1421.Full.Pdf

Vol. 1. 1421-1428. November 1995 Clinical Cancer Research 1421

Elevated Levels of 2’,5’-linked Oligoadenylate-dependent Ribonuclease L Occur as an Early Event in Colorectal Tumorigenesis’

Liming Wang,2 Aimin Zhou,2 Sandip Vasavada, cause double-stranded RNA is frequently produced during viral Beihua Dong, Huiqin Nie, James M. Church, infection, 2-SA accumulation is observed in some IFN-tneated and virus-infected cells (7). Effects of 2-SA in cells are transient Bryan R. G. Williams, Sipra Banerjee, and due to the combined activity of 2’,5’-phosphodiesterase and Robert H. Silverman3 5’-phosphatase activities that degrade 2-SA (8). 2-SA functions Departments of Cancer Biology [L. W., A. Z.. S. V., B. D., H. N., by activating the IFN-inducible endonibonuclease, RNase L B. R. G. W., S. B., R. H. S.], Urology [S. V.], and Cobonectal (also known as 2-SA-dependent RNase) (9-1 1), resulting in the Surgery [J. M. C.], Research Institute, The Cleveland Clinic Foundation, Cleveland, Ohio 44195 cleavage of single-stranded RNA predominantly after UpUp and UpAp sequences (12, 13). Although cellular levels of RNase L increase during IFN treatment, basal bevels of the enzyme are ABSTRACT found in most mammalian cells. Human RNase L is an Mr RNA decay in IFN-treated cells is controlled by 2’,S’- 83,539 protein of 741 amino acid residues that is converted to its linked oligoadenylate (2-5A)-dependent RNase (RNase L), a catalytically active, homodimenic form by binding to 2-SA (1 1, uniquely regulated endoribonuclease that requires short 5’- 14, 15, and references therein). The structure of RNase L in- phosphorylated, 2-5A for its activity. Because RNase L is eludes nine ankynin protein-protein interaction domains also also implicated in the regulation of cell proliferation, we found in all members of the 1KB family (inhibitors of transcnip- monitored its expression in colorectal adenocarcinomas and tion factor NF-KB), and in some cell cycle control proteins (5 noncancerous pobyps from familial adenomatous polyposis and references therein). The gene for RNase L (RNS4) maps to patients. Elevated bevels of RNase L mRNA and activity chromosome 1q25, a locus that is deleted or rearranged in many were found in 17 of 20 tumors compared with corresponding breast cancer cell lines, as well as in some gastric adenocarci- normal mucosa. An mAb against RNase L revealed elevated nomas and oral squamous cell carcinomas (16). Aberrant regu- amounts of this RNase in sections of the tumors, largely in lation of RNA stability in tumor cells is suggested by some the base of the villi. The occurrence of elevated levels of studies on the 2-5A system. Elevated endonibonuclease activity RNase L seems to be an early event in colorectal tumorigen- against nRNA has been reported in mononuclear cells from esis, suggesting that control of RNA turnover is an impor- patients with chronic myebogenous leukemia (17). Previously tant step in tumor progression. These data also indicate that we reported that the antipnoliferative activity of IFN (a plus 3) regulating RNase L activity may be a useful strategy in was suppressed in munine cells expressing a dominant-negative treating colorectal carcinomas. form of RNase L, suggesting a possible role for the 2-SA system in cell growth control (5). INTRODUCTION Cancer statistics indicate that cases of coborectab cancer in Control of gene expression requires an intricate balance men and women are increasing steadily in the United States as between RNA synthesis and decay (reviewed in Ref. 1). The well as worldwide, at least in part because of improvements in 2-5A4 system (2) is an RNA degradation pathway present in diagnosis (18). Cancer mortality due to cobonectal cancer ranks higher vertebrates that is implicated in some of the antivinal and second among men and third among women (18). Although the antiprolifenative activities of IFNs (3-5). IFN treatment of cells precise etiology of this disease is still unknown, exposure to activates genes encoding several synthetases that produce short environmental factors, in particular mutagens and carcinogens in the diet, are contributors. In addition, patterns of genetic 5 ‘-tniphosphonylated, 2’,5 ‘-linked obigoadenylates known as 2-SA from ATP in response to double-stranded RNA (6). Be- changes associated with human cobonectal cancer are emerging (19-22). Here we report high amounts of RNase L in a range of coborectab tumor types and in polyps. Results suggest that control of RNA turnover by RNase L is an early event in tumonigenesis. Furthermore, based on these findings, we suggest a Received 5/8/95; revised 7/10/95; accepted 7/13/95. therapeutic approach designed to activate RNase L preferen- I Supported by USPHS Grant CA44059 awarded by the Department of tially in tumors as compared with in normal intestinal mucosa. Health and Human Services, National Cancer Institute (R. H. S.). 2 L. W. and A. Z. have contributed equally to this work. 3 To whom requests for reprints should be addressed, at the Department MATERIALS AND METHODS of Cancer Biology, NN1-06, The Cleveland Clinic Foundation, 9500 Isolation of Tumors and Normal Mucosa and Prepara- Euclid Avenue, Cleveland, OH 44195. Phone: (216) 445-9650; Fax: tion of Cell Extracts. Adenocancinomas and benign pobyps (216) 445-6269. were surgically removed and immediately frozen and stored in 4 The abbreviations used are: 2-5A, p,.(A2’p),,A, where x - 1-3 and n 2; RNase L 2-5A-dependent RNase, 2-5A-cellulose, pp(A2’p)3A-ceblu- liquid nitrogen. The normal coborectal mucosa of the same lose; FAP, familial adenomatous polyposis; RT, reverse tnanscniptase. patients, distal to the tumors and free of metastatic lymph nodes,

