Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/17006752

Protein-codingcapacitiesofpolyadenylatedRNAsfromnormalandneoplasticratmammarytissues

ARTICLEinCANCERRESEARCH·MAY1983

ImpactFactor:9.33·Source:PubMed

CITATIONS

10

READS

8

3AUTHORS,INCLUDING:

TobyMHorn

CarnegieInstitutionforScience

10PUBLICATIONS114CITATIONS

SEEPROFILE

PradmanKQasba

U.S.DepartmentofHealthandHumanServices

123PUBLICATIONS2,792CITATIONS

SEEPROFILE

Availablefrom:TobyMHorn

Retrievedon:04February2016

  

1983;43:1819-1826. Published online April 1, 1983.Cancer Res Toby M. Horn, Joseph Sodroski and Pradman K. Qasba and Neoplastic Rat Mammary TissuesProtein-coding Capacities of Polyadenylated RNAs from Normal  

  

Updated Version http://cancerres.aacrjournals.org/content/43/4/1819

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

SubscriptionsReprints and

.pubs@aacr.orgDepartment atTo order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions.permissions@aacr.orgDepartment at

To request permission to re-use all or part of this article, contact the AACR Publications

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

[CANCER RESEARCH 43,1819-1826, April 1983]

Protein-coding Capacities of Polyadenylated RNAs from Normal and NeoplasticRat Mammary Tissues1

Toby M. Horn, Joseph Sodroski, and Pradman K. Qasba2

Laboratory of Pathophysiology, National Cancer Institute, NIH, Bethesda, Maryland 20205

ABSTRACT

Patterns of gene expression in normal and neoplastic ratmammary tissues were compared by cell-free translation of their

total polyadenylated RNAs and by dot blot hybridization of theRNA to cloned complementary DNA probes for six of the majormilk proteins, including: M, 42,000 and 25,000 caseins, X-casein,whey phosphoproteins, «-proteins, and «-lactalbumin. Distinct

but overlapping messenger RNA populations were evident fromthe translation patterns of normal virgin, pregnant, and lactatingmammary glands. Dot blot analysis showed that each milkprotein RNA had a different characteristic accumulation profileduring pregnancy and lactation. The MTW9 and MCCLX mammary tumor lines, which are transplantable, prolactin dependentfor growth, and produce a-lactalbumin, both showed high <*-

lactalbumin and M, 42,000 casein messenger RNA activity. Thetumors also had other milk protein RNA sequences, although indifferent proportions than at any stage during functional differentiation of normal adult mammary gland. Our results indicatethat normal pregnant mammary gland expresses all of the abundant milk protein genes prior to detectable milk secretion. Thepatterns of gene expression in the two mammary tumors do notappear to correspond to any particular stage of functional differentiation of the normal mammary gland.

INTRODUCTION

Comparisons of normal and neoplastic mammary tissues areimportant for understanding the neoplastic process and for providing a rational basis for cancer treatment. In the studies reported here, we have examined changes in the patterns of geneexpression in normal rat mammary gland during functional differentiation. Patterns of gene expression in 2 hormone-dependent

mammary tumor lines, MCCLX (8) and MTW9 (22), were thencompared.

The poly(A)+3 RNAs of the normal and neoplastic mammary

tissues were examined for their cell-free protein translation activ

ities and for milk protein sequence content. Accumulation ofa-LA (15) and casein (15,26) mRNA sequences at various stages

of pregnancy had previously been followed by molecular hybridization using radioactive cDNAs to the purified RNAs or bytranslation assay of the mRNAs (16). Recently, recombinantcDNA clones for several of the major rat milk proteins have been

1Part of this work was reported in abstract form (No. 2165) at the American

Society Biological Chemistry, June 1 to 5, 1980 (9).2To whom requests for reprints should be addressed, at Building 10, Room

5B56, NIH, Bethesda, Md. 20205.3The abbreviations used are: poly(A)+ RNA, polyadenylated ribonucleic acid; <«-

LA, «-lactalbumin; cDNA, complementary deoxyribonucleic acid; SOS, sodiumdodecyl sulfate; SSC, 1x = 0.15 M sodium chloride/0.015 M sodium citrate, pH7.2.

Received April 5.1982; accepted January 10,1983.

isolated and characterized in our (4, 5, 21 )4 and another labora

tory (23). We selected 6 different milk protein cDNA clones,representing the gene sequences for M, 42,000 casein, M,25,000 casein, Wp-proteins, «-LA, and 2 as yet unidentifiedproteins, "X" and K, to follow the accumulation of their corre

sponding RNA sequences in the mammary gland. By thesetechniques, we found that each milk protein mRNA had a characteristic pattern of appearance during functional differentiation.Both mammary tumors contained all of the milk protein RNAsthat we could probe but in proportions that differed from mammary gland.

