Metabolism in Human Colon Adenomatous Polyps and Cancers1

[CANCER RESEARCH 43, 3441-3446, July 1983]

Aberration of Poly(Adenosine Diphosphate-Ribose) Metabolism in Human Colon Adenomatous Polyps and Cancers1

Kiyoshi Mirai,2 Kunihiro Ueda, and Osamu Hayaishi3

Department of Medical Chemistry, Kyoto University Faculty ol Medicine, Yoshida, Sakyo-ku, Kyoto 606, Japan

ABSTRACT pears to be of great importance. In the present study, we examined the enzymatic activities for The activities of three principal enzymes engaged in the bio the biosynthesis and degradation of poly(ADP-ribose)4 in nuclei synthesis and degradation of poly(adenosine diphosphate-ri- isolated from normal mucosa, adenomatous polyps (tubular ad bose) [poly(ADP-ribose)] were examined in cell nuclei isolated enomas of familial polyposis coli, villous adenoma, and tubulo from adenomatous polyps (tubular adenomas of familial pol- villous adenoma) and colon cancers, and compared the results yposis coli, villous adenoma, and tubulovillous adenoma), can with morphological changes. Poly(ADP-ribose) is a homopoly- cers, and normal mucosa of human colon. The activities of mer synthesized from NAD"" by a chromatin-bound enzyme, poly(ADP-ribose) synthetase in adenomatous polyps [161 Â±46 poly(ADP-ribose) synthetase, and is ubiquitously distributed in (S.E.) pmol/min/mg DNA] and cancers (114 Â±32 pmol/min/mg almost all types of eukaryotic cells (16, 17, 33). The synthesis DNA) were, on an average, about 3 and 2 times, respectively, proceeds in a covalent association of ADP-ribose chains with higher than those in normal mucosa (52 Â± 24 pmol/min/mg various chromosomal proteins including histones, nonhistone DNA); the difference was statistically significant (p < 0.001 ). The proteins, and the synthetase itself (23). The protein-bound ADP- activity of poly(ADP-ribose) glycohydrolase was also significantly ribose chains thus comprise monomers, oligomers, and polymers high in adenomatous polyps (13.0 Â±3.4 nmol/min/mg DNA), but of various chain lengths up to >100 ADP-ribose units (16, 23). not in cancers (10.1 Â±2.5 nmol/min/mg DNA), compared with Recently, a branch structure was found in polymers synthesized normal mucosa (5.2 Â±1.4 nmol/min/mg DNA) (p < 0.001). The in vitro (26) and in vivo (21). The degradation of poly(ADP-ribose) activity of ADP-ribosyl protein lyase, in contrast, was lower in is catalyzed consecutively by 2 enzymes, i.e., poly(ADP-ribose) adenomatous polyps (152 Â± 40 pmol/min/mg DNA) than in glycohydrolase that cleaves the linkage between adjacent ADP- normal mucosa (345 Â±111 pmol/min/mg DNA) and cancers (288 ribose units in the polymer (16, 38), and ADP-ribosyl protein Â±80 pmol/min/mg DNA) (p < 0.001). lyase5 that cleaves the bond between ADP-ribose and histone Analyses of reaction products with snake venom phosphodi- (31). Biological functions of poly(ADP-ribose) have not yet been esterase digestion revealed that poly(ADP-ribose) synthesized in precisely understood, but a close correlation has been suggested nuclei of normal mucosa, adenomatous polyps, and cancers had between this polymer synthesis and DNA repair (1, 3, 12, 20), the average chain lengths of 2.9, 1.7, and 9.7 ADP-ribose units, DNA replication (7), cell differentiation (6, 18), or neoplastic respectively. Based upon these values and total amounts of transformation (25). ADP-ribose incorporated, the amount of poly(ADP-ribose) syn Our results to be presented herein revealed marked changes thesized per mg DNA in 30 min was calculated as 308, 1510, not only in the enzymatic activities for the biosynthesis and and 106 pmol in the above three types of colon tissues, respec degradation of poly(ADP-ribose) but also in the amount and chain tively. These results suggested that a larger amount of mono length of product polymers in adenomatous polyps, cancers, and mers and short oligomers of ADP-ribose was synthesized in normal mucosa. Therefore, it appears that poly(ADP-ribose) me adenomatous polyps, while a smaller number of longer polymers tabolism undergoes characteristic alterations upon transforma was produced in cancers as compared with normal mucosa. tion of normal colon mucosa to adenomatous polyp or cancer. Immunohistochemical analysis of these tissues using anti- As far as we know, this is the first report of quantitative studies poly(ADP-ribose) antibody supported this view. on the enzymatic activities for poly(ADP-ribose) metabolism in precancerous and cancerous tissues of human origin. INTRODUCTION MATERIALS AND METHODS The concept that a majority of human colon cancers develops from pre-existing adenomatous polyps is known as the ade Specimens. Fresh tissues of colon cancers and adenomatous polyps noma-carcinoma sequence. This concept has been supported were obtained from surgical specimens of patients who were operated by many clinical (13,15, 27, 28) and experimental studies (9, 29, on at Kyoto University-affiliated hospitals. Normal mucosal tissues were 34, 39), especially by those on familial polyposis coli (5,10, 24) taken from the uninvolved area adjacent to cancerous lesions of patients, and were ascertained for the histological normality microscopically. and villous adenoma (15, 28). In order to consolidate this hy Chemicals and Radiochemicals. Calf thymus DNA (type I) and NAD* pothesis and elucidate its mechanism, a comparison of morpho logical changes with biochemical ones at a molecular level ap- (Grade V) were purchased from Sigma Chemical Co. (St. Louis, Mo.). Snake venom phosphodiesterase and DNase I were obtained from 1This work was supported in part by Grants-in-Aid for Scientific Research and *The abbreviations used are: poly(ADP-ribose). poly(adenosine diphosphate- Cancer Research Fund from the Ministry of Education, Science, and Culture, ribose); PBS, phosphate-buffered saline [0.15 M NaCI containing 10 mm sodium Japan, and by a research grant from the Vitamin B Research Committee. phosphate (pH 7.4)]. 2 On leave from the Department of Surgery, Kyoto University Faculty of Medicine, 5The enzyme, formerly termed ADP-ribosyl histone splitting enzyme (31), was Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606, Japan. renamed in view of recent data on the substrate specificity and the product of this 3 To whom requests for reprints should be addressed. enzyme (H. Komura, T. Iwashita, H. Naoki, K. Nakanishi, J. Oka, K. Ueda. and 0. Received November 30,1982; accepted April 4,1983. Hayaishi, submitted for publication).

