Document wD3KZBxnN9RgqVqNJyGJrdN3E

f- -..- . BRIEF COMMUNICATION Relationship Between p53 Mutations and Inducible Nitric Oxide Synthase Expression in Human Colorectal Cancer Stefan Ambs, William P. Bennett, William G. Merriam, Mofolusara 0. Ogunfusika, Sean M. Oser, Anita M. Harrington, Peter G. Shields, Emanuela Felley-Bosco, S. Perwez Hussain, Curtis C. Harris Inducible Ca*'-independent nitric oxide synthase (NOS), also referred to as NOS2, which is expressed in a variety of human cancers ( 1 4 ) , can generate mutagenic concentrationsof nitric oxide (NO) in mice (5). NOS2 is the most active isoform among the three known nitric oxide synthases (6),which also include the neuronal (NOS 1) and endothelial (NOS3) isoforms. Only NOS2 is capable of producing sustained NO concentrations in the micromolar range (7). We investigated the hypothesis that NO generated by NOS2 is capable of inducing mutations in the p53 (also known as TP53) gene and contributesto human colon carcinogenesis. We analyzed 118 sporadic colon tumors for NOS2 expression and p53 gene mutations. Colon tumors and surrounding normal tissues were collected from the Cooperative Human Tissue Network highest average NOS2 activity. The and the Department of Pathology, Uni- NOS2 activity declined with advancing versity of Baltimore, with the approval tumor stage and was Seen at the lowest of local boards governing research on level in metastatic tumors (Fig. 1, A). human subjects, as described previously NOS2 activity correlated with NOS2 (3). The expression of NOS2 was in- protein expression. Immunohistochemicreased in various tumors (Fig. 1, A and cal analysis localized NOS2 protein 'B) throughout the right, left, and sig- mainly in tumor-infiltrating mono- moid colon. Adenomas showed the nuclear cells and less frequently in en- A lo001 0 B Adenoma Normal 0 0 0 0 0 00 I Carcinoma P0 I0 0 8 I Metastasis 0 0 80 0 CA, 0000 0 --9- ,N = 6 8 N = 2 6 A N=l6 6- c Dukes N-27 N-30 D N=13 N=6 Afiliarions of authors: S . Ambs, W. P. Bennett, W. G. Meniam, M.0.Ogunfusika, S. M.Oser, A. M. Harrington, P.G. Shields, S. P. Hussain, C. C. Harris, Laboratory of Human Carcinogenesis, Division of Basic Sciences, National Cancer Institute, Bethesda, MD; E. Felley-Bosco, Institute of Pharmacology and Toxicology, University of Lausanne, Switzerland. Correspondence to: Curtis C. Harris, M.D., Na- tional Institutes of Health. Bldg. 37, Rm.2C01.37 Convent Dr. MSC 4255, Bethesda, MD 20892- 4255 (e-mail: Curtis-Harris@nih.gov). See "Notes" following "References." 1 0 Oxford University Press L `I Fig. 1. Inducible Ca"-independent nitric oxide synthase (NOS?) expression in human colon tumors. A) Ca*'-independent nitric oxide synthase (NOS) activity, characteristic of the type NOS2, is high in colon adenomas, while it is low in the surrounding normal colon tissues. The activity decreases with the progression of colorectal cancer (Dukes' stages A through D) and is lowest in colon carcinoma metastases in the liver and lung (adenomas: Pc.001 and n = 20; carcinomas: P = .Ol and n = 48; both versus surrounding normal colon tissues from the same patient analyzed by the Wilcoxon signed rank test for two-tailed, paired analysis). Details of the NOS2 assay, together with the NOS activities for a subset of the tissues, were reported recently ( 3 ) . B) Immunohistochemical analysis of colon tumors for NOS2 protein. In panels a and b, focal clusters of tumor cells (arrows) have cytoplasmic staining with brown chromogen indicating NOS2. In panel E, scanning magnification shows extensive staining of tumor interstitium. Panel d is a higher magnification detail of panel c that shows heavy staining of mononuclear cells (small arrows) and endothelial cells (large arrows). Hematoxylin counterstain; original magnifications: panel a, ~400p, anel b, x630; panel c, x50; panel d, x400. NOS2 immunohistochemistry was performed as described previously (3). 