MJClin Pathol: Mol Pathol 1995;48:M264-M268

Proportionally constant quantitativetransmission of nucleolin and protein B23 incycling cancer cells

V Sirri, A Pession, D Trere, L Montanaro, M Derenzini

AbstractObjective-To investigate whether and towhat extent the two major AgNOR pro-teins, nucleolin and protein B23, are main-tained after one cell division inproliferating cells.Design-Using three asynchronouslygrowing human cancer cell lines, TG,SJNKP, and CHP 212 cells, nucleolin andprotein B23 were first identified on SDS-polyacrylamide separated nucleolar pro-teins, transferred to nitrocellulose andsilver stained for AgNOR proteins.Measurement of doubling time indicateda period very close to 24 h for each of thecell lines. To quantify the percentage ofnucleolin and protein B23 maintained indaughter cells after duplication, cells werelabelled with [35S]-methionine and a 24 hcold chase performed. Nucleolin and pro-tein B23 labelling was evaluated by den-sitometric analysis on nitrocelluloseautoradiograms.Results-The radioactivity relative to nuc-leolin and protein B23 bands maintainedin the daughter cells was a constant frac-tion ofthat present before cell duplication.In the three cell lines the percentage ofresidual radioactivity measured in thenucleolin bands was 42i2, 40-6, and 412and in protein B23 bands 48-0, 46-2, and441.Conclusions-After one cell division thenucleolin and protein B23 quantity presentin cells may be highly variable, dependingon the amount of the two proteins presentin the mother cell. This is important inrelation to the correct utilisation ofAgNOR protein quantity as an index forevaluating cell kinetics.( Clin Pathol: Mol Pathol 1995;48:M264-M268)

Keywords: Nucleolin, protein B23, cycling cancer cells.

In the interphase nucleolus there is a distinctset of acidic proteins which are selectivelystained by methods usually used for staining theproteins associated with metaphase nucleolarorganiser regions.' The nucleolar silver stainedproteins are therefore generally defined as in-terphase "AgNOR" proteins. In SDS gels ofnucleolar proteins, demonstrated by the samesilver staining procedure that is applied to cyto-logical preparations for the selective visu-alisation of the AgNOR proteins, two mainargyrophil bands are constantly identified ataround 100 kDa and 40 kDa. AgNOR proteins

are necessary for the transcription of ribosomalgenes.3 Nucleolin has been shown to inducechromatin decondensation by binding to theHi histone4 and to be associated with nascentpre-rRNA.5 Protein B23 is associated withrRNA maturation processes. In resting cellsstimulated to proliferate, the amount ofAgNORproteins begins to increase during earlyGI phase and reaches its greatest value at theend of S phase.6 The quantitative distributionofAgNOR proteins in cancer cells is thereforeused as an index of cell kinetics in tumourpathology.78On the other hand, as far as we know, no

information has been obtained about the quant-itative transmission ofAgNOR proteins in pro-liferating cells from mother to daughter cells.In fact we do not know whether after a celldivision the AgNOR proteins have to be com-pletely resynthesised or whether all the AgNORproteins derived from the previous cell cycleare still present in the daughter cells.This is a crucial point in relation to the

use of the AgNOR protein index in tumourpathology since in tumour lesions in vivo someofthe cancer cells are not continuously dividingbut have left the cycle after division. In thiscase the quantity of AgNOR proteins trans-mitted from the previous cell cycle may nolonger reflect the actual kinetic activity of thecells. To ascertain whether and to what degreeAgNOR proteins are transmitted to the daugh-ters of cycling cells we have evaluated thequantitative transmission of the two mainAgNOR proteins, nucleolin and protein B23,in three human cancer cell lines.For this purpose, we first identified nucleolin

and protein B23 in nucleolar proteins separatedon SDS gels and transferred to nitrocellulosemembrane, using the silver staining procedureaccording to Hozaik et al.9 Then we measuredthe percentage of nucleolin and protein B23transmitted after cell duplication to the daugh-ter cells by [35S]-methionine protein labelling.

