Cyp1b1 Exerts Opposing Effects on Intestinal Tumorigenesis Via Exogenous and Endogenous Substrates

Research Article

Cyp1b1 Exerts Opposing Effects on Intestinal Tumorigenesis via Exogenous and Endogenous Substrates

Richard B. Halberg,2 Michele Campaigne Larsen,1 Tammy L. Elmergreen,1 Alex Y. Ko,1 Amy A. Irving,1 Linda Clipson,2 and Colin R. Jefcoate1

Departments of 1Pharmacology and 2Oncology, University of Wisconsin, Madison, Wisconsin

Abstract familial adenomatous polyposis (FAP). These patients with this Cytochrome P450 1B1 (Cyp1b1) metabolism contributes to hereditary form of the disease develop hundreds to thousands of physiologic functions during embryogenesis but also to colonic tumors by the second decade of life. Apc maintains homeostasis in the intestine as an integral carcinogenic activation of polycyclic aromatic hydrocarbons (PAH).We generated Cyp1b1-deficient mice carrying the component of Wnt signaling (2), which disrupts the cytoplasmic h Min allele of the adenomatous polyposis coli gene.These complex of Apc with -catenin, axin, and glycogen synthase kinase h h h Cyp1b1-deficient Min mice developed twice as many tumors as 3 (GSK3 ). This prevents GSK-initiated proteolysis of -catenin, Min controls, which, however, remained similar in size and enhances translocation to the nucleus, and stimulates transcription d histology.Tumors from older (130 days) Cyp1b1-deficient Min of target genes, including c-Myc, cyclin D, MMP-9, PPAR , and h mice selectively exhibited focal areas of nuclear atypia Sox-9. -Catenin also directly contributes to epithelial cell h associated with less organized epithelia.The metabolism of adhesion. Loss of Apc activity and elevated -catenin drastically endogenous substrates by Cyp1b1, therefore, suppresses alters the intestinal epithelium, including expanding the stem cell tumor initiation but also affects progression.Treatment of population. c-Myc is a key mediator of these dramatic biological Min mice with 7,12-dimethylbenzanthracene (DMBA) doubled changes (3). Array analysis of several colon cancer models shows both tumor multiplicity and size within 20 days but not when that mouse and human tumors recapitulate embryonic develop- mice lacked Cyp1b1.This was paralleled by an abnormal mental signatures, irrespective of etiology (3). staining of crypts with B-catenin, phospho-IKB kinase, and Several mouse models of FAP have been developed. Mice RelA, which may represent an early stage of tumorigenesis carrying the Min allele of Apc express an altered form of Apc h similar to aberrant crypt formation.Cyp1b1 deletion did not protein, lacking the COOH-terminal domains that bind axin, - affect circulating DMBA and metabolites.Cyp1b1 expression catenin, and GSK3h (2). C57BL6/J mice carrying Min develop on was higher in the tumors compared with normal small average 100 tumors along the entire length of the intestinal tract. intestines.Increased tumorigenesis may, therefore, arise from Several genes involved in polyunsaturated fatty acid signaling, generation of DMBA metabolites by Cyp1b1 in the developing including secreted phospholipase A2, Cox1, Cox2, Lpl, PPARd, and tumors.Benzo( a)pyrene (BP), which is similarly activated by PPARc, affect the multiplicity of intestinal tumors in Min mice (4–6). Cyp1b1 in vitro, did not affect tumorigenesis in Min mice.By Carcinogens also affect intestinal tumorigenesis in Min mice. contrast, BP and DMBA each suppressed tumor multiplicity in Some carcinogens affect tumorigenesis in specific regions of the the absence of Cyp1b1.Cyp1b1 metabolism of DMBA and intestine. Treatment of Min mice with ethylnitrosourea (ENU), a endogenous oxygenation products may each affect a tumor- direct acting alkylating agent, increased the multiplicity of small promoting nuclear factor-KB activation, whereas Ah receptor intestinal tumors by a factor of 3 (7). The tumors from these mice activation by PAH affects suppression.Tumorigenesis often carried a point mutation that inactivated the wild-type allele may, therefore, depend on activation of PAH by Cyp1b1 and of Apc. Similarly, treatment of Min mice with 2-amino-1-methyl-6- on offsetting suppression by Cyp1b1 of endogenous tumor- phenylimidazol[4,5-b]pyridine (PhIP) increased the multiplicity of enhancing substrates. [Cancer Res 2008;68(18):7394–402] small intestinal tumors by inactivating the wild-type allele through a point mutation and in part through inducing genetic instability Introduction (8). Other carcinogens specifically affect tumorigenesis in the colon. Treatment of Min mice with azoxymethane (AOM) increased the Colorectal cancer is the second leading cause of cancer death in incidence and multiplicity of colonic tumors (9). The majority of the United States. This disease is the culmination of epigenetic and tumors from these mice exhibited loss of heterozygosity (LOH) at genetic alterations that initiate tumorigenesis and drive progres- the Apc locus. Thus, exogenous chemicals can affect intestinal sion. An early molecular event is the inactivation of the tumorigenesis in Min mice by promoting the inactivation of the adenomatous polyposis coli (Apc) gene (1). The role of Apc in wild-type allele of Apc through a variety of distinct mechanisms. maintaining homeostasis is best exemplified by the existence of Most carcinogens, including PhIP and AOM, must be metabolically activated by cytochrome P450s and other metabolizing enzymes before exerting their effect on tumorigenesis (10). In this article, we Note: R.B. Halberg and M.C. Larsen contributed equally to this work. treat Min mice with polycyclic aromatic hydrocarbons (PAH) for Dedicated to Margaret Thorne, whose valiant fight with colon cancer was an added the first time. These chemicals, to which humans are extensively motivation for this research. Requests for reprints: Colin R. Jefcoate, Department of Pharmacology, University exposed, cause many cancers through their conversion to reactive of Wisconsin School of Medicine and Public Health, 1300 University Avenue, Madison, dihydrodiol epoxides (11). WI 53706. Phone: 608-263-3975; Fax: 608-262-1257; E-mail: [email protected]. I2008American Association for Cancer Research. Cytochrome P450 1B1 (CYP1B1) is often more highly expressed doi:10.1158/0008-5472.CAN-07-6750 in neoplasms, notably those of the bladder, breast, colon, kidney,

