[CANCER RESEARCH 60, 1070–1076, February 15, 2000]

Insulin-like Growth Factor II Supply Modifies Growth of Intestinal Adenoma inApcMin/1 Mice1

A. Bassim Hassan2 and Julie A. HowellDepartment of Zoology, University of Oxford, Oxford OX1 3PS, United Kingdom

ABSTRACT

Insulin-like growth factor-II (IGF-II) is an embryonic growth promoterand cell survival factor. IGF-II supply is normally limited by gene expres-sion because transcription occurs predominantly from the paternal allelein mouse and man (maternal imprinting). Excess IGF-II has detrimentalsystemic and local effectsin vivo, promoting somatic overgrowth and anincreased frequency of tumors.IGF2 mRNA is overexpressed in colorectaland many other human cancers. In this paper, we show that alteredIGF-II supply modifies intestinal tumor growth. Mice genetically alteredin the IGF-II system were combined in crosses withApcMin/1 , a murinemodel of human familial adenomatous polyposis. Depending on geneticbackground, ApcMin/1 acquires multiple small intestinal adenoma beforebecoming moribund with anemia. Mice that express excess IGF-II deliv-ered using a bovine keratin 10 promoter (k10Igf2/1) develop a dispro-portionate overgrowth of colon, uterus, and skin. Combination withApcMin/1 leads to a 10-fold increase in the number and the diameter ofcolon adenoma (P< 0.0001) compared toApcMin/1 littermate controls(postnatal day 80), an increased susceptibility to rectal prolapse (41%),and a histological progression to carcinoma. Mice with reduced IGF-IIsupply, secondary to the disruption of the paternalIgf2 allele (Igf21m/2p),are 60% the weight of wild-type littermates. Combination with ApcMin/1

leads to a 3-fold reduction in small intestinal adenoma number(P < 0.0001) compared toApcMin/1 littermate controls (postnatal day150), and a significant decrease in adenoma diameter (P< 0.001). With insitu hybridization, we show that Igf2 was expressed in all adenomairrespective of IGF-II supply. This suggests that there is an increasedmaternal allele expression ofIgf2 (loss of imprinting) in adenoma whichform, despite paternal Igf2 allele disruption. We conclude that IGF-IIsupply is a modifier of intestinal adenoma growth, and we provide geneticevidence for its functional role in colorectal cancer progression.

INTRODUCTION

Mouse models of human cancer providein vivo systems for theidentification of polygenic modifiers that may account for the varia-tion in penetrance, rate of tumor progression, and clinical behavior oftumors. For example, our understanding of the genetics of highlypenetrant genes such asAPC has been complemented by the investi-gation of murine models,e.g.,ApcMin/1, a close genotypic and phe-notypic model of human familial adenomatous polyposis. In familialadenomatous polyposis, multiple colonic adenomas are frequent andcommonly progress to invasive carcinomas if not treated. The adeno-mas that develop in theApcMin/1 mouse can also progress to invasivecarcinoma, but the model differs from the human syndrome becausemultiple adenomas tend to be concentrated in the small intestine,which usually leads to anemia and intestinal obstruction (1). Thenumber of intestinal adenomas is dependent on the inbred strain,which has been exploited in mapping a major modifier,Mom1(Pla2g2a) (2, 3). Although the relevance ofPla2g2ato human colonic

carcinoma progression is still under investigation (4), its discoveryenforces the notion that the identification of genetic modifiers willdefine the key interactions between biochemical systems that deter-mine tumor progressionin vivo. We have taken a complementaryapproach, which has been to use mice with a genetic alteration of acandidate modifier known to be frequently disrupted in human cancer.In this paper, we show that genetic manipulation of IGF3-II supplymodifies the growth of intestinal adenoma inApcMin/1.

IGF-II is a paternally expressed, maternally imprinted, embryonicgrowth factor that is a potent modifier of growthin vivo (5). In mice,after disruption of the paternalIgf2 allele, total body weight is reducedby 40% at birth (6). Systemic IGF-II levels fall after birth, andpostnatal growth is then regulated by the related ligand, IGF-I (5).Increased systemic availability of IGF-II, either due to biallelic ex-pression, increased delivery from a transgene, or disruption of theIGF-II/M6P receptor, results in overgrowth phenotypes in mice (7–9).Similar effects occur in human overgrowth syndromes such as Beck-with-Wiedemann, where the biallelic expression ofIGF2 results fromeither LOI or unipaternal disomy (10). Increased expression of IGF-IIin transgenic mice not only increases tumor frequency in organs thatexpress the transgene, but also at distant sites, suggesting that bothlocal and systemic supply can promote tumor progression (11, 12).

IGF-II is well known for promoting a cell number increasein vitro.The mechanism may predominantly relate to cell survival rather thancell division (13, 14). IGF-II and the related ligand, IGF-I, exert cellsurvival and growth effects via heterotetrameric IGF-I and insulinreceptors, which mediate signal transduction through the PI3 kinase/Akt pathway also modified by the recently identified phosphatase andtensin homologue deleted on chromosome ten (PTEN) tumor suppres-sor (15). Serial analysis of gene expression has identified IGF-II as themost abundant mRNA overexpressed in human colorectal cell linesand tumors compared to normal tissue (16). IGF-II is also overex-pressed in Wilms tumor, rhabdomyosarcoma, neuroblastoma, germcell tumors, adrenocortical carcinoma, breast, and hepatocellular car-cinoma (reviewed in Ref. 17). Furthermore, increased maternal alleleexpression (allele ratio of,3:1 taken as LOI) has also been detectedat surprisingly high frequency in normal human leukocytes and co-lonic mucosa (12%), particularly in cases with microsatellite instabil-ity in associated tumors (91%; Ref. 18). This significant findingsuggests that increased local IGF-II supply may predispose to thedevelopment of early onset colorectal cancer before the appearance ofeither an adenoma or tumor, especially in individuals with defects inDNA mismatch repair (19).

