correlation between the detection of specific - cancer research

[CANCER RESEARCH 40, 2316-2322. July 1980]0008-5472/80/0040-OOOOS02.00

Correlation between the Detection of Specific Mouse Mammary Tumor

Proviral Sequences and the Presence of Pulmonary MÃ©tastasesin MiceBearing Spontaneous Mammary Tumors1

W. Drohan, R. D. Cardiff, J. K. Lund, and J. Schlom2

Laboratory of Viral Carcinogenesis. Division of Cancer Cause and Prevention. National Cancer Institute. NIH. USPHS. United States Department ol Health,Education and Welfare. Bethesda, Maryland 20205 Â¡W.O..J.S.I, and Department of Pathology. University of California School of Medicine. Davis. California 95616Â¡RD.C.. J.K.LI

ABSTRACT

Pulmonary mÃ©tastases in C3H/He mice bearing spontaneous mammary tumors were detected and characterized byhistological criteria and immunocytochemical staining formouse mammary tumor virus antigens. The same lungs containing mÃ©tastases were also positive when assayed for aspecific subset of mouse mammary tumor virus proviral DNAsequences. These sequences, termed tumor-associated se

quences, have previously been shown to be present in the DNAof spontaneous mammary tumors that arise before 1 yr of agein C3H/He mice but are absent in DNA's of apparently normal

tissues of C3H/He mice. Reconstruction experiments demonstrated that the nucleic acid hybridization method will detect atleast one mammary tumor cell/250 lung cells. While DNA from13 lungs of apparently normal C3H/He mice did not containsequences homologous to mouse mammary tumor virus tumor-associated-sequence RNA, DNA from lungs of 9 of 12 C3H/

He mice bearing spontaneous mammary tumors did containthese sequences. Since the entire DNA content of the lung canbe assayed as one sample, the hybridization method minimizesfalse negatives resulting from histological analysis of randombiopsy sampling. The hybridization procedure described herethus represents a sensitive and quantitative element as anadjunct for the detection of micrometastatic lesions in micebearing viral-mediated spontaneous mammary carcinomas.

INTRODUCTION

The mouse system has proved to be an excellent model forinvestigating the multiple etiological factors involved in mammary tumors. The roles of viruses, hormones, chemical andphysical carcinogens, genetics, diet, and the immune systemhave all been extensively investigated. The murine model alsorepresents an excellent system for the study of metastaticdissemination of mammary tumor cells. Pulmonary mÃ©tastaseshave been detected and characterized by numerous investigators in such mouse strains as C3H, BALB/cfC3H, BALB/cfRIII, Rill, (BALB/c x DBA) F,, and (C57BL x A/He) F, (1, 2,7, 11, 12, 15, 17, 18). Incidences of pulmonary mÃ©tastasesthat range from 2 to 95% of tumor-bearing animals have beenreported; the primary factor in this wide range has been reported to be the size of the primary mammary tumor (1).Several methods have been used to score for pulmonary mÃ©

tastases; these include gross and microscopic examination oflung tissues and bioassays by trocar implants of lung tissuesinto syngeneic mice.

The C3H/He mouse has been extensively used in studies ofboth the etiology and metastatic spread of mammary carcinoma. Recent nucleic acid hybridization results have demonstrated that sequences homologous to a part of the RNAgenome of the highly oncogenic MuMTV3 of C3H/He mice,i.e., MuMTV(CSH), can be detected in the DNA's of mammary

tumors that occur early in the life span of C3H/He mice (before1 year of age); these sequences are absent in the DNA's of

apparently normal livers from these same tumor-bearing ani

mals (5). We report here the use of a radioactive probe, whichcontains sequences homologous to these "TA" sequences,

that can detect specific MuMTV proviral sequences in lungsthat contain micrometastatic lesions of C3H/He mice bearingspontaneous mammary tumors. Pulmonary mÃ©tastases havealso been characterized by histochemical staining for MuMTVantigens. The sensitivity and specificity of the nucleic acidhybridization technique is also delineated.