Cancer Res 2008; 68: (18). September 15, 2008 7394 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2008 American Association for Cancer Research. Cyp1b1 Affects Intestinal Tumorigenesis in Mice ovaries, and prostate, than in the adjacent normal epithelium Materials and Methods (12–14). CYP1B1-mediated conversion of estradiol to 4-OH Mouse breeding and maintenance. Mice were maintained in the estradiol has been linked to a variety of human endocrine cancers Association for Assessment and Accreditation of Laboratory Animal Care– (13, 14). Certain allelic variants of CYP1B1 have been further accredited University of Wisconsin School of Medicine and Public Health associated with increased estradiol 4-hydroxylation and increased Animal Care Facility. All experiments were carried out in accordance with risk for colon cancer, and elevated CYP1B1 expression is a marker protocols approved by the Animal Care and Use Committee in accordance for more aggressive colon tumors (15). Mouse Cyp1b1, which to the NIH Guide for the Care and Use of Laboratory Animals. All mice were differs from human CYP1B1 by 20% (amino acid level), shows provided food and water ad libitum and were maintained on a 12-h light/ appreciably different PAH metabolism and much less estradiol dark cycle. The mice were fed 2019 Teklad Global 19% Protein Extruded metabolism (16). Rodent Diet, which is essentially free of flavonoids. The initial C57BL/6 ApcMin/+ founders were obtained from Dr. Amy Moser at the University of The Cyp1b1 gene is scarcely expressed in liver but is frequently Wisconsin (26). A colony was maintained by breeding wild-type C57BL6/J À À found in extrahepatic tissues (17). Cyp1b1 was initially purified females with C57BL6/J ApcMin/+ males. The initial C57BL6/J CYP1B1 / from mouse embryonic fibroblasts (17) and was first cloned from founders were obtained from Dr. Frank Gonzalez (Center for Cancer human keratinocytes (18). Cloning from the rodent tissue Research, National Cancer Institute, Bethesda, MD; ref. 21). A colony was À À established that the same form exhibited very diverse expression maintained by intercrossing CYP1B1 / offspring. To produce C57BL6/J À À À À (17). The gene has a very unusual structure, containing a promoter ApcMin/+ CYP1B1 / mice, C57BL6/J CYP1B1 / females were first bred to À À with multiple hormonal control elements, only two introns, and a C57BL6/J ApcMin/+ males and then C57BL6/J CYP1B1 / females were bred +/À Min/+ highly extended 3¶-untranslated region (3¶-UTR; 19). Cyp1b1 is to C57BL6/J CYP1B1 Apc males. Offspring were genotyped before regulated by a highly conserved 200-bp enhancer that responds to 21 d of age, in accordance to previously published protocols (7, 21). the Ah receptor (AhR), to cyclic AMP through a far upstream Treatments. BP, DMBA, and olive oil were purchased from either Sigma Chemical Co. or AccuStandard, Inc. Female mice were injected i.p. on days enhancer, and also through a miRNA that targets the 3¶-UTR. This 28and 35 with 0.1 mg of either BP or DMBA (or other doses as indicated) latter mechanism has been linked to the elevated expression in in olive oil (200 AL, 0.5 mg/mL) or with vehicle alone. Eight to 10 mice tumors (20). from three to four litters were treated over an extended period. In one The effect of Cyp1b1 on carcinogenesis in mice is tissue specific experiment, DMBA was injected as a single dose of 0.2 mg on day 28. Mice (21). Treatment of wild-type mice with 7,12-dimethylbenzanthra- were sacrificed on days 55, 85, or 130, as specified. cene (DMBA) induces tumors in several organs, including ovary, Tumor analysis. Mice were sacrificed by CO2 asphyxiation. The skin, uterus, and lung, which occur less frequently in Cyp1b1- intestinal tissues were removed and rinsed in PBS. Four-centimeter deficient mice. DMBA and benzo(a)pyrene (BP) bind with high segments of proximal, medial, and distal small intestine as well as the affinity to recombinant mouse Cyp1b1 forming typical substrate entire colon were placed on Whatman filter paper. Each tissue sample Â complexes (16). DMBA and BP are converted to dihydrodiol was opened longitudinally, rinsed with 1 PBS to wash away luminal contents, and fixed overnight in 4% paraformaldehyde solution at room epoxides by Cyp1b1 (22). Formation of tumors parallels formation temperature. After 16 h, the tissue samples were carefully removed from of DNA-DMBA dihydrodiol adducts in the responsive tissues (21). the filter paper and stored in 75% ethanol at room temperature. Tumor By contrast, in these same studies, Cyp1b1 suppressed DMBA- multiplicity was determined by two investigators, who were both blinded to induced lung tumors. This tissue specificity, therefore, may reflect a the genotype of the animal and treatment, using an Olympus dissecting cancer-inducing role for endogenous substrates of Cyp1b1. microscope at Â10 magnification. The maximum diameter of each tumor Physiologic substrates are implicated in the highly conserved was measured using a calibrated eyepiece reticule. expression during early development in mice, which plays a role Statistical analysis. Tumor multiplicity and size were compared among in embryonic patterning (23, 24). A lack of Cyp1b1 activity in mice different groups of mice using the Wilcoxon rank sum test (4). and humans causes aberrant eye development (23, 25). Histology. Intestinal tumors and adjacent normal tissue were analyzed We describe the multiplicity, size, and pathology of tumors by immunohistochemistry. The tissues were embedded in paraffin and serially cut into a sequential series of 10 micrometer sections. Sections were throughout the intestinal tract of Cyp1b1-deficient Min mice and stained with H&E, h-catenin (27), phospho-InB kinase (IKK; Cell Signaling Min controls, as well as the effects of treating both groups of mice Technology; ref. 28), and p65 (RelA; Cell Signaling Technology; ref. 29), with DMBA or BP, which are similarly mutagenic through following the manufacturer’s protocol. A minimum of four tumors from dihydrodiol epoxides in vitro (22). We use two i.p. injections at each treatment was subjected to detailed pathologic analysis. times that correspond to ongoing intestinal development. This In situ hybridization. Paraffin sections of intestinal tissue were mode of administration provides a sustained systemic exposure hybridized with either a sense or antisense 35S-radiolabeled Cyp1a1 and that can be compared with the previous studies. The reported Cyp1b1 riboprobe, as described previously (30). The Cyp1a1 probe spanned experiments provide evidence for opposing effects of Cyp1b1 on nucleotides 5562 to 6212 and Cyp1b1 spanned nucleotides 752 to 1489. After intestinal tumorigenesis, derived from the metabolism of endog- development, sections were counterstained with propidium iodide and enous substrates and the activation of PAH. The role of images were documented under dark-field and epifluorescent illumination. Using this method, strong expression of Cyp1a1 was observed in the liver endogenous substrates is particularly notable, owing to our recent and lung of DMBA-treated mice, as described previously (30). finding that a lack of Cyp1b1 also substantially affects adiposity3 and the manner in which fatty acid metabolism affects intestinal tumorigenesis in the Min mouse (4–6). A simple model is Results developed to account for these paradoxical findings that involves CYP1B1 is overexpressed in many types of neoplasms, including opposing contributions of Cyp1b1 to the tumorigenic process human colorectal tumors. We therefore sought to test whether through metabolism of the PAH and endogenous substrates. Cyp1b1 activity contributes to the development of intestinal tumors in Min mice. Cyp1b1-deficient Min mice and Min controls were sacrificed at 55 days of age and the intestinal tract was removed. Tumors were scored in three segments of the small intestine 3 M.C. Larsen and C.R. Jefcoate, unpublished data. corresponding to the first 4 cm nearest the stomach (proximal), www.aacrjournals.org 7395 Cancer Res 2008; 68: (18). September 15, 2008

