Stoichiometry of large T antigen and pp53 in complexes isolated from simian virus 40-transformed rat cells.

(1)

0022-538X/83/061061-05$02.00/0

Stoichiometry of

Large T

Antigen and pp53 in Complexes

Isolated from Simian Virus 40-Transformed Rat Cells

MARTINI. FREED, IRA LUBIN, AND DANIEL T. SIMMONS*

Schoolof Lifeand Health Sciences, Universityof Delaware, Newark, Delaware 19711

Received 30 December1982/Accepted 17March1983

Simian virus40-transformed cellssynthesize high-molecular-weight protein com-plexes (22to30S) that consist of the virus-coded large T antigen (81,500 daltons) and the cellular antigen pp53. These complexes were partially purified from lysatesof transformedratcells bysucrosevelocity sedimentation. The stoichiom-etry of the two proteins in the complex was studied by direct enzyme-linked immunosorbent assays, using alkaline phosphatase-conjugated T and anti-pp53 monoclonal antibodies.The resultsfromtheseexperiments indicate that the Tantigen-to-pp53 ratio in the complex is 0.87 ±0.27. Nostatistically significant differenceswerefound in this ratio for faster- andslower-sedimenting complexes. These results fromenzyme-linkedimmunosorbentassaysandprevious molecular weightestimates of the complex suggest that this complexis composed, on the

average, of four molecules ofTantigen and fourorfive molecules of pp53.

Cells transformed with simian virus40(SV40) containaproteincomplexmade upofthe virus-coded T antigen and a cellular phosphoprotein named pp53 (12,16). Thiscomplex sedimentsat

22 to 30S (7, 16) andappears to be resistant to

heat, partialreduction (9, 12), highsalt concen-trations (9, 15), and incubation with chelating

agentsandnonionic detergents (9). The cellular

pp53 protein has a molecular weight of50,000

(50K)to56K(dependinguponthe speciesfrom

which it is isolated) and wasoriginally detected inlysates ofSV40-transformedcellsby immuno-precipitation with hamster anti-T serum(3, 11-13,17, 24). Cells infected with SV40 (9, 17, 23) andsomelinesof normal, uninfected cells (5, 13, 22, 24) also accumulatepp53, although thelevels of the proteinaregenerallylower than in

trans-formed cells. Substantialamountsoftheprotein are made in cellstransformed with other papo-vaviruses (12, 22), adenovirus (21), retroviruses (4, 20), orchemicals (4). Interestingly, the ade-novirus tumorantigen (Elb product) (21) binds to

pp53

incells transformedbythese viruses. In SV40-transformed and -infectedcells, nearlyall ofthe pp53 is complexed with T antigen (7-9,

16), and in some transformed cell lines that

produce largeamountsofpp53, nearlyallofthe T antigen is present in this complex (7; this report).

Thefunction ofthecomplex in the cell is not known,andlittleis known about its structure. In previous work, ithas not beenpossibletoobtain areliable estimate ofthe ratio ofT antigen to

pp53 molecules in this aggregate (7, 9,15,16). In

the present study, we examined the

stoichiom-etry of the two proteins in the complex by di-rect enzyme-linked immunosorbent assays (ELISAs). We chose to use ELISAs because this method isverysensitive and givesaccurate estimates ofantigen (and antibody) concentra-tions(25). Furthermore, it circumvents the diffi-culties inherent inestimatingthe levels ofthese

two proteins in the cell from labeling

experi-ments(1).

It was firstnecessary to find aline of

SV40-transformed cells in which the majority of theT

antigenandpp53arecomplexedto oneanother.

Figure1 shows thatthis is thecaseforalineof

transformed rat cells (line 14B). Labeled

ex-tracts wereincubated with monoclonal

antibod-iesto Tantigen(PAblOl)or topp53 (PAb122)(8)

followed with fixed Staphylococcus aureus to

bring down the immune complexes (Fig.1,lanes aandf, respectively). After the reaction withthe first antibody, the cleared supernatants were incubated with eithernonimmune medium as a control (Fig. 1, lanes b andg), PAblOl (Fig. 1, lanescandh),PAb122(Fig. 1,lanes d andi),or hamsteranti-tumorserum(Fig.1,laneseandj). Theprecipitated proteins were subjectedto so-diumdodecyl sulfate-acrylamide gel electropho-resis(23),and thelabeledproteinsweredetected

in thegel by fluorography at -80°C. The X-ray

film was scannedwith a densitomerto obtain a quantitative estimate of the intensity of each band (data not shown). PAblOl (anti-T) and PAb122 (anti-pp53) precipitated both Tantigen

and pp53 from the cell extracts (Fig. 1, lanes a

and f). Greater than 95% of the pp53 in the

extract wasinitially precipitatedwithPAblO1;a

1061

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

very small fraction remained it camedown withPAb122(compa a and d). This indicated that ne

pp53 was complexed to T

antil

about85% of the labeled Tantige

was initially precipitated with P

mainder (15%) being brought d( tumorserum(compare Fig. 1,lano indicated thatalarge proportionc in this cell line was complexec additional point is that about !