MATERIALS AND METHODS

Animals and Tissues. Spontaneous mammary tumors andtissues were obtained from C3H/HeCrgL or C3H/HeJ retiredbreeder females. Some tumor-bearing C3H/HeJ mice were

kindly supplied by Dr. B. Diwan, Meloy Laboratories, Inc.,Springfield, Va. Mice were monitored for mammary tumors bypalpation. Tumor-bearing mice were held 3 to 4 weeks beforesacrifice. Autopsy of all tumor-bearing animals revealed that

the internal organs, including the lung, appeared normal withno gross evidence of mÃ©tastases.

Immunocytochemistry. The entire lung and mammary tumors were removed from 3 animals and were fixed and submitted for immunocytochemistry. Random biopsies, representing approximately one-fifth of the total tissue volume, were

taken from lungs and tumors of 4 animals for immunocytochemistry. The remaining tissue was frozen in liquid nitrogenfor biochemical analysis. Random cu mm samples of mammarytumor and lung of each animal were fixed in Tellyezniski's

solution, embedded in paraffin, sectioned at 4 /im, and mountedon precleaned glass slides. Following incubation at 56Â°over

night, the slides were deparaffinized in xylene. The sectionswere stained by the peroxidase-antiperoxidase technique (13,

16). Briefly, sections were preincubated in 10% normal goat1This work was supported in part by Contracts N01-CP-43223 and N01-CP-

61013 of the National Cancer Institute.2 To whom requests for reprints should be addressed, at Building 37. Room

1B19, National Cancer Institute. NIH. Bethesda. Md 20205.Received December 27. 1979; accepted March 28, 1980

1The abbreviations used are: MuMTV, murine mammary tumor virus(es); TA.

tumor-associated; PBS, phosphate-buffered saline consisting of (in g/liter) KCI(0.2), KH2PO4 (0.2). NaCI (8.0), and NajHPO,-HÂ¿,O(2.1); NaPB. sodium phosphate buffer; SLS, sodium laurel sarcosinate.

2316 CANCER RESEARCH VOL. 40

Research. on December 18, 2018. © 1980 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Detection of Mammary Cell MÃ©tastases

serum (Antibodies, Inc., Davis, Calif.) in 0.1 M PBS (pH 7.4) for30 min at ambient temperature. Rabbit anti-MuMTV, charac

terized previously (4,13), was absorbed with BALB/c lactatingmammary gland (13), diluted to 1:500 in PBS, and placed onsections for 60 min. After a 15-min wash in PBS, goat anti-

rabbit globulin (Antibodies, Inc.) that had been diluted to 1:20in PBS was placed on sections for 30 min followed by a 15-minwash in PBS. Rabbit antiperoxidase-peroxidase-soluble com

plex (Litton Bionetics, Kensington, Md.) that had been diluted1:50 in PBS was placed on sections for 45 min followed byPBS wash. Diaminobenzidine (0.06%) containing 0.01% H2O2was placed on the sections for 2 to 5 min. The slides wererinsed with PBS with hematoxylin. Substitution and deletioncontrols were routinely done (3). The slides were evaluatedand photographed with a Zeiss photomicroscope using a Kodak BG-12 filter and Panatomic-X filter.

Viruses. MuMTV(C3H) was obtained from supernatant fluidsof the C3H mouse mammary tumor cell line Mm5mt/cl (10).Propagation of cells and purification of MuMTV from supernatant fluids was as previously described (5). No type C virus wasdetected by electron microscopy, by radioimmunoassays, orby divalent cation preference of DNA polymerase (Mg2+:Mn2*

ratio of 23:1) using oligodeoxyguanylate:polyribocytidylate astemplate. Competition hybridization experiments were used todemonstrate that the virus produced in Mm5mt cells was alsofree of contaminating endogenous type C virus. These studieshave been detailed previously (5).

Purification of Viral RNA. The procedure for the isolationand iodination of MuMTV 60 to 70 S RNA was as describedpreviously (5). lodinated RNA had a specific activity of approximately 2 x 107 cpm//Â¿gand was 99% acid precipitable and

greater than 95% RNase sensitive.Purification of Cellular DNA. DNA was extracted from tissue

using a procedure previously described (5). DNA prepared inthis way had a 260:280 nm ratio of at least 1.70 and wasusually above 1.80. After sonication, the sedimentation coefficient of DNA as determined by alkaline sucrose sedimentationwas between 6 and 9S.