JULY 1983 3441

Worthington Biochemical Corp. (Freehold, N. J.). Fluorescein isothiocy- CI3CCOOH and was centrifuged at 20,000 x g for 20 min. The precipitate anate-labeled swine anti-rabbit immunoglobulin (fluorescein isothiocya- obtained was washed twice with 20% CI3CCOOH and once with diethyl nate/protein molar ratio, 2.3) was purchased from DAKO-immunoglobu- ether and then incubated in 100 /tl of 1 M NH4OH for 40 min at 25Â°.This lins, Ltd. (Copenhagen, Denmark). Other reagents of analytical grade procedure was performed to liberate poly(ADP-ribose) from protein. After were obtained from Nakarai Chemicals, Ltd. (Kyoto, Japan). [adenine-U- lyophilization, the residue was dissolved in a solution (61 .5 ^l) containing 14C]NAD+ (266 Ci/mol) was purchased from Amersham International, 50 mM potassium phosphate (pH 7.5), 10 HIM MgCI2, 5 mM AMP (added Ltd. (Amersham, England). [aden/ne-l/-14C]poly(ADP-ribose) was pre to inhibit a trace activity of phosphomonoesterase contaminating the pared from (14C]NAD+ as described by Ikai ef al. (19). The polymer had phosphodiesterase preparations), 2 /^g trypsin [added to degrade accep the average chain length of approximately 20 ADP-ribose units, [adenine- tor proteins thereby facilitating phosphodiesterase digestion of poly(ADP- C/-14C]ADP-ribosyl histone H2B was prepared by the method of Okayama ribose); this protease did not affect the phosphodiesterase activity under ef al. (30) and was kindly donated by Dr. J. Oka of this department. these conditions], and 0.55 unit of snake venom phosphodiesterase. The Anti-Poly(ADP-Ribose) Antibody. The antibody toward poly(ADP- mixture was incubated for 6 hr at 37Â°. The digest was applied to ribose) produced in a rabbit was a gift of Dr. K. Ikai of this department Whatman No. 3MM filter paper, and the paper was chromatographed in (19). The antibody was reactive with poly(ADP-ribose) larger than a the solvent system of isobutyric acidxoncentrated NH4OH:water:0.1 M trimer, and the reactivity increased as the chain length of poly(ADP- disodium EDTA (66:1 :32:1 ). Strips 1 cm wide were cut out and examined ribose) increased up to about 25-mer (19). for radioactivity. Average chain length was calculated by: Isolation of Nuclei. Surgically resected colon specimens were quickly cpm(AMP) + cpm(2'-(5"-phosphoribosyl)-5'-AMP) opened and freed of debris and mucus by repeated washings with 0.9% V ~" cpm(AMP) NaCI solution. The specimens were placed on a glass plate mounted on an ice block and were scraped with an edge of slide glass to detach the Branching was not corrected for, because the radioactivity associ normal mucosal layer from the underlying muscle layer. The scraped ated with a branch structure [2'-[1"-ribosyl-2"-(1'"-ribosyl)]adenosine- material was weighed and suspended at a final concentration of 20 mg/ 5',5",5'"-tris(phosphate)] (26), as estimated by paper chromatography ml in 1.9 M sucrose containing 1 HIM potassium phosphate (pH 7.5), 3 HIM CaCI2, and 0.2% (v/v) Triton N-101. The cells were broken by 5 (23), was <1% of total radioactivity in all of our samples. strokes with a Potter-Elvehjem homogenizer kept in an ice:water bath. Immunohistochemical Analysis. Tissues obtained at the operation The homogenate was filtered through nylon cloth, and the filtrate was room were immediately frozen in a dry ice:acetone mixture and were cut into 2-fjm-thick pieces on a cryostat. The pieces were placed on a slide centrifuged at 27,000 x g for 60 min. The pellet was resuspended in glass, dried with a fan for 1 hr at 4Â°,and fixed in cold 95% ethanol for 0.32 M sucrose containing 1 FTIMpotassium phosphate (pH 7.5), 3 mM 10 min. This fixation procedure was shown to barely affect poly(ADP- CaCI2, and 0.5 mw MgCI2 at the same concentration as above. The suspension was centrifuged at 3000 x g for 5 min. The nuclear pellet