Table 1 RNase L mRNA and activity in cobonecta 1 tumors

Tumor Increased bevels Fold increase in Tumor type Tumor site Distal metastasis RNase mRNA” levels of RNase”

22 adcc Sigmoid None - 1

25 adc Distal and transverse None - 1

28 adc Rectum None - 1 50 adc Rectosigmoid Liven, lymph nodes + 1.5 27 adc Rectum Liver, lymph nodes + 2 36 adc Sigmoid None + 2 29 adc Proximal Liver, ovaries, lymph nodes + 2.5 30 adc Transverse None + 2.5 46 adc Distal None + 2.5 38 adc Rectosigmoid Liver, lymph nodes + 4 45 adc Rectum None + + 4 56 adc Distal Lymph nodes + + 4 31 adc Sigmoid None + 4.5 6d 23 FAP P Desmoid None 39 adc Proximal Liven, ovaries + + 6 Si FAP P Rectosigmoid None + + 6 53 adc Sigmoid Liven, lymph nodes + + 6 40 adc Proximal and transverse None + + 8 41 adc Rectosigmoid None + + 8 52 adc Rectosigmoid Liven, lymph nodes + + + 10

“Approximate relative levels of mRNA encoding RNase L were determined by RT-PCR. -, no change; +, 1.5- to 3-fold; + +, 3- to 5-fold;

+ + + ‘ >5-fold increase of the cellular RNase mRNA compared with normal mucosa. b Numbers indicate the fold increase of levels of 2-5A binding to RNase L in extracts of tumors compared with extracts of corresponding normal

mucosa as determined by the 2-SA binding assay (see ‘‘ Materials and Methods”). C adc, adenocancinoma; FAP P, FAP polyps. d Compared with the average level of normal mucosa.

