| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Bristol-Myers Squibb Pharmaceutical Research Institute, Department of Oncology Drug Discovery, Princeton, New Jersey 08543 [D. K. B., R. B. R., C-P. H., B. P., J. D., M. S.], and University of Texas M. D. Anderson Cancer Center, Science Park-Research Division, Smithville, Texas 78957 [K. K., S. M., R. K., S. M. F.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
25% versus 100%, respectively, on an ICR background) and presented with significantly reduced tumor burdens (average, 0.03 versus 12.7 tumors/mouse, respectively, on a FVB background and 0.5 versus 7.1 tumors/mouse, respectively, on an ICR background). Mice fed a COX-2 inhibitor in utero and as weanlings up to the time of promotion also showed a significant resistance to tumor development. These results clearly raise questions regarding the role of COX-2 and elevated prostaglandin levels in skin tumor development. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Several potential mechanisms have been invoked to account for the tumor-promoting properties associated with COX-2 overexpression. The observations that the expression level of Bcl-2 is elevated in rat intestinal epithelial cells overexpressing COX-2 (12) and in human colon cancer cells treated with PGE2 (13) led to the suggestion that COX-2 expression may promote tumor growth by imparting resistance to apoptosis. Such protection could prolong the survival of abnormal cells, thereby allowing time for the accumulation of multiple genetic mutations, ultimately resulting in a transformed phenotype. COX-2 overexpression in colon cancer cells is also associated with increased expression of metalloproteinases and increased production of vascular endothelial growth factors (14 , 15) . The resulting cells possess both increased metastatic potential, as well as the ability to stimulate endothelial cell growth and promote angiogenesis. Finally, suppression of the immune system in response to elevated PGE2 levels could also contribute to tumor promotion (16 , 17) .
Further evidence implicating COX-2 in tumor development comes from the observation that NSAIDs, which are inhibitors of COX-1 and COX-2, can block tumor formation in several genetic and chemical models of tumor induction, including in mouse skin (11 , 18, 19, 20) . This observation, however, is somewhat confounded by data that clearly indicate that NSAIDs can exert their antineoplastic effects through both COX-2-dependent and -independent mechanisms (21 , 22) . In particular, NSAIDs inhibit the proliferation of cancer cells that do not express either COX isoform (23) , and sulindac sulfone, a metabolite of the NSAID sulindac, which lacks COX inhibitory activity, can still protect against both genetic and carcinogen-induced tumor formation (24, 25, 26) . Thus, genetic rather than pharmacological approaches are likely to be more informative in understanding the role of PGs in neoplasia.
To better understand the role of COX-2 in skin tumor development, we generated a line of transgenic mice in which the human K14 promoter was used to regulate tissue-specific expression of a murine COX-2 transgene. This allowed us to examine in vivo the relationship between elevated COX-2 activity, the morphology and proliferative activity of the epidermis, and its response to chemical carcinogenesis protocols. Here we report on the phenotype of these mice as well as the impact of COX-2 expression on tumor induction, using a chemical initiation/promotion protocol for multistage skin tumor carcinogenesis.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Production of K14.COX2 Transgenic Mice.
The plasmid pK14.COX2 was digested with KpnI, and the 5.7-kb fragment gel was purified with a kit according to the manufacturer’s protocol (Qiagen). The K14.COX2 transgene was microinjected into the pronucleus of one-cell fertilized B6D2F2 embryos. After overnight incubation, two cell embryos were transferred into the oviducts of pseudopregnant ICR female mice for full-term gestation. Progeny were genotyped by PCR analysis using the oligomers 5'-CTG GTT ATT GTG CTG TCT-3' (complementary to rabbit ß-globin intron sequences), and 5'-GCT GAG TTC CTT CGT GAG-3' (specific for the murine COX-2 cDNA). Founders were mated onto the ICR Swiss Webster outbred background to produce hemizygous animals for further characterization. The transgene was also crossed onto a FVB background by 10 generations of backcrossing with inbred FVB mice.
PGE2 Analysis.
Epidermal PGE2 levels were measured with a RIA kit (NEN Life Sciences, Boston, MA) as we have described previously (30)
.
Northern and Western Blot Analyses.