MATERIALS AND METHODS

Materials. Materials were obtained as follows.Sephadex G-25 and poly(U)-Sepharose (Pharmacia Fine Chemicals,

Piscataway, N. J.); guanidine-HCI and molecular weight marker kit (Be

thesda Research Laboratories, Rockville, Md.); creatine phosphokinase(Sigma Chemical Co., St. Louis, Mo.); [35S]methionine (carrier free;specific activity, 600 to 1000 Ci/mmol) and 32P-labeled nucleotides (Amer-sham, Arlington Heights, III.); [35S]cysteine (carrier free; specific activity400 to 1000 Ci/mmol) and En3Hance (New England Nuclear, New

Bedford, Mass.); nitrocellulose filter papers (Schleicher & Schuell, Inc.,Keene, N. H.); SDS (British Drug House); acrylamide: (Bio-Rad Laboratories, Richmond, Calif.); b/s-acrylamide(Eastman Kodak Co., Rochester,N. Y.; Pansorbin (fixed Staphylococcus aureus cells) (Calbiochem-Behr-ing Corp., La Jolla, Calif.); 2,5-diphenyloxazole (Research Products In

ternational, Inc., Elk Grove Village, III.).Animals. Mature virgin or primiparous Sprague-Dawley rats were used

for preparation of RNA from normal tissues. The pregnant rats wereexamined by laparotomy to ensure that they were at the proper stage ofpregnancy. The MTW9 mammary tumor was maintained by serial passage in W/Fu virgin females coimplanted with the MITW10 mammotropichormone-secreting pituitary tumor (22). MCCLX mammary tumor tissuewas generously provided by Drs. T. Kano-Sueoka and J. E. Errick of the

University of Colorado, Boulder, Colo. It had been maintained by serialpassage in 4- to 6-week-old male A x C rats coimplanted with a long-acting 25-mg estradici pellet (10).

Poly(A)+ RNA Extraction. RNA was extracted from mammary tissuesand livers that had been quickly frozen in liquid nitrogen and stored at-80°. The guanidine HCI/sodium acetate procedure described by Deeley

ef al. (6) was used, omitting the chloroform/butanol extraction. TotalRNA was then applied to a poly(U)-Sepharose column, and the poly(A)+RNA was eluted using 70% formamide buffered with 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid/EDTA (31). For maximal recovery, 90%

formamide is used. The RNA was precipitated 2 times from ethanol in0.2 M NaCI before using for cell-free translation or dot blot hybridization.The RNA was stored at -20° in water or 20 mw 4-(2-hydroxyethyl)-1 -

piperazineethanesulfonic acid, pH 7.6. The A26o/28owas typically >1.8.Cell-free Translation. Wheat germ extract was prepared according

to Roberts and Paterson (25) and Deeley ef al. (6). Just prior to use, theextract was treated with micrococcal nuclease for 10 min at 25°(18).

4E. Devinoy, A. M. Dandekar, and P. K. Qasba. cDNA clones for rat caseins,

submitted for publication.

APRIL 1983 1819

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

7. M. Horn et al.

Wheat germ tRNA was purified by the procedure described by Zubay(38).

Cell-free translation was assayed by the procedure of Roberts and

Paterson (25).Incorporation of [^Slmethionine or [^SJcysteine into protein was

monitored by a hot trichloroacetic acid method (6).Immunoprecipitations were carried out as described elsewhere (5)

using Pansorbin.Hybrid-selected mRNA translations were performed as described (5).

Gel Electrophoresis. SDS/polyacrylamide slab gels (0.75 mm thick)were prepared as described by Laemmli (11). Molecular weight standardswere 14C-labeled bacteriophage T4 (34), kindly provided by Dr. L. L.

Silver, or Bethesda Research Laboratories high-molecular-weight marker

mix. Gels were fixed for 0.5 hr in 50% trichloroacetic acid, fluorographedusing En3Hance, dried, and then exposed to Kodak XR-5 film at -70°

usually overnight for 20,000 to 50,000 cpm/lane of total translationproducts.

Two-dimensional gel electrophoresis was performed according to theprocedure of O'Farrell (17) with the following modifications. In a multiple

gel apparatus (2), the stacking gel in the second (SDS) dimension wasomitted, and the running gel was 12% polyacrylamide. Samples thatwere in Laemmli sample buffer (11) were diluted into 2 volumes of lysisbuffer (17) prior to isoelectric focusing. The gels were fixed in aceticacid/methanol, fluorographed in PPO in glacial acetic acid, washed, driedonto Whatman No. 1MM paper, and then autoradiographed.

Dot Blots. The procedure used was as described by Thomas (30).Using a Pipetman pipettor, the RNA (100 ng) in a 1-n\ aliquot was applied

to nitrocellulose filters that had been treated by soaking in water, washingtwice with 20x SSC, then drying under a heat lamp. Filters wereprehybridized overnight and typically hybridized to 32P-labeled clonedcDNA probes at 42°for 2.5 days, then washed in decreasing concentrations of SSC/0.5% SDS (from 2x SSC to 0.2x SSC) at 68°.The final 2washes were in 0.1 x SSC/0.1% SDS for 15 min each at 68°.The filters

were air dried and then exposed to Kodak day-pack film using 2 Cronexintensifies at -70°. Plasmid DNA of each clone was purified, and their

Psfl inserts were isolated and nick-translated as described elsewhere(5).6

An LKB microdensitometer with adjustable slit was used to analyzethe dot blot autoradiograms. The film exposure time chosen gave a linearresponse for dots containing known dilutions of radioactivity. The spotintensity of the hybridization dot for 12-day lactating mammary gland

RNA was within this range. The densitometer was then calibrated tomaximize response to the 12-day lactating mammary gland dot for a

particular probe, and all other dots on the filter were read at this setting.