86 BRIEF COMMUNICATION Journal of the National Cancer Institute, Vol. 91, No. 1, January 6,1999 otklial cells within the tumors and in ;e tumor cells (Fig. 1, B). The decline I NOS2 activity with advancing tumor tage may be attributed to the tumor ell-induced immunosuppression of iOS2 expression in tumor-infiltrating iononuclear cells of advanced tumors 3). We then determined the p53 mutation requency and mutation type in relation I the NOS2 activity levels in colon tuiors. We confined the mutational nalysis to the evolutionarily conserved :gion in the p53 gene. This genomic :@on contains about 90% of the known 53 mutations and all of the mutational otspots at CpG dinucleotide sites 9,IO). We found 11 mutations among 6 adenomas (mutation frequency = 4.6%, with one tumor containing three iutations) and 44 mutations among 92 arcinomas of Dukes' stages A through ) (mutation frequency = 47.8%). None f the carcinomas had multiple mutaons. There were 44 missense mutaons, six nonsense mutations, two inserons, two deletions, and one inversion. 'he predominant mutation was the G:C to A:T transition at CpG dinucleotides (n = 34, mutation frequency = 61.8%), and a significant association between t!me transitions and increased NOS2 activity was observed when compared with tumors with other types of mutations, e.g., transversions and frameshift mutations (P = .004; Mann-Whitney U rank sum test) (see Fig. 2, A). Further analysis demonstrated a convincing dose-response relationship between NOS2 activity and G:C to A:T transitions at CpG dinucleotides in carcinomas (P = .003; Mantel-Haenszel test for trend) (Fig. 2, B); the rates of all other mutations varied inversely with NOS2 activity. Most p53 transition mutations in colorectal carcinoma occur at CpG dinucleotides that contain 5-methylcytosine (9,1O), and our data support the hypothesis that NOS2 activity generates the high frequency of G:C to A:T mutations at 5-methylcytosine sites in the p53 gene. The formation of deaminating NO intermediates through up-regulation of NOS2 has been documented ( l l ) ,and exposure of Salmonella typhimurium, plasmid DNA, and the p53 complemen tary DNA to NO donors generatel mostly G:C to A:T transitions (12,13, In addition, the increased formation o N-nitrosamines in activated macro phages (14) and Corynebacterium pa? vum-treated rats (15) indicates that au toxidation of endogenously produce1 NO leads to electrophilic and nitrosatin agents such as N,O, in vivo. Endogenous NO production alsj causes oxidative DNA damage (16) as result of stoichiometric fluxes of N( and superoxidethat generateperoxynitrit (17). Our observation that detectable ni trotyrosine formation is restricted tl only a subset of NOS2-expressing tu mor-infiltrating mononuclear cells i colon tumors (3)suggests that the NO tl peroxynitrite pathway is not a dominar pathway in adenomas. Peroxynitrite ha also been shown to cause mainly G:C tl T:A and G:C to C:G transversions (18, which does not match the p53 muta tional spectrum of colon tumors with predominance of G:C to A:T transitio mutations. However, NO quenches bot superoxide and an oxidizing intermedi -F 5 F c I t B 0 0 Q9 0 0 0 0 0 0 COOtOlXm Other (n-21) 0 00 0 x 8 000 8 00 0 0 -I I G:C to A:T at CpG (n=34) 45 1 I 3 4- " Low (n= 30) 7 p53mutations 0QEA;T: All other mutations Medium (n= 31) NOS2 activity Hmh (n = 31) t ig. 2. lnducible Ca"-independent n k i c oxide synthase (NOS2) activity and 53 mutations in human colon tumors. A) NOS2 activity is significantly higher human colon tumors with G:C to A:T transitions at CpG sites in the p53 :ne when compared with that in tumors with mutations at other sites in this :ne (P = .004; Mann-Whitney U rank sum test). The specimens include lenomas, adenomas with carcinoma ip rihc, and carcinomas of Dukes` stages through D.B)The fraquency of G:C to A:T transitions at CpG dinucleotide tes is positively correlated with NOS2 activity (P = .003: Mantel-Haenszel SI for trend). NOS2 activity and p53 mutation frequency and mutation type ere analyzed in 92 colw carcinomas