Figure 1. A deficiency of Cyp1b1 increases tumorigenesis throughout the entire intestinal tract ofMin mice but leads to a suppression of tumorigenesis by DMBA and BP. Tumors were scored in the proximal, medial, and distal regions ofthe small intestine and the entire colon of Cyp1b1-deficient Min mice and Min controls (A–C) or the distal region ofCyp1b1-deficientmice treated with either oil, DMBA, or BP (D).

4 cm in the middle (medial), and the last 4 cm nearest the cecum Min tumors have been extensively examined in many laborato- (distal), as well as the entire colon. A lack of Cyp1b1 dramatically ries and are uniformly adenomas, which exhibit a relatively affected tumor multiplicity along the entire length of the intestinal uniform progression (4, 5, 7) that was seen here in each treatment tract in Min mice (Fig. 1A–C; Table 1). Cyp1b1-deficient Min group. We used antibody staining for h-catenin (Fig. 3, left), which developed on average 37 F 18intestinal tumors, whereas Min is increased by Apc LOH, and for phospho-IKK (Fig. 3, middle) controls developed on average 15 F 9. This difference is statistically and RelA (Fig. 3B, right), which each mediate nuclear factor-nB significant (P = 0.02, two-sided Wilcoxon analysis). Thus, a lack of (NF-nB) activation (28), a key participant in intestinal cancer (31). Cyp1b1 metabolism of endogenous substrates clearly enhances the h-Catenin is also a component of E-cadherin cell-cell adhesion early stages of tumorigenesis. Tumors in Cyp1b1-deficient Min mice complexes, and strong interepithelial h-catenin complexes are were similar in size to those in Min controls, either at 55 or 85 days prevalent in the luminal epithelia of normal villi and also on the of age (Fig. 2A). Many tumors from 85-day-old mice exceeded 1 mm surface of the tumors (brown laddering feature, B). h-Catenin is À À in maximum diameter, a size that requires angiogenesis (5). Thus, a low within these day 55 Min tumors but increases in Cyp1b1 / lack of Cyp1b1 activity does not affect tumor expansion. tumors (A). There was a general trend for increased levels of

À À Table 1. Tumor multiplicity in ApcMin/+ and ApcMin/+ Cyp1b1 / mice

Genotype Treatment No. mice Mean tumor count Total tumor count (mean F SD)

Proximal SI Medial SI Distal SI Colon

ApcMin/+ Oil 9 3 4 7 2 15 F 10 DMBA 10 3 7* 12* 1 23 F 11 BP 10 3 4 82 17 F 11 À À c ApcMin/+ Cyp1b1 / Oil 86 11 183 37 F 18 b b DMBA 83 7 12 224F 8 BP 6 5 86 1 20 F 8

NOTE: Mice received PAH injections (0.1 mg each) on days 28and 35 of age and were sacrificed at 55 d of age. Abbreviation: SI, small intestine. *Significantly different from oil-treated ApcMin/+ mice (P = 0.03). cSignificantly different from oil-treated ApcMin/+ mice (P = 0.02). À À bSignificantly different from oil-treated ApcMin/+ Cyp1b1 / mice (P < 0.05).

Cancer Res 2008; 68: (18). September 15, 2008 7396 www.aacrjournals.org

Figure 2. The size ofintestinal tumors is increased by treating Min mice with DMBA. The maximum diameter was determined for tumors in the distal region ofthe small intestine fromseveral different genetic backgrounds and treatment groups. A lack of Cyp1b1 activity did not affect tumor size in Min mice at 55 and 85 d ofage ( A), whereas treatment with DMBA (D) increased tumor size in Min mice at 55 d ofage, relative to oil-treated controls ( C), with the significant percentage having a maximum diameter between 0.5 and 1 mm (B and C). E, DMBA; n, BP; y, oil-treated control.