was also collected and stoned. The upper central parts of the renaturation buffer. The membranes were incubated with tumors were used to ensure that the majority of cell populations 2-5A probe (200,000 cpm/ml) and 0.02% sodium azide in were cancer cells. None of the patients were treated with IFN or renaturation buffer for 24 h in a cold room with shaking. other cytokines, radiation therapy, or chemotherapy before sun- After washing once with renaturation buffer and twice with gery. The frozen tumor on mucosa was grounded to a fine water, air-dried membranes were covered with Saran Wrap powder in liquid nitrogen, suspended in NP4O lysis buffer and exposed to X-ray film. supplemented with the protease inhibitor leupeptin (23), and Poly(U) Degradation Assay for RNase L. 2-5A-cellu- disrupted by vortex mixing. The total cell lysates were centni- lose, 100 jib 144 jiM, was washed three times with NP4O lysis fuged at 10,000 X g for 10 mm at 4#{176}C;the supernatant was buffer and suspended in 150 jib buffer. Tissue extracts, 100 jig collected; and the protein concentration was determined accord- protein, were added to 20 jib 2-5A-cebbubose in 100 jib NP4O ing to the Bradford method (Bio-Rad). buffer and were incubated on ice for 1 h, vortexing briefly every Assay for 2-5A Binding Activity. Covalent binding of is mm. The complexes were washed three times with 200 jib a radiolabeled and bromine substitute 2-SA probe, p(A2’p)2 NP4O buffer by centrifuging, discarding the supernatant, and (br8A2’p)2A3’-[32P]Cp, to RNase L was performed as de- resuspending. After the final centnifugation, the pellets were scnibed previously (24). Briefly, a cell extract containing 200 suspended in 30 jib buffer [20 msi Tnis-HC1 (pH 7.5), 10 mi jig protein from either tumor, polyp, or normal mucosa was incubated on ice with the 2-5A probe (50,000 cpm) for 1 h. magnesium acetate, 8 mri 3-mercaptoethanob, 90 mst KC1, and 10 jig/mb leupeptin] containing 0.2 mti poly(U)-[32P]pCp pre- Covalent cross-linking was achieved under a UV lamp (308 nm) on ice for 1 h; then the proteins were separated on pared as described (26) and incubated at 30#{176}Cfor 1 h. Reactions SDS-polyacnylamide (10%) gels. The gels were dried and were terminated with 15 jib stop buffer, boiled 3 mm, and exposed to X-ray film overnight. Quantitation was performed centrifuged briefly, and 4 jib were loaded to 6% pobyacnylamide/ with a Phosphorlmager (Molecular Dynamics). The denatur- urea gels. After electrophoresis, X-ray film was exposed to the ation-renatunation protocol was performed by a modification gels. (11) of the method of Singh et a!. (25). Briefly, after elec- Anabysis of RNase L on Western Blots Probed with a trophoresis of 300 jig cell protein/bane on SDS-pobyacryb- mAb. Extracts of tumors, pobyps, and normal mucosa, 50 jig amide (10%) gels, the proteins were transferred to nitroceb- protein each, were subjected to electrophoresis in SDS-poly- bubose membranes (Schleicher & Schuell). The membranes acrybamide gels, and the proteins were then transferred to nitno- were blocked with 3% nonfat milk in renaturation buffer [20 cellulose filters. Blots were probed with a 1 :5000 dilution of an mM Tnis-HC1, pH 7.5, 20 mri magnesium acetate, 50 msi KC1, ascites fluid containing mAb to recombinant human RNase L 1 mM EDTA, 50 mM 3-memcaptoethanol, and 5% (v/v) glyc- (15). Detection was with goat antimouse IgG-penoxidase erob] at room temperature for 1 h and washed briefly with (GIBCO) and enhanced chemiluminescence (Amersham).

45 46 51 53 23 MNTNTNTN TT KD

97.4 - - .

.‘ -‘*--RNase -.w

66.2 -

42.7 -

41 50 52 29 27 40 M NTNTNT N T NT NT KD

97.4 - . . ai ‘#{248}-RNase

66.2 -

42.7 -

Fig. I Comparison of RNase L bevels from coborectal polyps (samples 23 and 51) and tumors (samples 45, 46, 53, 41, 50, 52, 29, 27, and 40; both labeled T) to corresponding normal mucosa (N). A radioactive 2-5A probe was covalently cross-linked to RNase L under UV light in extracts of polyps, tumors, and normal mucosa prior to SDS-PAGE (see ‘ Materials and Methods’ ‘). An autoradiognam of a dried gel is shown. Tumor and polyp numbers are indicated for each pair of samples (see Table 1 for their identities). Left, relative molecular masses [in kibodabtons (KD)] of the protein markers; arrows, positions of the RNase L.

41 52 40 53 51 tieth of the amount of the first strand of eDNA synthesized was N T N T N T N TNT used for 28 rounds of PCR amplifying either RNase L on 3-actin lcD mRNA sequences. The PCR cycles included 1 mm denatunation at 94#{176}C, 1 mm annealing at 55#{176}C, and 1.5 mm extension at 97.4 - 72#{176}C,using a GeneAmp PCR Reagent Kit (Perkin Elmer/Ce- -RNase tus). A pain of primers: sense, GCCTTCATGGAAGCCGC, and

66.2 - antisense, CTFGAG1T1’GCCAATCATAG, based on the human RNase L eDNA sequence (1 1), was used to amplify mRNA sequences from nucleotides 382 to 1086, resulting in 705-bp 42.7 - PCR products. Another pair of primers: sense, TCTGGCAC-