ribose) synthetase activity in tissues (19). Tissue sections on a slide glass were preincubated for 30 min at 25Â°with a mixture containing 0.1 was weighed and resuspended in 3 volumes (v/w) of the same sucrose M Tris-HCI (pH 8.0), 10 HIM MgCI2, 1 mM dithiothreitol, and 0.2 mM NAD*. solution. The suspension was sonicated at 20 watts for 15 seconds with a Branson sonifier. All of these procedures were performed at 0-4Â°. The After the preincubation, the sections were incubated for 1 hr at 25Â°with rabbit anti-poly(ADP-ribose) antiserum diluted 10-fold with PBS. Control suspension thus prepared was used for enzyme assays and the deter mination of average chain length of product polymers. The recovery of was incubated with preimmune rabbit serum. The sections were washed with PBS and were incubated for 1 hr at 25Â°with fluorescein isothiocy- nuclei, as judged by the yield of DNA from the initial homogenate, was anate-labeled swine anti-rabbit immunoglobulin diluted 10-fold with PBS. approximately 70%. Enzyme Assays. The activity of poly(ADP-ribose) synthetase was The slides were washed as above and were viewed with a Nikon assayed by incubating the reaction mixture (50 Â¿il)containing 0.1 mM photomicroscope equipped with an epiillumination system (19). [aden/ne-14C]NAD+(20 cpm/pmol), 0.1 MTris-HCI(pH8.0), 10mMlvlgCI2, 1 mw dithiothreitol, 1.6 M (NHâ€ž)2SO4,2^9 calf thymus DNA, and a RESULTS sonicated nuclear preparation for 5 min at 37Â°.The radioactivity incor porated into 20% CI3CCOOH-insoluble material was determined as de Kinetics of Enzyme Reactions with Isolated Nuclei. Chart 1 scribed previously (30). The activity of poly(ADP-ribose) glycohydrolase shows the time course of poly(ADP-ribose) synthesis in typical was assayed in a mixture (100 ^0 containing 1.6 ÃŸM(as ADP-ribose samples of cell nuclei isolated from normal mucosa, adenoma- residues) [aden/ne-14C]poly(ADP-ribose) (5 cpm/pmol), 0.1 M potassium tous polyps, and cancer of human colon. With all of these phosphate (pH 7.5), 10 mM KF, 10 mM 2-mercaptoethanol, and enzyme sample. The mixture was incubated for 2 min at 37Â°.After the addition samples, the incorporation of ADP-ribose increased almost line- of 70 n\ of 50% CI3CCOOH, the mixture was centrifuged at 20,000 x g for 20 min, and the radioactivity in the supernatant was determined. The activity of ADP-ribosyl protein lyase was assayed in a mixture (50 n\) containing 10 ^M [acten/ne-14C]ADP-ribosyl histone H2B (6 cpm/pmol), 0.1 M potassium phosphate (pH 6.0), 5 mM dithiothreitol, and enzyme sample. The mixture was incubated for 15 min at 37Â°.The incubation o_ 20 was terminated by the addition of 40 p\ of 45% CI3CCOOH. The mixture l.5 was centrifuged at 20,000 x g for 20 min, and the radioactivity in the supernatant was determined. All enzyme activities were assayed in 1.0 duplicates, and the mean values, normalized for DNA contents, were presented. The difference between duplicate determinations was a. o 0.5 <Â±15% of the mean value. Statistical comparison was made by the use of Student's paired and unpaired f tests. DNA was determined by the method of Burton (4) using calf thymus DNA as the standard. IO 20 30 Determination of Average Chain Length of Poly(ADP-Ribose). A Time (min) sonicated nuclear preparation was incubated with a solution (75 ÃŸ\) Chart 1. Time course of poly(ADP-ribose) synthesis in isolated nuclei of normal containing 0.1 mM [aden/ne-'4C]NAD* (267 cpm/pmol), 0.1 M Tris-HCI mucosa (O), adenomatous polyps (A), villous adenoma with a small malignant change (A), and cancer (â€¢).A 5-fold scaled-up standard mixture was incubated (pH 8.0), 10 HIM MgCI2, 1 mM dithiothreitol, 1.6 M (NH4)2SO4, and 3 ng and, at intervals, aliquots were removed and examined for 20% CbCCOOH- of calf thymus DNA for 30 min at 37Â°.The mixture was made 20% with insoluble radioactivity.