MethodsCELL CULTUREThe three cell lines used in this study werederived from human tumours: CHP 212 andSJNKP cell lines were from neuroblastomas;the TG cell line was from a tubal carcinoma.All cell lines were maintained as monolayercultures in RPMI 1640 medium, supplementedwith non-essential amino acids, 100 U/ml ofpenicillin, 100 gg/ml of streptomycin, and 10%fetal calf serum (FCS). Cells were incubatedat 37°C in a humidified atmosphere of 5%CO2, 95% air.

Centro diPatologia Cellulare,Dipartimento diPatologiaSperimentale,Via San Giacomo 14,Bologna, ItalyV SirriA PessionD TrereL Montanaro

Istituto diMicroscopiaElettronica Clinica,Policlinico S Orsola,Universita diBologna, ViaMassarenti 9,40138 Bologna, ItalyM Derenzini

Correspondence to:Professor M Derenzini.

Accepted for publicationI June 1995

M264

on May 17, 2020 by guest. P

rotected by copyright.http://m

p.bmj.com

/C

lin Mol P

athol: first published as 10.1136/mp.48.5.M

264 on 1 October 1995. D

ownloaded from

Nucleolin and protein B23 transmission

CELL DOUBLING TIME MEASUREMENTInocula of 5 x 105 cells were grown in plasticflasks and the cell doubling times were de-termined by counting the asynchronouslygrowing cells in triplicate samples at 24, 36,and 48h intervals according to the methodsdescribed by Patterson.10 Cell doubling timeswere also measured by means of [3H]-thy-midine incorporation. Three flasks of each cellline were labelled for 30 min with 0 5 jCi/ml of[3H]-thymidine (specific activity 25 gCi/mmol)and used as controls; three other flasks werelabelled in the same way for 30 min, washedwith RPMI 1640 containing 1 mM cold thy-midine, and cultured for 24h in RPMI 1640supplemented with non-essential amino acids,100 U/ml of penicillin, 100 jtg/ml of strep-tomycin, and 10% FCS. For the evaluation ofDNA incorporation, cells were collected fromthe flasks in cold phosphate buffered saline(PBS) containing 1 mM thymidine and cent-rifuged for 10 min at 1500g. Pellets were re-suspended in PBS and acid precipitated with0-6 N perchloric acid. Pellets were treated with0*3N KOH to solubilise RNA, and after wash-ing twice with 0-2N perchloric acid, DNA wasextracted with 7% perchloric acid for 15 minat 70'C." DNA quantity was estimated ac-cording to Burton. 12 The radioactivity ofDNAsamples was measured after addition of scin-tillation fluid (Read Gel, Beckman).

in a RPMI 1640 medium without methioninecontaining 60,Ci/ml of [35S]-methionine.Medium with labelled methionine was addedto flasks for 6 h; after the incubation time halfof the cells were collected and used to evaluatethe level of amino acid incorporation into theproteins. The other half of the cells were leftin culture for another 24 h, in a medium con-taining cold methionine, and then collected.To measure the level of radioactivity in theindividual nucleolar proteins, SDS-PAGE-sep-arated nucleolar proteins were transblotted onnitrocellulose membranes as previously de-scribed for AgNOR protein detection. Nitro-cellulose membranes were exposed on a ,-MAX Hyperfilm (Amersham). For image ac-quisition of the autoradiographic signals anEPSON GT 8000 scanner was used. In-