RESULTS

Translation Patterns of Mammary Gland Poly(A)+ RNAs.Individual mRNAs are known to have varying requirements foroptimal translation (32, 36). Since our interest was to examine amixed population of mRNAs, we sought conditions of cell-free

protein synthesis that maximized the number of different mRNAsdetected. SDS/polyacrylamide gel electrophoresis was used tomonitor the diversity of translation products. With lactating RNAas the standard, optimum conditions used 0.8 u.g of poly(A)+RNA, 100 mw K+ and incubation at 25°for 2.5 hr.

Fig. 1 shows a representative autoradiogram of translationproducts of virgin, pregnant, and lactating mammary glandpoly(A)+ RNAs after SDS/polyacrylamide gel electrophoresis.On these gels, only the most active or abundant RNA productscan be visualized; however, dramatic changes in the patterns of

mRNA activity during functional differentiation were evident,while translation activity increased 3-fold.

Five-day lactating mammary gland RNA encoded a high pro

portion of abundant proteins (e.g., Bands 1 through 7). ProteinBands 7, 5, 4, and 1 contain M, 42,000, 29,000, and 25,000caseins and a-LA, respectively4'5 (Fig. 1C). In the region of protein

Bands 2 through 4, there are several abundant proteins, whichinclude X-casein, Wp-protein (4), and an unidentified x-protein.5

Analysis using 2-dimensional gels clarifies this region (see below).

Translation products larger than the M, 35,000 band (whichcorresponds to M, 42,000 casein) were not found under ourassay conditions even when the gels were exposed for longertimes.

In contrast, virgin poly(A)+ RNA generated a wide variety ofpolypeptides, some of which (Bands 1,3,5, 6) migrated to thesame position as in the lactating translation pattern. The mostabundant protein of the virgin RNA translation pattern was Band8. Its RNA was either not translatable or absent from lactatingRNA.

The translation pattern characteristic of lactation emergedduring pregnancy. M, 25,000 and 42,000 caseins (Bands 4 and7) became prominent as early as Day 5 of pregnancy; «-protein,Wp-protein, and M, 29,000 casein (Bands 2, 3, and 5) began to

increase somewhat later (between Days 8 and 10). Band 1increased slowly during pregnancy. Band 6 appeared to maintainits relative proportion at all stages of pregnancy and was prominent in virgin and lactating translation products as well. Activitiesof mRNAs for the higher-molecular-weight proteins (Bands 8 to

11) appeared to decrease as pregnancy progressed. However,if we applied radioactivity to the gels in proportion to the poly(A)+RNA concentration of the mammary gland at each stage, wefound that the products of these mRNAs in the translation assayactually increased during pregnancy (gel not shown). This suggests that activities of the mRNAs for the large proteins increased at a slower pace than the mRNAs encoding the abundant proteins of lactating mammary gland.

These patterns indicate the presence of at least 3 temporallyoverlapping abundant mRNA populations which can be classifiedby the reproductive stages in which we can detect their translation activity (Table 1).

Comparison of the Translation Patterns of Normal andNeoplastic Mammary Tissue. MTW9 of the W/Fu rat strain andMCCLX of the A x C strain are adenocarcinomas that producesignificant amounts of the milk protein, a-LA (8,22). Both tumorshad very high poly(A)-t- RNA contents, averaging 20 and 31

¿ig/gwet weight of tissue, respectively; however, translationactivities of their RNAs averaged only 30 to 50% of the RNAfrom lactating mammary gland.

As seen in Fig. 16, MTW9 and MCCLX poly(A)+ RNAs generated translation patterns that contained elements of both earlypregnancy and lactation in the normal gland. Both tumors express the mRNAs for the high-molecular-weight proteins (Bands

Table 1Temporally overlapping mRNA populations distinguished by cell-free translation

and SDS-polyacrylamide gel electroptioresisa

Reproductive stage Translation product(s) (band no.)

5 P. K. Qasba, A. M. Dandekar, E. Devinoy, T. Horn, and M. Siegel. Isolation

and characterization of low molecular weight rat lactoprotein cDNA clones, submitted for publication.

Virgin-pregnantPregnant-lactatingAll stages

8-111-5,76

a See Fig. 1.

1820 CANCER RESEARCH VOL. 43

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

Normal versus Neoplastia Rat Mammary RNAs

8 to 11). Bands 1, 4, 5, and 7 (containing a-LA and M, 25,000,

29,000, and 42,000 casein, respectively) are also prominent,indicating the presence of active mRNAs for milk proteins otherthan o-LA. However, as in early pregnant (e.g., 5 to 10 days)

mammary glands, both tumors showed reduced activities ofmRNAs encoding Bands 2 and 3 (containing K-, Wp-, and X-

proteins). In fact, there was a disproportionately large amount ofBand 1 («-LA;see also Fig. 3) produced in the tumors compared

to any stage of pregnant mammary gland. It thus appeared fromthe cell-free translation patterns that both MTW9 and MCCLX

tumors had aberrant patterns of poly(A)+ RNA expression.Analysis of a-LA and «-ProteinTranslation Products. Two

molecular forms of a-LA, which in rats is a glycoprotein, have

been isolated from W/Fu rat milk (20,21). By radioimmunoassay,rabbit antiserum to rat milk a-LA, recognized a-LA2 equally well