We examined the effect of IGF-II supply in crosses between micewith genetically altered IGF-II expression and theApcMin/1 model ofcolorectal cancer. Increased IGF-II delivery to the alimentary tractwas achieved using a bovine keratin 10 promoter-driven transgene(K10Igf2), which results in the phenotype of colon, skin, and uterineovergrowth (9). Although the predominant growth effect is in tissuesthat express the transgene, there are also subtle metabolic effects dueto increased circulating IGF-II (20). Decreased IGF-II supply in the

Received 10/21/99; accepted 12/14/99.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby markedadvertisementin accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by a grant from The Cancer Research Campaign United Kingdom 2390/0101(to A. B. H.). A. B. H. is a Cancer Research Campaign Senior Clinical Research Fellow.

2 To whom requests for reprints should be addressed, at the Department of Zoology,University of Oxford, South Parks Road, Oxford OX1 3PS, United Kingdom. Phone:44-1865-271227; Fax: 44-1865-271228; E-mail: bass.hassan@zoo.ox.ac.uk.

3 The abbreviations used are: IGF, insulin-like growth factor; IGF-IR, insulin-likegrowth factor-I receptor; LOI, loss of imprinting; SPF, specified pathogen-free; APC,adenomatous polyposis coli; TGF, transforming growth factor.

1070

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

alimentary tract was achieved using mice with a disruption of thepaternal allele ofIgf2 (Igf21m/2p) (6). Except for the leptomeningesof the brain, Igf2 mRNA expression from the maternal allele isnormally undetectable in these animals. Therefore, the source of anylocally delivered IGF-II to the alimentary tract should be derived fromincreased maternal allele expression.

MATERIALS AND METHODS

Mice. ApcMin/1 (C57Bl/6J) were obtained from the Imperial Cancer Re-search Fund. SPF colony (gift from W. Bodmer, mice established in 1992 fromA. Moser and W. Dove, McArdle laboratory, Madison, Wisconsin; Ref. 21)and bred in the host department. MaleApcMin/1 mice were backcrossed toinbred female C57Bl/6J (from Harlan, Bicester, Oxon, United Kingdom) undernon-SPF conditions for two generations before commencing experimentalcrosses.Entamoeba muriswas the only intestinal parasite detected in bothimportedApcMin/1 and C57BL/6J stock. SPF conditions do not significantlyalter adenoma number (22). Animals were housed with littermates under a 14 hlight/10 h dark cycle and were fed normal chow (3% total fat, Special DietaryServices, Witham, Essex, United Kingdom) and tap waterad libitum.

Mice with increased IGF-II supply (K10Igf2) were at.10th generationinbred onto 129/SvJ (9). Mice with disruption of the paternal allele ofIgf2(2p) but an intact maternal allele (1m, denotedIgf21m/2p) were a gift fromA. Efstratiadis and were also at.10th generation inbred onto the same129/SvJ (6). DNA was extracted from tails (day 7) and liver (at the time ofdissection). After incubation (12 h; 55°C) with 0.5mg/ml proteinase K in lysisbuffer [50 mM Tris (pH 8.0), 100 mM EDTA, 100 mM NaCl, 1% SDS] andRNase A (1 h; 37°C), DNA was extracted with phenol/chloroform, precipitatedwith ethanol, and resuspended in TE buffer [10 mM Tris (pH 7.4), 1 mM

EDTA]. Animals were genotyped for the presence ofApcMin/1, K10Igf2/1,and Igf21m/2p by established PCR protocols (2, 23). All breeding used male129/JSv IGF-II mutant mice and female C57Bl/6JApcMin/1. FemaleK10Igf2/1are poor mothers because they develop an imperforate uterus, andthe disrupted allele inIgf21m/2p is paternally inherited. All litters werecross-fostered to F1 mothers before postnatal day 7 (C57Bl/6J,CBA/Ha orC57Bl/6J,129/SvJ). All animal procedures were approved by the Home Officeof the United Kingdom government, departmental ethics committee and werecarried out in accordance with the United Kingdom Coordinating Committeeon Cancer Research guidance for the welfare of animals in experimentalneoplasia (second edition; Ref. 23).

Adenoma Scoring and Collection.Depending on age,ApcMin/1 becomemoribund as a result of chronic anemia and intestinal obstruction. After dailymonitoring of initial litters for signs of anemia and distress, the dates ofdissection of experimental crosses were refined so that animals did not sufferunduly. Small intestinal adenoma and colonic adenoma were therefore scoredfor number and diameter either at postnatal day 80 or 150 for the cross betweenApcMin/1 3 K10Igf2/1andApcMin/1 3 Igf21m/2p, respectively. The stomach,small intestine, and colon were dissected free of mesentery and opened alongthe longitudinal axis using a jig and blade designed by us to aid rapidprocessing. Intestinal contents were cleared with PBS, and the small intestinewas divided into three equal-length segments and laid open with the colon onthe absorptive side of Benchkote (Whatman). Intestines were fixed in 4% (4g/100 ml) paraformaldehyde in PBS (24 h) followed by 70% ethanol (v/v).Using a dissecting microscope (310–30) and calipers, adenoma number anddiameter were obtained for the entire length of the small intestine and colon.Adenoma analysis was performed without knowledge of genotype by oneperson (A. B. H) and confirmed independently by another (J. A. H.). Materialfor cryosections was either placed face down relative to the cutting surface forthe small intestine or rolled (for the colon), immediately embedded in Tis-sueTek (Sakura Fintek, Zoeterwoude, the Netherlands), and stored at240°C.Small intestine surface area was calculated by summation of the multiples oflength of each fixed segment by width, which was measured at the midpoint ofeach segment. Colon surface area was calculated by multiplying the length ofthe fixed material from the anorectal junction to the point of insertion of thesmall intestine, omitting the appendix, by the width at the midpoint. Statisticalanalysis is described in figure and table legends. Calculations were performedusing the Minitab 10Xtra (Minitab Inc.).