À À phospho-IKK in the day 55 Cyp1b1 / tumors, both in the normal and C) of intestinal tumors even within a mere 20 days of the luminal borders and tumor epithelia [Fig. 3A, middle and right second administration. DMBA stimulated twice as many tumors in (enlarged insets)]. Surprisingly, RelA, which is released from NF-nB the medial and distal regions of the small intestine as oil-treated complexes by IKK, was low in the border villi epithelia, irrespective Min animals, wild-type for Cyp1b1 (Min controls; 19 F 10 versus of Cyp1b1 status (B). We observed abundant punctuate, nuclear 11 F 9), but was ineffective in the proximal segment. The RelA in the vascular region of the normal villi (data not shown). difference is statistically significant (P = 0.03, one-sided Wilcoxon We next tested whether Cyp1b1 metabolism of DMBA affects analysis). DMBA doubled the average maximum tumor diameter at intestinal tumorigenesis in Min mice. Moser and colleagues (26) 55 days of age and greatly increased the proportion of large tumors found that a high dose of DMBA was toxic to Min mice, (Fig. 2B and C). In addition, more than 20% of the tumors in presumably because of the high tumor burden in the intestine. We DMBA-treated Min mice exceeded 1 mm in maximum diameter, determined the toxicity of DMBA across a range of doses at days whereas none reached this size in the Min controls (P < 0.0001). 28and 35 of age. When mice were administered a dose of 0.1 mg When the same total amount of DMBA (0.2 mg) was administered DMBA at 28and 35 days of age, the majority of Min mice (10 of as a single dose on day 28, the number and size of tumors in the 11) survived to 55 days of age, but typically failed to survive most responsive distal region were the same as in the control- beyond 75 days, even with lower doses (0.02 mg/kg). The mice treated Min mice (Fig. 2B). Consequently, a second DMBA exposure typically exhibited extensive peritoneal inflammation. Importantly, seems to be an essential part of the carcinogenesis process. Min all Cyp1b1-deficient Min mice survived to 100 days of age, mice similarly treated with BP showed no changes in multiplicity regardless of treatment. (Table 1) or size (Fig. 2C) of the tumors. BP and DMBA are similarly Treatment with DMBA (0.1.mg, days 28and 35) substantially toxic to a variety of cells in vitro. BP, however, is similarly less toxic affected both the multiplicity (Fig. 1D; Table 1) and size (Fig. 2B than DMBA in bone marrow (30). www.aacrjournals.org 7397 Cancer Res 2008; 68: (18). September 15, 2008

Figure 3. Effects of Cyp1b1 deletion and DMBA treatment on the histology of Min tumors at days 55 and 130. Mice were sacrificed at either 55 d of age (A and B)or 130 d ofage ( C). Tumors were isolated from day 55 oil-treated Min and Cyp1b1-deficient Min control mice (A), day 55 DMBA-treated Min and Cyp1b1-deficient Min mice (B), and 130-d-old DMBA-treated Cyp1b1-deficient Min mice (C). Sections were stained with H&E (column 3 insets)orh-catenin (left), phospho-IKK (p-IKK; middle), or RelA (right inset; each dark brown), as indicated. A,right, phospho-IKK staining is shown in the enlarged regions (4Â; boxes shown). B,right, 4Â enlargements are shown for boxed regions a to c stained with anti-p65, anti-h-catenin, anti-phospho-IKK, and H&E, as indicated. C,bottom row, 4Â enlargement of the anti-h-catenin staining (left, box shown), an antibody-deleted control (middle), and an H&E-stained section (right). B, anti-p65 (RelA) staining ofDMBA-treated ApcMin/+ (top) and DMBA-treated ApcMin/+ Cyp1b1À/À (bottom). h-Catenin–stained luminal surface epithelia as a laddering of cell junctional E-cadherin/h-catenin complexes (B,left images ofDMBA-treated Apc Min/+ and ApcMin/+ Cyp1b1À/À tissue). Typically, the proportion of h-catenin–stained cells within tumors (indicative of Apc-LOH) increased with tumor enlargement but were also higher in tumors ofsimilar size fromCyp1b1 À/À Min mice. DMBA treatment ofMin mice produced many enlarged tumors at day 55 (B) that were typical of Min tumors seen at about day 100. Tumors from DMBA treatment of Cyp1b1-deficent Min mice, again, were mostly typical ofnormal Min tumors at day 55 ( B). Tumors from DMBA treatment of Min mice showed strong staining for phospho-IKK, h-catenin, RelA, and hematoxylin in about a third ofthe crypts, although not specificallyin the tumors and with only partial overlap ofthese responses. The larger tumors seen at day 130 commonly showed regions ofdisorganized growth, exhibiting nuclear atypia, intense h-catenin staining with common nuclear localization (enlarged in C), and absence of phospho-IKK. These atypia regions were also seen in tumors from these Cyp1b1-deficient mice without DMBA treatment (data not shown) but notinthe equivalent day 55 tumors or in Cyp1b1+/+ Min mice.

Cancer Res 2008; 68: (18). September 15, 2008 7398 www.aacrjournals.org

We compared the effect of DMBA on the multiplicity and size RelA. These activated crypts were present outside the tumors. They distribution of intestinal tumors in Cyp1b1-deficient Min mice. were greatly diminished when Cyp1b1 was deleted (B), even at DMBA-treated Cyp1b1-deficient mice actually developed signifi- 130 days (C). The increases in IKK phosphorylation and RelA and cantly fewer tumors in the medial and distal region than vehicle- h-catenin expression only partially overlapped. This process may treated mice (19 F 8versus 29 F 15 tumors; Fig. 1A and B; Table 1). be related to the formation of aberrant crypt foci, which have The size of the tumors in the distal region was also unchanged been recognized as clusters of hematoxylin-stained and methylene (diameter, 0.54 F 0.23 mm versus 0.63 F 0.32 mm; Fig. 2B). blue–stained crypts in the colon of chemically treated mice and Together, these observations indicate that the stimulatory effect are implicated as an early step toward tumor formation (33). of DMBA on intestinal tumor size in Min mice is eliminated At day 55, the punctate nuclear h-catenin was most visible when Cyp1b1 activity is removed. The appreciable decrease in following DMBA treatment, consistent with the increased tumor tumor multiplicity produced by DMBA in Cyp1b1-deficient Min growth (B, right inset, box a), but was particularly evident in the À À mice indicates a strong DMBA suppression effect in the absence disorganized atypia of the 130-day DMBA-treated Cyp1b1 / Min of CYP1B1 metabolism. There was no net effect of Cyp1b1 mice (C). This loss of cell organization, which was particularly deficiency in the presence of DMBA (Table 1, line 2 versus line 5). evident with H&E staining (C, right), was also seen in equivalent À À This arises because the effect of Cyp1b1 deficiency on endogenous tumors from untreated Cyp1b1 / mice (data not shown) but regulation (line 4 versus line 1) is opposite to the effect of Cyp1b1 was not seen when Cyp1b1 was present. The enhanced h-catenin deficiency on DMBA activity (line 5 versus line 4). DMBA, in the staining seen at day 55 tumors with Cyp1b1 deletion (B) absence of metabolism by Cyp1b1, seems to prevent the stimulation corresponded mostly to epithelial adherens complexes with little produced by loss of endogenous metabolism. An equivalent BP punctate nuclear staining. À À treatment did not affect Min tumorigenesis, but, surprisingly, BP Unlike the tumor epithelia of day 55 Cyp1b1 / min mice, the was at least as effective as DMBA in decreasing tumor multiplicity atypia seen after 130 days were completely devoid of phospho-IKK. in Cyp1b1-deficient Min mice (Fig. 1D; Table 1). This suppression Low phospho-IKK seemed to be typical of the larger Min tumors, by DMBA and BP may derive from a tumor suppression effect irrespective of conditions (Fig. 3C). Surprisingly, RelA expression resulting from activation of the AhR, which is appreciable in PAH- was low in the border epithelia, although they exhibited high treated mouse colon and small intestine (32). BP is a much more phospho-IKK (Fig. 3B). Although the typical IKK phosphorylation effective activator of AhR than DMBA (30). of the inhibitory regulator (InB) releases RelA to the nucleus, h-Catenin staining was elevated in the larger Min tumors phosphorylation by IKKa leads to proteolysis of RelA (31). The produced by the DMBA treatment, but not when Cyp1b1 was antibody recognizes both IKKa/h forms. By contrast, atypia deleted, thus paralleling tumor numbers and size (Fig. 3B and C). expressed robust RelA expression, although rarely with punctate h-Catenin was strikingly evident in about a third of the crypts after nuclear distribution (data not shown). DMBA treatment (B, boxes b and c), which also stained more We examined whether Cyp1b1 deletion had a systemic effect on strongly with hematoxylin. These distinctive crypts, which were DMBA and its metabolism. In Cyp1b1-deficient Min mice, the blood present in clusters, likewise stained strongly for phospho-IKK and serum concentrations of DMBA and the active 3,4-dihydrodiol