INCN

CL

z

en(9

I *-- 105 kDa

DETECTION OF INDIVIDUAL AGNOR PROTEINSFor each cell line, cells were collected, washedin PBS, and lysed at 4°C in TKM buffer(10mM Tris-HCl pH 7 4, 10mM KCI, 3 mMMgCl2) containing 0 1% Triton X-100, 1 mMPMSF, 10 gg/ml aprotinin, 1 jig/ml pepstatin,and 1 jg/ml leupeptin. The nuclear pellets werethen washed twice with 0-25M sucrose inTKM buffer. Nucleoli were prepared by sonic-ation of the isolated nuclei (15 s cycle and 15 spause at 70 W). The nucleolar pellets werewashed with 0'88M sucrose in TKM buffer,suspended in Laemmli sample buffer,'3 boiledfor 5 min, and centrifuged for 30 s. The proteinconcentration of each sample was determinedaccording to the method described by Lowry'4using serum albumin as standard. Samples ofnucleolar proteins (15 jig) were electro-phoresed in 10% SDS-polyacrylamide gels.Size standards from 200 to 29 kDa (Sigma)were included in each gel. Polypeptides wereelectrotransferred to reinforced cellulose nitratemembranes (BA-S 85, Schleicher and Schuell)which were then cut into strips. For AgNORprotein detection, the membranes were stainedin plastic culture dishes. The membranes werepretreated twice with 20% ethanol for 10 minand the silver staining method with gelatincolloidal developer, selective for AgNOR pro-teins, was used as previously described.9

QUANTITATIVE ANALYSIS OF NUCLEOLARPROTEINS CONSERVED AFTER A CELL CYCLEProtein synthesis of the CHP 212, SJNKP, andTG cell lines was evaluated by incubating cells

_ _ _ ^ 4~~~-,d 39=si ~~~~~~38Figure 1 AgNOR staining on nitrocellulose membrane.The nucleolar proteins extractedfrom the three cell lines(CHP 212, S_JNKP, TG) after SDS-polyacrylamide gelseparation have been transferred to nitrocellulosemembrane and stained by the silver staining procedureaccording to Hozak et al.9 Only a few proteins arestained. The major positive bands correspond topolypeptides at 105, 39, and 38 kDa.

M265

on May 17, 2020 by guest. P

rotected by copyright.http://m

p.bmj.com

/C

lin Mol P

athol: first published as 10.1136/mp.48.5.M

264 on 1 October 1995. D

ownloaded from

Sirri, Pession, Trere, Montanaro, Derenzini

Population doubling time. Values are means (SD)

3H-thymidine labelling"

Countingmethod' (hours) Control 24 hour chase

CHP 212 23-7 770 (125) 351 (50)SJNKP 24-1 662 (152) 289 (22)TG 26-4 758 (- 92) 378 (39)

'Cells were counted in triplicate according to Patterson" and doubling time measured in hours.bCells were incubated with 'H-thymidine for 30 min and either immediately processed for theevaluation of DNA labelling by H-thymidine (control) or incubated without addition of 'H-thymidine for 24 h and then processed (24 h chase). Radioactivity is expressed in dpm/4g DNA.

corporation of [35S] -methionine in the proteinbands was quantified by measuring the peakarea of the densitometric tracings. The GelImage System (Pharmacia) was employed.

ResultsIDENTIFICATION OF NUCLEOLIN AND PROTEINB23 ON NITROCELLULOSE TRANSFERREDSDS-PAGE SEPARATED NUCLEOLAR PROTEINSNucleolar proteins extracted from the three celllines and separated on an SDS-polyacrylamidegel were transferred to nitrocellulose mem-branes and stained with the silver method se-lective for AgNOR proteins. No significantdifference was observed in the distribution pat-tern of the AgNOR proteins among the threecell lines (fig 1). In each line, two polypeptidesat 105 and 39 kDa, respectively appeared to bemarkedly stained by silver. As shown by Rousseland Hernandez-Verdun,'5 these silver stainedproteins correspond to nucleolin and proteinB23. A less intensely stained band was alsovisible at 38 kDa. This appeared to correspondto the isoform of protein B23.

QUANTITATIVE CONSERVATION OF NUCLEOLINAND PROTEIN B23To obtain precise information about the celldoubling time of the three cell lines two differ-ent procedures were employed: (1) growingcell were counted in triplicate at regular timeintervals; data indicated that the doubling timeof the three cell lines was similar, ranging from23.7 to 264h (table); (2) the three cell lineswere labelled with [3H]-thymidine for 30 minand then cultured for 24 h with cold thymidine;after this incubation time the residual radio-activity present in the DNA ranged from 44%to 50% of the control values (table).