(21).Immunoprecipitates of 35S-labeled translation products are

illustrated in Fig. 2A. When [35S]methionine was used as traceras here, 3 bands are precipitated, but when [35S]cysteine is

used, only the lowest band is recovered. The 2 upper bands,which do not contain cysteine, have now been identified as «-

proteins by hybrid selection translation assay, while the lowerband corresponds to a-LA (5). Thus, RNA activities for both a-LA and «-proteincould be monitored by using the a-LA antiserum

and the appropriate tracer.Figure 2ßshows the comparison of [35S]methionine-labeled

translation products and a-LA and «-proteinimmunoprecipitates

coded by the poly(A)+ RNA of the MCCLX and MTW9 tumorsand the mammary glands of MTW9-bearing rats. Both tumorshave significant mRNA activity for a-LA but little activity for «-

protein. On the other hand, MTW9 host mammary gland hasrelatively high mRNA activity for «-proteinand less for a-LA. TheMTW9 host mammary gland translation pattern is very similar tolate pregnant gland.

Analysis of Translation Products by 2-Dimensional Gel

Electrophoresis. To further examine the translation products,we performed 2-dimensional gel electrophoresis. This was useful

because 3 distinct cDNA clones containing coding sequencesfor X-caseins and Wp- (4) and «-proteins(included in the broaddiffuse Bands 2 and 3 on SDS gels; Fig. 1C) had been isolated.5

To simplify the identification of the primary translation prod-

uct(s) corresponding to the abundant milk protein RNAs, eachmRNA was purified by hybrid selection to the 6 cDNA clones (5)and translated using [35S]methionine as tracer (Fig. 3A). mRNAs

for a-LA, Wp-protein, and M, 25,000 and 42,000 casein eachprogrammed only one translation product. X-Casein RNA gen

erated a spectrum of translation products, which could be dueto multiple RNAs or to early termination. «-ProteinRNA gener

ated 2 spots of equal intensity that differed both in size and incharge. Fortuitously, the M, 18,400 molecular weight markermigrated close to the a-LA translation product.

Two-dimensional translation patterns of total poly(A)+ RNAsfrom normal lactating mammary glands of Sprague-Dawley rats

and the MTW9 tumor (of W/Fu rats) are shown in Fig. 3B. Sincewe were especially interested in comparing the a-LAs of the 2tissues, we used [35S]cysteine as tracer, so «-protein in the

lactating mammary gland translation products would not obscurethe a-LA region. Both the normal and neoplastic RNAs generated

one predominant spot with a trace amount of a more acidic spotin the a-LA region. This was also seen in 2-dimensional gels ofa-LA immunoprecipitates (gels not shown). This suggests con

servation of overall charge in the primary translation productsfor a-LA in the Sprague-Dawley and W/Fu strains, even thougha-LAs isolated from the milk of various rat strains show different

charge forms (13, 20,21).Tumor and lactating mammary gland translation patterns on

2-dimensional gels differ in the relative proportions of other, asyet unidentified, spots aside from the high-molecular-weight ones

(Fig. 3B). Some of these spots are present to the same extentin the MTW9 tumor as in 16-day pregnant mammary glandpoly(A)+ RNA translation products (Fig. 36, p).6 One of the p

proteins, which is the same size as the a-LA translation product,is present at much higher levels than a-LA in 16-day pregnanttranslation assays (Horn ef a/., Fig. 2, Spot e).6 This probably

contributes to the increase in Band 1 during pregnancy (Fig. 1A).RNA Dot Blot Analysis. In order to examine the large number

of poly(A)+ RNA samples for their content of specific milk proteinRNA sequences, we utilized the dot blot hybridization technique(30).

Filters containing dots of equal masses of total poly(A)+ RNAsfrom virgin, pregnant, and lactating mammary glands and liverswere hybridized individually to 6 lactating mammary gland cDNAclone probes. Fig. 4 shows the autoradiograms. These wereanalyzed by microdensitometry. Chart 1 shows plots of the spotintensities at each stage after normalizing to the values of 12-day lactating mammary gland RNA. We could not detect anyhybridization signal to liver RNA.

RNA sequences for M, 42,000 and 25,000 casein were alreadypresent in virgin mammary gland. X-Casein sequences could not

be detected until about Day 10 of pregnancy. During pregnancy,M, 42,000 casein RNA reached a steady state level of about50% of the sequence content in 12-day lactating mammary

gland, while M, 25,000 casein content increased more rapidlyand attained approximately lactating levels by Day 8 of pregnancy. By dot blot analysis, we could first detect the mRNAsequence for «-proteinat Day 5 of pregnancy (Fig. 4; Chart 1).

At this time, the RNA was also translationally active, since smallamounts of the 2 «-proteintranslation products of 5-day pregnantmammary gland RNA could be immunoprecipitated with a-LAantiserum in translation assays using [35S]methionine (notshown). «-ProteinRNA sequences increased continuously during

pregnancy, attaining relatively higher levels than during lactation.Wp-Protein RNA accumulated in a manner suggesting it is

unstable during pregnancy. It could not be detected until Day 5but then increased rapidly until Day 14. By Day 16, its leveldropped to less than one-half; then it increased rapidly again

during the last week of pregnancy.a-LA RNA showed even more dramatic variation during preg

nancy. Its sequences could be detected in one of the pools ofvirgin poly(A)+ RNA, and very low levels of the protein were alsofound in virgin mammary glands by radioimmunoassay.6 Significant amounts of a-LA RNA were present in one of the pregnant

mammary gland poly(A)+ RNA pools from Days 10 and 12. Thiswas reproducibly found, since filters washed free of a-LA andM, 42,000 casein probes both generated the same «-LApatternwhen hybridized to fresh a-LA probe.