3442 CANCER RESEARCH VOL. 43

arly for 5 min and then decreased gradually. This decrease in poly(ADP-ribose) synthesis appears to be due, partly, to partial 200 inactivation of the synthetase undergoing automodification (23) and, partly, to degradation of product polymers by poly(ADP- ribose) glycohydrolase in the nuclear preparation (14, 38). Chart 5ÃŒI50 2 shows the time course of poly(ADP-ribose) glycohydrolase reaction. With all samples tested, the release of ADP-ribose ft100 increased linearly for 2 min, at least, and then decreased. The SE decrease appeared to be due to exhaustion of the substrate Â¿5 50 poly(ADP-ribose). Chart 3 shows the time course of ADP-ribosyl a protein lyase reaction. The linearity of ADP-ribose release was maintained for about 15 min. Normal Adenomatous Mucosa Polyps Cancers In our following studies, all enzyme assays were performed Chart 4. Poly(ADP-ribose) synthetase activity in normal mucosa (O), adenoma within respective linear ranges. Under these conditions, the tous polyps (A), and cancers of colon (â€¢).Lines,data obtained with normal mucosa reactions were strictly proportional to the amounts of nuclear and cancers from same patients. C, macroscopically uninvolved, but microscopi cally hyperplastic (and partly adenomatous), mucosa from familial polyposis coli; suspension added (data not shown). A. villous adenoma with a small malignant change (carcinoma in situ); V, tubulov- Changes in Enzymatic Activities. Chart 4 shows the activities illous adenoma; oars, mean for each class. of poly(ADP-ribose) synthetase in nuclei isolated from normal mucosa, adenomatous polyps, and cancers. The activities of adenomatous polyps (161 Â±46 pmol/min/mg DNA) and cancers 15 (114 Â±32 pmol/min/mg DNA) were, on an average, about 3 and 2 times, respectively, higher than that of normal mucosa (52 Â± 24 pmol/min/mg DNA). The difference was statistically significant 10 at the level of p < 0.001. The mucosa of familial polyposis coli o: E that appeared macroscopically uninvolved, but microscopically had many hyperplastic glands and small adenomas (Chart 4, O) Â°-E Ã› C

Normal AdenomaTousÂ» Mucosa Polyps Cancers

Chart 5 Poly(ADP-ribose) glycohydrolase activity in normal mucosa (O), ade nomatous polyps (A) and cancers of colon (â€¢).Lines, data obtained with normal mucosa and cancers from same patients; C, macroscopically uninvolved, but microscopically hyperplastic (and partly adenomatous). mucosa from familial pol yposis coli; A, villous adenoma with a small malignant change (carcinoma in situ); bars, mean for each class.

exhibited a considerably higher activity (132 pmol/min/mg DNA) than that of normal mucosa. This value was not included in the statistical analysis. Among 5 specimens from a single patient of Chart 2. Time course of poly(ADP-ribose) degradation in isolated nuclei of familial polyposis coli, there was apparently no correlation be normal mucosa (O), adenomatous polyps (A), villous adenoma with a small malig nant change (A), and cancer (â€¢).A5-fold scaled-up standard mixture was incubated tween the enzyme activity and the grade of histological atypism and. at intervals, aliquots were removed and examined for 20% CI3CCOOH-soluble (data not shown). In villous adenoma, a small malignant change radioactivity. (carcinoma In situ) was found, and the residual region showed varying degrees of atypical histology; this specimen (Chart 4, A) showed the highest activity of poly(ADP-ribose) synthetase among all samples tested. By comparison of the pairs of normal mucosa and cancer from the same individuals, qualitative paral lelism in the enzyme activity between the 2 tissues was notice able. The activity of poly(ADP-ribose) glycohydrolase was signifi cantly higher in adenomatous polyps (13.0 Â±3.4 nmol/min/mg DNA) than in normal mucosa (5.2 Â±1.4 nmol/min/mg DNA) (p < 0.001) (Chart 5). The activity of cancers (10.1 Â±2.5 nmol/min/ mg DNA) was, on the average, 2 times higher than that of normal mucosa, but the difference did not reach a statistical level at p < 0.05 because of wide deviation among samples. Chart 3. Time course of ADP-ribosyl histone splitting in isolated nuclei of normal Chart 6 shows ADP-ribosyl protein lyase activities in nuclei mucosa (O), adenomatous polyps (A), villous adenoma with a malignant change isolated from the same 3 types of colon tissues. The activity in (A), and cancer (â€¢).A 5-fold scaled-up standard mixture was incubated and, at intervals, aliquots were removed and examined for 20% CI3CCOOH-soluble radio adenomatous polyps (152 Â±40 pmol/min/mg DNA) was signifi activity. cantly lower than that in normal mucosa (345 Â±111 pmol/min/