Total RNA was isolated using TriReagent (Molecular Research Corp., Cincinnati, OH), according to the manufacturer’s protocol. Northern blots were prepared and probed as we have described previously (11)
. Immunoblot analysis for COX-1 and COX-2, using 25 µg of protein was performed as described (11)
. Authentic COX-2 protein (Caymen Chemical, Ann Arbor, MI) was included as a positive control. Chemiluminescence detection was performed as directed by the manufacturer (Amersham Pharmacia Biotech, Piscataway, NJ).
Histologies.
Skin samples were fixed in 10% formalin before being embedded in paraffin. Sections (5 µm) were stained with H&E, using standard protocols. Immunohistochemical staining was also performed, using anti-COX-2 antibody (Caymen) at a 1:1000 dilution and a fluorescein-conjugated secondary antibody (Amersham) at a 1:2000 dilution. The presence of apoptotic cells was determined using the ApopTag kit (Intergen, Purchase, NY) according to the manufacturer’s protocol.
Tumor Experiments.
Groups of 20–30 adult mice were initiated by topical application of DMBA. Two weeks later, all mice were treated twice weekly with TPA for up to 20 weeks. For mice on a FVB background, 100 µg of DMBA and 2.5 µg of TPA were used; for ICR mice, 50 µg of DMBA and 5 µg of TPA were used. The doses of DMBA and TPA used for ICR and FVB mice were those that have previously been shown to produce a reasonable number of tumors (31
, 32)
. Mice were observed daily, and the number of tumors present was determined weekly. Tumor data were calculated in terms of incidence (percentage of mice bearing tumors) and yield (average number of tumors per mouse). For one experiment, pairs of FVB K14.COX2 mice were fed 500 ppm celecoxib (LKT Laboratories, St. Paul, MN) in AIN76 diet beginning 1 day after mating and continuing through weaning, at which time pups were placed on diets containing celecoxib, as shown in the schematic below. Celecoxib was removed from the diet 1 week after initiation. The nontransgenic controls consisted of littermates of the K14.COX2 mice on celecoxib.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In this construct, the COX-2 cDNA is flanked by the rabbit ß-globin intron and SV40 poly(A) sequences to ensure proper post-transcriptional processing of the transgene. Of the seven founder mice identified, two demonstrated germ-line transmission of the transgene, and progeny from two founders demonstrated a gross phenotype with apparent differences in severity. Although both lines exhibited similar phenotypes, the line showing the highest level of expression was chosen for study.
|
. The levels of COX-1 expression were essentially the same in wild-type and transgenic mice (Fig. 1B)
. Western blot analysis also showed that the COX-2 transgene was expressed at the protein level (Fig. 1C)
. COX-2 transgene expression was also observed in the forestomach, uterus, gall bladder, prostate, and esophagus of transgenic animals (data not shown), consistent with the K14 promoter driving transgene expression in several epithelial tissues (33)
. COX-2 expression in transgenic mice was further verified by immunofluorescence (Fig. 1D)
. Fluorescence intensity was greatest in the basal epithelial layer of the skin of K14.COX2 mice, including those cells lining the outer root sheath of the hair follicle. This staining pattern was in concordance with the localization of K14 expression. Only background fluorescence was observed in the epidermis of wild-type littermates. Assays performed on both wild-type and transgenic skin (Fig. 1E)
demonstrated a 2-fold increase in the levels of PGE2 in the skin of K14.COX2 mice. This observation strongly suggests that the COX-2 protein produced as a result of transgene expression is a functionally active enzyme. Because PGE2 is readily degraded by PG dehydrogenase, which is expressed in murine epidermis, its level is not necessarily expected to be concordant with the level of COX-2 expression (data not shown).
Gross and Histological Phenotypes of the K14.COX2 Transgenic Mice.
