC, it remains to be determined which Hh target genes contribute to the enhanced proliferation and survival of tumor cells in BCC (
), but additional molecular mediators are likely to be involved in the formation and expansion of BCC in sun-exposed skin.
Rendering Malignant Cells Susceptible to Apoptosis
As several cancer therapeutics kill tumor cells by inducing apoptosis, cancer cells that resist apoptosis can protect themselves from such therapeutic agents (
). Many groups have attempted to enhance the effectiveness of various chemotherapy agents and ionizing radiation by focusing on methods to inhibit cell survival signals provided by NFκB (
). Almost immediately after TNF-α was discovered and cloned, many investigators hoped that TNF-α would be the “magic bullet” to cure cancer because TNF-α could kill so many different types of tumor cells (
). When two different preparations of TRAIL were examined, including a recombinant, soluble, native-sequence version (amino acids 114–281; NT-Apo2L/TRAIL) and a version of the ligand fused to trimerizing leucine zipper (LZ-Apo2L/TRAIL), it was observed that NT-Apo2L/TRAIL did not induce apoptosis of normal KC but could induce apoptosis of HaCaT cells (
To explore potential therapeutic strategies that could more effectively kill SCCs but spare normal epidermal cells such as KC, several SCC cell lines were exposed to various treatment protocols. The SCC cell lines studied were SCC-9, SCC-13, and SCC-25 (including SCC-25 cells infected by a retrovirus containing either linker alone or IκBαDN). Initially, cells were exposed to a number of pro-apoptotic stimuli including anti-Fas monoclonal antibody (CH-11; 100 ng per ml; Upstate Biotechnology, Lake Placid, NY), TNF-α (103 U per ml), LZ-Apo2L/TRAIL (300 ng per ml), UV light (30 mJ per cm2), or IFN-γ plus TNF-α. As can be seen, varying levels of apoptosis were induced depending on the stimulus and the cell type (Table I). Beginning with the SCC-9 cell line, it was relatively resistant to the apoptotic effects of anti-Fas antibody treatment or after exposure to TNF-α, IFN-γ plus TNF-α, or UV light; however, it was moderately susceptible to apoptosis triggered by LZ-Apo2L/TRAIL. By contrast, SCC-13 cells were markedly susceptible to both LZ-Apo2L/TRAIL and UV light, being relatively resistant to the other indicated death stimuli. As regards SCC-25 cells, they were relatively resistant to all stimuli except for a slight increase triggered by UV light. Thus, these cells were selected to determine if they could be rendered more susceptible to apoptosis by interfering with their NFκB signaling pathway. Indeed, whereas infection with a linker only containing retroviral construct did not consistently change their phenotype, disrupting NFκB using the IKBαDN-containing retrovirus did render the SCC-25 cells between two and three times more sensitive to apoptosis induced by TNF-α, LZ-Apo2L/TRAIL, or UV light.
Values represent % apoptotic cells (determined by Sub-G0DNA content using PI/FACS), 18 hrs after exposure to indicated death ligand. This Table portrays representative results from one of 3 independent experiments.
This SCC line was infected with a retrovirus containing either linker alone, or a cDNA with the DN mutant of κBα; and then treated with the indicated death ligand.
This SCC line was infected with a retrovirus containing either linker alone, or a cDNA with the DN mutant of κBα; and then treated with the indicated death ligand.
None
8.1
4.0
3.8
5.3
6.9
Anti-F as Ab
11.0
6.6
3.5
9.3
14.2
TNF-α
6.0
11.8
3.8
5.7
20.2
LZ-TRAIL
21.4
55.9
5.8
9.6
19.7
IFN-γ+TNF-α
14.5
7.0
5.6
7.8
23.9
UV-Light
13.8
71.9
11.0
12.0
35.8
* Values represent % apoptotic cells (determined by Sub-G0DNA content using PI/FACS), 18 hrs after exposure to indicated death ligand. This Table portrays representative results from one of 3 independent experiments.
** This SCC line was infected with a retrovirus containing either linker alone, or a cDNA with the DN mutant of κBα; and then treated with the indicated death ligand.