JULY 1983 3443

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1983 American Association for Cancer Research. K. Mirai et al. nuclei in the upper differentiated cells (Fig. 16). Immunofluores- cence was prominent also in nuclei of adenomatous potyp with 400 some variation among the cells (Fig. 1C). The immunofluores- < cence was most prominent in nuclei of cancer cells (Fig. 1D). o 300 9 E _ DISCUSSION E 200

o. E Familial polyposis coli is an inherited disease known to be -- 100 predisposed to malignancy (5). Villous adenoma of colon has also been established to be precancerous (27, 28). What these predisposed states mean and how these states develop cancers Normal Adenomatous_ have not yet been understood at a molecular level. One of the Mucosa Polyps Cancers possible mechanisms is that, during rapid replication and/or Chart 6. ADP-ribosyl protein lyase activity in normal mucosa (O), adenomatous active repair of the gene, which are characteristic of these colon polyps (A), and cancers of colon (â€¢).Lines,data obtained with normal mucosa and cancers from same patients; C, macroscopically uninvolved, but microscopically polyps (8), fortuitous gene rearrangements or mutations may hyperplastic (and partly adenomatous), mucosa from familial polyposis coli; A, take place and lead to malignant transformation. Although pre villous adenoma with a small malignant change (carcinoma in situ); bars, mean for cise mechanisms of transformation and DNA repair in mammalian each class. cells have not been elucidated, poly(ADP-ribose) synthesis [or mg DNA) or cancer (288 Â±80 pmol/min/mg DMA) (p < 0.001). poly(ADP-ribosyl)ation of chromatin proteins] has been noted The difference between normal mucosa and cancers was not recently for its close correlation to carcinogenesis and DNA repair significant at p < 0.05. (37). For example, mouse fibroblasts transformed by SV40 ex Changes in Chain Length and Amount of Products. In order hibit a 2 to 10 times higher activity of poly(ADP-ribose) synthe- to investigate the significance of changes in various enzyme tase than do parental normal cells (25); leukemic granulocytes, activities on poly(ADP-ribose) metabolism, we analyzed the prod unlike normal granulocytes, retain a potent activity of poly(ADP- ucts synthesized in nuclei of various colon tissues. We first ribose) synthetase even at the stage of appearing in peripheral analyzed the average chain length of poly(ADP-ribose) produced blood (18); various DNA-damaging agents induce a marked in a relatively long (30 min) incubation, i.e., at a near equilibrium increase in poly(ADP-ribose) synthesis (2,12,20,22,35); various of synthesis and degradation (Chart 1) (14). The products syn inhibitors of poly(ADP-ribose) synthetase inhibit the repair of thesized in nuclei of normal mucosa, adenomatous polyps, and damaged DNA at the stage of rejoining (11, 12); and, finally, an cancers had the average chain length of 2.92 Â±0.42 (n = 2), inhibitor of poly(ADP-ribose) synthetase enhances the carcino 1.69 Â±0.06 (n = 2), and 9.69 Â±1.41 ADP-ribose units (n = 3), gen-induced occurrence of hyperplastic nodules, suggesting pro respectively. From the amount of ADP-ribose incorporated and motion of chemical carcinogenesis (36). The present study re the average chain length of products, the amount of poly(ADP- vealed a 3-fold increase in the mean activity of poly(ADP-ribose) ribose) (as whole molecule) synthesized in 30 min per mg DNA synthetase in nuclei of colon adenomatous polyps compared was calculated as follows: with normal mucosa. The nuclei of colon cancers exhibited a Total amount of ADP-ribose incorporated lower, but 2 times higher than normal, activity of this enzyme. Amount of polymer = Av. chain length of poly(ADP-ribose) Average chain lengths and amounts of nuclear products were also very different among normal mucosa, adenomatous polyps, The amounts of polymer thus calculated for normal mucosa, and cancers. adenomatous polyps, and cancers were 308, 1510, and 106 The activity of poly(ADP-ribose) synthetase in isolated nuclei pmol, respectively. These results, together with the data of has been claimed to be a poor measure of the in vivo activity (2), average chain lengths, suggested that a large amount of mono because the procedure of nuclear isolation produces, more or mers and short oligomers of ADP-ribose was synthesized in less, strand breaks in DNA by endogenous DNases, and such adenomatous polyps, while a relatively small number of long breaks in DNA markedly affect poly(ADP-ribose) synthetase ac polymers was mainly synthesized in cancers, as compared with tivity (33). The possibility of these artifacts was carefully ruled normal mucosa. out, Â¡nour present study, as follows, (a) We examined the size Imrnunohistochemical Analysis. Fig. 1 shows the profiles of of DNA in intact cells (scraped tissue specimen) and isolated immunofluorescent staining of normal mucosa, adenomatous nuclei using 0.6% agarose gel electrophoresis (pH 8.0). No polyp, and cancer with anti-poly(ADP-ribose) antibody. All tissue appreciable difference in size was observed between the 2 sections were preincubated with NAD+ to allow poly(ADP-ribose) samples. Electrophoresis also indicated that the brief sonication synthesis in situ; without this preincubation, the sections stained at a low power used for homogenization of the nuclear suspen also positively but with lesser intensity (data not shown). The sion did not introduce detectable strand breaks in DNA. (b) specificity of immunofluorescent staining for poly(ADP-ribose) Preincubation (37Â°for 30 min) of nuclei with DNase I (800 ^g/ was confirmed by disappearance of the fluorescence in sections ml), that is known to fully activate the enzyme activity by DNA stained with control (preimmune) serum or the antiserum preab- breakage (2), increased both activities of normal mucosa and sorbed with poly(ADP-ribose) (data not shown). The immunoflu- cancers markedly and almost equally, i.e., 2.8- and 3.0-fold, orescence appeared to be exclusively localized in the nucleus in respectively. This result suggests that the DNAs of both nuclei all tissues examined. In normal mucosa, potent immunofluores- had been damaged equally or, more probably, not very signifi cence was seen diffusely in nuclei of lower crypt cells undergoing cantly, (c) We examined the effect of addition of exogenous DNA proliferation and differentiation (Fig. ~\A), whereas weaker im- in the assay system. There was essentially no change in the munofluorescence was detectable only in the marginal area of ADP-ribose incorporation by all samples examined, suggesting