Histopathology. Distal colon samples and small intestinal segments wereparaffin-embedded and sectioned (5mm), and every fifth section was stainedwith H&E. Stained sections were viewed without knowledge of genotypes(A. B. H) and checked by an independent histopathologist (D. Rowlands,Department of Histopathology, University of Birmingham, United Kingdom).Sections for immunohistochemistry were cleared with xylene and rehydratedin an ethanol series to PBS, and endogenous peroxidases were quenched with3% H2O2 in PBS (v/v; 15 min). Sections were incubated at 4°C with primaryantibodies in PAT (PBS/0.1% BSA/0.1% Tween 20) for 16 h. The followingantibodies were used: antihuman APC (C-20) rabbit polyclonal to the COOH-terminus of human APC (amino acid 2824–2843), antihuman APC (N-15)rabbit polyclonal to the amino terminus of human APC (amino acid 2–16), andIGF-IRb (sc-713) antihuman rabbit polyclonal to a carboxy-terminal peptide,all from Santa-Cruz Biotechnology Inc. (Santa Cruz, CA). Proliferation wasassessed with the HsMCM2/BM28 rabbit polyclonal antibody to humanMCM2, a cell cycle protein marker specific for G1-S-phase cells, a gift fromI. Todorov (24, 25). Sections were blocked with 2% (v/v) horse serum in PAT(30 min), incubated with biotinylated antirabbit antibody from horse (VectorLaboratories Inc., Burlinghame, CA), and visualized with a peroxidase ABCVector Elite kit with 3,39-diaminobenzidine substrate. Sections were dehy-

Fig. 1. Effect of IGF-II supply on whole body weight (g). Male and female mice wereweighed at different times after birth until the time of dissection, and results were pooledfor each genotype (error bars,6 SEM). A, increased IGF-II supply usingK10Igf2/1transgene combined withApcMin/1. B, decreased IGF-II supply usingIgf21m/2p com-bined withApcMin/1. n 5 number of mice per genotype.

1071

IGF-II SUPPLY AND INTESTINAL ADENOMA GROWTH

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

drated, counterstained with methyl-green, washed in acetone plus 1% (v/v)acetic acid, and mounted.

In Situ Hybridization. DNA containing the distal coding region and 39 un-translated region ofIgf2 exon 6 was amplified from a pGEM plasmid containingIgf2 cDNA (derived from a gift from P. Rotwein). Primers BH1, 59-NNG AGCTCA GCC TCT TCG GAG ATG TC-39 (position 541–558, M14951) and BH2,

59-NNG GTA CCA ACA GCC TGA TGT GGG GA-39 (position 738–721,M14951) amplified a 198-bp fragment. Sequence analysis confirmed the fragmentwas mouseIgf2. EngineeredSacI andKpnI restriction sites in primers BH1 andBH2, respectively (italicized) were used to clone digested gel-purified product intopBluescript KS-. Linearized plasmid was used to transcribe sense and antisenseprobe with digoxigenin-UTP (Roche Diagnostics Ltd, Essex, United Kingdom).Ten-mm cryosections were mounted on adhesive-treated slides (3-aminopropyl-triethoxy-silane), were dried with a gentle heat from a hair dryer for 2–3 min, andwere immediately placed in fresh 4% (w/v) paraformaldehyde/PBS (diethylpyro-carbonate-treated) at 4°C for 20 min. To quench the endogenous alkaline phos-phatase, incubation in 2N HCl for 20 min at 4°C significantly reduced background(26). Slides were washed once in PBS, passed through a PBS/ethanol dilutionseries to 100% ethanol, and air dried. Slides were hybridized with 10 ng ofdenatured probe/section at 25°C for 16 h in filtered hybridization buffer (10%;w/v) dextran sulfate, 23 SSC, 50% (w/v) formamide, 10% (w/v) SDS, 1 mM DTT,10mg/ml salmon sperm, 10mg/ml tRNA, and 20 units of RNase inhibitor (Sigma,United Kingdom). Sections were washed with 13 SSC [150 mM NaCl, 15 mM

sodium citrate (pH 7.0)], O.13 SSC/50% (v/v) formamide (23 5 min), andincubated with 20mg/ml RNase A for 30 min at 37°C. Digoxigenin RNA wasvisualized with an antidigoxigenin-alkaline phosphate Fab fragment (Roche Di-agnostics Ltd, Essex, United Kingdom) using nitro-blue tetrazolium chloride and5-bromo-4-chloro-3-indoyl-phosphate staining with 0.5mg/ml levamisole inbuffer [0.1M Tris-HCl (pH 9.5), 0.1M NaCl, 50 mM MgCl2] containing 10% (w/v)polyvinyl alcohol (30–70 kDa) for 16 h at 25°C. Sections were counterstainedwith 0.02% fast green and viewed with a JVC color charge coupled device cameraattached to a Leica upright microscope. Control experiments using a sense probe,excess unlabeled antisense probe, predigestion with RNase A, tissue fromIgf21m/

2p, and omission of alkaline phosphatase Fab fragment all resulted in a back-ground signal. Sections from anIgf2 transgene-derived mammary tumor acted asa positive control in every experiment.

RESULTS

Minor Modifier of Apc Min/1 in 129/SvJ. We first established thetotal adenoma count in litters from crosses between C57Bl/6JApcMin/1

and inbred stock 129/SvJ mice (coisogenic). Mean counts revealed a2-fold reduction in adenoma number at 100 days (14 adenoma;n 5 5)compared to C57Bl/6JApcMin/1 controls (31 adenomal;n 5 7), whichis consistent with the presence of a suspected semidominant modifierin the 129 strain (27).

Increased IGF-II Supply Increases Number, Diameter, andMalignant Progression of Colon Adenoma.The bovine keratin 10promoter used to deliverIgf2 mRNA (K10Igf2/1) was previouslyfound to target transgene expression to the suprabasal layers of theskin, alimentary canal, and uterus (9, 28). To our knowledge, this isthe only transgene available that increases IGF-II supply in the colon,

Fig. 2. Effect of IGF-II supply on number of adenoma. The number of adenoma in thesmall intestine and colon and the combined total adenoma count per mouse were pooled foreach genotype after dissection (day 80 and day 150 forApcMin/1 3 K10Igf2/1andApcMin/

1 3 Igf21m/2p cross, respectively). Only genotypes that developed adenoma are shown forclarity (i.e.,with ApcMin/1). Results are expressed asbox plots:box,interquartile range;cross,median;verticle line,95% confidence interval.n 5 number of animals.NS,not significant.p,P , 0.05;pp, P , 0.01;ppp, P , 0.001;pppp, P , 0.0001 (Mann-Whitney).

Table 1 The effect of IGF-II supply on intestinal growth and adenoma number

Intestines were dissected free, opened along the longitudinal axis, cleaned in PBS, and fixed (4% paraformaldehyde). Surface area was calculated by multiplying length by widthat midpoint of fixed small intestine and colon. Adenoma were visualized by using a dissecting microscope (310–30). Values are means6SD. Statistical comparison between genotypeswith normal and altered IGF-II supply utilized Mann Whitney test.