Both ICR and FVB K14.COX2 mice appeared healthy at birth and presented with no obvious abnormalities or reproductive problems, although the transgenic mice were slightly smaller than their wild-type littermates. Differences between transgenic and wild-type littermates became apparent when hair growth began, starting as early as 1 week after birth. Hair development occurred at a much reduced rate in the K14.COX2 mice compared with wild-type animals, manifesting in alopecia in adult animals (Fig. 2A
, middle). At the microscopic level, slight differences in skin thickness were noted between transgenic animals and wild-type mice (Fig. 2B)
such that the skin of ICR K14.COX2 animals appeared thinner and showed signs of atrophy; this was less evident in FVB K14.COX2 skin. Additionally, the hair shafts had an altered morphology in both ICR and FVB transgenic animals (Fig. 2C)
and contained cells with structurally deformed nuclei lining the hair follicles. The presence of such pyknotic nuclei, a characteristic of cells undergoing apoptosis, was infrequently observed in skin from wild-type animals. However, we were unable to identify apoptotic cells by the terminal deoxynucleotidyl transferase (Tdt)-mediated nick end labeling assay. Also notable in the transgenic mice was the sizable enlargement of the sebaceous gland, which frequently was larger than the hair follicle (Fig. 2C)
.
|
, right). These findings support the suggestion that the alopecia results from improper or incomplete development of the hair shafts brought on by high PG levels within the epidermis of the K14.COX2 mice. Several animals were followed up to several months after discontinuation of treatment. The coats appeared to remain stable for at least 12 weeks, in that no apparent hair loss was observed in these animals, although it is not known whether they underwent another round of the hair growth cycle during this timeframe. By 3 months, some hair loss was noted in all of the treated animals (data not shown). Administration of celecoxib to pregnant animals and subsequent weanlings resulted in normal hair development and normal-appearing hair follicles and sebaceous glands in the weanlings that was indistinguishable from wild-type mice (Fig. 2D)
. These animals were used in a subsequent tumor experiment.
COX-2 Overexpression Reduces DMBA/TPA-induced Skin Tumor Incidence.
To determine the effect of COX-2 overexpression on skin tumor development, transgenic and wild-type mice were subjected to a DMBA/TPA treatment protocol. This was carried out with both the ICR and FVB K14.COX2 mice to determine whether tumor responses were mouse strain specific. On the basis of previously reported evidence implicating an important role for COX-2 in promoting skin tumor formation (11)
, we expected to find either greater tumor incidence and multiplicity or larger tumor size in the transgenic mice. However, although 100% of the wild-type ICR mice developed tumors, papillomas formed in only 25% of the ICR K14.COX2 transgenic mice (data not shown). The ICR K14.COX2 animals that developed skin tumors also had lower tumor multiplicity than observed for littermates (0.5 versus 7.1 tumors/mouse), and those tumors that did arise were significantly smaller, having a high lymphatic infiltrate (data not shown). The FVB K14.COX2 mice were also resistant, with tumor incidences of only 3.3% compared with 97% in the wild-type FVB controls (Fig. 3A)
. There were also dramatic differences in tumor multiplicity, with the wild-type FVB producing an average of 12.7 papillomas/mouse, whereas the FVB K14.COX2 produced only 1 papilloma on 1 of 30 mice (0.33/mouse). There was no difference in the response of wild-type and FVB K14.COX2 epidermis to TPA with regard to degree of epidermal thickening (Fig. 3B)
or labeling index (data not shown). Because both FVB and ICR transgenic mice showed a similar resistance to tumor development, it is unlikely that this phenomenon is a stain-of-mouse effect.
|
and normal hair growth. As shown in Fig. 3A
, the FVB K14.COX2 mice retained most, but not all of their resistance to tumor development. Tumor multiplicity was 0.9 at 22 weeks, an increase over a multiplicity of 0.03 in the original experiment. The largest change was in tumor incidence, which was 55% compared with 3.3% in the original experiment. Because the use of celecoxib should have allowed for normal initiation, given that the hair follicles and sebaceous glands were normal, it was concluded that elevated expression of COX-2 confers resistance primarily to the promotion stage of multistage carcinogenesis, although the initiation stage is also affected. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many of the keratinocytes lining the hair follicles of K14.COX2 transgenic mice display pyknotic nuclei, indicating the possibility that these cells may be undergoing apoptosis. We were, however, unable to detect any differences in the rates of apoptosis by standard methods. It is not clear whether the pyknotic nuclei represent apoptosis or an incompletely differentiated cell. Nonetheless, such a defect in keratin-ocyte development would likely have a negative impact on the ability of the hair follicles to support hair growth, consistent with the alopecia phenotype.