Molecular Hybridization. Hybridization conditions used havebeen described previously (5). Cellular DNA was annealed toMuMTV(C3H) 126I or 3H 60 to 70S RNA at 68Â° under the

following conditions. Cellular DNA (first boiled in 0.01 M NaPB,pH 6.8, for 1 min in a bath of ethylene glycol), at a concentrationof 1.5 mg/ml in 0.4 M NaPB (pH 6.8) and 0.05% sodiumdodecyl sulfate, was mixed with labeled MuMTV RNA andincubated for various times to obtain the desired C0f. C0t valuesare corrected to 0.12 M NaPB. Aliquots of 0.067 ml (100 fig ofDNA and 1000 cpm of MuMTV RNA) were diluted with 2.0 mlof 2x 0.1 5 M NaCI and 0.015 M sodium citrate to obtain a finalDNA concentration of 50 /ig/ml. RNase's A and T, were added

to one-half of the sample at final concentrations of 15 /ig/mland 3 units/ml, respectively. Both tubes were incubated at 37Â°

for 30 min. The samples were chilled for 10 min, and trichlo-

roacetic acid was added to a final concentration of 10%. Afterbeing kept on ice for 30 min, the acid-insoluble material wascollected on 0.45-/im nitrocellulose filters (Gelman, Inc.). Azero time control, which was boiled for 1 min in a bath ofethylene glycol prior to the addition of RNase, was included inall experiments. The cpm of radioactivity present in the RNase-

treated portion of this sample was deducted from the amountobtained in all other samples. The percentage of hybridization

was determined by dividing the cpm present in the RNase-

treated sample by the cpm present in the untreated sample.Isolation of TA Sequences. The method used is a modifi

cation of that of Shoyab ef al. (14) and has been detailedpreviously using MuMTV (5). Briefly, 300,000 cpm ofMuMTV(C3H)-labeled 60 to 70S RNA were hybridized to 30

mg of apparently normal C3H/He liver DNA to a C0f of 15,000using the hybridization conditions described above. Liver wasused since it has been reported to be free of many MuMTVmarkers (9). The sample was then diluted to 150 ftg of DNAper ml, heated to 60Â°, and applied to a column of 150 ml ofpacked hydroxylapatite at 60Â°. The sample that eluted was

reloaded, and the column was rinsed with 0.01 M NaPB (pH6.8) containing 0.01% SLS until no material eluted whichabsorbed at 260 nm. Single-stranded RNA was then eluted

with a solution of 0.14 M NaPB (pH 6.8) in 0.01% SLS and wasthen dialyzed against 3 changes of 6 liters of water containing0.01 % SLS. After dialysis, 1 to 2 mg of yeast carrier RNA wereadded, and the "TA" RNA was precipitated with 2 volumes of

alcohol. The RNA was then pelleted, dissolved in 1 ml of 0.001M NaPB, and stored at -20Â° until use.

Thermal Stability of DNA-RNA Hybrids. The technique forassaying the thermal stability of the DNA-RNA hybrids is as

described previously (5). The hybridization mixture contained1 mg of cellular DNA and 6000 cpm of radioactive MuMTVRNA. After hybridization to a C0r of 20,000, the DNA wasdiluted to 50 /Â¿g/ml,and the solution was applied to a 1.0-mlcolumn of packed hydroxylapatite maintained at 60Â°in 0.006

M NaPB (pH 6.8). These are the correct conditions establishedby Martinson and Wagenaar (8) for monitoring thermal stabilityof nucleic acids on hydroxylapatite. Using these conditions,approximately 95% of the cellular DNA remains bound to thecolumn. While at 60Â°,the column was washed twice with 10-

ml samples of 0.12 M NaPB (pH 6.8) containing 0.01% SLS.This procedure was repeated at 5Â°increments until a temperature of 100Â° was attained. The thermal dissociation of the

DNA-RNA hybrids was measured by monitoring the absorb-

ance at 260 nm in the effluent fractions, and that of theradioactivity-labeled RNA-DNA hybrids was measured by determining the trichloroacetic acid-precipitable cpm in the ef

fluent fractions.