Fig. 2 Southwestern blot assay for RNase L with radiolabeled 2-5A as CACACCFTCTAC, and antisense, AAGGCTGGAAGAGT- probe. Protein in extracts from tumors (samples 41, 52, 40, and 53) and GCCTCA (covering exons 2 and 3 of the -actin gene) was used polyp (sample 51; both labeled 1) and corresponding normal mucosa to amplify a partial 3-actin mRNA sequence (27), resulting in (N), as indicated, were subjected to denaturing SDS-PAGE, blotted to 530-bp products. PCR products were separated on 1.2% aganose nitrocellulose, and probed with a radiobabebed 2-5A analogue (see

‘‘ Materials and Methods’ ‘). This method removes endogenous bound gels and analyzed. The RNase L PCR fragments were tnans- 2-5A from RNase L, which may compete with the 2-5A probe. Label- fenned to a Nytran membrane (Schleicher & Schuell), and hy- ing is as described in the legend to Fig. 1. An autoradiogram of the bnidized with 106 cpm/ml 32P-labeled RNase L eDNA (nude- nitrocebbubose is shown. otides 63-637) labeled by random priming as described (28). Amounts of eDNA were measured with a Phosphorimagen. As a control for DNA contamination, no amplified product was detected Detection of RNase L mRNA. Total RNA was prepared in the absence of RT in PCR reactions with total cellular RNA. from tumor on normal tissue as described (20). Total RNA, 0.1 Immunoperoxidase Staining. A total of four frozen jig, was used for eDNA synthesis by RT as described by the samples [FAP (sample 51), tumor (sample 52), and correspond- supplier (GIBCO-Bethesda Research Laboratories). One-twen- ing normal tissuesi were cut into 5-jim cryostat sections and

Patient No.: 41 52 40 53 51

NT N TNTNTNT Purified RNase 12 3 4 5 6 7 8 9 10 11 12 13 14 214.2-

1 1 1 .4

RNase 1I 74.3 - : ‘ 5 - ..*

29.5 -

Fig. 3 RNase L amounts as determined in a Western blot probed with a mAb. Lanes 1-10, extracts (50 pg protein/lane) of a polyp (sample 51), tumors (samples 41, 52, 40, and 53; both labeled 7’), and normal mucosa (N), and purified recombinant human RNase L (Lanes 11-14, 10, 3, 1, and 0.3 ng, respectively) were subjected to SDS-PAGE, Western blotting, and probing with a munine mAb against human RNase L (15).

placed onto poly-L-lysine-coated microscope slides and air dried The tissues represented a range of different stages and types of for 24 h at room temperature. Immediately prior to beginning cobonectal tumors. For comparison, normal cobonectab mucosa, the immunohistochemistry procedure, the sections were fixed in distal to the tumors, from the same patients was collected. cold acetone for 10 mm. Sections were then rinsed in PBS and To measure RNase L levels in the tumors, we performed incubated in normal horse serum (Vector Laboratories) for 1 h covalent cross-linking of a bromine-substituted and 32P-babeled followed by incubation in the ascites preparation of the mAb (a 2-5A probe to the RNase under UV bight (24). Based on this titer of 1 :1000) for 1 h at 24#{176}C.Subsequently, the sections were highly specific assay, 17 of 20 tumors and polyps showed incubated in secondary biotinybated antibody (1:200; Vector elevated levels of RNase L compared with normal mucosa Laboratories, ABC Elite kit) for 30 mm followed by a 30-mm removed from the same patients (Fig. 1 and Table 1). Nine incubation in Vectastain Elite ABC reagent (Vector Labonato- tumors contained 4- to 10-fold elevated amounts, and six nies). Sections were then washed in buffer, and the reaction was showed 1.5- to 2.5-fold increased amounts, whereas two benign developed in 3-amino-9-ethyl canbazole for 10 to 30 mm. See- pobyps from FAP patients showed 6-fold elevated amounts of tions were then counterstained with hematoxylin, mounted with RNase L. An autoradiognam showing the labeling of RNase L aqueous mounting solution, and examined. by the 2-SA probe from 10 sets of tumor and normal samples