MTW9 and MCCLX poly(A)+ RNA contained higher proportions of the mRNA sequences for M, 42,000 casein and a-LA

6T. M. Horn, F. H. Grantham, and P. K. Qasba. Effects of ovariectomy and

perphenazine treatment on milk protein gene expression in rat mammary glands,submitted for publication.

APRIL 1983 1821

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

T. M. Horn et al.

20 10 20DAYS

PREGNANCY LACTATION

Chart 1. Microdensitometric analysis of the dot blots. The films shown in Fig. 4were directly analyzed using an LKB microdensitometer with laser beam. Theintegrated values of dot intensities were normalized to the values of the 12-daylactating mammary gland RNA preparations for each dot blot, k, molecular weight,in thousands; cas, casein; x-prot, x-protein; wp-prot, Wp-protein.

than for M, 25,000 casein, X-protein, and «-protein.mRNA sequences of X-protein and «-proteinwere barely detectable in

MTW9 poly(A)+ RNA, which also contained very low levels ofWp-protein mRNA. In one pool of primary tumors induced by

dimethylbenz(a)anthracene, the poly(A)+ RNA had low but detectable amounts of M, 42,000 casein and a-LA sequences.

DISCUSSION

During pregnancy, a complicated series of changes occurs inthe mother's mammary glands. The cellular composition is grad

ually transformed due to proliferation of epithelial and stromalcells at the expense of the fat pad. Growth and differentiation ofthese cells are regulated by many hormones (34), the levels ofwhich change dramatically during pregnancy (1, 12, 19, 27, 35,37). The RNA/DNA ratio of the gland also increases progressively(15, 33). Many enzyme activities change markedly during pregnancy and lactation (3), and the subunits of certain enzymes(including lactose synthetase) accumulate differentially (7, 28).

This study shows the interesting result that mRNAs corresponding to 6 distinct abundant lactating mRNAs all becometranslationally active at some time during pregnancy, and theRNAs show different temporal patterns of accumulation. Thisemergence of milk protein RNA precedes the morphologicaldifferentiation of secretory cells and the detection of significantmilk protein in the gland.

Both by translation activity and dot blot hybridization to clonedcDNA probes, the mRNA sequences for M, 42,000 and 25,000caseins were seen to increase earlier than the mRNA sequencesfor other milk proteins and then to attain a steady state. Shortlybefore parturition, relative proportions of the sequences increased again. a-LA mRNA can be present, at very low levels,

even in virgin mammary gland, but it increases transiently between Days 8 and 12 of pregnancy, reappearing in a uniformmanner after the 14th day of pregnancy. These results are similarto the ones obtained earlier, using molecular hybridization touncloned cDNA probes to purified M, 42,000 casein and a-LA

mRNA sequences (15).

mRNAs for X-casein and Wp-protein become detectable byDay 8 of gestation, and «-proteinRNA is present, although at

low levels, as early as Day 5. Appearance of these sequencesmight result from the emergence of new cell types or changes ingene expression of the existing cells triggered by changes in thehormonal milieu of pregnancy.

In the present study, only the accumulated mRNA sequenceswere measured. Rates of transcription or degradation of theindividual mRNA sequences may also change during differentreproductive stages, which could result in the transient appearance of a-LA and Wp-protein RNAs during pregnancy.

Comparison of MTW9 and MCCLX tumors indicated that thetumors had quite similar gene expression patterns. In both ofthe tumors, the relative proportions of RNA sequences for M,42,000 casein and a-LA were much higher than the other milk

protein mRNA sequences examined. By translation assay, thetumors contained the mRNA sequences coding for proteins ofmass greater than 35,000 which were detectable only in virginand pregnant gland. The tumor RNAs also generated the Band6 protein (Fig. 1) common to all the normal mRNA populations.On the other hand, the tumors had comparatively very low levelsof X-casein, «-protein,and Wp-protein RNAs. MCCLX appeared

to have higher translation activity encoding Band 8 and lessencoding Mr 29,000 casein than MTW9 (Fig. Mi). However, bothof their patterns of expression of nonmilk and milk proteinmRNAs did not precisely conform to any stage of pregnancy.

The translation patterns characteristic of MCCLX and MTW9tumors could simply be due to the peculiar hormone environments in which they are maintained (nonpregnant rats havingabnormally high prolactin levels). However, RNA from the mammary glands of MTW9-bearing rats generates a translation pat

tern closely resembling late pregnant mammary gland (Fig. 2).Thus, the differences found in the tumors are intrinsic to them,and not due to the unusual hormonal status of their hosts.

The very low levels of X-casein, «-protein, and Wp-protein

mRNAs in MTW9 and MCCLX tumors may be the result of oneor more of the following possibilities: (a) «-and Wp-proteins are

produced by the adipose cells, a constituent of normal mammarygland not present in the adenocarcinomas; (£>)adenocarcinomasoriginated from a specific, though as yet unidentified, subpopulation of mammary epithelial cells; (c) expression of the «-andWp-protein genes at high levels requires interaction with cell

types no longer present in the tumors; or (d) alterations in thecoding or the regulatory gene sequences of K- and Wp-proteins

are associated with neoplasia. Such possibilities are presentlybeing investigated in our laboratory.