antigenin the extractwasrecogni;

(compare Fig. 1,lanesaande),sh

monoclonal antibody binds to r

notall)of the Tantigen molecules

with the results ofGurney et al. To carry out the ELISAs, mc bodies PAblOl and PAb122 were

hybridoma medium by protein

(Pharmacia Fine Chemicals) cl

accordingtothe method describe

(6) exceptthat3MNaSCNwas u

antibodies.PAblOl andPAb122z

types immunoglobulinG2a and G

ly (Hybridoma Profiles, August

CancerInstitute).The purified ml

bodieswereconjugatedtobovine

phatase (Sigma Chemical Co.; t:

cordingto the method of Voller(

same amountof each antibody (1

a b cd ef g h

_ _

FIG. 1. ImmunoprecipitationofI and pp53 from SV40-transformed

transformed rat(line 14B) cells wer(

with [35S]methionineataconcentrati

The cells werelysed, andimmunopi tionswerecarriedoutinantibodyexc

describedby Gurneyetal. (8). The reacted witheitherPAblOl(lane a)o0 The supernatants (lanes b-e for PA PAb122)wereused in asecond imrr with nonimmune medium (lanes b (lanes c andh), PAb122 (lanes d ar anti-tumor serum (lanes e and j). proteinsweredenatured with sodiun andappliedtoasodiumdodecylsulfa (23). The labeledproteinsin thegeli

fluorography.

n solution and pledto2.5mgofenzyme(900 U/mg)sothatthe

reFig. 1,lanes two conjugated antibodies had similar specific

-arly all of the activities. Extracts of SV40-transformed rat

gen.

Similarly,

cells were

subjected

to sedimentation

through

nin theextract gradients of5to20% sucrosetopartiallypurify

Ab122, the re- thecomplexes. Gradientfractions were

collect-own with anti- ed, and serial twofold dilutions (starting with

esf andj).This 1:4)weremade inprechilled disposable,96-well Afthe Tantigen (flat-bottomed) polystyrene microtiter plates

I to pp53. An (Linbro)with thecoating buffer of Volleretal.

90% of the T (25) as the diluent. The plates were incubated zed

by

PAblOl overnight at 4°C and washed three times with

Lowing

that this buffer A (0.05 M Tris-hydrochloride [pH 7.8],

nost (although 0.001 M MgCl2, 1% bovine serum albumin,

s,inagreement 0.01% NaN3) (10). After these washes,200 ,ulof

(8).

alkaline phosphatase-conjugated PAblOl or )noclonal anti- PAb122immunoglobulin G(1.25 ,ug ofantibody

e purifiedfrom permlin bufferA)wasaddedtoeach well.After

A-Sepharose

an overnightincubation at 4°C, the plateswere

hromatography extensivelywashedwithphosphate-buffered

sa-edby Ey etal. line(0.0027 MKCI, 0.0015 M KH2PO4,0.14M

sedtoelute the NaCl, 0.02 M Na2HPO4 [pH 7.4]) containing

areof

antibody

0.05%Tween20(25), and 200puloffreshlymade

i2b, respective-

substrate solution (1 mgofp-nitrophenyl

phos-1982,

National phate [Sigma] per ml in 9.7% diethanolamine, onoclonal anti- 0.0049 M MgCl2, 0.02% NaN3 [pH 9.8]) was alkalinephos- addedtoeach well.Theplateswereincubated in ype

VII-T)

ac- the dark at room temperature, and the

absor-etal.

(25).

The banceat405 nmwas measured withan MR580 I mg) wascou- Microelisa Autoreader (Dynatech Instruments). Thereactionswereallowedtocontinue untilan absorbance ofgreaterthan 1.0 was obtained in themostactive sample.

Figure 2 shows a plot of enzyme activity

versus fraction number for the reactions with anti-Tand

anti-pp53

antibodies. Thetwoprofiles - T A are similar, peaking around fractions 7 or 8.

9 Althoughthepeaksappeartobequite broad,the

trueantigen distribution is undoubtedly

quanti-- 53

tatively

different because, under these

condi-tions,

a twofold drop in antigen concentration doesnotresult inatwofolddropinactivity(see

below).

The peak fractions correspond to a

sedimentation coefficient of 27 to 29S. This

value lies within estimates of the sedimentation

ratcellsn (STVA)

coefficientof thecomplexobtainedbyothers(2,

r

ateled

for 3 h 7,16)andbyus(datanotshown)asdetermined

ion of 20

uCi/mlr

by immunoprecipitation reactions of gradient

recipitationreac- fractions with anti-T and

anti-pp53

antibodies.