3444 CANCER RESEARCH VOL. 43

that DNase action in colon tissues was negligible, if not non 2. Berger, N. A., Sikorski, G. W., Petzold, S. J., and Kurohara, K. K. Association of poly(adenosine diphosphoribose) synthesis with DNA damage and repair in existent, during incubation and, probably, isolation of nuclei. For normal human lymphocytes. J. Clin. Invest., 63: 1164-1171,1979. the sake of safety against possibly contaminating DNases, ex 3. Berger, N. A., Sikorski, G. W., Petzold, S. J., and Kurohara, K. K. Defective ogenous DMA was included in all assays of poly(ADP-ribose) poly(adenosine diphosphoribose) synthesis in xeroderma pigmentosum. Bio chemistry, 19.: 289-293,1980. synthetase. 4. Burton, K. A study of the conditions and mechanism of the diphenylamine The notion that in vitro reactions in isolated colon nuclei reaction for the colorimetrie estimation of deoxyribonucleic acid. Biochem. J., reflected faithfully in vivo states with respect to both enzymatic 62:315-323,1956. 5. Bussey, H. J. R. Familial Polyposis Coli. Baltimore: The Johns Hopkins Univer activities and profiles of products was in accord with the results sity Press, 1975. of Â¡mmunohistochemical analyses of tissue sections. The inten 6. Caplan, A. I., and Rosenberg, M. J. Interrelationship between poly(ADP-Rib) sity of immunofluorescence of poly(ADP-ribose) was in the order synthesis, intracellular NAD levels, and muscle or cartilage differentiation from mesodermal cells of embryonic chick limb. Proc. Nati. Acad. Sei. U. S. A., 72: of cancers > adenomatous polyps > normal mucosa. Referring 1852-1857,1975. to the findings that the activity of poly(ADP-ribose) synthetase 7. Colyer, R. A., Burdette, K. E., and Kidwell, W. R. Poly ADP-ribose synthesis and DNA replication in synchronized mouse L-celis. Biochem. Biophys. Res. was highest in adenomatous polyps followed by cancers (Chart Commun., 53: 960-966, 1973. 4) and that the reactivity of anti-poly(ADP-ribose) antibody in 8. Deschner, E. E., and Lipkin, M. Proliferation and differentiation of gastrointes creased with increasing chain lengths of poly(ADP-ribose) (19), tinal cells in health and disease. In: M. Lipkin and R. A. Good (eds.), Gastroin testinal Tract Cancer, pp. 3-27. New York: Plenum Publishing Corp., 1978. this result was compatible with the view that a larger number of 9. Druckrey, H., Preussmann, R., Matzkies, F., and Ivankovic, S. Selektive shorter oligomers and a smaller number of longer polymers were Erzeugung von Darmkrebs bei Ratten durch 1,2-Dimethyl-Hydrizin. Naturwis senschaften, 54: 285-286, 1967. mainly produced in adenomatous polyps and cancers, respec 10. Dukes, C.E. The hereditary factor in polyposis intestini, or multiple adenomata. tively, than in normal mucosa. This is exactly the profile of in Cancer Rev., 5: 241-256,1930. vitro products estimated from total ADP-ribose incorporation and 11. Durkacz, B. W., Irwin, J., and Shall, S. The effect of inhibition of (ADP-ribose)* biosynthesis on DNA repair assayed by the nucleoid technique. Eur J. chain length analyses (see above). Biochem., 727:65-69,1981. The activity of poly(ADP-ribose) synthetase has been shown 12. Durkacz, B. W., Omidiji, O., Gray, D. S., and Shall, S. (ADP-riboseV participates to correlate with the capacity of proliferation in intestinal mucosa; in DNA excision repair. Nature (Lond.), 283: 593-596,1980. 13. Fenoglio, C. M., and Lane, N. The anatomic precursor of colorectal carcinoma. upper (surface) differentiated and nondividing cells have much J. A. M. A., 237. 640-642, 1975. less activity than do lower (crypt) dividing cells6 (32). As judged 14. Gill, D. M. Poly(adenosine diphosphate ribose) synthesis in soluble extracts of by the patterns of immunohistochemistry of poly(ADP-ribose) animal organs. J. Biol. Chem., 247: 5961-5971, 1972. 15. Grinneii, R. S., and Lane, N. Benign and malignant adenomatous polyps and (Fig. 1, A and 8), the epithelial cells of normal colon mucosa papillary adenomas of the colon and rectum. An analysis of 1856 tumors in appear to undergo a similar change in the synthetase activity 1335 patients. Surg. Gynecol. Obstet. 706: 519-538,1958. 16. Hayaishi, 0., and Ueda, K. Poly(ADP-ribose) and ADP-ribosylation of proteins. during differentiation. The higher synthetase activity of adenom Annu. Rev. Biochem., 46: 95-116, 1977. atous polyps and cancers, compared with normal mucosa, how 17. Hilz, H., and Stone, P. Poly(ADP-ribose) and ADP-ribosylation of proteins. Rev. ever, does not correlate with their proliferation rates. Although Physiol. Biochem. Pharmacol., 76:1-58,1976. 