ACKNOWLEDGMENTS

We thank M. Siegel, B. Stubblefield, S. Parmenter, and B. Nervik for theirexcellent technical assistance; Dr. H. Cooper for guidance and use of 2-dimensionalgel equipment; Drs. A. M. Dandekar and S. Safaya for assistance with the cDNAprobe preparation; F. Grantham, D. Hill, and B. Strong for obtaining the tissuesamples. Dr. L. Silver, Laboratory of Molecular Pharmacology (NIAID), kindlyprovided the 14C-labeled bacteriophage T4. We also thank Dr. B. K. Vonderhaar for

her constructive criticism of this manuscript. We are especially grateful to Dr. PietroGullino for his suggestions and encouragement.

REFERENCES

1. Amenomori, Y.,Chen, C. L., and Meites, J. Serum prolactin levels in rats duringdifferent reproductive states. Endocrinology, 86: 506-510, 1970.

2. Anderson, N. L., and Anderson, N. G. Analytical techniques for cell fractionsXXI, XXII. Anal. Biochem., 85. 331-340, 341-354, 1978.

1822 CANCER RESEARCH VOL. 43

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

Normal versus Neoplastia Rat Mammary RNAs

3. Baldwin, R. L, and Yang, Y. T. Enzymatic and metabolic changes in thedevelopment of lactation. In: B. L. Larson and V. R. Smith (eds.), Lactation: AComprehensive Treatise, Vol. 1, pp. 349-411. New York: Academic Press,

Inc., 1974.4. Dandekar, A. M., Robinson, E. A., Appella, E., and Qasba, P. K. Complete

sequence analysis of cDNA clones encoding rat whey phosphoprotein: Ho-mology to a protease inhibitor. Proc. Nati. Acad. Sci. U.S.A., 79: 3987-3991,

1982.5. Dandekar, A. M.. and Qasba, P. K. Rat a-lactalbumin has a 17-residue-long

COOH-terminal hydrophobe extension as judged by sequence analysis of thecDNA clones. Proc. Nati. Acad. Sei. U. S. A., 78: 4853-4857, 1981.

6. Deeley, R. G., Gordon, J. I., Burns, A. T. H., Mullinix, K. P., Bina-Stein, M.,and Goldberger, R. F. Primary activation of the vitellogenin gene in the rooster.J. Biol. Chem., 252: 8310-8319,1977.

7. Hill, R. L., and Brew, K. Lactose synthetase. Adv. Enzymol. Relat. Areas Mol.Biol., 43. 411-490, 1975.

8. Horn, T. M., and Kano-Sueoka, T. Effects of hormones on growth and «-lactalbumin activity in the transplantai*? rat mammary tumor MCCLX. CancerRes., 39: 5028-5035,1979.

9. Horn, T. M., Sodroski, J., and Qasba, P. K. Changing patterns of poly(A)+RNA-coding capacity of normal and neoplastic mammary tissues. Fed. Proc.,39:2017,1980.

10. Kano-Sueoka, T., and Hsieh, P. A rat mammary carcinoma in vivo and in vitro'.

establishment of clonal lines of the tumor. Proc. Nati. Acad. Sei. U. S. A., 70:1922-1926, 1973.

11. Laemmli, U. K. Cleavage of structural proteins during the assembly of the headof bacteriophage T4. Nature (Lond.), 227. 680-685,1970.

12. Linkie, D. M., and Niswender, G. D. Serum levels of prolactin, luteinizinghormone, and follicle stimulating hormone during pregnancy in the rat. Endocrinology, 9f: 632-637,1972.

13. McKenzie. R. M., and Larson, B. L. Purification and characterization of rat «-lactalbumins: apparent genetic variants. J. Dairy Sci., 67: 714-722, 1978.

14. Murphy, G., Ariyanayagam, A. D., and Kühn,N. J. Progesterone and themetabolic control of the lactose biosynthetic pathway during lactogenesis inthe rat. Biochem. J., 736: 1105-1116,1973.

15. Nakhasi, H. L., and Qasba, P. K. Quantitation of milk proteins and their mRNAsin rat mammary gland at various stages of gestation and lactation. J. Biol.Chem., 254. 6016-6025, 1979.

16. Nardacci, N. J., Lee, J. W. C., and McGuire. W. L. Differential regulation of «-lactalbumin and casein messenger RNAs in mammary tissue. Cancer Res.,38: 2694-2699,1978.

17. O'Farrell, P. H. High resolution two-dimensional electrophoresis of proteins. J.

Biol. Chem., 250: 4007-4021, 1975.18. Pelham. H. R. B., and Jackson, R. J. An efficient mRNA-dependent translation

system from reticulocyte lysates. Eur. J. Biochem., 67. 247-256, 1976.19. Pepe, G. J., and Rothchild, I. A comparative study of serum progesterone

levels in pregnancy and in various types of pseudopregnancy in the rat.Endocrinology, 95: 275-279,1974.

20. Prasad, fl., and Ebner, K. E. Charge forms of Wistar rat «-lactalbumin J. Biol.

Chem., 255. 5834-5837,1980.

21. Qasba, P. K., and Chakrabartty, P. K. Purification and properties of two formsof rat «-lactalbumin. J. Biol. Chem., 253. 1167-1173,1978.