Competition binding controls were performed with recom- are shown in Fig. 1 . There was no apparent correlation between binant human RNase L produced from a bacubovirus vector in the site of the tumor and the levels of RNase L. Indeed, even the SF21 insect cells (14). The control insect cell extracts or the benign pobyps contained enhanced amounts of the protein. insect cell extracts containing recombinant human RNase L (each Therefore, the increase in bevels of RNase L is a very early event at 1.25 mg/ml) were added to the mAb and allowed to pneincubate in tumonigenesis. for 2 h at 24#{176}Cprior to application to the tissue sections for 1 h. The To determine whether endogenous 2-SA was present in purpose of this control was to inhibit the reaction competitively to sufficient amounts to reduce binding of the 2-SA probe to the confirm that staining was specific for RNase L RNase L, denaturing gel ebectrophoresis was performed, fob- bowed by a Southwestern blotting procedure (Fig. 2; Ref. 1 1). RESULTS This method, which removes 2-SA that may be pnebound to RNase L Levels in Human Coborectal Tumors and Pob- RNase L, also showed that the relative bevels of RNase L were yps. The collection of colorectal tumor samples used in this significantly higher in the tumor and polyp extracts than in the study included 18 adenocarcinomas isolated from different sites, corresponding normal mucosa (Fig. 2). Therefore, the elevated 8 of which were from patients with distal metastasis (Table 1). amounts of 2-SA binding activity in the tumor and polyp In addition, polyps from two patients with FAP were analyzed. extracts were due to increased levels of RNase L itself.

41 52 29 50 N TNT NT NT

I.- RNase 705bp

1: ------.10-- 3-actin 530bp

Fig. 5 Relative bevels of RNase L mRNA and 3-actin mRNA from coborectab tumors (7) and corresponding normal mucosa (N) as determined by RT-PCR. Arrows, RT-PCR products of RNase L mRNA (705 hp) and 3-actin mRNA (530 bp).

, .

S mutant forms of RNase L (5), RNase assays were performed with RNase L, which was isolated from the tumor and normal mucosa extracts on the activating affinity matrix, 2-5A-cellulose (26). Previously, we showed that poly(U)-3’[32P]pCp was

‘ . S. degraded to discrete obigomens of unidylate by the human ne- combinant RNase L (Fig. 4, lane 8; Ref. 14). A similar result was seen with RNase L isolated from three separate tumors and activated with 2-5A-cebbubose (Fig. 4). There was significantly less RNase L activity from the corresponding normal tissues. Therefore, both the 2-5A binding and the RNase activities of the iif#{149} RNase L were elevated in the tumors. RNase L mRNA Levels Are Elevated in the Tumors. To determine whether the increased amounts of RNase L were due to an accumulation of its mRNA, RT-coupbed PCRs were performed (Table 1 and Fig. 5). For comparison, we also am- 1234567! plified cDNA from 3-actin mRNA. Enhanced amounts of RNA 40N 40T 52N 52T 53N 53T RNase RNase L mRNA were observed in the same tumors that showed Fig. 4 RNase activities of RNase L isolated from tumors and normal elevated amounts of RNase L activity (Table 1 and Fig. 5). mucosa. The RNase L isolated on an activating affinity matrix, 2-5A- Therefore, we conclude that in these tumors either increased cellulose, was used to cleave nadiolabebed poly(U) (see ‘‘ Materials and Methods’ ‘). Lane 1, the input poly(U) alone; Lanes 2-7. tumor and transcription of the RNase L gene and/on enhanced stability of normal mucosa samples as indicated; Lane 8, recombinant human the RNase L mRNA were responsible for the increased amounts RNase L. An autoradiogram of a sequencing gel is shown. of RNase L in the tumors. Preliminary analysis of the RNase L gene by Southern blotting did not show any measurable alter- ations, suggesting that rearrangement of the gene was probably not involved in the overexpnession observed in the tumors (data Amounts of RNase L Were Elevated in the Tumors as not shown). Determined with a mAb. Because activity and absolute Immunobocalization of RNase L in Colorectal Tumors amounts of enzymes are not always directly proportional, for and Normal Mucosa. Samples of polyp, tumor, and normal instance, as a result of posttranslational controls, we also mea- mucosa were sectioned and subsequently probed with the mAb sured RNase L amounts on Western blots with an mAb (15). (Fig. 6). Adjacent normal tissue had minimal staining confined Levels of RNase L as measured with the antibody were signif- to the cytoplasm, mostly in the areas of the base of the cnypts of icantly elevated in four of five sets of tumors compared with several villi (Fig. 6, A and C). This result seems to be consistent matched sets of normal tissue (Fig. 3). In general, these results with the basal levels of endogenous RNase L activity in normal were consistent with the 2-SA binding data, although in tumor samples. The polyp and tumor specimens (Fig. 6, B and D, 53, the amount of RNase in the tumor appeared to be greaten as respectively) revealed significantly more pronounced staining in determined by the 2-5A binding assays than in the immunobbot the base of the villi as well as other glands in the section, yet (compare Figs. 1 and 3). The purified, recombinant RNase L inconsistent staining was apparent in the tumor cells themselves. was included in the same Western blot for comparison (Fig. 3, The tumor stained was histopathologically identified to be a Lanes 11-14). These findings show that the amounts of RNase well-differentiated neoplasm, and the polyp was a FAP. L protein were elevated in the tumors. As a control for specificity, the staining was prevented by The RNase L in the Tumor Extracts Was Catalytically pneincubating the antibody with an extract of insect cells ex- Active in the Presence of 2-.5A. Because the 2-SA binding pressing recombinant human RNase L (Ref. 14; Fig. oF). In assay may not always discriminate between wild-type and some contrast, an extract of control insect cells lacking recombinant