Genotype IGF-II supply n

Surface area (cm2) Adenoma (cm2)

Small intestine Colon Small intestine Colon

ApcMin/1 3 K10Igf2/1Apc1/1, 1/1 Normal 22 27.06 3.8 7.76 2.1Apc1/1,K10Igf2/1 Increase 18 28.96 3.3a 8.96 1.6b

ApcMin/1,1/1 Normal 24 25.66 2.8 6.16 1.8 0.536 0.3 0.126 0.2ApcMin/1,K10Igf2/1 Increase 17 29.76 3c 8.96 1.4c 0.636 0.4a 1.36 0.6c

Ratio (ApcMin/1,K10Igf2/1:ApcMin/1,1/1) 1.16 1.46 1.18 10.84ApcMin/1 3 Igf21m/2p

Apc1/1,Igf21m/1p Normal 16 27.46 3.2 6.16 0.8Apc1/1,Igf21m/2p Reduced 8 19.56 2.1d 4.86 0.6d

ApcMin/1,Igf21m/1p Normal 18 26.46 3.7 5.86 0.9 1.216 0.8 0.46 0.5ApcMin/1,Igf21m/2p Reduced 14 216 1.8c 4.66 0.5c 0.46 0.2c 0.56 0.6a

Ratio (ApcMin/1,Igf21m/2p: ApcMin/1, Igf21m/1p) 0.79 0.79 0.33 1.25a Not significant.b P , 0.05.c P ,0.001.d P ,0.01.

1072

IGF-II SUPPLY AND INTESTINAL ADENOMA GROWTH

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

although a similar phenotype has recently been described for anactin-IGF-I transgene (29). Overgrowth of the colon can result inrectal prolapse inK10Igf2/1 mice that are.6 months old, yet nointestinal epithelial tumors have been observed, even in the highestexpressing line used (“Blast” line). However, small raised pale polypsin the distal colon can be seen in these animals (average of 3.5 polyps,.2 mm in diameter/mouse colon at postnatal day 80;n 5 18).Histological analysis revealed mucosal collections of lymphocytes(not shown). Therefore, all colon polyps were checked by H&E-stained paraffin-embedded sections and cryosections, and only non-lymphoid adenoma counts were reported. AlthoughK10Igf2transgeneexpression has not been resolved with respect to the four separate celltypes of the crypt (seein situhybridization), there is evidence for bothsmooth muscle thickening and increased crypt depth in the colon,indicating overgrowth in both compartments (30). Apart from sys-temic delivery from the blood stream, a further source of IGF-II maybe release from the stomach into the lumen and distal delivery to thesmall intestine and colon. Both IGF-I and IGF-II may increase mu-cosal cell growth after intraluminal supply (31).

Mice that develop intestinal adenoma and increased IGF-II supply inthe colon (ApcMin/1,K10Igf2/1) lose weight (Fig. 1A), rapidly becomeanemic, and commonly develop rectal prolapse by 80 days (41%, 7/17compared toK10Igf2/1alone 6%, 1/18). To counter the possibility thataltered adenoma growth was simply a reflection of an alteration of smallintestine and colon growth, we corrected for surface area measured infixed tissue. Adenoma number and diameter were expressed either with-out correction (Fig. 2) or with correction for surface area (Table 1). Thenumber of adenoma increased in the colon (P, 0.0001; Fig. 2), evenwhen correcting for increased colon growth (P , 0.001; Table 1). Thediameter of the adenoma also increased disproportionately relative to theincreased colon growth (Table 2). Examination of dissected colons re-vealed large distal adenomas (Fig. 3A). Histopathological features withineach adenoma showed a spectrum of changes, with increased progressionto carcinoma in situand invasion in a significant proportion (Fig. 3B;Table 3). Only one polyp-like lesion was seen in the uterus of femaleApcMin/1,K10Igf2/1 (n 5 7 animals); no mammary, skin, or stomachtumors were observed.

The majority of small intestinal and colonic adenoma.2 mm showedreduced staining for Apc using a COOH-terminal monoclonal antibody(25/31 adenoma from 13 animals),e.g., Fig. 4D. NH2-terminal APCantibody stained all adenoma. There were no differences in Apc stainingrelated to IGF-II supply. Smaller adenoma tended to retain Apc stainingin some cells located within the adenoma rather than on its surface (Fig.4A). Proliferation visualized with an anti-MCM2 antibody (BM-28) wasconfined to adenoma and basal crypts (Fig. 4,B and E). Again, nodifferences in ratio of labeled:unlabeled nuclei (proliferation index) ineach high power field were detected between adenoma of different

genotypes (not shown). The distribution of IGF1Rb was concentrated inthe smooth muscle layer, the upper luminal surface of villi, and the upperzone of small intestine and colonic crypts (Fig. 4C). Large .2-mmcolonic adenoma generally appeared to have a ramifying network ofstaining confined to the stromal cell compartment, rather than stainingexclusively confined to adenoma cells (Fig. 4F). However, although aproportion of adenoma had increased staining (18/31 adenoma from 13animals), some showed low level staining with anti-IGF1Rb. No differ-ences between genotypes were detected.

Reduced IGF-II Supply Reduces Number and Diameter of SmallIntestine Adenoma. Mice with reduced IGF-II supply (Igf21m/2p) were60% of the weight of wild-type littermates throughout postnatal life, aspreviously described (6, 32; Fig. 1). There was significant reduction inadenoma number and diameter inApcMin/1, Igf21m/2p in the smallintestine at 150 days, even allowing for the reduction in small intestinalgrowth (P, 0.001; Fig. 2; Tables 1 and 2). The reduction in adenomasize was most pronounced when comparing adenoma with diameter.2mm, suggesting that reduced IGF-II supply limits early adenoma pro-gression as well as total number (Table 2). There were too few adenomain the colon to detect a similar trend. Histological comparison of smallintestinal adenoma fromApcMin/1,Igf21m/2p with ApcMin/1,Igf21m/1p

revealed similar histological features, with no differences independent ofadenoma size (not shown).