22. Qasba. P. K., and Cullino, P. M. n-Lactalbumin content of rat mammarycarcinomas and the effect of pituitary stimulation. Cancer Res., 37: 3792-

3795, 1977.23. Richards, D. A., Rodgers, J. R., Supowit, S. C., and Rosen, J. M. Construction

and characterization of the rat casein and «-lactalbumin cDNA clone. J. Biol.Chem., 256: 526-532, 1981.

24. Rigby, P. J. W., Dieckmann. M., Rhodes. C., and Berg, P. Labeling deoxyri-

bonucleic acid to high specific activity in vitro by nick translation with DNApolymerase I. J. Mol. Biol., 773. 237-251, 1977.

25. Roberts, B. E., and Paterson, B. M. Efficient translation of tobacco mosaicvirus RNA and rabbit globin 9S RNA in a cell-free system from commercialwheat germ. Proc. Nati. Acad. Sei. U. S. A., 70: 2330-2334. 1973.

26. Rosen, J. M., Woo, S. L. C., and Comstock, J. P. Regulation of casein mRNAduring the development of the rat mammary gland. Biochemistry, 74: 2895-

2903, 1975.27. Shiu, R. P. C., Kelly, P. A., and Friesen, H. G. Radioreceptor assay for prolactin

and other lactogenic hormones. Science (Wash. D. C.), 780: 968-971, 1973.28. Smith, S., and Ryan, P. Asynchronous appearance of two enzymes concerned

with medium chain fatty acid synthesis in developing rat mammary gland. J.Biol. Chem., 254: 8932-8936,1979.

29. Taylor, J. M., and Tse, T. P. H. Isolation of rat liver albumin messenger RNA.J. Biol. Chem., 257. 7461-7467,1976.

30. Thomas. P. S. Hybridization of denatured RNA and small DNA fragmentstransferred to nitrocellulose. Proc. Nati. Acad. Sei. U. S. A., 77. 5201-5205,1980.

31. Topper, Y. J., and Freeman, C. S. Multiple hormone interactions in thedevelopmental biology of the mammary gland. Physiol. Rev., 60: 1049-1106,1980.

32. Tse, T. P. H., and Taylor, J. M. Translation of albumin messenger RNA in acell-free protein-synthesizing system derived from wheat germ. J. Biol. Chem.,252: 1272-1278, 1977.

33 Tucker. H A., and Reece, R. P. Nucleic acid content of mammary glands ofpregnant rats. Proc. Soc. Exp. Biol. Med.. 772: 370-372, 1963.

34. Vanderslice. R. W., and Yegian, C. D. The identification of late bacteriophageT4 proteins on sodium dodecyl sulfate polyacrylamide gels. Virology, 60: 265-

275, 1974.35. Voogt, H. K., Sar. M., and Meites, J Influence of cycling, pregnancy, labor,

and suckling on corticosterone-ACTH levels. Am. J. Physiol., 276: 655-658,

1969.36. Weber, L. A., Mickey, E. D., Maroney, P. A., and Baglioni, C. Inhibition of

protein synthesis by chloride. J. Biol. Chem., 252: 4007-4016, 1977.37. Yoshinaga, K., Hawkins, R. A., and Stocker, J. F. Estrogen secretion by the

rat ovary in vivo during the estrous cycle and pregnancy. Endocrinology, 85:103-112,1969.

38. Zubay, G. The isolation and fractionation of soluble ribonucleic acid. J. Mol.Biol.. 4. 347-356, 1962.

Fig. 1. Profiles of translation products of poly(A)+ RNAs of normal and neoplastic rat mammary tissues. Poly(A)+ RNAs were translated using wheat germ extractand [^Slmethionine as tracer in the standard assay. Approximately equal amounts of [^Slmethionine as incorporated into protein were applied to each well of an 11%Laemmli gel so that the relative distributions of individual translation products could be monitored. A, translation profiles during pregnancy. 5Õ.,5-day lactating Sprague-Dawley mammary gland, 50.000 cpm. V, virgin, 34,000 cpm. Days of pregnancy: 5, 26,000 cpm; 8, 50,000 cpm; 10, 36,000 cpm; 12, 50,000 cpm; 14, 46,500 cpm; 16to 20, 50,000 cpm. T4, "C-labeled bacteriophage T4; mol wt, molecular weight. B, MTW9 and MCCLX tumors (of the W/Fu and AxC inbred strains, respectively), 50,000

cpm/slot.To visualize the [^Sjmethionine-labeled translation products, the processed dried gel was exposed to X-ray film overnight. At this exposure time, only the M, 47,000

head protein of the "C-labeled T4 was detectable. To visualize the other T4 proteins, the gel was reexposed for 5 days several months later, by which time the|MS]methionine had undergone several half-lives of decay.

C, assignment of cDNA clones and corresponding translation products of 5-day lactating mammary gland RNA. Caseins (cas): M, 42,000, 29,000, 25.000. and X.Whey proteins: «-LA,Wp-protein (ivp-prof). »-Proteinhas not as yet been identified in rat milk. Correspondence of cDNA and translation product was obtained by hybridselection purification of mRNA and cell-free translation (5). k, molecular weight, in thousands; p, plasmid number.