of Dukes' stages A through D containg 44 mucrrions. The NOS2 activity of the 92 carcinomas was divided into rtiles of tbe distribution, described as low (undetectable NOS2 activity), edium (range from 0.1to 3.5 pmol citrullinelmin per milligram prolein as mC ghest NOS2 activity hmd in this group), and h g h (range from 3.7 pmd citrulline/min per milligram protein as the lowest NOS2 activity found in thi group to 48.8 pmol citrullindmin per milligram protein). The frequency of bot G:C to A:T mutations at CpG dinucleotides and all other mutations is show as the percentage of tumors with a mutation for each tertile. The category "d other mutations" includes transversions. transitions at sites other than CpC and frameshift mutations. The frequency of G:C t o A:T muta tions at CpG dinucleotide sites is highest in the g r w p with high NOS2 activit) whereas the frequency of all other p53 gene mutations is inversely correlate with NOS2 activity. For p53 sequencing, paraffinzmbedded tumor sample were dewaxed and microdmxted from 5O-prnsections. DNA was isolated b; use of &um dodecyl sulfatelprotknase K trrrtmeat and phenoVchlomforn extraction, and the p53 coding seq- was am@i!icd as described previousl: (25) and sequenced with the T7 sequenase kit (Amrsham Life Science Inc. Arlington Heights, IL). *'# &dw~piai(t1e7)and may, thereFore,pmwt against superoxide toxicity. This paaicular NO chemistry can explain our finding that NOS2 expression is inversely correlated with the frequency of mutations other than G:C to =-Starhi I, Vergari WA, et al. Role of &ric oxide in angiogenesis and tumor pro-ion in h d and neck cancer. J Natl Can- CCT h s t 1998;90:587-%. (5) Gal A, Wogan GN.Mutagenesis associated with nitric oxide production in transgenic Sn mice, Roc Natl Acad Sci U S A 1996 (17) Wit& DA,Hanbauer I, Grisham MB, Lavd F,N i RW.Laval J. et al. Chemical biology d nitric oxide: regulation and protective and tonic mechanisms. Cum Top Cell Regul 19%,34:159-87. (18) Juedes ui, Wogan GN. Peroxynitriteinduced mutation spectra of pSP189 follow- A:T at CpC dinucleotides (Fig. 2. B). 93:15102-7. ing replicationin bacteria and in human cells. Our investigation of primary human colon tumors establishes a strong positive relationship between the presence of NOS2 in the tumors and the fre- quency of G:C to A:T transitions at CpG dinucleotides. These mutations also are (6) Nathan C, Xie QW. Nitric oxide synthases: roles, tolls, and controls. Cell 1994;78: 915-8. (7) Laurent M, Lepoivre M, Tenu JP. Kinetic modelling of the nitric oxide gradient generated in virro by adherent cells expressing inducible nitric oxide synthase. Biochem J Mutat Res 199634951-61. (19) Levine AJ, Momand J, Finlay CA. The p53 tumour suppressor gene. Nature 1991:351: 453-6. (20) Sibghat-Ullah. Gallinari P. Xu Yz, Goodman MF,Bloom LB, Jiricny J, et al. Base analog and neighboring base effects on substrate common in lymphoid, esophageal, head 19%,314:109-13. specificity of recombinant human G:T mis- and neck, stomach, brain, and breast cancers (9,10,19).Increased NOS2 expression has been demonstrated in four of these cancers (1-4). Tumor-associated NO production may modify DNA (8) Alleva DG,Burger CJ, Elgert KD. Tumor- induced regulation of suppressor macrophage nitric oxide and TNF-alpha production. Role of tumor-derived IL-10, TGF-beta. and pros- taglandin E2. J Immunol 1994;153:1674-86. (9) Hollstein M, Sidransky D. Vogelstein B, match-specific thymine DNA-glycosylase. Biochemistry 1996,35:1292632. (21) Messmer UK, Brune B. Nitric oxide-induced apoptosis: p53-dependent and p53-independent signalling pathways. Biochem J 1996;319(Pt 111299-305. directly, or it may inhibit DNA repair Hanis CC. p53 mutations in human cancers. (22) Forrester K, Ambs S, Lupoid SE, Kapust RB, activities (17), such as the recently described human thymine-DNA glycosylase, which has been shown to repair G:T mismatches at CpG dinucleotides (20). Because NO production also induces accumulation of wild-type