Igf2 Is Expressed in ApcMin/1 Adenoma, and Maternal Igf2 AlleleIs Expressed in ApcMin/1 ,Igf21m/2p. Igf2 in situhybridization of adultwild-type C57Bl/6J villi, crypts, and smooth muscle layers of smallintestine and colon showed only background signal.In situhybridizationin K10Igf2/1(Blast) revealedIgf2 expression in the upper two thirds ofthe crypts of the stomach and colon and associated low level signal insmooth muscle layers (Fig. 4I).4 In situ hybridization in adenoma re-vealed an increased signal inApcMin/1,Igf21m/1p (17/21 small intestinaladenoma from six animals),ApcMin/1, K10Igf2/1(10/12 colon adenomafrom six animals), andApcMin/1, Igf21m/2p (9/13 small intestinal ade-noma from four animals; Fig. 4,G-I). Signal intensity appeared similarirrespective of genotype in parallel processed slides. We presume thatmRNA degradation accounts for the failure to detect signal in all ade-noma because frozen section samples kept for longer than 3 months (at220°C) frequently showed background signal.

DISCUSSION

Our results demonstrate that intestinal adenoma inApcMin/1 ex-press IGF-II mRNA and IGF1Rb. Genetic manipulation of IGF-IIsupply significantly modified adenoma growth inApcMin/1. IncreasedIGF-II supply led to a disproportionate increase in adenoma number,

4 Unpublished data.

Table 2 The effect of IGF-II supply on adenoma diameter

Intestines were processed as for Table 1 and adenoma visualized with a dissecting microscope (310–30) and size determined using fine calipers. Genotypes withApcMin/1 areshown only. Values are means6 SD. Statistical comparison between genotypes with normal and altered IGF-II supply was done with the Mann Whitney test.

Genotype IGF-II supply n

Adenoma small intestine Adenoma colon

,2 mm 2–4 mm .4 mm ,2 mm 2–4 mm .4 mm

ApcMin/1 3 K10Igf2/1ApcMin/1,1/1 Normal 24 9.56 5.7 3.66 3.3 0.66 0.7 0.36 0.5 0.26 0.5 0.26 0.5ApcMin/1,K10Igf2/1 Increase 17 12.86 9.7a 4.96 3.9a 1.06 1.2a 5.86 4.6b 3.66 3.5b 2.36 2.7b

Ratio (ApcMin/1,K10Igf2/1:ApcMin/1, 1/1) 1.35 1.36 1.67 19.34 18.0 11.5ApcMin/13Igf21m/2p

ApcMin/1,Igf21m/1p Normal 18 10.56 6.7 15.36 15.3 5.96 5.3 0.66 0.7 0.66 1.4 0.66 1.2ApcMin/1,Igf21m/2p Reduced 14 4.56 3.5c 2.16 2.2b 0.76 0.9c 0.56 0.8a 0.66 1.0a 0.86 2.4a

Ratio (ApcMin/1,Igf21m/2P:ApcMin/1,Igf21m/1p) 0.43 0.14 0.12 0.83 1 1.3a Not significant.b P , 0.0001.c P , 0.001.

1073

IGF-II SUPPLY AND INTESTINAL ADENOMA GROWTH

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

suggesting either an increased rate of adenoma initiation or an in-creased rate of early adenoma progression soon after initiating muta-tion. The effect was greater in the colon rather than the small intestine,reflecting the distribution of transgene expression used to deliver extraIGF-II (9). K10Igf2/1 transgene expression is relatively low in thesmall intestine compared to the colon, and small intestinal adenoma,number, and size were not disproportionately increased in mice withthe combined genotype (ApcMin/1,K10Igf2/1). This observationmakes it unlikely that systemic IGF-II levels significantly alteredadenoma growth. InApcMin/1,K10Igf2/1 mice, rectal prolapse wasprominent. We note that rectal prolapse has also been observed inmice that also develop colorectal cancer due to a homozygous dis-ruption of Smad3 (33) and in clinical cases where rectal prolapse maylead to a 4-fold increased risk of colorectal carcinoma (34).

Decreased IGF-II supply limited the number of adenoma in the smallintestine, again suggesting either a reduced rate of adenoma initiation or adecrease in early adenoma progression. The disruption of theIgf2 paternalallele had a clear effect on the number of large adenoma, suggesting thatIGF-II influences the growth of established adenoma. The fact that adenomaappeared in the absence of paternalIgf2 expression may be explained byeither the selection for the autocrine expression ofIgf2 from the maternal

allele or by the expression of an alternative growth factor. It is improbablethat adenoma growth will depend on a single growth factor such as IGF-II.However, results fromIgf2 in situhybridization show adenoma-specific Igf2expression and support selection for increased maternal allele expression.Similar findings have been described by Christoforiet al. (35, 36) in apancreatic tumor model using SV40 T-antigen expression from a rat insulinpromoter (RIP-Tag) and in both TGFa and SV40 T-antigen-induced hepa-tocellular carcinoma models (37, 38). However, although SV40 T-antigen-

Table 3 Histological evaluation of colonic adenoma from (ApcMin/1 3 K10Igf2/1)cross

Histopathology classification

Genotype

ApcMin/1,1/1na 5 8% (a/A)

ApcMin/1,K10Igf2/1n 5 17% (a/A)

Adenoma 100 (11/11) 100 (46/46)1 .50% carcinoma in situb 18 (2/11) 41 (19/46)1 Stromal expansion/invasionc 9 (1/11) 28 (13/46)a n, number of animals; a, number of adenoma with feature; A, total adenoma number.b Area occupying a representative 5mM section through center of adenoma.c Area visualized in any section through an adenoma.

Fig. 3. A, dissected, cleared, and fixed colons with genotypesApcMin/1

(Min/1),k10Igf2/1 (K10Igf2/1) and ApcMin/1,K10Igf2/1 (K10Igf2/Min). Ar-row, multiple distal adenoma (all subsequently confirmed by histology).B,H&E-stained section of adenoma fromA showing disruption of normal glandulararchitecture and basement membrane, with early invasion (arrow).Bar (B),1 mm.Inset,100mm.