18. Ikai, K., Ueda, K., Fukushima, M., Nakamura, T., and Hayaishi, 0. Poly(ADP- the proliferation rate, as judged by a turnover rate, of adenom ribose) synthesis, a marker of granulocyte differentiation. Proc. Nati. Acad. atous polyps varies considerably, from faster to slower than Sei. U. S. A., 77: 3682-3685, 1980. 19. Ikai, K., Ueda, K., and Hayaishi, O. Immunohistochemical demonstration of normal, with individuals and loci, the proliferation of cancers is poly(adenosine diphosphate-ribose) in nuclei of various rat tissues. J. Histo- much slower than that of normal mucosa (8). It has also been chem. Cytochem., 28: 670-676,1980. shown that the activity of poly(ADP-ribose) synthetase in other 20. Jacobson, M. K., Levi, V., Juarez-Salinas, H., Barton, R. A., and Jacobson, E. L. Effect of carcinogenic N-alkyl-W-nitroso compounds on nicotinamide tissues does not always parallel the proliferation rate (17). adenine dinucleotide metabolism. Cancer Res., 40:1797-1802,1980. The change in activities of poly(ADP-ribose)-degrading en 21. Juarez-Salinas, H., Levi, V., Jacobson, E. L., and Jacobson, M. K. Poly(ADP- zymes was smaller than that of the synthesizing enzyme, but it ribose) has a branched structure in vivo. J. Biol. Chem., 257: 607-609,1982. 22. Juarez-Salinas, H., Sims, J. L.. and Jacobson, M. K. Poly(ADP-ribose) levels was noticeable. A higher activity of poly(ADP-ribose) glycohydro- In carcinogen-treated cells. Nature (Lond.), 282: 740-741,1979. lase and a lower activity of ADP-ribosyl protein lyase in adenom 23. Kawaichi, M., Ueda, K., and Hayaishi, O. Multiple autopoly(ADP-ribosyl)ation of rat liver poly(ADP-nbose) synthetase. Mode of modification and properties atous polyps than in normal mucosa and cancers may possibly of automodified synthetase. J. Biol. Chem., 256: 9483-9489, 1981. contribute to the increase in short-chain oligo(ADP-ribose) in the 24. Lockhart-Mummery, J. P. Cancer and heredity. Lancet, /: 427-429,1925. former tissue. The significance of degrading enzymes in the 25. Miwa, M., Oda, K., Segawa, K., Tanaka, M., Ine, S., Yamaguchi, N., Kuchino, regulation of poly(ADP-ribose) metabolism remains to be inves T., Shiroki, K., Shimojo, H.. Sakura, K., Matsushima, T., and Sugimura, T. Cell density-dependent increase in chromatin-associated ADP-ribosyl-transferase tigated. activity in simian virus 40-transformed cells. Arch. Biochem. Biophys., 787: These results altogether indicate that not only the activity of 313-321,1977. 26. Miwa, M., Saikawa, N., Yamaizumi, Z., Nishimura, S., and Sugimura, T. poly(ADP-ribose) synthetase but also the profile of products Structure of poly(adenosine diphosphate ribose): Identification of 2'-[1"-ribo- undergo dramatic changes at both stages from normal mucosa syl-2" (or 3"-H1"'-ribosyl)]adenosine-5',5",5"'-tris(phosphate) as a branch linkage. Proc. Nati. Acad. Sei. U. S. A., 76: 595-599, 1979. to adenomatous polyp and from the polyp to cancer. It seems 27. Morson, B. C., and Dawson, I. M. P. Adenomas and the adenoma-carcinoma plausible that these changes reflect alterations in chromatin sequence. In: B. C. Morson and I. M. P. Dawson (eds.). Gastrointestinal structure and/or activities of, for example, response to DNA Pathology, pp. 615-647. Oxford: Blackwell Scientific Publications, 1979. 28. Muto, T., Bussey, H. J. R., and Morson, B. C. The evolution of cancer of the damage, and such alterations may play a role in the state of colon and rectum. Cancer (Phila.), 36: 2251-2270, 1975. predisposition or carcinogenesis. 29. Narisawa, T., Magadia, N. E., Weisburger, J. H., and Wynder, E. L. Promoting effect of bile acids on colon carcinogenesis after intrarectal instillation of N- methyl-A/'-nitro-N-nitrosoguanidine in rats. J. Nati. Cancer Inst., 53: 1093- REFERENCES 1097, 1974. 30. Okayama, H., Edson, C. M., Fukushima, M.. Ueda, K., and Hayaishi, 0. 1. Benjamin, R. C., and Gill, D. M. A possible role for poly ADP-ribose in DNA Purification and properties of poly(adenosine diphosphate ribose) synthetase. repair. In: P. G. Hanawalt, E. C. Friedberg, and C. F. Fox (eds.), DNA Repair J. Bid. Chem., 252: 7000-7005,1977. Mechanisms, pp. 337-340. New York: Academic Press, Inc., 1978. 31. Okayama, H., Honda, M., and Hayaishi, O. Novel enzyme from rat liver that cleaves an ADP-ribosyl (listone linkage. Proc. Nati. Acad. Sei. U. S. A., 75: 6 K. Mirai and K. Ueda. unpublished results. 2254-2257, 1978.