), we next determined if another potential therapeutic strategy for rendering SCC cells more susceptible to TRAIL could also involve modulation of decoy receptor expression. Thus, the response of SCC-25 cells to adriamycin was studied. Adriamycin was selected because a previous report demonstrated that adriamycin could sensitize tumor cells to TRAIL-induced apoptosis (
). SCC-25 cells constitutively expressed both death (TR-1, TR-2) and decoy receptors (TR-3, TR-4) as shown in Figure 7. The presence of the decoy receptors may explain the resistance of this cell line to both NT-Apo2L/TRAIL and LZ-Apo2L/TRAIL; however, after exposure to adriamycin, although the death receptors could still be detected, there was no detectable cell surface expression for either decoy receptor (Figure 7, upper right panel). Whereas adriamycin by itself did not induce apoptosis, the combination of adriamycin plus either NT-Apo2L/TRAIL or LZ-Apo2L/TRAIL dramatically induced the apoptotic response. This was somewhat unexpected, because a previous report suggested that adriamycin sensitized cells to TRAIL by increasing DR5 (
). Furthermore, the ability of adriamycin to reduce decoy receptors is independent of NFκB signaling as there is no reduction in NFκB activity (based on luciferase reporter assays) following exposure to adriamycin, and because KC infected with the IκBαDN retrovirus (designed to block NFκB activity) were equally influenced by adriamycin as evidenced by a decrease in decoy receptor expression (data not shown). Taken together, these results support a novel mechanism by which adriamycin can influence the cell survival phenotype of SCC cells by reducing levels of expression of decoy receptors.
Figure 7TRAIL death (TR1, TR2) and decoy receptor (TR3, TR4) cell surface expression before and 24 h after adriamycin (1 μg per ml) exposure in SCC-25 cells. Semi-confluent monolayers of KC were exposed to trypsin/ethylenediamine tetraacetic acid to produce a single cell suspension and then stained to detect cell surface levels of the respective death/decoy receptors. Ten micrograms per ml of each primary monoclonal antibody (obtained from Immunex) were incubated for 30 min at 4°C followed by washing and exposure to a fluorescein isothiocyanate labeled goat antimouse secondary antibody. After washing, cells were analyzed on an Epics V Flow cytometer (Hialeah, Fl) as previously described (
). Pooled mouse IgG was used as a negative control antibody. Apoptotic response of SCC-25 cells before and after exposure to either NT-Apo2L/TRAIL or LZ-Apo2L/TRAIL (lower left panel), or in the presence/absence of adriamycin (lower right panel). The extent of apoptosis was determined as described for Figure 6.
Although it is obvious that maintenance of a physiologic epidermal thickness and creation of a stratum corneum requires a properly regulated balance between life and death signaling pathways, many molecular determinants remain unknown, as do the interrelationships amongst various pathways. Moreover, the malignant cell transformation process in skin clearly involves alterations and/or abrogation of signaling pathways that impact cell death/survival, but clear-cut distinctions between physiologic and pathologic events have yet to be discerned by investigative skin biologists. Nevertheless, considerable progress in identifying and dissecting specific components of the biochemical machinery governing life and death events in epidermis has been made in the past 5 y. It is likely not only that new advances will aid our understanding of the mechanisms underlying creation of the stratum corneum, but that such discoveries will also facilitate development of tumor-selective death ligands (
) that will not trigger premature apoptosis of adjacent normal, nonmalignant epidermal cell types. Because the skin is chronically exposed to sunlight, special attention to the interaction between the new drugs and UV light will be particularly important for dermatologists as they seek novel therapies to use to combat skin cancer.
ACKNOWLEDGMENTS
The authors thank Ms. Stephanie Russo for preparation of this manuscript. Genentech Inc. and Immunex Corp. provided indicated TRAIL preparation and reagents to detect TRAIL-related death and decoy receptors. This work was supported by NIH grants AR 40065, AR 47307, AR 47814 (B.J.N.), and CA 83784 (M.F.D.).
References
Aggarwal B.B.
Kohr W.J.
Hass P.E.
et al.
Human tumor necrosis factor. Production, purification, and characterization.
Caspase activation and disruption of mitochondrial membrane potential during UV radiation-induced apoptosis of human keratinocytes requires activation of protein kinase C.
Expression of a dominant-negative mutant inhibitor -κBα of nuclear factor κB in human head and neck squamous cell carcinoma inhibits survival, proinflammatory cytokine expression, and tumor growth invivo.
The UV (ribotoxic) stress response of human keratinocytes involves the unexpected uncoupling of the Ras-extracellular signal-regulated kinase signaling cascade from the activated epidermal growth factor receptor.
Keratinocyte apoptosis induced by ultraviolet B radiation and CD95 ligation – differential protection through epidermal growth factor receptor activation and Bcl-xL expression.
Differential phosphorylation of mitogen-activated protein kinase families by epidermal growth factor and ultraviolet B irradiation in SV40-transformed human keratinocytes.
Human keratin-1 bcl-2 transgenic mice aberrantly express keratin 6, exhibit reduced sensitivity to keratinocyte cell death induction, and are susceptible to skin tumor formation.
52920-5&id=gr1.jpg){kind=link}
52920-5&id=gr2.jpg){kind=link}
52920-5&id=gr3.jpg){kind=link}
52920-5&id=gr4.jpg){kind=link}
52920-5&id=gr5.jpg){kind=link}
52920-5&id=gr6.jpg){kind=link}
52920-5&id=gr7.jpg){kind=link}