1074

IGF-II SUPPLY AND INTESTINAL ADENOMA GROWTH

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

induced hepatocellular carcinoma showed reduced tumor size in combinationwith Igf21m/2p, increased maternal allele expression was rarely found. In-creased IGF-II supply in tumors with intactIgf2alleles appeared to be due toselection for paternal allele disomy and maternal-specific LOH (38). Studiesof intestinal adenoma growth in mice with homozygous disruption ofIgf2arein progress (ApcMin/1,Igf22m/2p).

The effects of increased and decreased IGF-II supply support our viewthat IGF-II supply is a modifier of adenoma number and progression.How IGF-II alters adenoma number is not known, but the mechanismscould include either enhanced survival of cells that have lost the normalApc allele (ApcMin/2, LOH) or via modification of an increased cryptfission rate detected in the developing intestine ofApcMin/1 (39). Themitogenic and apoptotic functions ofC-MYC,transcriptionally up-regu-lated as a result ofAPC dysfunction (40), often require addition ofsurvival factors, such as IGF-II, particularly in c-myc-induced, p53-dependent, cell death transduced by p19ARF (14, 41). Mechanisms ofhow IGF-II acts as a survival factor include phosphorylation and inacti-vation of Bad, which normally antagonize Bcl-2 blockage of cytochromec release (42). IGF-IR-mediated cell survival functions may also beinfluenced by systemic levels of the growth hormone-controlled ligand,IGF-I, because an increased frequency of colonic carcinoma can occur inacromegalic patients with excess IGF-I (43).

Patients with microsatellite instability in colorectal tumors developLOI of IGF2, which may promote growth of colon tumors. However, it

is not known whether LOI in this circumstance has significant functionalconsequences in terms of increasing the probability of developing earlyonset colorectal cancer. However, our experiments highlight the potentialimportance of this observation and of increased local IGF-II expressiondue to LOI in normal human colonic tissue (18). It is not known whetherincreased IGF-II supply in normal colon mucosa predisposes to colonicadenoma without mutation of APC or whether IGF-II supply contributesto the development of polyclonal adenoma via a paracrine/communityeffect (44). We found no obvious increase in the proportion of adenomawith normal Apc staining in mice with increased IGF-II supply.

Excess IGF-II expression is not the only perturbation of growthfactor pathways in colorectal cancer. Frequent mutations can occur ingrowth factor receptor genes in human tumor-associated mismatchrepair defects,e.g., the TGFb type II receptor (45) and the IGF-II/M6P receptor (46). In addition to mutation of the type II TGFbreceptor, Smad3 and Smad4 transducers of the TGFb pathway arealsomutated in human tumors and result in increased malignant progressionof intestinal tumors after disruption of murine genes (33, 47).

Igf2 expression and subsequent autocrine/paracrine growth effectsmust offer adenoma cells a selective advantage. Our data provide exper-imental support for mathematical models concerning natural selection ofexpanding tumor cell clones expressing autocrine cell survival factors(48). It is clear that IGF-II supply is tightly regulated in normal tissue,with expression predominantly from one allele during embryonic growth

Fig. 4. Analysis of adenoma tissue.A-F, immunohistochemistry of a small intestinal early adenoma (A-C) and colon adenoma (D-F) fromApcMin/1. Apc staining is commonly seenwithin small adenoma but absent from large adenoma (arrow;D), except cells forming edge of adenoma (arrow head;D). Proliferation occurs in adenoma as assessed by an antibodyto human MCM2 (BandE; arrow). Crypt cells also label (arrowhead;B). IGF-IR (anti-IGF1Rb) labeling occurs in crypts, villi, and smooth muscle (arrowhead;F). Labeling alsooccurs in cells at the edge of adenoma (arrowhead;C) and as a ramifying network within adenoma (arrow;F). G-I, in situhybridization with theIgf2 antisense probe and sense probe(left lower insets). Adenoma from the small intestine ofApcMin/1 (arrow; G); ApcMin/1,Igf21m/2p(two adenoma; arrows;H); andApcMin/1,k10Igf2/1(large arrow;I) all show a strongsignal. Colon crypts, and to a lesser extent, smooth muscle, show a signal inApcMin/1,k10Igf2/1(small arrow; I). Bars,100 mm.

1075

IGF-II SUPPLY AND INTESTINAL ADENOMA GROWTH

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

in both the human and mouse (6, 49). The addition of a single-expressedIgf2 allele results in overgrowth, and reduced IGF-II supply results inreduced embryonic growth (240%). We and others have shown thatIGF-II is an important regulator of murine tumor growth, both in earlyadenoma and in the progression to carcinoma (35–38). However, this isthe first demonstration of the influence of IGF-II supply on intestinaltumor growth in a murine model that closely mimics a human colorectalcancer syndrome and is independent of the SV40 T antigen. We concludethat IGF-II supply is a potent modifier of intestinal tumor growth and thatIGF-II may subsequently prove to be an important target for humancolorectal cancer therapy.

ACKNOWLEDGMENTS

We thank Chris Graham (advice and reading the manuscript), Silvio Zaina(advice), David Rowlands (histopathology), Viv Clarke and Jenny Corrigan(technical assistance), Jane Morrice (plasmids), and Julie Bee (establishingMin mouse).

REFERENCES

1. Shoemaker, A. R., Gould, K. A., Luongo, C., Moser, A. R., and Dove, W. F. Studiesof neoplasia in the Min mouse. Biochim. Biophys. Acta,1332: F25–48, 1997.

2. Dietrich, W. F., Lander, E. S., Smith, J. S., Moser, A. R., Gould, K. A., Luongo, C.,Borenstein, N., and Dove, W. Genetic identification of Mom-1, a major modifier locusaffecting Min-induced intestinal neoplasia in the mouse. Cell,75: 631–639, 1993.

3. Cormier, R. T., Hong, K. H., Halberg, R. B., Hawkins, T. L., Richardson, P., Mulherkar,R., Dove, W. F., and Lander, E. S. Secretory phospholipase Pla2g2a confers resistance tointestinal tumorigenesis (see comments). Nat. Genet.,17: 88–91, 1997.

4. Tomlinson, I. P., Beck, N. E., Neale, K., and Bodmer, W. F. Variants at the secretoryphospholipase A2 (PLA2G2A) locus: analysis of associations with familial adenomatouspolyposis and sporadic colorectal tumours. Ann. Hum. Genet,60: 369–376, 1996.