In order to evaluate whether and to whatextent nucleolin and protein B23 were main-tained after a cell cycle, we measured the radio-activity in relation to the 105 and 39 kDa bandsafter [35S]-methionine labelling. Growing cellswere labelled with [35S]-methionine for 6 h: halfwere used as controls while the remainder wereleft in culture with cold methionine for a further24 h. The [35S]-methionine incorporation into105 and 39 kDa proteins was evaluated bydensitometric analysis on autoradiographic tra-cings (fig 2). After 6 h of incorporation thepeak area corresponding to 105 and 39 kDaproteins showed values of 0-15 and 0-55 forCHP 212 cells, 0-21 and 0-51 for SJNKPcells, and 0414 and 0 55 for TG cells. Theautoradiographic tracings corresponding to the

CHP212 SJNKP

1-4

!..

... ..:S,.

J 1 2*e-.;.....U.S - t_4.-, i

4- N.ucleI:zin

_~~~~Il "WWJ_ ',.t:,r otei-_I 1W 2

Figure 2 Autoradiography of SDS-PAGE separatednucleolar proteins transblotted on nitrocellulose membrane.The three cell lines, CHP 212, S7NKIP, and TG, havebeen labelled for 6 h with /"S]-methionine, then half ofthe cells (lanes 1) were collected for nucleolar proteinextraction and half were left in culture for another 24 h(lanes 2) in a medium containing cold methionine. Afterthis time the cells were collected for nucleolar proteinextraction.

[35S]-methionine still present in the cells after24 h showed that the area corresponding to105 and 39 kDa proteins was 0-065 and 0-27for CHP 212 cells, 0 09 and 0-25 for SJNKPcells, and 0-06 and 0-25 forTG cells. Measure-ment of autoradiographic tracings relative to[35S]-methionine incorporation into histones,still present after 24 h, indicated that the radio-activity maintained in the three cell lines wasfairly constant, ranging from 51% to 53% ofcontrol values.

DiscussionOur results have shown that, in proliferatingcells, the two main AgNOR proteins, nucleolinand protein B23, are transmitted to the daugh-ter cells and the quantity transmitted is a con-stant fraction of the amount of the two proteinspresent in the mother cells.

Identification of nucleolin and protein B23was obtained using nitrocellulose transblottedSDS-PAGE-separated proteins from purifiednucleoli of three cancer cell lines stained by the

M266

2

* H -!st,,.)ri

on May 17, 2020 by guest. P

rotected by copyright.http://m

p.bmj.com

/C

lin Mol P

athol: first published as 10.1136/mp.48.5.M

264 on 1 October 1995. D

ownloaded from

Nucleolin and protein B23 transmission

silver method selective for AgNOR proteins.9Only two bands, at 105 and 39 kDa re-spectively, were markedly stained. Silver stain-ing ofSDS-PAGE-separated nucleolar proteinshas constantly revealed bands around 100 kDaand 40 kDa.2 Roussel and Hernandez-Verdun'5showed that in western blots of SDS-PAGE-separated nucleolar proteins from CHO cellstwo polypeptides of 105 kDa and 39 kDa werestained by a silver method specific for AgNORproteins. The identification of these two majorsilver stained bands as nucleolin and proteinB23 was demonstrated recently in western blotsusing antibodies and purified proteins.16 It istherefore reasonable to conclude that the silverstained nucleolar proteins at 105 kDa and39 kDa correspond to nucleolin and proteinB23.The quantity of 105 kDa and 39 kDa pro-