:essessentiallyas Thenearcoincidence of the anti-T andanti-pp53

lysates

werefirst reactionprofilesinFig. 2indicatesthat the two rPAb122(lane f). antigens are similarly distributed in the gradient.

blO1and g-jfor Since the majority of the T antigen and pp53

rlunoprecipitation molecules are complexed to one another in this

and g), PAblOl cellline(Fig. 1),it is reasonableto assumethat

The precipitated the profilesinFig. 2 are, for the most part, those

Thdodecyl sulfate forT antigen and

pp53

in complexed form.

Lte-acrylamide gel To determine the molar ratio of T antigento

weredetectedby

pp53

in thecomplex, itwasnecessaryto

calcu-late the relative antigen concentration of these

http://jvi.asm.org/

[image:2.488.53.245.398.514.2]

(3)

0

w 0 w

ct

0

1.2 I.0

0.8 0.6 0.4 0.2

0.0

2 4 6 8 10 12 14 16 18 20 FRACTION

FIG. 2. ELISA of T antigen-pp53 complexes. SV40-transformed rat (line 14B)cells were grownto confluency in roller bottles. The cells were washed

three times with ice-cold phosphate-buffered saline and lysed in 4ml of0.01 M Tris-hydrochloride (pH 9.0)-0.14M NaCI-0.001 M dithiothreitol-0.5% Noni-det P-40. The lysatewasclarified by centrifugation at 4°Cfor10minat16,000xgandconcentratedto0.6 ml by dialysis against dry Sephadex beads at 4°C. The concentratewaslayeredon a5to20% sucrose

gradi-ent in 0.01 M Tris-hydrochloride (pH 7.4)-0.14 M NaCI-0.001Mdithiothreitol and centrifugedat21,200 rpm for 15 h at 4°C in a Beckman SW41 rotor. Fractions (0.6 ml) were collected in the cold and assayed with enzyme-conjugated anti-T (0) and anti-pp53 (0) antibodies as described in the text. The plotted absorbances at 405 nm (A405) were derived fromassaysofa1:8 dilution of each fraction andare

correctedbytheplatereader forextraneousnoise due

todirt orocclusions ontheplastic. Sedimentation is

fromrighttoleft, and the position of HeLa cellrRNAs isnotedatthetop.

twoproteins in various fractions of the gradient.

These valueswereestimated by plotting

absorb-ance (enzyme activity) versus antigen dilution

for the fractions (Fractions 6 through 11)

con-taining most of the complex (Fig. 3). These

curves describe the reactions of T antigen and pp53 with their corresponding monoclonal

anti-bodies and, for most of these fractions, are

sigmoidal in shape. To estimate relative antigen

concentrations, the point on the curve that is

mostsensitivetochanges in antigen

concentra-tion was determined. This point is called the

50% point and can be used as an estimate of

antigen concentration. It is definedasthat point

which is halfway between the reaction plateau

values athigh and low antigen concentrations.

The low plateau values were obtained directly

from the figure. The high plateau values were

estimated by extrapolation of thecurvesshown

in Fig. 3. These extrapolations were somewhat

difficultfor fraction 9 andarelikelytobe slightly

different from the true values. However, the

anti-T and anti-pp53 reaction curves for this

fraction are nearly superimposable, and

there-fore theirplateau valuesareprobablyvery

simi-lar. Table 1 lists the relative antigen.

concentra-tions of both T antigen and pp53 at the

calculated 50% point for each of fractions 6

through 11 of the complex peak. The ratio of

these two values is also shown in Table 1 and

represents the deduced T antigen-to-pp53 ratio

in each of these six fractions. The average

antigen ratio (and hence molar ratio since the

antibodies are monoclonal) is 0.87 ± 0.27.

Ex-tending these calculations for the fractions

be-yond 6 and 11 had little effect on the average

molar ratio. Sincenoconsistent differenceswere

observed between the calculated molar ratios in

differentfractions, itappearsthat sedimentation

did not separate complexes with different

rela-tiveamounts of Tantigen and pp53.

