The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
Arsenic is a naturally occurring element that is present ubiquitously in the environment in both inorganic and organic forms. Human exposure to the generally more toxic inorganic arsenic compounds occur in occupational or environmental settings, as well as through medicinal arsenical use.1 In recent years, arsenic contamination of drinking water has become a major public health concern worldwide, especially in Asia.2, 3 A primary concern for chronic arsenic exposure in human populations is its carcinogenic potential. Inorganic arsenic has been classified as a known human carcinogen based on strong epidemiological data.1 Arsenic exposure in humans is associated with a marked increase in skin cancer.4, 5 This is potentially due to the high affinity of arsenic for sulfhydryl groups, which leads to arsenic accumulation and retention in keratin-rich skin tissue.6 Characteristic arsenic-induced skin lesions are verrucous hyperkeratosis and pigmentation disorders.3 Other proliferative skin lesions associated with human arsenic exposure include Bowen's disease and squamous cell or basal cell carcinoma.1, 7, 8
It is estimated that in the United States 1.3 million new cases of skin cancer are diagnosed annually.9 Ultraviolet (UV) radiation from sunlight is the primary carcinogen in development of human skin cancer and can act as both an initiator and promoter.9 Recent work in mouse skin models of carcinogenesis shows that arsenic acts as a co-carcinogen or co-promoter to induce tumors.8, 10 In particular, oral arsenic greatly enhances the incidence and progression of skin malignancies induced by ultraviolet radiation.10 Arsenic alone, however, does not induce skin tumor development in these model systems.8, 10 This points toward some sort of arsenic-induced enhancement of the carcinogenic potential of other agents, such as UV radiation, as a likely mechanism. Transplacental inorganic arsenite exposure in mice acts alone to produce various internal cancers including liver and lung tumors,11, 12 but does not initiate skin cancers.12 Events in the skin associated with arsenic carcinogenesis may be distinct from other target tissues and deserve additional investigation.
Several lines of evidence indicate that tumorigenesis is a multistep process involving progressive transformation of normal cells into malignant phenotypes along with the aberrant accumulation of such abnormal cells.13, 14 Agents that impact the carcinogenic process may act during malignant transformation or aberrant cell accumulation. In this regard, acquired resistance to apoptosis is a hallmark of most cancers.13 Under normal circumstances apoptosis eliminates DNA-damaged or mutated cells.15 Acquired resistance to apoptosis may be a critical molecular event during carcinogenesis, and disruption of apoptosis has been shown to play a major role in tumor formation and malignant progression.15 During long-term low level arsenite exposure both rodent liver cells and human prostate epithelial cells acquire a remarkable generalized resistance to apoptosis.16, 17 This includes resistance to potent chemotherapeutics as well as other agents that induce apoptosis as the primary mode of cell death. Whether chronic arsenic exposure can induce acquired apoptotic resistance in skin cells is undefined. Because the skin is an important target of arsenic carcinogenesis, and is exposed to a variety of other carcinogenic stimuli, such as UV radiation, acquired apoptotic resistance could be a critical event in carcinogenesis. It is recognized that apoptosis in skin cells is a key response for elimination of genetically damaged cells with the potential for tumor development.9 Thus, a ubiquitous environmental agent, like arsenite, that could reduce apoptosis in skin cells could have clear impact on skin cancer.
In our present study human keratinocyte cell line (HaCaT) was continuously exposed to a low level (100 nM) of inorganic arsenite for 28 weeks. The HaCaT cell line was originally derived from normal human adult skin, exhibits normal differentiation and is nontumorigenic.18 This cell line has been used successfully to elucidate various stages in the carcinogenic transformation process. Our results showed chronic inorganic arsenite exposed HaCaT cells developed generalized resistance to apoptosis that included resistance to UV radiation-induced apoptosis, without becoming resistant to UV-induced DNA damage. The results of our study provide important insights into the initial molecular response to arsenite in potential target cells of arsenic carcinogenesis and are consistent with the concept that arsenic is a co-carcinogen in skin.10
As-TL, arsenic-tolerant cells; PDK-1, phosphoinositide-dependent kinase-1; PI3K, PI3 kinase; PKB, protein kinase B; P-PKB, phosphorylated PKB; PTEN, phosphatase and tensin homolog deleted on chromosome 10; ROS, reactive oxygen species; UV, ultraviolet; UVA, ultraviolet A.
Material and methods
Sodium arsenite (NaAsO2), LY294002, Wortmannin, cis-diaminedichloroplatinum (II) (cisplatin), doxorubicin (Adriamycin) and etoposide were obtained from Sigma (St. Louis, MO). Annexin V-FITC kits were purchased from Trevigen (Gaithersburg, MD).
The human keratinocyte cell line HaCaT is a spontaneously immortalized human epithelial cell line developed by Boukamp et al.18 and generously provided by Prof. N. Fusenig (DKFZ, Heidelberg, Germany). The line has specific mutations in p53, but is not tumorigenic after inoculation into mice. The cells were cultured in DMEM supplemented with 10% FBS, 100 U of penicillin/ml, and 100 μg of streptomycin/ml. Cultures were maintained at 37°C in a humidified 5% CO2 atmosphere. For chronic arsenic exposure, cells were maintained continuously in medium containing 100 nM of sodium arsenite. The arsenite concentration used in our study is comparable to human blood arsenic levels found in chronic arsenosis patients in Inner Mongolia, China, where arsenic-induced skin lesions are common.3
Acute cytotoxicity assay
A minimum of 5 replicates of 10,000 cells per well were plated in 96-well plates and allowed to adhere to the plate for 24 hr, at which time the media was removed and replaced with fresh serum-free media containing the cisplatin, etoposide or doxorubicin. Cells were then incubated for an additional 24 hr and cell viability was determined using Non-Radioactive Cell-Proliferation Assay Kit (Promega, Madison, WI). Measurements expressed as percent of untreated control of appropriate cells. The LC50 values were determined from analysis of the log-linear phase of the curves.
The cells were treated as described previously.19 Briefly, the medium was removed and cells were washed twice with PBS. After the addition of PBS, the cells were irradiated with fluorescent lamps (Houvalite F20T12BL-HO PUVA, National Biological Corp., Twinsburg, OH) with the dish lid on. The UVA dose was monitored with a Goldilux UV meter equipped with a UVA detector (Oriel Instruments, Stratford, CT). Control samples were kept in the dark under the same conditions. After treatment, fresh medium containing 1% serum was added and the cells were kept at 37°C in a humidified 5% CO2 atmosphere.
The pattern of DNA cleavage was analyzed by agarose gel electrophoresis. Briefly, cell pellets were resuspended in lysis buffer (5 mM Tris-HCl pH 8.0; 20 mM EDTA; 0.5% Triton X-100) and incubated on ice at 4°C overnight. After incubation at 56°C for 1 hr with RNase A (100 μg/ml) and then 1 hr with proteinase K (200 μg/ml), the cell lysate was extracted with phenol/chloroform/isopropyl alcohol (25:24:1, v/v). DNA was precipitated with ethanol and subsequently washed with 70% ethanol. DNA samples, dissolved in TE buffer, were separated by horizontal electrophoresis on 1.5% agarose gels, stained with ethidium bromide and visualized under UV light.
Determination of apoptosis by flow cytometry
Cells were seeded in 25 cm2 flasks and grown to ∼80% confluence. After the cells were exposed to arsenite for 24 hr or 18 hr post-UV-A exposure, floating and attached cells were harvested for apoptosis analysis. Detection of phosphatidylserine on the outer leaflet of apoptotic cells was carried out using Annexin V and propidium iodide according to the manufacturer's recommendations. For each sample, 10,000 cells were examined by flow cytometry using a Becton Dickinson FACSort (Becton Dickinson, San Jose, CA). The percent of apoptotic cells was determined by statistical analysis of the various dot plots using CellQuest software.20
Preparation of protein extracts
Cells were treated with various chemicals as detailed in the respective figure legends. After washing 3 times with ice-cold PBS, whole cell extracts were obtained by using Cell Lysis Buffer (Cell Signaling, Technology, Inc. Beverly, MA) with 0.5% of Protease Inhibitor Cocktail (Sigma, St. Louis, MO) and 1% of Phosphatase Inhibitor Cocktail I (Sigma, St. Louis, MO). Nuclear and cytosolic fractions were separated by TransFactor Extraction Kit (BD Biosciences Clontech, Palo Alto, CA) according to the manufacturer's recommendation. All the protein fractions were stored at −70°C until use and the protein concentrations were determined by Bio-Rad protein assay (Bio-Rad, Hercules, CA) with BSA as a standard.
Western blot analysis
Proteins were separated by Novex Tris-Glycine Gel (Invitrogen, Carlsbad, CA) and transferred onto nitrocellulose membranes. The blots were probed with primary antibodies (1:500–2,000) followed by incubation with horseradish peroxidase-conjugated secondary antibodies. Antibody incubations were carried out in Blocker BLOTTO in TBS (Pierce, Rockford, IL). Immunoreactive proteins were detected by chemiluminescence using ECL reagent (Amersham Pharmacia, Piscataway, NJ) and subsequent autoradiography. Quantitation of the results was carried out by Bio-Rad Gel Doc 2000 Systems with Bio-Rad TDS Quantity One software. After the blots were stripped using Restore Western Blot Stripping Buffer (Pierce, Rockford, IL), the blots were probed for β-actin (Cell Signaling Technology, Inc. Beverly, MA), and the level of β-actin was used to normalize for sample loading. Antibodies used include: Phospho-PKB (Ser 473), Phospho-PKB (Thr 308), PKB, caspase-3, Phospho-PTEN, and Phospho-PDK1 from Cell Signaling Technology, Inc.; PI3 Kinase (PI3K) from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA),
Detection of reactive oxygen species generation
Intracellular reactive oxygen species (ROS) generation was monitored with the fluorescent probe 5-(and-6)-chloromethyl-2′, 7′-dichlorodihydrofluorescein diacetate, acetyl ester (CM-H2DCFDA, Molecular Probes, Inc., Eugene, OR), which is a chloromethyl derivative of H2DCFDA and exhibits much better retention in live cells. Cells were grown in 60 × 15 mm dish for 24 hr and then exposed to UVA or arsenite. After treatments, fresh serum-free DMEM containing 2 μM of CM-H2DCFDA was added and incubated for 30 min. The cells were rinsed 3 times with PBS and were then trypsinized followed by washing. Cells were examined by flow cytometry using a Becton Dickinson FACSort.
Determination of 8-oxo-dG level in nuclear DNA
The level of 8-oxo-dG in nuclear DNA was determined using the enzymatic hydrolysis procedure according to Adachi et al.21 After treatment with arsenite and UVA, cells were washed with PBS and harvested by trypsinization. Separate samples of 8-oxo-dG and 2′-dG were used as standards. DNA hydrolysates were analyzed by an HPLC system consisting of a Waters 2690 Separation Module, a Waters 490E Programmable Multiwavelength Detector (Waters Co., Milford, MA), an ESA (Chelmsford, MA) Coulochem II 5200A electrochemical detector (guard cell: 700 mV, standard analytical cell model 5010: working electrode E1 at 300 mV), and a Supelcosil LC-18-S column (250 × 4.6 mm, 5 μm grain; Supelco, Switzerland) equipped with a 2-cm guard column. Twenty microliter aliquots of the DNA hydrolysates were chromatographied isocratically at a flow rate of 1 ml/min, using 100 mM sodium acetate-orthophosphoric acid, pH 5.2/methanol (92:8) as the eluent. The system was controlled and chromatograms were acquired and integrated by a Millennium Chromatography Manager.
Data are expressed as mean ± SEM (n = 3–6) in all cases. For comparisons of between groups, a Student's t-test was carried out. One-way ANOVA followed by a Tukey's or Dunnett's test was used to compare all pairs of groups or selected groups to control. A p-value of ≤0.05 was considered significant in all cases.
Arsenic tolerant cells show generalized apoptotic resistance
To examine whether the low level, chronic arsenite exposure can cause acquired apoptotic resistance in human keratinocytes, HaCaT cells were continuously exposed to a low level (100 nM) of arsenite (as sodium arsenite) for 28 weeks. The arsenite-exposed cells were then exposed to various stimuli to induce apoptosis compared to passage-match untreated (control) HaCaT cells. These chronically exposed cells became very tolerant to apoptosis, including arsenite- and UVA-induced apoptosis, and are henceforth termed arsenic-tolerant (As-TL) cells. In our preliminary experiments, we treated the HaCaT cells with both arsenite (20, 100, 500 nM) and arsenate (100, 500, 2,500 nM) and checked the tolerance to arsenite at 10, 20 and 28 weeks. We found the 100 nM arsenite-treated cells acquired a stronger resistance to subsequent high doses of arsenite than other treatments, so we selected the As-TL cells (100 nM arsenite for 28 weeks) to conduct the current study. The proliferation of As-TL cells is slightly slower than control HaCaT cells both under normal or conditions of decreased serum concentration. There were no significant differences in cell cycle (G1 check point), number of passages and cell viability between As-TL and control cells.
Arsenite can effectively induce apoptosis in many kinds of cells. In our present study, arsenite caused dose-dependent apoptotic cell death in both control and As-TL cells. As-TL cells were clearly tolerant to arsenite-induced apoptosis compared to control cells (Fig. 1a). These results were supported by caspase-3 analysis (Fig. 1b). Western blot analysis indicated that the As-TL cells exhibited much less activated caspase-3, a key enzyme in cellular dedication to apoptotic cell death, than control cells after acute high dose arsenite exposure.
For UV radiation-induced apoptosis, As-TL and control cells were exposed to environmentally relevant doses of UVA radiation (25 J/cm2) and the number of apoptotic cells was determined by flow cytometry. As shown in Figure 2a, 44.9% of the control cells underwent apoptosis after UVA exposure, compared to only 19.5% of the As-TL cells. This difference in UVA-induced apoptosis between the 2 cell lines was confirmed by DNA fragmentation (Fig. 2b). In UVA-irradiated control cells, a more pronounced DNA laddering, characteristic of apoptosis, was detected than that in the As-TL cells.
To further determine whether the As-TL cells processed generalized apoptotic resistance, As-TL and control cells were exposed to the chemotherapeutic compounds cisplatin, etoposide and doxorubicin, all of which induce apoptosis as their primary mode of cell death. Cytolethality assays showed that As-TL cells were clearly more resistant to all 3 compounds. The LC50 value (concentration estimated to kill 50% of the cells) was 92.1 ± 4.35 μM in As-T vs. 63.9 ± 1.54 μM in controls for cisplatin (p < 0.05), 96.6 ± 4.53 μM in As-TL cells vs. 63.7 ± 1.85 μM control cells for etoposide (p < 0.05), and 3.78 ± 0.37 μM in As-TL cells vs. 2.52 ± 0.03 μM control cells for doxorubicin (p < 0.05).
Apoptotic resistance in As-TL cells seemed to represent a stable phenotypic change accompanying chronic arsenite exposure. In support of this contention apoptotic resistance to arsenite and the other various treatments persisted in As-TL cells even after they had been maintained in arsenite-free medium for up to 6 weeks (data not shown).
ROS production and oxidative DNA damage
UV radiation, as a carcinogen likely acts, at least in part, through generation of potentially genotoxic ROS. UVA increased ROS production, as assessed by the fluorescent probe CM-H2DCFDA (that detects hydrogen peroxide and other ROS), in both control and As-TL cells (Table I). There was no significant difference between As-TL and control cells in the levels of ROS produced in response to UVA treatment. Arsenite at high doses can produce ROS and compared to control cells, the As-TL cells produced noticeably less ROS after high dose arsenite exposure (Table I). As-TL cells have a generalized acquired arsenite resistance that includes reduced apoptosis and ROS. Although As-TL cells are resistant to UV-induced apoptosis this resistance does not extend to UV-induced ROS production.
Table I. UVA Radiation and Acute High Dose of Arsenic Exposure Increased ROS Production in As-TL and Control Cells1
Cells were grown in 60 × 15 mm dish for 24 hr and then exposed to UVA or arsenite at the doses indicated, and analyzed as described in Material and Methods. Quantitative results of H2O2 levels in As-TL and control cells are expressed by fluorescence intensity; n=3–12 determinations.
p < 0.05 compared with treatment-matched control cells.
UV-induced ROS would be expected to result in oxidative damage of critical bio-molecules, such as DNA. UVA exposure induced a significant increase of 8-oxo-dG, a marker of oxidative DNA damage, in As-TL and control cells (Fig. 3). Interestingly, UVA exposure produced the same levels of oxidative DNA damage in As-TL and control cells, despite the resistance to UVA-induced apoptosis in As-T cells. There is the distinct possibility that more As-TL cells survive with UV-induced DNA damage. High dose acute arsenite exposure also induced oxidative DNA damage but this response was significantly perturbed in As-TL cells (Fig. 3), again indicating a resistance to arsenite on multiple levels.
Mode of apoptotic resistance in As-TL cells
Cellular apoptosis is dependent on the balance between proapoptotic and survival signals and PKB is a key molecule for cell survival.22 The activation mechanism of PKB remains to be fully characterized but occurs downstream of PI3K. PKB activation requires phosphorylation on both Thr-308 and Ser-473. The phosphorylation of Thr-308 is apparently catalyzed by phosphoinositide-dependent kinase-1 (PDK-1), whereas Ser-473 by an as-yet-unidentified kinase.22 PTEN (phosphatase and tensin homolog deleted on chromosome 10) inhibits PI3K-dependent activation of PKB and deletion or inactivation of PTEN results in constitutive PKB activation.23 In our study, neither chronic arsenite nor acute high dose arsenite treatment affected the cellular levels of PI3K, phosphorylated PDK1, phosphorylated PTEN, PKB or P-PKB (Ser 473) (data not shown). Interestingly, in contrast to the cellular levels of P-PKB, high dose acute arsenite or UVA markedly decreased nuclear P-PKB levels detected by anti-P-PKB (Ser473) and Thr 308 before the apoptosis in the control cells, whereas the As-TL cells exhibited an increased stability of nuclear P-PKB (Fig. 4).
Inhibition of PI3K pathway abolishes the acquired resistance to apoptosis in As-TL cells
PKB phosphorylation is mediated by PI3K, which is LY294002- or Wortmannin-sensitive. Pretreatment of the As-T cells with LY294002 (10 μM) or Wortmannin (0.5 μM) completely blocked arsenite-induced apoptotic resistance in As-TL cells, suggesting that PKB plays a major role in this resistance (Fig. 5). The concentrations of LY294002 and Wortmannin used in our study are concentrations that induce obvious cell death in control and As-TL cells.
Apoptosis can be initiated by a number of stimuli, including ionizing radiation and chemotherapeutic agents.24 As-TL cells acquired a remarkable resistance to apoptosis induced by various agents including UVA radiation and an acute high dose of arsenite. UVA radiation from sunlight, which represents more than the 95% of solar UV light reaching the Earth's surface,19 is clearly associated with human skin cancer. UVA can deeply penetrate into the skin tissue reaching the basal layers of actively dividing dermis and, therefore, the mutagenic potential of UVA is mechanistically significant.19 Part of the mutagenic potential of UV likely is due to the fact that it induces oxidative DNA damage. Arsenite exposure and UVA radiation are associated with skin cancer in humans, suggesting a potential synergistic effect of UV radiation and arsenite exposure might exist. The work of Rossman et al.10 indicates oral inorganic arsenite exposure greatly enhances the onset, growth and aggressiveness of skin tumors induced by UV radiation in mice. Apoptosis is an important process for eliminating damaged cells that potentially could become cancerous22 and acquired resistance to apoptosis is typical of many developing tumors. Apoptosis functions normally to control the integrity of cell populations by eliminating aberrant clones and failure of apoptosis likely is a key contributor to tumor initiation, progression and drug resistance in skin cancers9 and cancer in general. Arsenite-induced acquired apoptotic resistance may be an important event in skin cancer development by allowing damaged cells to escape normal cell population control mechanisms. In our present study, although UV-induced apoptosis was greatly perturbed in As-TL cells, UV induced oxidative DNA damage to the same extent in As-TL and control cells. Oxidative DNA damage may lead to mutation and, eventually, tumors. The fact that a reduction in UV-induced apoptosis occurred but UV-induced genotoxicity is unchecked may well allow damaged cells to survive with the potential to form malignancies. This could help explain the co-carcinogenic effects of arsenite and UV in mouse skin model systems22 and would potentially be a very important aspect of the role of arsenic in human skin cancer. Furthermore, the As-TL cells become cross-tolerant to several chemotherapeutic compounds, such as cisplatin, etoposide and doxorubicin, indicating the As-TL cells acquired a generalized apoptotic resistance. The resistance to these chemotherapeutics may indicate that skin cancers formed during arsenic exposure could potentially be resistance to chemotherapy. Recently, Lee et al.25 examined the acute effects and interactions of low doses of arsenite and UVB on apoptosis in normal human keratinocytes. A 24-hr pretreatment with 1 μM of arsenite did not affect significantly the apoptosis induced by followed-UVB radiation.25 The major difference between the study of Lee et al.25 and our work is the time of pretreatment by arsenite. The apoptotic resistance in As-TL cells represents a stable phenotypic change accompanying chronic (≥26 week) arsenite exposure.
ROS are important mediators of UVA-induced cellular toxicity and DNA damage. UVA and high doses of inorganic arsenite increased ROS production in both As-TL and passage-matched control cells. The As-TL cells produced markedly less ROS after acute high dose arsenite exposure than control cells, but UVA treatment induced a similar level of ROS regardless of cell type. Consistent with these results, similar levels of 8-oxo-dG, a marker of oxidative DNA damage, were found in As-TL and control cells after UVA irradiation, whereas lower levels of 8-oxo-dG were observed in As-TL cells after high doses of arsenite treatment. The difference between arsenite-induced ROS production and oxidative DNA damage in As-TL and control cells may be caused by increased arsenic efflux in As-TL cells, which is common in acquired resistance to arsenic26 and would perhaps account for arsenic specificity. Whatever the basis of resistance to arsenite-induced ROS and oxidative DNA damage is in As-TL cells, it does not seem to provide the same protection against similar UV-induced molecular lesions.
The susceptibility of cells to apoptosis seems to be dependent on the balance between proapoptotic and antiapoptotic (survival) signals. The acquired apoptotic resistance observed in the present work was associated with increased stability of nuclear P-PKB. Furthermore, pretreatment of the As-TL cells with PI3K inhibitors LY294002 or Wortmannin completely blocked the acquired apoptotic resistance. The serine/threonine kinase PKB, identified first as an oncogene, has a number of downstream substrates that may contribute to malignant transformation.27 PKB is a major downstream target of growth factor receptor tyrosine kinases that signal via PI3K. PKB is a well-established survival factor, exerting anti-apoptotic activity by preventing release of cytochrome c from mitochondria, inactivating the pro-apoptotic factors BAD, pro-caspase-9, and forkhead transcription factors known to induce expression of pro-apoptotic factors such as Fas ligand.22 Among these PKB-associated antiapoptotic factors, the forkhead transcription factors including FKHR, FKHRL1, AFX and several others are known PKB nuclear targets and PKB negatively regulates their transcriptional activities.28, 29 Phosphorylation of these forkhead transcription factors by PKB can induce their relocalization to the cytoplasm, and impairment of their transcriptional activity.28 In contrast, as a result of decreased phosphorylation of the forkhead transcription factors, proapoptotic genes are upregulated causing cell death.29 In our study, high dose arsenite- and UVA-induced decrease of nuclear P-PKB in normal cells likely induce activation of the forkhead transcription factors resulting in apoptosis. The increased stability of nuclear P-PKB observed in As-TL cells likely accounts, at least in part, for the acquired generalized resistance to apoptosis. Another known nuclear target of PKB is p53. The HaCaT cell has a p53 gene mutation, however, and p53 has been shown not to be involved in the induction of apoptosis by arsenite.30 Overexpression of PKB is thought to contribute to malignant phenotype27 and although the mechanism by which PKB contributes to neoplastic transformation remains unclear, increased nuclear P-PKB observed in As-TL cells and increased stability of nuclear P-PKB after apoptosis stimulation may be an important signature of this transformation. These findings provided potentially new insight into the mechanisms of arsenic carcinogenesis.
In conclusion, our study shows that long-term exposure to low levels of arsenite can cause generalized apoptotic resistance in a human skin cell line. This is particularly significant because it occurred using a cell line analogous to a potential in vivo target site of arsenic carcinogenesis. The induction of generalized resistance to apoptosis with chronic low level arsenite may provide a potential mechanism for the observed co-carcinogenic effect on dermal cancers of combined arsenite and UV irradiation.10 As a ubiquitous environmental contaminant that could potentially enhance UV-induced skin cancers, arsenite could play an important role in this all too common malignancy, although the in vivo applicability of the present findings will require additional confirmation.
Research was funded in part by National Cancer Institute under Contract NO1-CO-12400. The authors are grateful to Drs. J.M. Reece and E. Leslie for their excellent contribution to our study. We also thank Drs. Y. Xie, J. Shen and L.K. Keefer for their critical comments and assistance during the preparation of this manuscript.