teins which remained in the three cell linesafter one cell cycle was then evaluated. In orderto do this we first determined the doublingtime of the three cell lines by counting the cellsat regular time intervals'0 and also measuredthe residual radioactivity of previously in-corporated [3H]-thymidine. Using the countingprocedure the doubling time ranged from 23-7to 26&4 hours and the evaluated quantity ofradioactivity relative to [3H]-thymidine rangedfrom 44% to 50% of that present in the threecell lines 24 hours previously. These data takentogether show that the doubling time of thethree cell lines was very close to the time (24hours) determining the quantity of nucleolinand protein B23 left after a cell cycle using[35S]-methionine labelling. This was also sup-ported by the finding that the radioactivitypresent in the histone band (histones are trans-mitted in a semiconservative way) ranged from51% to 52% of that present 24 hours before.The quantity of radioactivity remaining after24 hours and relative to [35S]-methionine in-corporated into the 105 kDa and 39 kDa pro-teins of the three cell lines ranged, respectively,from 42% to 43% and from 45% to 49%, ofthat measured in the previous cell generation.These values showed that the quantity of nuc-leolin transmitted from mother to daughtercells was similar in the three cell lines, and thesame was also true for protein B23.The quantity of AgNOR proteins can be

used as an index in the evaluation of cell kineticsin routine cytological or histological pre-parations of tumours.78 Cancer cells are char-acterised by a higher quantity of AgNORproteins than the corresponding hyperplasticand normal tissues,17 and the quantitative dis-tribution ofAgNOR proteins is directly relatedto the state of cancer cell proliferation.78 Thefact that after division a cell maintains thequantity of AgNOR proteins which is relatedto the previous proliferating activity in-dependent of whether the cell will divide againcasts some doubt on the validity of the evalu-ation of cell kinetics of tumours by quan-tification of the mean AgNOR protein.Measurement of the mean AgNOR proteinvalue in tumour sections might not preciselyreflect the cell kinetic activity of all cancer cells,since some might have left the cell cycle and

yet still have a high AgNOR protein quantity"inherited" from the previous cell cycle. How-ever, this drawback can be eliminated by meas-uring the coefficient of variation (CV) relativeto the AgNOR protein value. The CV value isinfluenced by the standard deviation, which inturn is influenced by the different content ofAgNOR proteins in proliferating cells. Indeed,the quantity ofAgNOR proteins increases pro-gressively when the cell enters the mitotic cycle,from the GI to the S phase.6 Therefore, theCV relative to the AgNOR protein content canindicate the proliferation state of cancer cells,whatever the quantity ofAgNOR proteins pres-ent in the cell at the beginning of the cycle and"inherited" from the previous cell cycle.The present data are also interesting as far

as the relationship between AgNOR proteinquantity and cell doubling time is concerned.Cell proliferation is conditioned by the syn-thesis of the substances necessary for pro-gression through the mitotic cycle and a suitablequantitative ribosomal biogenesis is its pre-requisite. Accumulation of nucleolin in adultbovine aortic endothelial cells stimulated bybasic fibroblastic growth factor has been re-ported to precede the increase of rDNA tran-scription.'8 As far as protein B23 is concerned,after receptor mediated induction of cellularproliferation by various mitogens the synthesisof this protein increases rapidly at early Glphase. However, cell activation in relation tostimuli which do not induce synthesis of DNAfails to stimulate the synthesis of protein B23.'9It appears therefore that the synthesis of theseproteins represents a crucial event occurringvery early after mitogenic signals.20 A differentquantitative availability of these proteins atearly Gl phase could control cell proliferationin continuously dividing cells such as cancercells. The fact that the quantity of nucleolinand protein B23 transmitted to the daughtercells is a constant fraction of the amount ofthese proteins synthesised during the previouscell cycle may represent an important factorfor maintaining the cell duplication time con-stant, for a continuously dividing cell line.

This work was supported by grants from Ministero della Un-iversita e della Ricerca Scientifica e Tecnologica (MURST)40% and 60%, Pallotti's Legacy for Cancer Research, RegioneEmilia-Romagna (DGR 4243/1991) and AIRC (Milan).

1 Howell WM. In: Busch H, Rothblum L, eds. The cell nucleus,vol 11. New York: Academic Press, 1982:89-142.

2 Olson MOJ. In: Straus PR, Wilson SH, eds. The eukaryoticnucleus, vol 2. New York: Telford Press, 1990:519-59.

3 Miller OJ, Miller DA, Dev VG, Tantravahi R, Croce CM.Expression of human suppression of mouse nucleolusorganizer activity in mouse-human somatic cell hybrids.Proc Nad Acad Sci USA 1976;73:4531-5.