Fig. 2. A, Immunoprecipitates of cell-free translation products labeled with [^SJmethionine or ["Slcysteine. Aliquots of poly(A)+ RNA were translated in the wheatgerm extract using [^SJmethionine (Met) or ["SJcysteine (Cys) as amino acid tracer. Equal amounts of incorporated radioactivity were applied to each well of an 11.25%Laemmli gel. Approximately 1 x 106 cpm of total translation products (T) were then immunoprecipitated with anti-«-LAantiserum at 1/200 final dilution using Pansorbin(5). Immunoprecipitation in the presence (1C)or absence (IP) of 5 (¿gof unlabeled rat milk «-LA,.Arrows, n, «-LAtranslation product. The 2 upper bands (present only inmethionine-labeled immunoprecipitates) correspond to «-protein.5Note that some bands in the T-lanes are less intensely labeled when [MS]cysteine is used as tracer.

B, comparison of «-LAand »-protein translation products of normal and neoplastic poly(A)-t- RNAs. The translation assay was performed as described, using|35S]methionine as tracer. Equal amounts of radioactivity (1 x 106 cpm) of total translation products were immunoprecipitated with anti-«-LAantiserum and Pansorbin inthe absence or presence of 5 //g of unlabeled «-LA.All T-lanes contain 50,000 cpm of hot-trichloroacetic acid-precipitable material. Hosf, translation assays of poly(A)+RNA from MTW9-bearing rats; MTW9. MTW9 mammary tumor; MCCLX, MCCLX mammary tumor.

APRIL 1983 1823

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

T. M. Horn et al.

no DAY OF PREGNANCY MTW9

145- __T, RNA5L V 5 8 10 12 14 16 18 20 5L

4\MCCLX 5L

Mol wt

Protein

—42k cas

i^ 29k cas— 25k cas—ix-cas

wp-prot—lx-prot"

cDNAClone

p303

p530p305P52P94p 18

1

MET CYS

5L 20P HOST MTW9 MCCLXT IP 1C T IP 1C T IP 1C T IP 1C T IP 1C

1C IP T T 1C IP

1824 CANCER RESEARCH VOL. 43

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

Normal versus Neoplastia Rat Mammary RNAs

-43

42k-cas—-I

25k-cas

I wp-prot

x-casa-LA-

'£•-k

-18.4

3A

SDS

ILactating

Mammary Gland -68MTW9

TUMOR

-43

42k- 42k—».

25k v25k

Iwp

a-LA-

18.4

W/Fu/

»—t

3BFig. 3. Two-dimensional gel electrophoresis patterns of cell-free translation products. A, [^Slmethionine-labeled translation products of mRNAs purified by hybrid

selection to cloned milk protein cDNA plasmids. The individual mRNAs were translated separately, and their protein product(s) were fractionated on separate 2-dimensional gels. Each sample contained about 20,000 cpm of translation product(s) and an equal amount of Bethesda Research Laboratories 14C-labeled molecularweight marker mix (see "Materials and Methods"). The individual autoradiograms were aligned using the molecular weight markers as landmarks and then photographed.18.4, 43, and 68 indicate the molecular weight markers [xlO3 (k)]. Fig. 1C identifies the particular plasmids used to hybrid select the mRNAs. wp-prot, Wp-protein; \-prot, x-protein. B, [^SJcysteine-labeled total translation products of lactating mammary gland and MTW9 tumor poly(A)+ RNAs. Gels were exposed so that «-LAspots

were of equal intensity in the 2 samples. The lactating mammary gland sample also contained the Bethesda Research Laboratories molecular weight markers (18.4, 43,and 68 locations are indicated), p, translation products coded by both the MTW9 tumor and 16-day pregnant Sprague-Dawley mammary gland poly(A)+ RNAs (notshown). (, spot markedly elevated in the MTW9 tumor but not 16-day pregnant or lactating mammary gland. Sprague-Dawley and W/Fu indicate spots that may result

from genetic differences between the 2 strains.

APRIL 1983 1825

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from

T. M. Horn et al.

A. Virgin Pregnant Lactating Virgin Pregnant Lactating

42kcasein

Mabc

wp-

proteinM

abc

25kcasein

•••»•* *

X-

casein * *

x-

Mprotein

abc

M

••

a-lactal-

b bumincL

Mabc

Mab

Fig. 4. Changes ¡nRNA during functional differentiation of mammary glands. Replicate dot blots were made containing 100 ng in 1 ^l of each poly(A)+ RNA preparationfrom mammary glands of mature virgin, pregnant, and lactating rats (M). Liver poly(A)+ RNAs from castrated, virgin, 5- to 12-day pregnant, and 12-day lactating rats arealso spotted (L). The dots on the extreme right (a, b, and c) indicate poly(A)+ RNA (100 ng each) from pooled tumors MCCLX, MTW9, dimethylbenz(a)anthracene,respectively. The blots were hybridized separately for 2.5 days to radioactive probes of 6 cloned cDNAs of lactating mammary gland poly(A)+ RNA, washed to highstringency (0.1 x SSC, 0.1%SDS at 68°),then air dried, and autoradiographed. These are 4.5-hr exposures using 2 Cronex intensifier screens.

Abscissa (left to right), mature virgin; pregnant (5, 8, 10, 12,14, 16, 18, 20 days); lactating (5,12, 19 days). Ordinate vertical axis: each spot is a separate preparationof poly(A)+ RNA.

1826 CANCER RESEARCH VOL. 43

American Association for Cancer Research Copyright © 1983 on July 13, 2011cancerres.aacrjournals.orgDownloaded from