Figure 4. Intestinal tumors from Min mice exhibit elevated expression of Cyp1b1 and Cyp26 forms relative to surrounding tissues, but no Cyp1a1. Tumor sections were hybridized with antisense or sense probes against Cyp1b1 (A and B) or Cyp1a1 (B). Microarray analyses ofthree Min tumors relative to normal intestinal tissue. The arrays confirm increased expression of Cyp1b1 and also show increases in Cyp26a1 (two of three tumors) and Cyp26b1.

www.aacrjournals.org 7399 Cancer Res 2008; 68: (18). September 15, 2008

Figure 5. A model for participation of Cyp1b1 metabolism ofboth endogenous substrates and PAHs in Min tumorigenesis. Increased NF-nB activation increases tumorigenesis through antiapoptotic and proinflammatory effects. By contrast, Cyp1b1 activates PAHs to mutagenic dihydrodiol epoxides. In the absence ofCyp1b1, PAHs become elevated in the intestine and exhibit enhanced AhR activity. This suppresses the NF-nB activity, thus reversing the effect ofendogenous reactive oxygen species ( ROS), which can also increase in Cyp1b1-deficient cells. Cyp1b1 may, additionally, suppress tumorigenesis by converting retinaldehyde to retinoic acid, an active Min tumor suppressor (36).

metabolite were 6.3 F 3.1 pmol/mL and 2.1 F 0.9 pmol/mL, Cyp1b1 mediates a rapid stimulation of these adenomas by DMBA respectively, compared with 5.3 F 2.1 and 2.9 F 1.0 in the wild-type within 3 weeks at exposure levels that are only moderately higher mice. Because Cyp1b1 deletion does not affect circulating DMBA or than can be obtained from environmental PAH exposures (37). We metabolites, the effect on DMBA-induced tumorigenesis derives report a large difference according to the PAH (DMBAJBP) that from enhanced metabolism in the small intestine. parallels differences that we have characterized for bone marrow CYP1B1 is expressed at a very low level in the normal intes- toxicity (30). tinal epithelium of humans but is elevated in tumors (15). Cyp1b1 Cyp1b1 metabolism of endogenous substrates can limit tumori- expression was very low in microsomes isolated from normal genesis by either generating a suppressor or removing an acti- mouse intestine and was only detected after high exposures to PAHs vator. Participation of Cyp1b1 in the generation of retinoic acid, (34, 35). In situ hybridization shows that Cyp1b1 expression was a known tumor suppressor, plays a conserved functional role in clearly elevated in intestinal tumors relative to adjacent normal early embryogenesis as a contributor to retinoic acid synthesis tissue, particularly in the region of the crypts and associated stroma (24). Cyp1b1 is expressed at substantially higher levels in intestinal (Fig. 4A). By contrast, Cyp1a1 was not expressed (Fig. 4B). Previous tumors than in surrounding normal tissue (Fig. 4), notably in the work has shown the expression of multiple cytochrome P450s in region that is associated with stroma and the crypts. Cyp26b1 and, the mouse intestine (35). Microarray analysis of mRNA isolated to a lesser extent, Cyp26a1, which specifically remove retinoic from three Min tumors and their adjacent epithelia confirmed acid, are also overexpressed in Min tumors (Fig. 4). Thus, deletion that Cyp1b1 expression was substantially elevated in each tumor of the Cyp1b1 may cooperate with enhanced Cyp26 expression (Fig. 4C). These changes were very selective because only 3 other to lower retinoic acid restraints on early tumor development. cytochrome P450s among over 30 represented on the array Human colonic tumors have elevated CYP3A4, CYP2S1, CYP2U1, increased. Two additional cytochrome P450s, Cyp2b9 and Cyp2b10, and CYP51 and also express CYP26A1 (13). Suppression of local which are commonly found in liver, increased 5- to 10-fold (data retinoic acid, including increases in Cyp26a1 mRNA in Min tumors not shown). Importantly, Cyp26b1, which removes retinoic acid in in human FAP adenomas, consistent with h-catenin regulation, embryos, was substantially elevated in all three tumors. The highly has previously been shown (36). Human CYP1B1 also metabolizes expressed Cyp26a1 was elevated in two of three tumors, consistent estradiol to genotoxic quinones but much more effectively than with modest increases previously reported for Min tumors (36). rodent Cyp1b1 (16, 38). The estradiol in these female Min mice appreciably suppresses Min tumors (39), and therefore, any elevation by removal of Cyp1b1 should lower tumorigenesis. Discussion Min tumors and human colonic tumors each exhibit high We have shown that removal of Cyp1b1 more than doubles clusterin expression, consistent with activation of NF-nB (27). intestinal tumorigenesis by 55 days of age in the Min mouse cancer Activation of the NF-nB complex by IKKh releases dimeric NF-nB model (Fig. 1; Table 1) but with only minor effects on tumor size, transcription factors, including RelA, which generate antiapoptotic histology, or distribution (Figs. 2 and 3; Table 1). This exclusive and proinflammatory transcripts that enhance colon tumorigenesis À À effect on tumor multiplicity indicates that the metabolism of (29, 31, 40). In Cyp1b1 / Min small intestines, phospho-IKK endogenous substrates by Cyp1b1 suppresses early stages of increased modestly in the normal villi and tumor epithelia (Fig. 3A intestinal tumorigenesis. This expands on previous work, which and B), consistent with NF-nB activation. RelA also increased but has shown overexpression of Cyp1b1 in many tumors (12), without specific nuclear localization associated with IKK activa- including human colon cancers (13). Paradoxically, we show that tion. A similar constitutive activation of NF-nB in the lungs of