4 Erard MS, Belenguer P, Caizergues-Ferrer F, Pantaloni A,Almaric F. A major nucleolar protein, nucleolin, induceschromatin decondensation by binding to histone HI. EurJf Biochem 1988;175:525-30.

5 Herrera AH, Olson MOJ. Association of protein C23 withrapidly labelled nucleolar RNA. Biochemistry 1988;25:6258-64.

6 Pession A, Farabegoli F, Trere D, Novello F, MontanaroL, Sperti S, et al. The Ag-NOR proteins and transcriptionand duplication of ribosomal genes in mammalian cellnucleoli. Chromosoma 1991;100:242-50.

7 Derenzini M, Ploton D. Interphase nucleolar organizerregions in cancer cells. Int Rev Exp Pathol 1991;32:150-92.

8 Derenzini M, Trere D. Importance of interphase nucleolarorganizer regions in tumor pathology. Virchows Arch B1991;61:1-8.

9 Hozak P, Roussel P, Hemandez-Verdun D. Procedure forspecific detection of silver-stained nucleolar proteins onwestern blots. Jf Histochem Cytochem 1992;40:1089-96.

M267

on May 17, 2020 by guest. P

rotected by copyright.http://m

p.bmj.com

/C

lin Mol P

athol: first published as 10.1136/mp.48.5.M

264 on 1 October 1995. D

ownloaded from

Sirri, Pession, Trere, Montanaro, Derenzini

10 Patterson MJ. Measurement of growth and viability of cellculture: in: Jokoby WB, Pastan IH, eds. Methods in en-zymology, vol 58. New York: Academic Press, 1979:141-52.

11 Montecuccoli G, Novello F, Stirpe F. Effect of a-amanitinpoison on the synthesis of deoxyribonucleic acid of proteinin regenerating rat liver. Biochim Biophys Acta 1973;319:199-207.

12 Burton B. A study of the condition and mechanism of thediphenylamine reaction for the colorimetric estimation ofdeoxyribonucleic acid. Biochem 1956;62:315-33.

13 Laemmli UK. Cleavage of structural proteins during theassembly of the heat of bacteriophage T4. Nature 1970;227:680-5.

14 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Proteinmeasurement with the folin phenol reagent. Biol Chem1951;193:265-75.

15 Roussel P, Hernandez-Verdun D. Identification of Ag-NORproteins, markers of proliferation related to ribosomalgene activity. Exp Cell Res 1994;214:465-72.

16 Roussel P, Sirri V, Hernandez-Verdun D. Quantification ofAg-NOR proteins using Ag-NOR staining on Westernblots. 7 Histochem Cytochem 1994;42:1513-7.

17 Crocker J. Nucleolar organizer regions. Curr Top Pathol1990;82:91-149.

18 Bouche G, Gas N, Prats H, Baldin V, Tauber JP, Teissie J,et al. Basic fibroblast growth factor enters the nucleolusand stimulates the transcription of ribosomal genes inABAE cells undergoing GO-61 transition. Proc Natl AcadSci USA 1987;84:6770-4.

19 Feuerstein N, Chan PK, Mond JJ. Identification of nu-matrin, the nuclear matrix protein associated with in-duction of mitogenesis, as the nucleolar protein B23.Biol Chem 1988;263:10608-12.

20 Feuerstein N, Spiegel S, Mond JJ. The nuclear matrixprotein, numatrin (B23), is associated with growth factorinduced mitogenesis in Swiss 3Ts fibroblasts and with Tlymphocytes proliferation stimulated by lectins and byanti T-cell antigen receptor antibody. Cell Biol 1988;107:1629-42.

M268

on May 17, 2020 by guest. P

rotected by copyright.http://m

p.bmj.com

/C

lin Mol P

athol: first published as 10.1136/mp.48.5.M

264 on 1 October 1995. D

ownloaded from