5. Nielsen, F. C. The molecular and cellular biology of insulin-like growth factor II.Prog. Growth Factor Res.,4: 257–290, 1992.

6. DeChiara, T. M., Efstratiadis, A., and Robertson, E. J. A growth-deficiency pheno-type in heterozygous mice carrying an insulin-like growth factor II gene disrupted bytargeting. Nature (Lond.),345: 78–80, 1990.

7. Eggenschwiler, J., Ludwig, T., Fisher, P., Leighton, P. A., Tilghman, S. M., andEfstratiadis, A. Mouse mutant embryos overexpressing IGF-II exhibit phenotypicfeatures of the Beckwith-Wiedemann and Simpson-Golabi-Behmel syndromes.Genes Dev.,11: 3128–3142, 1997.

8. Lau, M. M., Stewart, C. E., Liu, Z., Bhatt, H., Rotwein, P., and Stewart, C. L. Lossof the imprinted IGF2/cation-independent mannose 6-phosphate receptor results infetal overgrowth and perinatal lethality. Genes Dev.,8: 2953–2963, 1994.

9. Ward, A., Bates, P., Fisher, R., Richardson, L., and Graham, C. F. Disproportionategrowth in mice with Igf-2 transgenes. Proc. Natl. Acad. Sci. USA,91: 10365–10369,1994.

10. Joyce, J. A., Lam, W. K., Catchpoole, D. J., Jenks, P., Reik, W., Maher, E. R., andSchofield, P. N. Imprinting of IGF2 and H19: lack of reciprocity in sporadicBeckwith-Wiedemann syndrome. Hum. Mol. Genet.,6: 1543–1548, 1997.

11. Bates, P., Fisher, R., Ward, A., Richardson, L., Hill, D. J., and Graham, C. F.Mammary cancer in transgenic mice expressing insulin-like growth factor II (IGF-II)(see comments). Br. J. Cancer,72: 1189–1193, 1995.

12. Rogler, C. E., Yang, D., Rossetti, L., Donohoe, J., Alt, E., Chang, C. J., Rosenfeld,R., Neely, K., and Hintz, R. Altered body composition and increased frequency ofdiverse malignancies in insulin-like growth factor-II transgenic mice. J. Biol. Chem.,269: 13779–13784, 1994.

13. Biddle, C., Li, C. H., Schofield, P. N., Tate, V. E., Hopkins, B., Engstrom, W.,Huskisson, N. S., and Graham, C. F. Insulin-like growth factors and the multiplicationof Tera-2, a human teratoma-derived cell line. J. Cell. Sci.,90: 475–484, 1988.

14. Harrington, E. A., Bennett, M. R., Fanidi, A., and Evan, G. I. c-Myc-induced apoptosisin fibroblasts is inhibited by specific cytokines. EMBO J.,13: 3286–3295, 1994.

15. Downward, J. Mechanisms and consequences of activation of protein kinase B/Akt.Curr. Opin. Cell. Biol.,10: 262–267, 1998.

16. Zhang, L., Zhou, W., Velculescu, V. E., Kern, S. E., Hruban, R. H., Hamilton, S. R.,Vogelstein, B., and Kinzler, K. W. Gene expression profiles in normal and cancercells. Science (Washington DC),276: 1268–1272, 1997.

17. Toretsky, J. A., and Helman, L. J. Involvement of IGF-II in human cancer. J. Endo-crinol., 149: 367–372, 1996.

18. Cui, H., Horon, I. L., Ohlsson, R., Hamilton, S. R., and Feinberg, A. P. Loss ofimprinting in normal tissue of colorectal cancer patients with microsatellite instability(see comments). Nat. Med.,4: 1276–1280, 1998.

19. Boland, C. R., Thibodeau, S. N., Hamilton, S. R., Sidransky, D., Eshleman, J. R., Burt,R. W., Meltzer, S. J., Rodriguez Bigas, M. A., Fodde, R., Ranzani, G. N., and Srivastava,S. A National Cancer Institute Workshop on Microsatellite Instability for cancer detectionand familial predisposition: development of international criteria for the determination ofmicrosatellite instability in colorectal cancer. Cancer Res.,58: 5248–5257, 1998.

20. Da Costa, T. H., Williamson, D. H., Ward, A., Bates, P., Fisher, R., Richardson, L.,Hill, D. J., Robinson, I. C., and Graham, C. F. High plasma insulin-like growth

factor-II and low lipid content in transgenic mice: measurements of lipid metabolism.J. Endocrinol.,143: 433–439, 1994.

21. Wasan, H. S., Novelli, M., Bee, J., and Bodmer, W. F. Dietary fat influences on polypphenotype in multiple intestinal neoplasia mice. Proc. Natl. Acad. Sci. USA,94:3308–3313, 1997.

22. Dove, W. F., Clipson, L., Gould, K. A., Luongo, C., Marshall, D. J., Moser, A. R.,Newton, M. A., and Jacoby, R. F. Intestinal neoplasia in the ApcMin mouse:independence from the microbial and natural killer (beige locus) status. Cancer Res.,57: 812–814, 1997.

23. UKCCCR. United Kingdom Coordinating Committee on Cancer Research(UKCCCR) Guidelines for the welfare of animals in experimental neoplasia (2nd ed.).Br. J. Cancer,77: 1–10, 1998.

24. Kearsey, S. E., and Labib, K. MCM proteins: evolution, properties, and role in DNAreplication. Biochim. Biophys. Acta,1398: 113–136, 1998.

25. Todorov, I. T., Werness, B. A., Wang, H. Q., Buddharaju, L. N., Todorova, P. D., Slocum,H. K., Brooks, J. S., and Huberman, J. A. HsMCM2/BM28: a novel proliferation markerfor human tumors and normal tissues. Lab. Invest.,78: 73–78, 1998.

26. Kiyama, H., and Emson, P. C. Anin situ hybridization histochemistry method for theuse of alkaline phosphatase-labeled oligonucleotide probes in small intestine. J. His-tochem. Cytochem.,39: 1377–1384, 1991.

27. Gould, K. A., Luongo, C., Moser, A. R., McNeley, M. K., Borenstein, N., Shed-lovsky, A., Dove, W. F., Hong, K., Dietrich, W. F., and Lander, E. S. Geneticevaluation of candidate genes for the Mom1 modifier of intestinal neoplasia in mice.Genetics,144: 1777–1785, 1996.