RESULTS

Immunocytochemistry. Lungs and spontaneous mammarytumors from C3H/He mice were examined by conventionalhistological techniques and by immunoperoxidase staining forMuMTV antigens. The 8 mammary tumors from 7 animalsexamined had immunoperoxidase-detectable MuMTV antigensas indicated by the red-brown deposit in the presence of anti-

MuMTV sera (Fig. 1). Both apical and cytoplasmic distributionof antigen were present. Four of the tumors were type A, and4 were type B. The distribution of MuMTV antigens in the typeB tumors was variable (13). Pulmonary mÃ©tastaseswere foundin each of the lungs of 3 animals whose entire lungs wereavailable for examination. MÃ©tastaseswere detected, however,in only 1 of 4 lungs sampled by random biopsy; 3 of these 4identical lung samples, on the other hand, were positive whenassayed by the nucleic acid hybridization procedure describedbelow. Pulmonary mÃ©tastasesappeared as masses within thepulmonary arteries or as expansile masses within the pulmo-

JULY 1980 2317



W. Drohan et al.

nary parenchyma (Figs. 1 and 2). The metastatic cells frequently extended through the wall of the blood vessel (Fig. 1).There was no evidence of individual tumor cells lodged in bloodvessels. No cells outside the tumor mass were positive forMuMTV antigens. Multiple metastatic foci were present in mostpositive sections. The tumor masses, when present, occupiedas much as 10 to 20% of the cross-sectional area of the

section. The histological pattern of the mÃ©tastases closelyresembled that of the primary tumor. MuMTV antigens wereidentified by the immunoperoxidase technique in each of themÃ©tastasesobserved.

Hybridization of MuMTV(CSH) Radioactive 60 to 70S RNAto Murine DNA's. We have previously shown that MuMTV(C3H)

radioactive 60 to 70S RNA hybridized approximately 55% tothe DNA of spontaneous C3H/He mammary tumors and 49%to the DNA s of apparently normal livers of these same animalsor other C3H/He mice free of mammary tumors (5). In anextension of these studies, the DNA's from the livers and lungs

of 4 apparently normal C3H/He mice and 4 C3H/He micebearing mammary tumors were hybridized to 1?i>l-labeled 60 to

70S RNA to MuMTV(CSH). Using this probe (Table 1), onecannot distinguish between DNA's obtained from lungs of

tumor-bearing and apparently normal animals or between liversand lungs from tumor-bearing animals.

Hybridization of Radioactive MuMTV(C3H) TA-SequenceRNA to Murine DNA's. Previous studies have shown that

radioactive MuMTV(CSH) TA-sequence RNA hybridizes approximately 55% to the DNA's of spontaneous mammary tu

mors of C3H/He mice but does not hybridize above the 6%background level to livers of C3H/He mice bearing mammarytumors or to livers of apparently normal C3H/He mice (5). Inan extension of these studies, it is observed that '"(-labeled

TA-sequence RNA also failed to hybridize to hearts, lungs,

kidneys, brains, and livers of each of 5 apparently normalC3H/He mice (Table 2.) DNA's of Animals 1,2,3, and 4 were

the same as those shown in Table 1, thus ruling out thepossibility that the lack of hybridization was due to the qualityof the DNA's. As a further control, this same '?;>l-labeled TA-

sequence RNA preparation hybridized approximately 50% toDNA s of spontaneous C3H mammary tumors (Table 3).

When 126l-labeled TA-sequence RNA was annealed to DNA's

of various internal organs of C3H/He mice bearing spontaneous mammary tumors, 6 of 8 lungs gave a positive hybridization (Table 3), whereas all hearts, kidneys, brains, and liversscored negative. The lack of detection of TA sequences in theDNA's of other highly vascularized organs, such as liver, pro-

Table1Hybridization of MuMTV 60 to 70S RNA to livers and lungs of C3H/HeJ mice

Hybridization conditions are as described in Materials and Methods. I25I-

MuMTV(C3H) 60 to 70S RNA (1000 cpm) was hybridized to 100 jig of DNA.Numbers refer to average hybridization to duplicate points hybridized to a C0f of25.000.