The most significant observation was the resistance to skin tumor development conferred by COX-2 overexpression. We considered several possible explanations related to the defects in follicular development. It is possible that cells initiated by exposure to DMBA underwent apoptosis before promotion with TPA and thus were unable to contribute to tumor formation in these animals. Alternatively, the subpopulation of cells that can be initiated by DMBA might be altered or absent from the skin of transgenic mice. To address these issues, the COX-2 selective inhibitor celecoxib was administered to impregnated females and to the subsequent weanlings, resulting in histologically normal hair follicles and hair growth. These mice were maintained on the COX-2 inhibitor up through initiation with DMBA at 6 weeks of age; this was also done to remove the possibility that COX-2 might alter DMBA metabolism (36 , 37) . These mice, however, were still significantly resistant to tumor development, strongly suggesting that the tumor resistance phenotype is not attributable to hair follicle abnormalities or perturbation of carcinogen metabolism. The slight decrease in tumor multiplicity observed in the wild-type mice that had been fed celecoxib was possibly attributable to residual celecoxib, which would have inhibited COX-2 induced by TPA.
The precise role(s) that induction of COX-2 expression, and the consequently increased levels of PGs, may play in the tumor promotion process remain unclear. In studies using the DMBA/TPA protocol in NMRI mice, the nonselective COX inhibitor indomethacin was shown to significantly reduce tumor multiplicity (38)
. This effect could be reversed by topical application of PGF2
, but not PGE2 (38)
. The effect of indomethacin, however, is mouse-strain dependent, because indomethacin and another nonselective COX inhibitor, flurbiprofen, were found to enhance, rather than inhibit, promotion in SENCAR mice (39)
. This latter study could be interpreted as showing that one or more or the PGs inhibit tumor promotion, an idea that is consistent with the tumor resistance phenotype of the K14.COX2 mice. Also in support of this concept, a study on the effect of topical application of various PGs during tumor promotion showed that PGE2, but not PGF2, inhibited tumor development (40)
. Recently, exogenous PG administration was also unexpectedly shown to significantly reduce the number of intestinal tumors in Min/+ mice (41)
. Additionally, overexpression of COX-2 protein has been reported to induce cell cycle arrest independently of PGs, which led to induction of apoptosis (42)
. Thus, evidence exists in several systems that indicates that PGs may inhibit tumor development, at least under some circumstances.
The observation that topically applied PGE2 can inhibit tumor promotion (40) suggests that through its signaling pathway(s), PGE2 abrogates responsiveness to TPA. Topically applied PGE2 alone induces cAMP, which remains elevated for 90 min (43) . Although TPA elicits PG synthesis (3) , this occurs as a result of TPA signaling; PG signaling before TPA application, as would occur in the transgenic mouse, may interfere with TPA signaling. Further studies are needed to verify this speculation
There are, however, numerous observations that support a critical or enhancing role for PGs in the development of skin tumors. First, it has been well established that tumor promoters induce COX-2 and that the resulting tumors have constitutively up-regulated COX-2, which correlates with elevated PG synthesis (3) . Second, inhibitors of COX-2 have been shown to reduce skin tumors induced by either the DMBA/TPA protocol or by UV irradiation (11 , 44) . Finally, mice that are deficient in COX-2 show a dramatically reduced tumor response to DMBA/TPA (45) .
The tumor resistance phenotype in our K14.COX2 mice is also surprising in light of the recent report that overexpression of COX-2, under the control of mouse mammary tumor virus, in murine mammary glands causes spontaneous tumor development. We did not observe mammary tumors in the K14.COX2 mice, although the level of expression of COX-2 in our mice appears to be lower than in the mouse mammary tumor virus transgenic mice (46) .
This study also raises the question of identification of the cells that are the targets of the PGs synthesized by the keratinocytes. PGs are secreted and can act in either an autocrine or a paracrine manner through binding to specific receptors. The effects of PGE2, for example, are mediated primarily by the EP receptor family, of which there are four major subgroups. The expression of specific EP receptors, which are linked to different signal transduction pathways, is cell-type dependent; the specific receptors expressed on murine keratinocytes are not known at present. The more recent identification of nuclear receptors, the peroxisome proliferator-activated receptors, that can be activated by some PGs is intriguing (47)
. The 
isoform of peroxisome proliferator-activated receptor, which is expressed in murine keratinocytes (48)
, has been associated with cell cycle inhibition and apoptosis (49)
. The extent to which this is occurring in the K14.COX2 mice is unknown, but it offers avenues for future investigation.