,. 4’ ,

,y . , -.

. #{149}1 .:#{149}‘ ;:‘r t .

RNase L did not inhibit reactivity with the antibody (Fig. 6, 2-SA-antisense strategy selectively ablates target RNA in intact compare E and G). Addition of normal horse serum in place of cells (31). Because pobyps and coborectrab tumor cells have high the mAb (negative control) showed no evidence of staining (Fig. bevels of RNase L, these cells may be especially sensitive to this 611). approach.

DISCUSSION ACKNOWLEDGMENTS Here, we have measured elevated amounts of RNase L We thank Paul F. Tonnence (Bethesda, MD) for the gift of mRNA and protein and activity in coborectab tumors developed p(A2’p)2(br8A2’p)2A. in various regions of the colon with and without metastasis at distal organs. Tumors were from both men and women between REFERENCES the ages of 35 and 80 years. There was no clean correlation 1. In: J. Belasco and G. Brawerman (eds.). Control of Messenger RNA between the amounts of RNase L and tumor site. Because high Stability. New York: Academic Press, 1993. amounts of RNase L were also observed in two benign pobyps, 2. Kerr, I. M., and Brown, R. E. pppA2’p5’A2’p5’A: an inhibitor of the elevation in RNase L levels may be important in tumonigen- protein synthesis synthesized with an enzyme fraction from interferon- esis and may be a relatively early event. The implication is that treated cells. Proc. Natb. Acad. Sci. USA, 75: 256-260, 1978. there may be a requirement for enhanced RNA turnover in the 3. Chebath, J., Benech, P., Revel, M., and Vigneron, M. Constitutive rapidly growing cells that develop into cancer cells. The robe of expression of (2-5’) obigo A synthetase confers resistance to piconna- virus infection. Nature (Land.), 330: 587-588, 1987. the 2-SA-system in controlling RNA turnover in cobonectal 4. Rysiecki, G., Gewert, D. R., and Williams, B. R. G. Constitutive tumors is unknown. Perhaps RNA turnover rates are limiting for expression of a 2’,5’-oligoadenylate synthetase cDNA results in in- cell growth in normal cobonic epithebial cells. For instance, creased antiviral activity and growth suppression. J. Interferon Res., 9: enhanced breakdown of RNA could lead to a rapid replenish- 649-657, 1989. ment in pools of nucleotides for the synthesis of mRNA. Alter- 5. Hassel, B. A., Zhou, A., Sotomayon, C., Maran, A., and Silverman, nately, perhaps the RNase L is responsible for selectively en- R. H. A dominant negative mutant of 2-SA-dependent RNase sup- hancing the turnover of mRNA encoding tumor suppressors and presses antiprolifenative and antivinal effects of interferon. EMBO J., 12: 3297-3304, 1993. other growth-restricting proteins. Regardless of the physiological significance, the high bevel 6. Hovanessian, A. G., Brown, R. E., and Kern, I. M. Synthesis of low molecular weight inhibitor of protein synthesis with enzyme from in- of RNase L bed us to suggest a novel therapeutic approach for tenferon-treated cells. Nature (Lond.), 268: 537-539, 1977. polyps and colorectal tumors. Stable 2-5A analogues with the 7. Williams, B. R. G., Gobghen, R. R., Brown, R. E., Gilbert, C. S., and ability to be both taken up by cells and to activate RNase L Kern, I. M. Natural occurrence of 2-SA in interferon-treated EMC could cause the regression of pobyps and tumors, which contain virus-infected L cells. Nature (Lond.), 282: 582-586, 1979. high bevels of the enzyme. In contrast, normal cobonic mucosa, 8. Williams, B. R. G., Kern, I. M., Gilbert, C. S., White, C. N., and Ball, which contain only low bevels of RNase L, may be spared on L. A. Synthesis and breakdown of pppA2’p5’A2’pS’A and transient only transiently affected. For example, it was shown that facib- inhibition of protein synthesis in extracts from interferon-treated and control cells. Eur. J. Biochem., 92: 455-562, 1978. itated uptake of 2-SA by human cells has a potent antignowth 9. Clemens, M. J., and Williams, B. R. G. Inhibition of cell-free protein affect (29). Anticellular properties of 2-SA are presumably due synthesis by pppA2’p5’A2’p5’A: a novel obigonucleotide synthesized to the induction of general, rapid RNA decay, including that of by interferon-treated L cell extracts. Cell, 13: 565-572, 1978. rRNA. Alternatively, 2-SA-antisense chimenas could be used, 10. Slatteny, E., Gosh, N., Samanta, H., and Lengyeb, P. Interferon, which selectively direct RNase L to mRNA-encoding oncogenic double-stranded RNA, and RNA degradation: activation of an endonu- proteins on growth factors, which promote the growth of cob- clease by (2’-5’)A,. Proc. Natl. Acad. Sci. USA, 76: 4778-4782, 1979. rectal carcinomas (30). Previously, we have shown that the 11. Zhou, A., Hassel, B. A., and Silverman, R. H. Expression cloning of 2-5A-dependent RNase: a uniquely regulated mediator of interferon action. Cell, 72: 753-765, 1993. 12. Wreschner, D. H., McCauley, J. W., Skehel, J. J., and Kern, I. M. Interferon action-sequence specificity of the ppp(A2’p)nA-dependent nibonuclease. Nature (Lond.), 289: 414-417, 1981.