% ofhybridizationMouse

noNormalTumor-bearing12345678Liver4545474642444443Lung4144464546434446

Table 2Hybridization between radioactive MuMTV TA-sequence RNA and individual

organs ot apparently normal C3H/He miceHybridization conditions are as described in 'Materials and Methods." '"I-

MuMTV TA-sequence RNA (1000 cpm) was hybridized to 100 /ig of murine DNA.Numbers refer to average hybridization to duplicate points hybridized to a Cof of25.000. Hybridization to the DNA of a pool of C3H livers was 5.6%. to bovineovary DNA it was 3.8%. and to DNA from the spontaneous C3H mammary tumorcell line MmSmt was 63%.

% of hybridization to individualmiceDNAHeartLungsKidneyBrainLiver14.43.85.15.23.924.25.75.44.63.135.24.15.83.66.044.35.14.14.43.953.9NT31.52.95.1

NT, not tested.

Table 3Hybridization between radioactive MuMTV TA-sequence RNA and individual

organs ot mammary tumor-bearing C3H/He miceHybridization was performed as described in Materials and Methods' and in

legend to Table 2.

% of hybridization to individualmiceDNAHeartLungKidneyBrainLiverTumor14.61.34.04.24645.123.910.63.35.15.553.635.110.64.95.75.256.945.214.73.43.25.754.255.928.25.72.64.660.365.26.65.25.35.351.975.225.24.64.84.449385.228.44.53.83.246.9

vides evidence that the hybridization to lung DNA is not simplydetecting circulating tumor cells.

Thermal analysis of hybrids formed between radioactive TA-sequence RNA and DNA of positive lungs revealed a 7"m(melting temperature) value of 77Â° for 125I-RNA-DNA duplexes(Chart 1). This value and the A7Â°Tmvalue observed between125I-RNA-DNA hybrid and DNA-DNA duplex dissociation are

the same as those previously reported for hybrids formedbetween 125l-TA-sequence RNA and DNA from spontaneous

C3H/He mammary tumors (5). These findings provide evidencethat the DNA sequences detected in lungs are identical orextremely similar to those sequences detected in primary C3H/He mammary tumors.

Kinetic Analysis of MuMTV TA Sequences in Lung DNA's.

Kinetics of hybrid formation was analyzed in an attempt todetermine the frequency of proviral sequences in DNA's of

lungs positive for sequences homologous to MuMTV TA-sequence RNA. One hundred fig of DNA from a positive lungwere hybridized with 1000 cpm of 125l-TA-sequence RNA and

were incubated for various time intervals. As is seen in Chart2, an accurate C0fi,2 could not be obtained, since the percentage of hybridization was still rising at a C0f of 50,000. Thisdoes mean, however, that the minimum C0fi/2 that could beattained (i.e., a maximum hybridization percentage of 28%) is6500. For comparison, 125l-labeled polyadenine-selected nor

mal murine RNA (5) was hybridized to the same DNA underidentical conditions; a C0fi/2 of 3000 was obtained (Chart 2,arrow; Ref. 5). When compared to this value for "unique"

sequence RNA, it can be estimated that the DNA of these lungtissues contains less than one copy of TA-sequence proviralDNA per haploid genome. When compared to the C0fi/2 of 380obtained between 125l-TA-sequence RNA and mammary tumor

DNA (5), it can be estimated that there is a maximum value of





60 70 80 90TEMPERATURE I Â°)

100

Chart 1. Thermal stability of hybrids formed between MuMTV TA-sequenceRNA and DNA from a pool of 4 lungs of mammary tumor-bearing C3H/HeJ miceshown to be positive for hybridization to MuMTV TA-sequence RNA. Hybridizationconditions and thermal elution from hydroxylapatite columns were as describedin Materials and Methods. â€¢DNA-DNA profiles as monitored by absorbanceat 260 nm. G, '25l-labeled RNA-DNA hybrid dissociation as measured by acid-

precipitable radioactivity in each fraction.