Consideration should also be given to the possibility that COX-2 overexpression and activity results in a reduction in the synthesis of lipoxygenase products. Several lipoxygenases are expressed in murine epidermis, including the 8-, 12-, and 15-lipoxygenases (50) . This family of enzymes, which synthesizes HETEs, compete with COX-1 and COX-2 for arachidonic acid, the precursor for both the PGs and HETEs. High levels of COX-2 may lead to metabolism primarily to PGs, at the expense of HETEs. The most abundant HETE, 12-HETE, has been implicated in tumor development in several systems because of its ability to cause angiogenesis and alter cell adhesion (51 , 52) .
In summary, we have shown that constitutive expression of COX-2 in murine epidermis confers both alopecia and a skin tumor resistance phenotype. These unexpected findings suggest that the role of PGs in normal skin physiology and skin tumor development is complex and not well understood. Clearly, much more work is needed to resolve the many questions raised by this and other studies.
|
|
FOOTNOTES |
|---|
1 This work was supported by NIH Grants CA-34443 and CA-16672. 
2 To whom requests for reprints should be addressed, at the University of Texas M. D. Anderson Cancer Center, Science Park-Research Division, P. O. Box 389, Smithville, TX 78957. E-mail: smfischer{at}mdanderson.org
3 The abbreviations used are: PG, prostaglandin; COX, cyclooxygenase; NSAID, nonsteroidal anti-inflammatory drug; K14, keratin 14; DMBA, 7,12-dimethylbenz[a]anthracene; TPA, 12-O-tetradecanoylphorbol-13-acetate; HETE, hydroxyeicosatetraenoic acid.
Received 11/21/01. Accepted 2/27/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
release via a mechanism involving IL-10. J. Immunol., 154: 3383-3390, 1995.[Abstract]
induces cell cycle withdrawal: inhibition of E2F/DP DNA-binding activity via down-regulation of PP2A. Genes Dev., 11: 1987-1998, 1998.This article has been cited by other articles:
|
|
 |
|
  K. M. Ansari, J. E. Rundhaug, and S. M. Fischer Multiple Signaling Pathways Are Responsible for Prostaglandin E2-Induced Murine Keratinocyte Proliferation Mol. Cancer Res., June 1, 2008; 6(6): 1003 - 1016. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Zhang, J. Li, L. Song, W. Ouyang, J. Gao, and C. Huang A JNK1/AP-1-Dependent, COX-2 Induction Is Implicated in 12-O-Tetradecanoylphorbol-13-Acetate-Induced Cell Transformation through Regulating Cell Cycle Progression Mol. Cancer Res., January 1, 2008; 6(1): 165 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wan, A. Saghatelian, L.-W. Chong, C.-L. Zhang, B. F. Cravatt, and R. M. Evans Maternal PPAR{gamma} protects nursing neonates by suppressing the production of inflammatory milk Genes & Dev., August 1, 2007; 21(15): 1895 - 1908. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Meng, S. Ceryak, Z. Aratsu, L. Jones, L. Epstein, and B. Bouscarel Biphasic regulation by bile acids of dermal fibroblast proliferation through regulation of cAMP production and COX-2 expression level Am J Physiol Cell Physiol, September 1, 2006; 291(3): C546 - C554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-Y. Tang, A. Shih, H. J. Cao, F. B. Davis, P. J. Davis, and H.-Y. Lin Resveratrol-induced cyclooxygenase-2 facilitates p53-dependent apoptosis in human breast cancer cells. Mol. Cancer Ther., August 1, 2006; 5(8): 2034 - 2042. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Xu, S. Choudhary, O. Voznesensky, M. Mehrotra, M. Woodard, M. Hansen, H. Herschman, and C. Pilbeam Overexpression of Cox-2 in Human Osteosarcoma Cells Decreases Proliferation and Increases Apoptosis. Cancer Res., July 1, 2006; 66(13): 6657 - 6664. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Aziz, D. L. Wheeler, B. Bhamb, and A. K. Verma Protein Kinase C {delta} Overexpressing Transgenic Mice Are Resistant to Chemically but not to UV Radiation-Induced Development of Squamous Cell Carcinomas: A Possible Link to Specific Cytokines and Cyclooxygenase-2 Cancer Res., January 15, 2006; 66(2): 713 - 722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Rigas and K. Kashfi Cancer Prevention: A New Era beyond Cyclooxygenase-2 J. Pharmacol. Exp. Ther., July 1, 2005; 314(1): 1 - 8. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. D. Dwyer-Nield, M. C. Srebernak, B. S. Barrett, J. Ahn, P. Cosper, A. M. Meyer, L. R. Kisley, A. K. Bauer, D. C. Thompson, and A. M. Malkinson Cytokines differentially regulate the synthesis of prostanoid and nitric oxide mediators in tumorigenic versus non-tumorigenic mouse lung epithelial cell lines Carcinogenesis, July 1, 2005; 26(7): 1196 - 1206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pozzi, X. Yan, I. Macias-Perez, S. Wei, A. N. Hata, R. M. Breyer, J. D. Morrow, and J. H. Capdevila Colon Carcinoma Cell Growth Is Associated with Prostaglandin E2/EP4 Receptor-evoked ERK Activation J. Biol. Chem., July 9, 2004; 279(28): 29797 - 29804. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-S. Chun, H.-H. Cha, J.-W. Shin, H.-K. Na, K.-K. Park, W.-Y. Chung, and Y.-J. Surh Nitric oxide induces expression of cyclooxygenase-2 in mouse skin through activation of NF-{kappa}B Carcinogenesis, March 1, 2004; 25(3): 445 - 454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. KUWANO, S. NAKAO, H. YAMAMOTO, M. TSUNEYOSHI, T. YAMAMOTO, M. KUWANO, and M. ONO Cyclooxygenase 2 is a key enzyme for inflammatory cytokine-induced angiogenesis FASEB J, February 1, 2004; 18(2): 300 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Raju and R. P. Bird Energy Restriction Reduces the Number of Advanced Aberrant Crypt Foci and Attenuates the Expression of Colonic Transforming Growth Factor {beta} and Cyclooxygenase Isoforms in Zucker Obese (fa/fa) Rats Cancer Res., October 15, 2003; 63(20): 6595 - 6601. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Klatt and M. Serrano Engineering cancer resistance in mice Carcinogenesis, May 1, 2003; 24(5): 817 - 826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. N. DuBois Evaluation of the Whole Prostaglandin Biosynthetic Pathway in Lung Cancer Clin. Cancer Res., May 1, 2003; 9(5): 1577 - 1578. [Full Text] [PDF]
|
|
|
|
|
|
M. M. Hardy, E. A. G. Blomme, A. Lisowski, K. S. Chinn, A. Jones, J. M. Harmon, A. Opsahl, R. L. Ornberg, and C. S. Tripp Selective Cyclooxygenase-2 Inhibition Does Not Alter Keratinocyte Wound Responses in the Mouse Epidermis after Abrasion J. Pharmacol. Exp. Ther., March 1, 2003; 304(3): 959 - 967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Zeng, L. K. McCauley, and C.-Y. Wang Hepatocyte Growth Factor Inhibits Anoikis by Induction of Activator Protein 1-dependent Cyclooxygenase-2. IMPLICATION IN HEAD AND NECK SQUAMOUS CELL CARCINOMA PROGRESSION J. Biol. Chem., December 13, 2002; 277(51): 50137 - 50142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Willis and N. H. Colburn Molecular Targets for Cancer Prevention: A Meeting Review of the Third American Cancer Society-Schilling Research Conference Cancer Epidemiol. Biomarkers Prev., October 1, 2002; 11(10): 972 - 978. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Muller-Decker, G. Neufang, I. Berger, M. Neumann, F. Marks, and G. Furstenberger Transgenic cyclooxygenase-2 overexpression sensitizes mouse skin for carcinogenesis PNAS, September 17, 2002; 99(19): 12483 - 12488. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||