Fig. 6 (opposite page) Immunobocalization of RNase L in polyp, 13. Floyd-Smith, G., Slattery, E., and Lengyel, P. Interferon action: tumor, and normal mucosa sections. A munine mAb against recombinant RNA cleavage pattern of a (2-5’) obigoadenylate-dependent endonucle- ase. Science (Washington DC), 212: 1020-i032, 1981. human RNase L was used to localize RNase L (see ‘‘ Materials and Methods”; Ref. 15). Immunolocalized RNase L appears as reddish- 14. Dong, B., Xu, L., Zhou, A., Hassel, B., Lee, X., Torrence, P., and brown material. A, normal cobonic specimen with minimal staining Silverman, R. H. Intrinsic molecular activities of interferon-induced (sample 51); B, FAP specimen demonstrating cytoplasmic staining 2-SA-dependent RNase. J. Biol. Chem., 269: 14153-14158, 1994. (sample 51); C, normal colonic specimen with minimal staining (sample 15. Dong, B., and Silverman, R. H. 2-5A-dependent RNase molecules 52, normal); D, tumor demonstrating cytoplasmic staining in epithelial dimenize during activation by 2-SA. J. Biob. Chem., 270: 4133-4137, components yet no staining in adjacent lamina propnia (sample 52); E, 1995. tumor specimen (sample 52) stained with a mAb preincubated with a control insect cell extract revealing a similar degree of staining as 16. Squire, J., Zhou, A., Hassel, B. A., Nie, H., and Silverman, R. H. previous tumor samples (a control for F); F, tumor specimen (sample Localization of the interferon-induced, 2-5A-dependent RNase gene 52) stained with a mAb preincubated with an extract of insect cells (RNS4) to human chromosome 1q25. Genomics, 19: 174-175, 1994. expressing high levels of recombinant human RNase L (a control to 17. Hubbell, H. R., Kaniko, K., Suhadolnik, R. J., and Brodsky, I. demonstrate the specificity of the mAb); G, well-differentiated tumor RNase L and increased endonibonuclease activities in the mononuclear with cytoplasmic staining (sample 52); H, tumor specimen (sample 52) cells of patients with chronic myelogenous leukemia. Anticancer Res., without the addition of a mAb. A-D and H, X72.S; E-G, X 145. 14: 341-346, 1994.