JULY 1983 3445

32. Porteous, J. W., Fumeaux, H. M., Pearson, C. K., Lake, C. M., and Morrison, 36. Takahashi, S., Ohnishi, T., Denda, A., and Konishi, Y. Enhancing effect of 3- A. Poly(adenosine diphosphate ribose) synthetase activity in nuclei of dividing aminobenzamide on induction of -y-glutamyl transpeptldase positive foci in rat and non-dividing but differentiating intestinal epithelial cells. Biochem. J., 180: liver. Chem. Biol. Interact., 39. 363-368,1982. 455-463,1979. 37. Ueda, K., and Hayaishi, O. Poly(ADP-ribose) in medicine. Leukemia and cancer. 33. Pumell, M. R., Stone. P. R., and Whish, W. J. D. ADP-ribosylation of nuclear In: O. Hayaishi and K. Ueda (eds.), ADP-Ribosylation Reactions. Biology and proteins. Biochem. Soc. Trans., 8: 215-227,1980. Medicine, pp. 561-572. New York: Academic Press, Inc., 1982. 34. Spjut, H. J., and Noate, M. W. Colonie neoplasms induced by 3,2'-dimethyl-4- 38. Ueda, K., Oka, J., Narumiya, S., and Hayaishi, O. Poly ADP-ribose glycohydro- aminobiphenyl. In: W. J. Burdette (ed.), Carcinoma of the Colon and Antecedent lase from rat liver nuclei, a novel enzyme degrading the polymer. Biochem. Epithelium, pp. 280-288. Springfield, III.: Charles C Thomas, Publisher, 1970. Biophys. Res. Commun., 46. 516-523, 1972. 35. Sudhakar, S., Tew, K. 0., and Smulson, M. E. Effect of 1-methyl-1-nitrosourea 39. Walpole, A. L., Williams, H. C., and Roberts, D. C. The carcinogenic action of on poly(adenosine diphosphate-ribose) polymerase activity at the nucleosomal 4-aminobiphenyl and 3,2'-dimethyl-4-aminobiphenyl. J. Ind. Med., 9:255-263, level. Cancer Res., 39: 1405-1410,1979. 1952.

t.. *-â€¢â€¢Â»A

Fig. 1. Localization of poly(ADP-ribose) synthesis in nuclei of normal mucosa (A, upper crypt cells; B, lower crypt cells), adenomatous polyp (C), or cancer (D) of human colon. Sections were preincubated with NAD* and stained with anti-poly(ADP-ribose) rabbit antiserum and fluorescein isothiocyanate-labeled swine anti-rabbit immunoglobulin. x 860.

3446 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1983 American Association for Cancer Research. Aberration of Poly(Adenosine Diphosphate-Ribose) Metabolism in Human Colon Adenomatous Polyps and Cancers

Kiyoshi Hirai, Kunihiro Ueda and Osamu Hayaishi

Cancer Res 1983;43:3441-3446.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/7/3441

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/43/7/3441. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.