Cancer Res 2008; 68: (18). September 15, 2008 7400 www.aacrjournals.org

Cyp1b1-deficient mice could explain the elevation of DMBA- ing, or mutation through DNA damage (49). The spontaneous LOH induced lung tumors in these mice (21). for Apc occurs through somatic recombination in these C57BL6/J Linoleic and arachidonic oxygenation products are plausible Min mice and is much higher in the distal intestine, whereas Cyp1b1 substrates. Thus, they modulate NF-nB activities, including tumorigenesis from allelic silencing predominates at the ileocecal inflammatory recruitment of macrophage (40), and can be junction (49). The increased multiplicity produced by Cyp1b1 activators (20-HETE and leukotriene B4; ref. 41). Prostaglandins deletion occurs equally in each region, suggesting enhanced formed by Cox1 and Cox2 also stimulate Min tumorigenesis by somatic recombination. DMBA, however, is most effective in the mediating angiogenesis when tumors exceed a critical size of 1 mm distal segment (Table 1). Cyp1b1 is preferentially expressed in the (5). However, Cyp1b1 deletion increased Min tumor initiation small intestine relative to the colon (35). rather than enlargement. Cyp1b1 deletion does not affect circulating levels of PAHs or Loss of Apc activates the Min tumorigenesis process by their dihydrodiol metabolites. The net effect of Cyp1b1 on enhancing h-catenin signaling. This results in enhanced expansion tumorigenesis in the mouse intestine should, therefore, depend of a mutated stem cell population at the bottom of the crypts. on local metabolism of PAHs or the dihydrodiol precursors by Adenomas in Min mice chimeric for ROSA26 expression show Cyp1b1 (Fig. 5). The elevation of expression of Cyp1b1 seen here distinct transformation events in adjacent crypts, suggesting that (Fig. 4) and in many human tumors, including in the colon (13, 14), initiation in one crypt facilitates initiation in an adjacent crypt (42). may occur early in the transformation process. This is likely Stimulation of NF-nB activity commonly increases the release of because DMBA metabolism is completed in a moderately short chemokines, which may enhance this intercrypt proximity effect. time period after each administration. By contrast, the effects of Adenomas also expand through a crypt fission process, which Cyp1b1 on the endogenous substrates are continuously present starts when stem cells reach a threshold level (43). This process is during the tumor development. Single i.p. doses of DMBA (but not enhanced by loss of Apc activity and by damage to the crypts, BP) produce major changes in bone marrow hematopoiesis that including the effects of reactive chemicals. are eventually transmitted to other sites (30). Importantly, these The metabolism of DMBA by Cyp1b1 doubles the number and cells derived from bone marrow migrate to the intestine and size of intestinal tumors in Min mice in a mere 20 days, as participate in tumorigenesis (40). The finding that repeat evidenced by the reversal with Cyp1b1 deletion (Fig. 2B; Table 1). administration is necessary is consistent with previous studies BP was ineffective despite effective activation in vitro (17, 22). We with AOM and ENU (7, 10). This is consistent with a model in have reported this difference for Cyp1b1-mediated bone marrow which an activated crypt, such as may be indicated by phospho- toxicity, apparently due to activation of AhR-dependent protective IKK and RelA staining, facilitates activation of an adjacent crypt. mechanisms by this potent AhR ligand. The stimulation was This clustering of transformed crypts was evident in the histology selective to the medial and distal segments and accompanied by a following DMBA treatment. novel large stimulation of h-catenin, RelA, and phospho-IKK in Although a lack of Cyp1b1 activity increased the number of about a third of intestinal crypts, including outside the tumors tumors in Min mice, these mice persisted beyond 130 days of age. (Fig. 3). This crypt activation may also be associated with the The atypia that are evident in these large Cyp1b1-deficient tumors generation of aberrant crypt foci, an early step in colon cancer indicate that Cyp1b1 additionally affects tumor progression (Fig. 3). that can derive from chemical activation of crypt fission (33, 43). These regions of atypia strongly express nuclear h-catenin but not The staining of adjacent crypts (Fig. 3B, enlargements) may arise the intermembrane adhesion complexes typical of organized in this way. epithelia (39). The absence of phospho-IKK contrasts with the Tumor multiplicities in mice lacking Cyp1b1 were actually elevation in early tumors associated with Cyp1b1 deletion (Fig. 3) decreasedbyDMBAandstillmorebyBP(Fig.1D). This but also reflects a trend in other larger tumors. The tumor paradoxical reversal of the Cyp1b1-deficient phenotype by DMBA enhancement provided by Cyp1b1 deletion has no noticeable effect and BP can be captured by a simple model in which Cyp1b1 on mortality despite many very large tumors. This altered deficiency not only increases endogenous chemicals that activate development of Min tumors in the absence of Cyp1b1, shown by NF-nB transcription (shown in Fig. 5) but uncovers an opposing prevalent atypia, may contribute to particularly benign tumors. suppression through stimulation of AhR by PAHs that can The link established here between Cyp1b1 and tumorigenesis is predominate in the absence of their metabolic activation. AhR is particularly relevant to human colon cancer (12). CYP1B1, in substantially activated in the mouse colon and small intestine by combination with nuclear h-catenin, provides an effective bio- PAHs (32), and BP is a much more effective activator than DMBA marker to screen for colorectal carcinoma (15), and the common (30). AhR inhibition is realized through direct interaction with the allelic variants may represent risk factors (13, 14). Rel components of the NF-nB transcription complex and has been reported for lung responses to cigarette condensate in vivo (44). Disclosure of Potential Conflicts of Interest AhR activity in the crypts of mouse small intestine is also shown by the AhR-Cre recombinase method for Apc deletion (45, 46). The No potential conflicts of interest were disclosed. procarcinogenic effect produced by PAHs may, therefore, depend on the balance between metabolite activation (high for DMBA) and Acknowledgments AhR suppression (high for BP). The absence of alfalfa and soybean Received 12/19/2007; revised 6/27/2008; accepted 7/9/2008. meal from the mouse diet thereby excludes phytochemicals, such Grant support: NIH grants R01 CA016265 (C.R. Jefcoate), R01 CA87609 (C.R. Jefcoate), and University of Wisconsin Clinical Cancer Center Grant 5P30-CA014520-34 as flavonoids (47) and indole derivatives, which are potential (Director, G. Wilding/C.R. Jefcoate, member). Cyp1b1 substrates (48) and AhR inducers (46). The costs of publication of this article were defrayed in part by the payment of page Cyp1b1 metabolism of endogenous substrates or PAHs may charges. This article must therefore be hereby marked advertisement in accordance with 18U.S.C. Section 1734 solely to indicate this fact. contribute to loss of the wild-type Apc allele, which arises by three We thank Dr. William Dove for his critical reading of this manuscript and Joe principal mechanisms: somatic recombination, epigenetic silenc- Hardin for his contribution to tumor histology. www.aacrjournals.org 7401 Cancer Res 2008; 68: (18). September 15, 2008