p53 Science 1991;253:49-53. (10) Greenblatt MS, Bennett WP, Hollstein M, Harris CC. Mutations in the p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res 1994: 544855-78. (1I ) deRojas-Walker T, Tamir S. Ji H, Wishnok Spillare EA, Weinberg WC. et al. Nitric oxide-induced p53 accumulation and regulation of inducible nitric oxide synthase expression by wild-type p53. Proc Natl Acad Sci U S A 1996;93:2442-7. (23) Ambs S, Hussain SP, Harris CC. Interactive effects of nitric oxide and the p53 tumor sup- (21,22), the resulting growth inhibition JS, Tannenbaum SR. Nitric oxide induces pressor gene in carcinogenesis and tumor can provide an additional strong selection pressure for nonfunctional, mutant p53 (23). NO may, therefore, act as both an endogenous initiator and a promoter in human colon carcinogenesis.Specific inhibitors of NOS2, as demonstrated re- oxidative damage in addition to deamination in macrophage DNA. Chem Res Toxicol 1995;8:473-7. ( 1 2 ) Wink DA, Kasprzak KS, Maragos CM, Elespuru RK, Misra M, Dunams TM, et al. DNA deaminating ability and genotoxicity of nitric oxide and its progenitors. Science progression. FASEB J 1997;11:443-8. (24) Thomsen LL, Scott JM, Topley P, Knowles RG, Keerie AJ, Frend AJ. Selective inhibition of inducible nitric oxide synthase inhibits tumor growth in vivo: studies with 1400W. a novel inhibitor. Cancer Res 1997; 57:3300-4. cently in an animal tumor model (24), 1991;2541001-3. (25) Lehman TA, Bennett WP, Metcalf RA, may have important chemopreventive potential in human colorectal cancer. (13) Murata J, Tada M, Iggo RD. Sawamura Y, Shinohe Y, Abe H. Nitric oxide as a carcinogen: analysis by yeast functional assay of in- Welsh JA, Ecker J, Modali RV, et al. p53 mutations, ras mutations, and p53-heat shock 70 protein complexes in human lung carci- REFERENCES activating p53 mutations induced by nitric oxide. Mutat Res 1997;379211-8. noma cell lines. Cancer Res 1991;51:409(M. (1) Thomsen LL, Miles DW, Happerfield L, Bo- (14) Marlerta MA, Yoon PS,Iyengar R, Leaf CD, Wishnok JS. Macrophage oxidation of L- NOTES brow LG, Knowles RG, Moncada S. Nitric arginine to nitrite and nitrate: nitric oxide is oxide synthase activity in human breast can- an intermediate. Biochemistry 1988;27: We thank Dr. Marc Krasna, Dr. Joshua Sonett, cer. Br J Cancer 1995;72:414. 8706-1 1. and Ms. Audrey Salabes for their assistance in the (2) Ellie E, Loiseau H, Lafond F, Arsaut J, De- (15) Wu Y,Brouet I, Catmels S, Bartsch H, Oh- collection of colon tissues from the University of motes-Mainard J. Differential expression of shima H. Increased endogenous N-nitro- Maryland at Baltimore, The Baltimore VA Medi- -. -a inducible nitric oxide synthase mRNA in hu- m i n e and nitrate formation by induction of cal Center, St. Agnes Hospital, Sinai Hospital, man ,brain tunours. Neuroreport 1995;7: nitric oxide synthase in rats with acute he- Northwest Hospital, and the Office of the Chief 2943. patic injury caused by Propionibacterium ac- Medical Examiner: Ms. Dorothea Dudek for her (3) Ambs S, Memam WG, Bennett WP, Felley- nes and lipopolysaccharide administration. editorial assistance: Dr. Jeffrey Weidner at the Bosco E, Ogunfusika MO, Oser KM, et al. Carcinogenesis 1993;14:7-10. Merck Research Laboratories for the anti-human Frequent nitric oxide synthase-2 expression (16) Kennedy LJ, Moore K Jr., Caulfield JL, Tan- NOS2 antibody, and Dr. Ray Jones at the Univer- in human colon adenomas: implication for nenbaum SR, Dedon PC. Quantitation of sity of Maryland at Baltimore for his assistance 1 tumor angiogenesis and colon cancer pro- 8-oxoguanine and strand breaks produced by with immunohistochemistry. gression. Cancer Res 1998;58:334-41. four oxidizing agents. Chem Res Toxicol Manuscript received June 9, 1998; revised Oc- (4) Gallo 0,Masini E, Morbidelli L, Franchi A, 1997;10386-92. tober 14, 1998; accepted November 3, 1998. 88 BRIEF COMMUNICATION Journal of the National Cancer Institute, Vol. 91, No. 1, January 6, 1999