Chart 2. Kinetic analysis of hybridization between MuMTV radioactive TA-sequence RNA and DNA from organs of tumor-bearing C3H/He mice Hybridization conditions are as described in "Materials and Methods Each pointcontained 100 ^g of murine DNA and 1000 cpm of TA-sequence MuMTV RNAand was incubated to the indicated C0f. A. hybridization to DNA from a pool of 4C3H/HeJ lungs from tumor-bearing animals shown to be positive for hybridizationto MuMTV TA-sequence RNA; â€¢hybridization to DNA from a pool of C3H/HeJlivers from apparently normal mice. Arrow, Cot, 2 for the hybridization between'"(-labeled polyadenine-selected normal murine RNA (selected on polyuridylic

acid-Sepharose 48) (5) and murine DNA under identical conditions of molecularhybridization.

1 mammary cell per 17 normal lung cells in the involved lung(i.e., a C0t of 6500 divided by 380).

In an attempt to determine the approximate sensitivity of the

hybridization method to detect mammary tumor cells in pulmonary tissue, mixing experiments were performed. Lungsfrom apparently normal C3H/He mice were trypsinized, andindividual lung cells were counted and mixed in various ratioswith C3H mammary tumor cells. Cells from a mammary tumorcell line were used to eliminate possible errors in numbers ofmammary tumor cells due to the existence of normal mammarystroma and infiltrate. DNA's were extracted from the various

mixtures of lung and mammary tumor cells, and these werethen hybridized to MuMTV(C3H) TA-sequence 125I-RNA. As

seen in Chart 3, hybridization above background levels couldbe detected in ratios of at least 1 mammary tumor cell/250normal lung cells.

DISCUSSION

Pulmonary mÃ©tastasesare the most frequent form of distalmÃ©tastases detected in patients with breast cancer, beingfound in 60 to 65% of autopsies. The murine model presentsan excellent system to study metastatic spread of mammarycells to lung tissue in that various studies have reported thatbetween 2 and 95% of animals with spontaneous mammarytumors possess pulmonary mÃ©tastases. The tumor-bearingC3H/He females studied here had typical MuMTV-positive

mammary adenocarcinomas and intrapulmonary masses.Since the masses were intravascular, MuMTV antigen positive,and histologically similar to the mammary cancer, they mostprobably originated in the mammary tumor. The lesions wereexpansile and invasive, indicating active growth in situ ratherthan passive entrapment of circulating tumor cells. Further,individual antigen-positive tumor cells were not found in the

lung. The pathological diagnoses of pulmonary mÃ©tastasesfrom primary mammary tumors were reinforced by the obser-

60

D

N 40goci\

20

~r

0 10 100 1000

RATIO OF C3H LUNG CELLS TO MAMMARY TUMOR CELLS

Charts. Hybridization between ''"'l-labeled MuMTV TA-sequence RNA and

DNA s of mixtures of lungs and mammary tumor cells. Lungs from apparentlynormal C3H/He mice were trypsinized. and the individual lung cells were disrupted and counted. Cells from the C3H/He mammary tumor cell line. MmSmt,were also counted. Mammary tumor cells and lung cells were mixed in proportionsstated on the abscissa A total of 4 x 108 cells were mixed at each dilution, theDNA was extracted, and 100 /ig of the DNA mixture were hybridized to "bl-labeled MuMTV TA-sequence RNA as described in "Materials and Methods."

Hybridization was at 68Â° in 0.4 M NaPB to a Cot of 25.000. The hybridizationbetween MuMTV TA-sequence RNA and undiluted DNA from C3H mammarytumor cells (MmSmt) is 62% (ratio of 0). Hybridization of MuMTV TA-sequenceRNA to undiluted lung cell DNA was 4%.

JULY 1980 2319



W. Drohan et al.

vation of TA DNA Â¡nthe lungs of only tumor-bearing mice and

by the lack of TA DNA sequences in other heavily vascularizedorgans such as the liver (Table 3).

The clear evidence of blood-borne mÃ©tastases found heredeserve mention. Although rarely emphasized, blood-borne

mÃ©tastasesare a major source of distant mÃ©tastasesin humanbreast cancer (6). Thus, this aspect of the human disease hasa biological counterpart in the murine mammary tumor model.