18. Boring, C. C., Squires, T. S., Tong, T., and Montgomery, S. Cancer 25. Singh, H., LeBowitz, J. H., Baldwin, A. S., and Sharp, P. A. statistics. CA Cancer J. Clin., 3-17, 1994. Molecular cloning of an enhancer binding protein: isolation by screen- 19. Feanon, E. R., and Vogelstein, B. A genetic model for coborectal ing of an expression library with a recognition site DNA. Cell, 52: tumonigenesis. Cell, 61: 759-767, 1990. 415-423, 1988. 20. Wang, L., Patel, U., Ghosh, L., and Banenjee, S. DNA pobymenase 26. Silverman, R. H. Functional analysis of 2-SA-dependent RNase and 2-SA using 2’5’-oligoadenylate-cellubose. Anal. Biochem., 144: 450- 13mutations in human coborectal cancer. Cancer Res., 52: 4824-4827, 460, 1985. 1992. 21. Hemminki, A., Pebtomaki, P., Mecklin, J-P., Jarvinen, H., 27. Ng, S-Y., Gunning, P., Eddy, R., Ponte, P., Leavitt, J., Shows, T., Sabovaara, R., Nystrom-Lahti, M., de ba Chapeble, A., and Aaltonen, and Kedes, L. Evolution of the functional human beta-actin gene and its L. A. Loss of the wild type MLHJ gene is a feature of hereditary multi-pseudogene family: conservation of noncoding regions and chro- nonpolyposis colorectab cancer. Nat. Genet., 8: 405-410, 1994. mosomab dispersion of pseudogenes. Mob. Cell. Biol., 5: 2720-2732, 1985. 22. Liu, B., Nicolaides, N. C., Markowitz, S., Willson, J. K. V., Par- sons, R. E., Jen, J., Papadopolous, N., Peltomaki, P., de Ia Chapelle, A., 28. Maniatis, T., Sambrook, J., and Fnitsch, E. F. Molecular Cloning, a Hamilton, S. R., Kinzler, K. W., and Vogebstein, B. Mismatch repair Laboratory Manual, Ed. 2. Cold Spring Harbor, NY: Cold Spring gene defects in sporadic coborectal cancers with microsateblite instabil- Harbor Laboratory, 1989. ity. Nat. Genet., 9: 48-55, 1995. 29. Hovanessian, A. 0., and Wood, J. N. Anticellular and antivinal 23. Silverman, R. H., and Krause, D. Interferon-regulated 2-SA and effects of pppA(2’pS’A)n. Virology, 101: 81-90, 1980. protein kinase systems. in: M. J. Clemens, A. G. Morris, and A. J. 30. Tomrence, P. F., Maitna, R. K., Lesiak, K., Khamnei, S., Zhou, A., Gearing (eds.), Lymphokines and Interferons: Practical Approach, pp. and Silverman, R. H. Targeting RNA for degradation with a (2’- 149-193. Oxford: I. R. L. Press, 1987. 5’)obigoadenylate-antisense chimera. Proc. NatI. Acad. Sci. USA, 90: 24. Nolan-Sorden, N. L., Lesiak, K., Bayard, B., Tomrence, P. F., and 1300-1304, 1993. Silverman, R. H. Photochemical crosslinking in oligonucleotide-protein 31. Maran, A., Maitna, R. K., Kuman, A., Dong, B., Xiao, W., Li, G., complexes between a bromine-substituted 2-5A analogue and 2-SA- Williams, B. R. G., Tornence, P. F., and Silverman, R. H. Blockage of dependent RNase by ultraviolet lamp or laser. Anal. Biochem., 184: NF-kB signaling by selective ablation of an mRNA target by 2-5A- 298-304, 1990. antisense. Science (Washington DC), 265: 789-792, 1994.

Downloaded from clincancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research. Elevated levels of 2',5'-linked oligoadenylate-dependent ribonuclease L occur as an early event in colorectal tumorigenesis.

L Wang, A Zhou, S Vasavada, et al.

Clin Cancer Res 1995;1:1421-1428.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/1/11/1421

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://clincancerres.aacrjournals.org/content/1/11/1421. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.