References cytochrome P-450EF, representative of a new 1B subfamily azoxymethane-induced colon carcinogenesis. Carcino- of cytochrome P-450s. Cloning, sequence determination, genesis 2000;21:1567–72. 1. Fearon ER, Vogelstein B. A genetic model for and tissue expression. J Biol Chem 1994;269:14905–11. 34. Zhang QY, Dunbar D, Kaminsky LS. Characterization colorectal tumorigenesis. Cell 1990;61:759–67. 18. Sutter TR, Yang YM, Hayes CL, et al. Complete cDNA of mouse small intestinal cytochrome P450 expression. 2. Clarke AR. Wnt signalling in the mouse intestine. sequence of a human dioxin-inducible mRNA identifies Drug Metab Dispos 2003;31:1346–51. Oncogene 2006;25:7512–21. a new gene subfamily of cytochrome P450 that maps to 35. Uno S, Dragin N, Miller ML, et al. Basal and inducible 3. Kaiser S, Park YK, Franklin JL, et al. Transcriptional chromosome 2. J Biol Chem 1994;269:13092–9. CYP1 mRNA quantitation and protein localization recapitulation and subversion of embryonic colon 19. Tang YM, Wo YY, Stewart J, et al. Isolation and throughout the mouse gastrointestinal tract. Free Radic development by mouse colon tumor models and human characterization of the human cytochrome P450 Biol Med 2008;44:570–83. colon cancer. Genome Biol 2007;8:R131. CYP1B1 gene. J Biol Chem 1996;271:28324–30. 36. Shelton DN, Sandoval IT, Eisinger A, et al. Up- 4. Cormier RT, Hong KH, Halberg RB, et al. Secretory 20. Tsuchiya Y, Nakajima M, Takagi S, et al. MicroRNA regulation of CYP26A1 in adenomatous polyposis coli- phospholipase Pla2g2a confers resistance to intestinal regulates the expression of human cytochrome P450 deficient vertebrates via a Wnt-dependent mechanism: tumorigenesis. Nat Genet 1997;17:88–91. 1B1. Cancer Res 2006;66:9090–8. implications for intestinal cell differentiation and colon 5. Takeda H, Sonoshita M, Oshima H, et al. Cooperation 21. Buters J, Quintanilla-Martinez J, Schober W, et al. tumor development. Cancer Res 2006;66:7571–7. of cyclooxygenase 1 and cyclooxygenase 2 in intestinal CYP1B1 determines susceptibility to low doses of 7,12- 37. Phillips DH. Polycyclic aromatic hydrocarbons in the polyposis. Cancer Res 2003;63:4872–7. y dimethylbenz[a]anthracene-induced ovarian cancers in diet. Mutat Res 1999;443:139–47. 6. Reed KR, Sansom OJ, Hayes AJ, et al. PPAR status and mice: correlation of CYP1B1-mediated DNA adducts 38. Rahman M, Hayes Sutter C, Emmert GL, et al. Apc-mediated tumourigenesis in the mouse intestine. with carcinogenicity. Carcinogenesis 2003;24:327–34. Regioselective 2-hydroxylation of 17h-estradiol by rat Oncogene 2004;23:8992–6. 22. Shimada T, Murayama N, Okada K, et al. Different cytochrome P4501B1. Toxicol Appl Pharmacol 2006;216: 7. Shoemaker AR, Moser AR, Dove WF. N-ethyl-N- mechanisms for inhibition of human cytochromes P450 469–78. nitrosourea treatment of multiple intestinal neoplasia 1A1, 1A2, and 1B1 by polycyclic aromatic inhibitors. 39. Javid SH, Moran AE, Carothers AM, et al. Modulation (Min) mice: age-related effects on the formation of Chem Res Toxicol 2007;20:489–96. of tumor formation and intestinal cell migration by intestinal adenomas, cystic crypts, and epidermoid 23. Choudhary D, Jansson I, Sarfarazi M, et al. Physio- estrogens in the ApcMin/+ mouse model of colorectal cysts. Cancer Res 1995;55:4479–85. logical significance and expression of P450s in the cancer. Carcinogenesis 2005;26:587–95. 8. Andreassen A, Vikse R, Mikalsen A, et al. 2-Amino-1- developing eye. Drug Metab Rev 2006;38:337–52. 40. Greten FR, Eckman L, Greten TF, et al. IKKh links methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) induces 24. Chambers D, Wilson L, Maden M, et al. RALDH- inflammation and tumorigenesis in a mouse model of genetic changes in murine intestinal tumours and cells independent generation of retinoic acid during verte- colitis-associated cancer. Cell 2004;118:285–96. with ApcMin mutation. Mutat Res 2006;604:60–70. brate embryogenesis by CYP1B1. Development 2007;134: 41. Pidgeon GP, Lysaght J, Krishnamoorthy S, et al. 9. Mollersen L, Paulsen JE, Alexander J. Loss of 1369–83. Lipoxygenase metabolism: roles in tumor progression heterozygosity and nonsense mutation in Apc in 25. Libby RT, Smith RS, Savinova OV, et al. Modification and survival. Cancer Metastasis Rev 2007;26:503–24. azoxymethane-induced colonic tumours in min mice. of ocular defects in mouse developmental glaucoma 42. Thliveris AT, Halberg RB, Clipson L, et al. Polyclon- Anticancer Res 2004;24:2595–9. models by tyrosinase. Science 2003;299:1578–81. ality of familial murine adenomas: analysis of mouse 10. Sohn OS, Faila ES, Requeijo SP, et al. Differential 26. Moser AR, Mattes EM, Dove WF, et al. ApcMin,a chimeras with low tumor multiplicity suggest short- effects of CYP2E1 status on the metabolic activation of mutation in the murine Apc gene, predisposes to range interactions. Proc Natl Acad Sci U S A 2005;102: the colon carcinogens azoxymethane and methylazox- mammary carcinomas and focal alveolar hyperplasias. 