Three of the random biopsies studied here had no histologi-

cally identifiable tumor cells. Since the same lungs containedTA DNA, occult tumor cells were present. These observationspoint out the sampling error inherent in random biopsies andunderline the usefulness of the hybridization technique described here.

The nucleic acid hybridization studies reported here conceptually add a new dimension in the detection of metastaticspread. To our knowledge, it is the first attempt to use thetechnique of molecular hybridization to such an end. Thereare, of course, several pros and cons to the hybridizationmethodology. The disadvantages include: (a) the availability ofa specific radioactive probe to detect unique nucleic acidsequences in tumor cells; (b) mammary tumors from differentanimals may vary in their MuMTV proviral sequence copynumber; (c) the need of highly trained personnel and specialized reagents to conduct nucleic acid hybridization experiments; and (d) like any other biochemical detection technique,hybridization only detects the presence of a marker (DNA)associated with cancer; it does not reveal the biological behavior or potential of the cells carrying the DNA. Some of theadvantages of this technique include: (a) the high degree ofsensitivity; (b) the specificity of the hybridization reaction; and(c) the large amounts of a given tissue that can be tested inone assay. This latter property reduces the possibility of missing metastatic cells in some tissues due to sampling error whenonly a part of a given tissue is examined histologically. Thetime factor involved in the molecular hybridization procedure(approximately 10 days from start to finish due to incubationtimes; see "Materials and Methods") is of course much slower

than histological or immunocytochemical techniques, but it ismuch more rapid than transplant bioassay techniques. In summary, the results reported here provide an adjunct and predictive element to the study of metastatic spread in a definedanimal model; furthermore, these studies conceptually indicatethat, with the proper nucleic acid probe, molecular hybridiza

tion may be used as a sensitive and specific adjunct for thestudy of metastatic lesions.

ACKNOWLEDGMENTSWe thank L. Dreyer, D. Thompson. K. Barnett, and E. Clark for their assistance.

REFERENCES

1. Anderson. J. C.. Fugmum, R. A., Stolfi, R. L.. and Martin, D. S. Metastaticincidence of a spontaneous murine mammary adenocarcinoma. CancerRes. ,34: 1916-1920, 1974.

2. Brooks. R. E. Mouse mammary tumor mÃ©tastases in lung: an electronmicroscopic study. Cancer Res., 30. 2156-2165, 1970.

3. Cardiff, R. D.. and Lund. J. K. Distribution of dengue-2 antigens by electronimmunocytochemistry. Infect. Immun. 3. 1699-1709. 1976.

4. Cardiff, R. D., Puentes, M. J.. Young, L. J. T.. Smith, G. H., Teramoto. Y. A.,Altrock. B. W., and Pratt. T. S. Serological and biochemical characterizationof the mouse mammary tumor virus with localization of p10. Virology. 85157-167. 1978.

5. Drohan, W.. Kettmann, R., Colcher, D., and Schlom, J Isolation of themouse mammary tumor virus sequences not transmitted as germinal provirusin the C3H and RIMmouse strain. J. Virol.. 21: 986-995, 1977.

6. Haagensen. C. D. The natural history of breast carcinoma. In: Diseases ofthe Breast, pp. 426-432. Philadelphia, Pa.: W B. Saunders Co., 1971.

7. Keydar. !.. Mesa-Tejada, R.. Ramanarayanan. M., Ohno. T.. Fenoglio. C..Hu, F.. and Spiegelman. S. Detection of viral proteins in mouse mammarytumors by immunoperoxidase staining of paraffin sections Proc. Nati. Acad.Sei. U. S. A., 75. 1524-1528, 1978.

8. Martinson, H. G., and Wagenaar, E. B. Thermal elution chromatography ofnucleic acids on hydroxylapatite. Biochim. Biophys. Acta. 474. 445-455.1978.

9. Mandi, S., and McGrath, C. M. Mammary neoplasia in mice. Adv. CancerRes., 17: 353-414, 1973.

10 Owens, R. B., and Hackett, A. J. Tissue culture studies of mouse mammarytumor cells and associated viruses. J. Nati. Cancer Inst.. 49: 1321-1332.1972.