Indetermining theaccuracyof the calculated

Tantigen-to-pp53 ratio,weshould consider that

someof the anti-Tactivity (upto15%) shownin

Fig. 2 may be due to a reaction with small

amounts of uncomplexed T antigen (Fig. 1)

present in fractions 6 through 11. This point

alone might indicate that the calculated ratio is

toohigh. However, notall(about 90%) ofthe T

antigen molecules are recognized by PAblOl

(see above), and consequently the T antigen

concentration measuredby ELISAissomewhat

lower than it should be. This would have an

approximately equal effectin theopposite

direc-tion, and thereforewethink thatavalue of 0.87

± 0.27 isreasonablyaccuratefor this ratio. The

simplest interpretation of these data is that T

antigen andppS3are, ontheaverage,present in

approximately a one-to-one molar ratio in the

complex.Thisinterpretationdoesassume,

how-ever,that the two proteinsof thecomplex are, on the average, equally available for antibody

binding. This assumption is not unreasonable

because PAblOl and PAb122 precipitate the

complex with about equal efficiency

(unpub-lished data).

It is possible to obtain an independent

esti-mate of the composition of the complex from

published data. Bradleyetal.(2) determinedthe

molecular weights of monomers, dimers, and

tetramersofTantigen by gel exclusion

chroma-tography. The values they obtained

[image:3.488.45.239.59.189.2]

corre-sponded, within ±10%,to the values estimated

TABLE 1. Calculation of Tantigen/pp53ratios

Relativeantigenconcn

Fraction

no. T antigen pp53 Tantigen/

pp53a

6 1/28 1/24 0.85

7 1/27 1/30 1.11

8 1/22 1/20.5 0.93

9 1/16 1/12 0.75

10 1/46 1/19.5 0.42

11 1/23.5 1/27.5 1.17

aAverage =0.87+ 0.27.

28S 18S 4S

4

46,

http://jvi.asm.org/

[image:3.488.252.445.569.666.2]

(4)

0 O

0

S

1.20

FRACTION 9 | FRACTION 10 FRACTION II

0 0.

80.8

-

L

0 0

0.4-1/10 1/100 1/1000 1/10 1/100 1/1000 1/10 1/100 1/1000

ANTIGEN

CONCENTRATION

FIG. 3. Antigen dilution curves of ELISAs of T antigen-pp53 complexes. Cell lysateswere subjected to sedimentation in sucrose gradientsasdescribed in the legend to Fig. 1. Fractions were collected and tested by ELISAs with monoclonal anti-T(O)andanti-pp53(0)antibodies asdescribedin the text. Corrected absorbances at405 nm(A405)wereplotted versus antigen concentration for fractions 6 through 11 of the gradient.

directly from sedimentation coefficients. With

their values for the sedimentation coefficients of

theTantigen-pp53complex and of theTantigen

tetramer, the relationship described by Martin

and Ames (14) gives a calculated molecular weight of 560K ± 10% for the complex. This calculation makes the assumption that the mo-lecular weight values of the complex as deter-minedby gel exclusion and by sedimentationare within 10% of one another. Second, it seems likely that the complex contains a tetramerofT

antigen, becauseMeyersetal. (18) have

report-ed that the tetrameric form of the T antigen analog, D2T, preferentially binds, in vitro, to

SV40 DNA at the origin of replication, and

Reich and Levine (19) recently have shown that

the T antigen-pp53 complex displays the same

SV40 DNA-binding specificity as uncomplexed

(withoutpp53) Tantigen. Consequently, a com-plex with an approximate molecular weight of

560K andtetramericTantigencomponentmight

beexpectedtocontain fourorfive molecules of pp53(basedonmolecularweights of81.5KforT antigenand 53Kforpp53).

Thedeductionsmade above about the compo-sition of the complex and our own estimates derivedfrom ELISAresults areentirely

consis-tent with one another. That four molecules of

pp53are associatedwith theprotein complex is supported by observations of McCormicketal.

(15) which indicate that the pp53 isolated from F9embryonalcarcinoma cells is inatetrameric form and binds to purified T antigen in vitro. However, a complex consisting of four mole-cules ofT antigen and five molecules of pp53

cannotbeexcluded. Althoughnotenough

infor-mationis availabletopredicttheactual arrange-ment of the individual protein subunits, addi-tional information on the structure of the

complex mightprovide insights into its function

in thecell and itspossible role in cellular trans-formation with SV40 and otherpapovaviruses.

We thank E. Gurney for the hybridoma cells and David Usherforhelpfulsuggestions.

Thiswork was supported by Public Health Service grant CA25942 from the National Cancer Institute.

LITERATURECITED

1. Benchimol, S., D. Pim,and L. Crawford. 1982. Radio-immunoassay of the cellular protein p53 in mouseand humancelllines. EMBO J. 1:1055-1062.

2. Bradley,M.K., J.D.Griffin,andD. M.Livingston. 1982. Relationship ofoligomerizationtoenzymaticand DNA-bindingpropertiesofSV40largeTantigen.Cell 28:125-134.

3. Chang, C.,D. T.Simmons,M. A.Martin,and P.T. Mora. 1979. Identification and partial characterization ofnew

antigens from simian virus 40-transformedmousecells.J. