| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Epidemiology and Prevention |
1 Departments of Pharmacology and Toxicology, 2 Medicine, 3 Pathology, and 4 the Norris Cotton Cancer Center, Dartmouth Medical School, Hanover, New Hampshire; Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; and 5 Pfizer Global Research and Development, Groton, Connecticut
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Retinoids, natural and synthetic derivatives of vitamin A, are useful pharmacologic tools to uncover mechanisms for cancer chemoprevention, as reviewed (2
, 3)
. Retinoids act by binding to the nuclear retinoic acid receptors or heterodimeric retinoid X receptors that, in turn, affect target gene expression (3)
. Retinoid-induced alterations in gene expression mediate diverse changes in cellular proliferation, differentiation, and apoptosis, among other changes. To understand better the mechanisms responsible for retinoid-mediated chemoprevention, cellular models for this chemoprevention were developed. We reported previously that acute N-nitrosamine-4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) treatment of the immortalized BEAS-2B HBE cell line caused cellular transformation to occur with increased proliferation, anchorage-independent growth, and tumor formation in athymic mice (ref. 4
; as summarized Fig. 1
). RA-treatment opposed these changes and also established a cellular model to assess in the preclinical setting mechanisms engaged during lung cancer chemoprevention.
|
This study examined changes in gene expression that occurred during retinoid chemoprevention of carcinogen-exposed HBE cells. Affymetrix-based microarray analyses of immortalized, carcinogen-transformed, and RA-chemoprevented HBE cells were used to uncover target genes having increased expression during chemoprevention. One of these augmented species was the ubiquitin-activating enzyme E1-like (UBE1L), a potential tumor suppressor in lung cancers (12) . Immunohistochemical analyses revealed that UBE1L was abundantly expressed in the histologically normal human bronchial epithelium where low levels of cyclin D1 expression were known to be evident (8) . This finding raised the prospect that UBE1L was mechanistically involved in the repression of cyclin D1, a target for cancer chemoprevention based on in vitro studies (4, 5, 6) as well as an animal model for chemoprevention (13) and clinical observations (8 , 14) . To examine this possibility further, cotransfection assays were done to determine whether UBE1L expression would trigger cyclin D1 repression. The findings that will be presented directly implicate UBE1L in lung cancer chemoprevention by causing repression of cyclin D1.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Gene Expression Profiles.
Trizol reagent (Life Technologies, Inc. Gaithersburg, MD) was used to isolate independently total cellular RNA from immortalized (BEAS-2B), transformed (BEAS-2BNNK), and chemoprevented (BEAS-2BNNK/RA) cells, and total cellular RNA was converted to cDNA according to the manufacturer’s procedures (Affymetrix, Santa Clara, CA). The Affymetrix Hu6800 and customized 8600 microarrays were described previously (15)
. Probe arrays were hybridized, washed, and scanned according to established techniques. Affymetrix Microarray suite 4.0 software was used to process primary data, determine average difference value, and assess signal intensity for each probe set. The global scaling method recommended by Affymetrix was used to normalize microarrays. These data were used to determine fold changes.
Expression Analyses.
The RNeasy Protect Mini kit (Qiagen, Valencia, CA) was used to isolate total cellular RNA. DNA-free kit (Ambion, Austin, TX) was used to remove contaminating DNA. Random hexamers and Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) were used to synthesize the cDNA from 3 µg total RNA. Taq polymerase (Invitrogen) was used for PCR assays, and products were resolved on a 1% agarose gel stained with Gelstar nucleic acid stain (BioWhittaker Molecular Applications, Rockville, MD). The primers for inhibitor of DNA binding 1 (Id1), insulin-like growth factor-binding protein 6 (IGFBP6), tumor necrosis factor, 
-induced protein 2 (TNFAIP2), and E74-like factor 3 (ELF3) have been described (15)
as have primers for UBE1L (16)
. The remaining primers used were as follows: 2'-5'-oligoadenylate synthetase 1 (OAS1) forward primer, 5'-CTGCAGAAAGAGGGCGAGTT-3', and OAS1 reverse primer, 5'-TTCCAAGACCGTCCGAATC-3'; retinoic acid-induced gene, G protein (RIG-G) forward primer, 5'-CAGAAGCCCAGACTTACCTG-3', and RIG-G reverse primer, 5'-ATAGGCAGAGATCGCATACC-3'; interferon (IFN), -inducible protein 27 (IFI27) forward primer, 5'-TGGCCAGGATTGCTACAGTT, and IFI27 reverse primer, 5'-CCAGTTGCTCCCAGTGACT-3'; glyceraldehyde 3-phosphate dehydrogenase (GAPDH) forward primer, 5'-GCCTGCTACCTTACCAACCA-3', and GAPDH reverse primer, 5'-GAAGGTGAAGGTCGGAGTCA-3'; midkine (MK) forward primer, 5'-CCTGCAACTGGAAGAAGGAG-3', and (MK) reverse primer, 5'-CTTTCCCTTCCCTTTCTTGG-3'; myxovirus resistance 1 (MXA) forward primer, 5'-GACAGGACCATCGGAATCTT-3', and MXA reverse primer, 5'-GTGATGAGCTCGCTGGTAA-3'; and IFN-stimulated protein, 15 kDa (ISG15) forward primer, 5'-GGTGGACAAATGCGACGAAC-3', and ISG15 reverse primer, 5'-ATGCTGGTGGAGGCCCTTAG-3'.
Established techniques were used to do Northern and Western analyses (15) . For Northern analyses, the radiolabeled cDNA probes used were as follows: a 1 kbp NcoI/EcoRI fragment of UBE1L cDNA and a 497 bp fragment containing the coding region of ISG15 isolated from pcDNA-ISG15 by BamHI/XbaI restriction endonuclease digestions. For Western analyses, the antibodies were described previously for UBE1L (16) and IGFBP6 (15) . A polyclonal antibody that recognized IRF7 (Santa Cruz Biotechnology, Santa Cruz, CA) was also used. An ABI Prism 7700 sequence detector (PE Biosystems, Foster City, CA) with SYBER Green PCR Master Mix (PE Biosystems) was used to perform real-time PCR assays.
Expression Analyses of Paired Normal and Malignant Lung Tissues.
To detect UBE1L expression in normal and malignant lung tissues, immunoblot analyses for UBE1L were done using total protein independently isolated from paired histologically normal lung tissues and malignant lung tissues obtained from a lung tissue bank developed at Dartmouth-Hitchcock Medical Center. The Institutional Review Board reviewed and approved these studies. Independent polyclonal antibodies recognizing the amino- or carboxyl-termini of UBE1L, respectively, were used in these immunoblot analyses. Total protein levels were normalized to -tubulin expression, as assessed by immunoblot analyses. The monoclonal -tubulin antibody was purchased (Oncogene Research Products, San Diego, CA).
Immunohistochemistry.
Histologically normal bronchial tissues were fixed, paraffin-embedded, sectioned, and processed for immunohistochemical analyses using standard techniques (8)
. Sections were lightly counterstained with hematoxylin. Polyclonal antibodies individually recognizing the carboxyl- or amino-terminal regions of UBE1L were used, and preimmune sera (16)
also served as a negative control. Polyclonal rabbit antibody recognizing cyclin D1 was purchased (Biocare Medical, Walnut Creek, CA). Blocking experiments were accomplished by preincubating the antibody with the appropriate immunogenic peptide before immunohistochemical detection was done.
Cotransfection Assays.
Exponentially growing BEAS-2B HBE cells were transfected using Effectene transfection reagent (Qiagen, Valencia, CA) according to procedures established previously (7)
. The plasmids used were pSG5-UBE1L, pCMV-HA-Cyclin-D1 and the pCMV-ß-galactosidase reporter plasmid. ß-Galactosidase activity and amounts of total protein each served as controls. Modified radioimmunoprecipitation assay buffer was used 24 hours after transfection to obtain total protein. Western analyses were done using a murine monoclonal antibody that recognized the HA-epitope (Berkeley Antibody Company, Richmond, CA) or a goat polyclonal antibody that recognized ß-actin (Santa Cruz Biotechnology).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Fold changes between the immortalized and either the chemoprevented or the transformed HBE cell lines were determined. Species induced at least 4-fold were examined in an in-depth manner. It is notable that only a small subset of species was induced to this extent. Eleven (0.099%) and 15 (0.136%) individual species from these arrays fulfilled this stringent criteria in chemoprevented as compared with immortalized (group I) or transformed (group II) HBE cells, as depicted in Tables 1
and 2
.
|
|
Highlighted species included known RA target genes as well as previously unrecognized RA-regulated species in HBE cells, as depicted in Tables 1
and 2
. To determine whether these species were RA-regulated, reverse transcription (RT)-PCR assays were performed for representative species after RA-treatment (1 µmol/L) of BEAS-2B HBE cells. As shown in Fig. 2
, some species were rapidly induced within 24 hours of RA-treatment. Species that were acutely induced after RA-treatment of these cells had been uncovered previously as regulated by RA-treatment of BEAS-2B cells (15)
, as shown in Tables 1
and 2
. In contrast, other species displayed in Fig. 2
were increased only after 10 days of RA-treatment and had not been highlighted previously as RA-regulated species in HBE cells, as displayed in Tables 1
and 2
. These species included OAS1, MXA, RIG-G, UBE1L, MK, and ISG15. Notably, these species included IFN-stimulated genes (ISGs).
|
that shows reduced UBE1L immunoblot expression in subsets of non–small-cell lung cancers as compared with adjacent normal lung tissues. To examine further the expression profile of UBE1L, an immunohistochemical assay was developed to detect UBE1L expression in human tissues. Attention focused on UBE1L expression in the normal human bronchial epithelium, because UBE1L expression was expected to be evident in this tissue. Independent polyclonal antibodies that recognized the amino- or carboxyl-terminus of UBE1L (16)
, respectively, were used to examine the immunohistochemical expression of UBE1L.
|
. We detected a similar immunostaining profile using the polyclonal antibody that recognized the carboxyl-terminus of UBE1L (data not shown). Preincubation of each of these anti-UBE1L antibodies with the relevant blocking peptide markedly inhibited UBE1L immunohistochemical staining and immunoblot detection (Fig. 4B
and data not shown). This result confirmed the specificity of these antibodies. Higher magnification revealed cytoplasmic, nuclear, and apical membrane staining in the normal human bronchial epithelium, as shown in Fig. 4C
. Preimmune sera served as an additional negative control, as displayed in Fig. 4D
.
|
. An inverse relationship was found to exist in Fig. 4E and F
between UBE1L and cyclin D1 immunohistochemical expression. This raised the prospect that UBE1L might directly cause cyclin D1 repression.
This possibility was explored through a cotransfection assay that was developed in BEAS-2B HBE cells. Cotransfection of HA-tagged cyclin D1 with UBE1L repressed cyclin D1 expression in a manner that depended on the dosage of transfected UBE1L, as displayed in Fig. 5A
. In marked contrast, cotransfection of a truncated UBE1L (UBE1L-T) that was known to inactivate UBE1L (16)
was no longer able to repress cyclin D1 expression. A reporter plasmid containing the ß-galactosidase gene was cotransfected with UBE1L as a control to confirm similar transfection efficiencies in each experiment. This transfected species was unaffected by UBE1L transfection and ß-actin expression was also unaffected (data not shown). These results directly linked UBE1L expression to repression of cyclin D1, a chemopreventive target in the lung (2
, 17)
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study used an established in vitro model of lung cancer chemoprevention. Our prior work revealed that acute NNK-treatment of HBE cells resulted in malignant transformation with increased proliferation, anchorage-independent growth, and ability to form tumors in athymic mice (4)
. Chronic RA-treatment resulted in chemoprevention and inhibited this transformation, as summarized in Fig. 1A
. In this study, microarray analyses of the immortalized (BEAS-2B), transformed (BEAS-2BNNK), and chemoprevented (BEAS-2BNNK/RA) HBE cell lines showed that RA-treatment inhibited gene expression changes associated with transformation (data not shown). These results underscore that chronic RA-treatment inhibited changes associated with NNK-mediated cellular transformation.
This model has already proven useful to identify degradation of G1 cyclins as a target for retinoid- or rexinoid-mediated chemoprevention (4, 5, 6
, 18)
and to validate the epidermal growth factor receptor as a target for lung cancer chemoprevention (7)
. In this study, further analyses were done to identify species augmented during this chemoprevention. Among the highlighted species were several identified previously as RA-target genes in HBE cells (15)
and others that had not been highlighted in prior work, as summarized in Tables 1
and 2
. To investigate whether these species were regulated in the BEAS-2B cell line, expression of representative species after acute (1 day) and prolonged (10 days) RA-treatment was assessed by RT-PCR analyses. Figure 2
shows that some species were increased acutely with RA-treatment whereas other species required longer RA-treatment for efficient induction. Interestingly, some species increased only after prolonged RA exposure included those identified previously as ISGs. The requirement for prolonged RA-treatment for induction of these species suggested that additional signaling mechanisms preceded this induction. Future work should explore the nature of these signals as well as potential cross-talk with the IFN signaling pathway.
One of these species, UBE1L, had been highlighted as a candidate tumor suppressor gene in lung carcinogenesis (12)
and was shown in this study as having repressed immunoblot expression in some examined lung cancers as compared with adjacent normal lung tissue, as shown in Fig. 3
. Immunohistochemical analyses revealed abundant UBE1L expression in the normal human bronchial epithelium where reduced expression of cyclin D1 was also detected, as shown in Figs. 4E and F
. To explore further the role of UBE1L expression in chemoprevention, the ability of UBE1L to regulate the expression of cyclin D1, a marker of retinoid chemoprevention response (2
, 5
, 6
, 14
, 18)
, was studied. Ectopic UBE1L repressed expression of transfected cyclin D1 in HBE cells, as shown in Fig. 5
. This activity depended on transfection of wild-type UBE1L because a mutant UBE1L species was unable to cause this repression. How UBE1L confers this repression of cyclin D1 is the subject of future work.
UBE1L is the activating enzyme for the IFN stimulated gene 15 kDa (ISG15; 19 ). ISG15 belongs to the family of ubiquitin-like proteins and is conjugated to different substrates after IFN treatment, microbial challenge, and genotoxic stress, among other stimuli, as reviewed (20) . Work conducted with the ISG15 deconjugating enzyme UBP43/USP18, has linked ISG15 conjugation to increased IFN response (20) . UBE1L was associated with these effects because restoration of UBE1L expression in IFN-resistant K562 cells could restore IFN-induced ISG15 conjugation and IFN-signaling, as reviewed (20) . RA- and IFN-treatments are each tightly associated with reduced cellular proliferation as well as the frequent induction of apoptosis (21 , 22) . The proposed involvement of UBE1L expression and ISG15 conjugation in these biological effects is an attractive hypothesis to explore in future work.
A role for UBE1L in cancer chemoprevention has not been established previously. This study revealed that UBE1L and ISG15 were each induced by 10 days of RA-treatment, as shown in Fig. 2
. This coordinate induction occurred along with the induction of other ISGs. Prior work has confirmed an interaction between UBE1L and ISG15 during retinoid-induced differentiation (23)
. A potential role for UBE1L in lung cancer chemoprevention was consistent with the inverse relationship found between immunohistochemical expression of UBE1L and cyclin D1 in the human bronchial epithelium, as shown in Fig. 4E and F
. A coincidental relationship between these regulated species was excluded by the results from engineered expression of UBE1L in HBE cells where exogenous cyclin D1 was repressed in these cells. Whether UBE1L, ISG15, or the ISG15 deconjugase, UBP43, are themselves pharmacologic targets remains to be shown.
The current study extends prior work by showing that RA-treatment activated expression of ISGs during chemoprevention. It is perhaps not a surprise that IFN target genes are activated during cancer chemoprevention because IFNs can trigger growth suppression and cell cycle arrest (22)
. This would permit repair of carcinogenic damage in a manner similar to that proposed previously for retinoids, as reviewed (21)
. Notably, UBE1L expression has been affected by both retinoid and IFN treatments (23)
. Prior work revealed that RA degraded the oncogenic translocation product, promyelocytic leukemia/retinoic acid receptor , and UBE1L, a retinoid target gene, substituted for RA in triggering this effect (16)
. This study extended that work by showing a cell cycle regulator was also susceptible to UBE1L-mediated repression.
In summary, retinoids are useful tools to uncover therapeutic and chemopreventive pathways. This study used an in vitro model and gene profiling experiments to discover target genes activated during HBE cell chemoprevention. Few studies have used a genomic approach to identify retinoid target genes (21) . The application of gene expression arrays to highlight chemopreventive targets represents a new application for this powerful technology. The challenge for future work is to learn whether species highlighted through these gene profiling experiments are engaged during clinical cancer chemoprevention.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: Ethan Dmitrovsky, Department of Pharmacology and Toxicology, Remsen 7650, Dartmouth Medical School, Hanover, NH 03755. Phone: 603-650-1667; Fax: 603-650-1129; E-mail: ethan.dmitrovsky{at}dartmouth.edu
Received 12/16/03. Revised 7/26/04. Accepted 8/26/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
is frequent in resectable non-small cell lung cancer, but does not predict tumor progression. Clin Cancer Res 1997;3:515-22.[Abstract]
degradation and apoptosis in acute promyelocytic leukemia. Proc Natl Acad Sci USA 2002;99:3806-11.This article has been cited by other articles:
|
|
 |
|
  S. Romagnoli, E. Fasoli, V. Vaira, M. Falleni, C. Pellegrini, A. Catania, M. Roncalli, A. Marchetti, L. Santambrogio, G. Coggi, et al. Identification of Potential Therapeutic Targets in Malignant Mesothelioma Using Cell-Cycle Gene Expression Analysis Am. J. Pathol., March 1, 2009; 174(3): 762 - 770. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Feng, D. Sekula, Y. Guo, X. Liu, C. C. Black, F. Galimberti, S. J. Shah, L. F. Sempere, V. Memoli, J. B. Andersen, et al. UBE1L causes lung cancer growth suppression by targeting cyclin D1 Mol. Cancer Ther., December 1, 2008; 7(12): 3780 - 3788. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Shah, S. Blumen, I. Pitha-Rowe, S. Kitareewan, S. J. Freemantle, Q. Feng, and E. Dmitrovsky UBE1L represses PML/RAR{alpha} by targeting the PML domain for ISG15ylation Mol. Cancer Ther., April 1, 2008; 7(4): 905 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Reu, S. I. Bae, L. Cherkassky, D. W. Leaman, D. Lindner, N. Beaulieu, A. R. MacLeod, and E. C. Borden Overcoming Resistance to Interferon-Induced Apoptosis of Renal Carcinoma and Melanoma Cells by DNA Demethylation J. Clin. Oncol., August 10, 2006; 24(23): 3771 - 3779. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Freemantle, K. H. Dragnev, and E. Dmitrovsky The retinoic Acid paradox in cancer chemoprevention. J Natl Cancer Inst, April 5, 2006; 98(7): 426 - 427. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Dmitrovsky D-Type Cyclins as Novel Targets for Chemoprevention Am. Assoc. Cancer Res. Educ. Book, April 1, 2006; 2006(1): 59 - 63. [Full Text] [PDF]
|
|
|
|
|
|
F. R. Hirsch and S. M. Lippman Advances in the Biology of Lung Cancer Chemoprevention J. Clin. Oncol., May 10, 2005; 23(14): 3186 - 3197. [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 |
). B, cotransfection of wild-type but not an inactive mutant UBE1L (pSG5-UBE1L-T, UBE1L-mut) species decreased transfected cyclin D1 expression in these HBE cells. Transfected cyclin D1 levels were unaffected by cotransfection of pSG5-UBE1L-T or by cotransfection of pSG5 vector alone. UBE1L transfection did not affect activity in these cells of a transfected ß-galactosidase reporter plasmid that served as a control for transfection efficiency and normalization (data not shown).