28. Bailleul, B., Surani, M. A., White, S., Barton, S. C., Brown, K., Blessing, M., Jorcano, J.,and Balmain, A. Skin hyperkeratosis and papilloma formation in transgenic mice ex-pressing a ras oncogene from a suprabasal keratin promoter. Cell,62: 697–708, 1990.

29. Wang, J., Niu, W., Nikiforov, Y., Naito, S., Chernausek, S., Witte, D., LeRoith, D.,Strauch, A., and Fagin, J. A. Targeted overexpression of IGF-I evokes distinctpatterns of organ remodeling in smooth muscle cell tissue beds of transgenic mice.J. Clin. Invest.,100: 1425–1439, 1997.

30. Zaina, S., and Squire, S. The soluble type 2 insulin-like growth factor (IGF-II)receptor reduces organ size by IGF-II-mediated and IGF-II-independent mechanisms.J. Biol. Chem.,273: 28610–28616, 1998.

31. MacDonald, R. S. The role of insulin-like growth factors in small intestinal cellgrowth and development. Horm. Metab. Res.,31: 103–113, 1999.

32. DeChiara, T. M., Robertson, E. J., and Efstratiadis, A. Parental imprinting of themouse insulin-like growth factor II gene. Cell,64: 849–859, 1991.

33. Zhu, Y., Richardson, J. A., Parada, L. F., and Graff, J. M. Smad3 mutant mice developmetastatic colorectal cancer. Cell,94: 703–714, 1998.

34. Rashid, Z., and Basson, M. D. Association of rectal prolapse with colorectal cancer.Surgery,119: 51–55, 1996.

35. Christofori, G., Naik, P., and Hanahan, D. A second signal supplied by insulin-like growthfactor II in oncogene-induced tumorigenesis. Nature (Lond.),369: 414–418, 1994.

36. Christofori, G., Naik, P., and Hanahan, D. Deregulation of both imprinted andexpressed alleles of the insulin-like growth factor 2 gene during beta-cell tumorigen-esis. Nat. Genet.,10: 196–201, 1995.

37. Harris, T. M., Rogler, L. E., and Rogler, C. E. Reactivation of the maternallyimprinted IGF2 allele in TGFa induced hepatocellular carcinomas in mice. Onco-gene,16: 203–209, 1997.

38. Haddad, R., and Held, W. A. Genomic imprinting and Igf2 influence liver tumori-genesis and loss of heterozygosity in SV40 T antigen transgenic mice. Cancer Res.,57: 4615–4623, 1997.

39. Wasan, H. S., Park, H. S., Liu, K. C., Mandir, N. K., Winnett, A., Sasieni, P., Bodmer,W. F., Goodlad, R. A., and Wright, N. A. APC in the regulation of intestinal cryptfission. J. Pathol.,185: 246–255, 1998.

40. He, T. C., Sparks, A. B., Rago, C., Hermeking, H., Zawel, L., da Costa, L. T., Morin,P. J., Vogelstein, B., and Kinzler, K. W. Identification of c-MYC as a target of theAPC pathway (see comments). Science (Washington DC),281: 1509–1512, 1998.

41. Zindy, F., Eischen, C. M., Randle, D. H., Kamijo, T., Cleveland, J. L., Sherr, C. J.,and Roussel, M. F. Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization. Genes Dev.,12: 2424–2433, 1998.

42. Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., and Greenberg, M. E.Akt phosphorylation of BAD couples survival signals to the cell-intrinsic deathmachinery. Cell,91: 231–241, 1997.

43. Ron, E., Gridley, G., Hrubec, Z., Page, W., Arora, S., and Fraumeni, J. F., Jr.Acromegaly and gastrointestinal cancer. Cancer (Phila.),68: 1673–1677, 1991.

44. Gurdon, J. B. A community effect in animal development. Nature (Lond.),336:772–774, 1988.

45. Markowitz, S., Wang, J., Myeroff, L., Parsons, R., Sun, L., Lutterbaugh, J., Fan, R. S.,Zborowska, E., Kinzler, K. W., Vogelstein, B., Brattain, M., and Willson, J. K. V.Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatelliteinstability (see comments). Science (Washington DC),268: 1336–1338, 1995.

46. Souza, R. F., Appel, R., Yin, J., Wang, S., Smolinski, K. N., Abraham, J. M., Zou,T. T., Shi, Y. Q., Lei, J., Cottrell, J., Cymes, K., Biden, K., Simms, L., Leggett, B.,Lynch, P. M., Frazier, M., Powell, S. M., Harpaz, N., Sugimura, H., Young, J., andMeltzer, S. J. Microsatellite instability in the insulin-like growth factor II receptorgene in gastrointestinal tumours. Nat. Genet.,14: 255–257, 1996.

47. Takaku, K., Oshima, M., Miyoshi, H., Matsui, M., Seldin, M. F., and Taketo, M. M.Intestinal tumorigenesis in compound mutant mice of both Dpc4 (Smad4) and Apcgenes. Cell,92: 645–656, 1998.

48. Tomlinson, I. P., and Bodmer, W. F. Failure of programmed cell death and differ-entiation as causes of tumors: some simple mathematical models. Proc. Natl. Acad.Sci. USA.,92: 11130–11134, 1995.

49. Giannoukakis, N., Deal, C., Paquette, J., Goodyer, C. G., and Polychronakos, C.Parental genomic imprinting of the human IGF2 gene. Nat. Genet.,4: 98–101, 1993.

1076

IGF-II SUPPLY AND INTESTINAL ADENOMA GROWTH

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

2000;60:1070-1076. Cancer Res   A. Bassim Hassan and Julie A. Howell 

MiceMin/+ApcIntestinal Adenoma in Insulin-like Growth Factor II Supply Modifies Growth of

  Updated version

  http://cancerres.aacrjournals.org/content/60/4/1070

Access the most recent version of this article at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/60/4/1070.full#ref-list-1

This article cites 47 articles, 18 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/60/4/1070.full#related-urls

This article has been cited by 29 HighWire-hosted articles. Access the articles at:

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

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

.http://cancerres.aacrjournals.org/content/60/4/1070To request permission to re-use all or part of this article, use this link

Research. on March 21, 2021. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from