6960–5. ymethanol. Cancer Res 2001;61:8435–40. Proc Natl Acad Sci U S A 1993;90:8977–81. 43. Yen TH, Wright NA. The gastrointestinal tract stem 11. Kleiner HE, Vulimiri SV, Hatten WB, et al. Role of 27. Chen X, Halberg RB, Ehrhardt WM, et al. Clusterin as cell niche. Stem Cell Rev 2006;2:203–12. cytochrome p4501 family members in the metabolic a biomarker in murine and human intestinal neoplasia. 44. Thatcher TH, Sanjay BM, Baglole CJ, et al. Aryl activation of polycyclic aromatic hydrocarbons in Proc Natl Acad Sci U S A 2003;100:9530–5. hydrocarbon receptor-deficient mice develop height- mouse epidermis. Chem Res Toxicol 2004;17:1667–74. 28. Shibata W, Maeda S, Hikiba Y, et al. The InB kinase ened inflammatory responses to cigarette smoke and 12. Murray GI, Taylor MC, McFadyen MCE, et al. Tumor- (IKK) inhibitor, NEMO-binding domain peptide, blocks endotoxin associated with rapid loss of the nuclear specific expression of cytochrome P450 CYP1B1. Cancer inflammatory injury in murine colitis. J Immunol 2007; factor-nB component RelB. Am J Pathol 2007;170: Res 1997;57:3026–31. 179:2681–5. 855–64. 13. Kumarakulasingham M, Rooney PH, Dundas SR, 29. Steinbrecher KA, Harmel-Laws E, Sitcheran R, et al. 45. Sansom OJ, Reed KR, Hayes AJ, et al. Loss of Apc et al. Cytochrome P450 profile of colorectal cancer: Loss of epithelial RelA results in deregulated intestinal in vivo immediately perturbs Wnt signaling, differenti- identification of markers of prognosis. Clin Cancer Res proliferative/apoptotic homeostasis and susceptibility ation, and migration. Genes Dev 2004;18:1385–90. 2005;11:3758–65. to inflammation. J Immunol 2008;180:2588–99. 46. Ito S, Chen C, Satoh J, et al. Dietary phytochemicals 14. SissungTM,PriceDK,SparreboomA,etal. 30. Galvan N, Teske DE, Zhou G, et al. Induction of regulate whole-body CYP1A1 expression through an Pharmacogenetics and regulation of human cytochrome CYP1A1 and CYP1B1 in liver and lung by benzo(a)py- arylhydrocarbon receptor nuclear translocator-depen- P450 1B1: implications in hormone-mediated tumor yrene and 7,12-dimethylbenz[a]anthracene do not affect dent system in gut. J Clin Invest 2007;117:1940–50. metabolism and a novel target for therapeutic inter- distribution of polycyclic hydrocarbons to target tissue: 47. Chang TKH, Chen J, Yeung EYH. Effect of Ginkgo vention. Mol Cancer Res 2006;4:135–50. role of AhR and CYP1B1 in bone marrow cytotoxicity. biloba extract on procarcinogen-bioactivating human 15. ChangH,SuJM,HuangCC,etal.Usinga Toxicol Appl Pharmacol 2005;202:244–57. CYP1 enzymes: identification of isorhamnetin, kaemp- combination of cytochrome P450 1B1 and h-catenin 31. Karin M, Greten FR. NF-nB: linking inflammation ferol, and quercetin as potent inhibitors of CYP1B1. for early diagnosis and prevention of colorectal cancer. and immunity to cancer development and progression. Toxicol Appl Pharmacol 2006;213:18–26. Cancer Detect Prev 2005;29:562–9. Nat Rev Immunol 2005;5:749–59. 48. Prince M, Campbell CT, Robertson TA, et al. 16. Savas U, Carstens CP, Jefcoate CR. Recombinant 32. Campbell SJ, Carlotti F, Hall PA, et al. Regulation of Naturally occurring coumarins inhibit 7,12-dimethyl- mouse Cyp1b1 expressed in Escherichia coli exhibits the CYP1A1 promoter in transgenic mice: an exquisitely benz[a]anthracene DNA adduct formation in mouse selective binding by polycyclic hydrocarbons and sensitive on-off system for cell specific gene regulation. mammary gland. Carcinogenesis 2006;6:1204–13. metabolism which parallels C3H10T1/2 cell micro- J Cell Sci 1996;109:2619–25. 49. Haigis KM, Hoff PD, White A, et al. Tumor somes, but differs from human recombinant Cyp1b1. 33. Papanikolaou A, Wang QS, Papanikolaou D, et al. regionality in the mouse intestine reflects the Arch Biochem Biophys 1997;347:181–92. Sequential and morphological analyses of aberrant crypt mechanism of loss of Apc function. Proc Natl Acad 17. Savas U, Bhattacharyya KK, Christou M, et al. Mouse foci formation in mice of differing susceptibility to Sci U S A 2004;101:9769–73.

Cancer Res 2008; 68: (18). September 15, 2008 7402 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2008 American Association for Cancer Research. Cyp1b1 Exerts Opposing Effects on Intestinal Tumorigenesis via Exogenous and Endogenous Substrates

Richard B. Halberg, Michele Campaigne Larsen, Tammy L. Elmergreen, et al.

Cancer Res 2008;68:7394-7402.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/68/18/7394

Cited articles This article cites 49 articles, 22 of which you can access for free at: http://cancerres.aacrjournals.org/content/68/18/7394.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/68/18/7394.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/68/18/7394. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.