11. Picket!, P. G., Pitelka, D. R.. Hamamoto, S. T.. and Misfeldt, D. S. Occludingjunctions and cell behavior in primary cultures of normal and neoplasticmammary gland cells. J. Cell Biol. 66. 316-322, 1975.

12. Pitelka. D. R. Cell recognition by malignant mammary epithelium. J. CellBiol. 77. 66a, 1976.

13. St. George. J. A.. Cardiff, R. D.. Young, L. J. T.. and Faulkin. L. J. Theimmunocytochemical distribution of mouse mammary tumor virus antigensin BALB/cfC3H mammary epithelium. J. Nati. Cancer Inst., 63. 813-820,1979.

14. Shoyab, M.. Evans. R. M., and Baluda. M. A. Presence in leukemic cells ofavian myeloblastosis in virus-specific DNA sequences absent in normalchicken cells. J. Virol.. 14: 47-49. 1974.

15. Squartini, F., and Bistocchi, M. Bioactivity of C3H and RIM mammary tumorviruses in virgin female BALB/c mice: brief Communication. J. Nati. CancerInst., 58. 1845-1847. 1977.

16. Taylor. C. Immunoperoxidase techniques. Practical and theoretical aspects.Arch. Pathol. Lab. Med., 102: 113-121, 1978.

17. Weiss, L.. and Holyoke. E. D. Some effects of hypervitaminosis A onmÃ©tastasesof spontaneous breast cancer in mice. J. Nati. Cancer Inst.. 43:1045-1054, 1969.

18. Wexler, H., Minton, J. P., and Ketcham, A. S. A comparison of survival timesand extent of tumor mÃ©tastases in mice with transplanted, induced, andspontaneous tumors. Cancer (Phila.) 18: 985-995. 1965.

Fig 1 a, photomicrograph of primary mammary tumor stained for MuMTV antigens using the immunoperoxidase technique and counterstained with hematoxylin.The tumor has a mixture of well-differentiated, lobuloalveolar type A regions (At and poorly differentiated, solid type B regions (B). Note the dark deposits of reactionproducts indicating MuMTV antigens on the apical edge of the alveoli in the type A tumor (arrows). The general lack of reaction product in the type B regions act asan internal control, x 320. b, photomicrograph of lung from the same mouse showing multiple intravascular mÃ©tastases(arrows), x 50. c. higher magnification of b,showing a tumor mass in a peribronchial vessel. The muscular wall of the vessel wall is clearly outlined (arrow). The tumor is attached to the vessel and is growingthrough the vessel musculature (V) Note the scattered dark areas of immunoperoxidase reaction product within the tumor indicating the presence of MuMTVantigens, x 250.

Fig. 2. A, photomicrograph of primary mammary tumor stained for MuMTV antigens using the immunoperoxidase technique and counterstained with hematoxylin.The tumor is moderately well-differentiated with many poorly formed alveolar structures. Note the dark immunoperoxidase reaction product indicating apicalconcentrations of MuMTV antigens (arrows), x 320. b. photomicrograph of lung from the same mouse demonstrating 2 tumor foci (arrows) which are expansile andare compressing surrounding tissue, x 50. c, higher-magnification photomicrograph of pulmonary mÃ©tastasesillustrated in b. Compare the histological structure andthe immunoperoxidase staining pattern (arrows) to the primary tumor illustrated in a. x 320.





ÃŽ â€¢â€¢-I Â«Â»-

JULY 1980 2321



W. Drohan et a/.

C>V-^Ã¬

"â€¢â€¢'

.?_&â„¢ ; - ' 'Â»â€¢â€¢.'

H g

Â«&1 ' , .;â€¢.'â€¢Â»''

OCM

â€¢ T

Â»*

.r â€¢r â€¢ Â»t^ S. . .\ A .* .




1980;40:2316-2322. Cancer Res W. Drohan, R. D. Cardiff, J. K. Lund, et al. Metastases in Mice Bearing Spontaneous Mammary TumorsTumor Proviral Sequences and the Presence of Pulmonary Correlation between the Detection of Specific Mouse Mammary

Updated version

http://cancerres.aacrjournals.org/content/40/7/2316

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] 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/40/7/2316To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



correlation between the detection of specific - cancer research

Documents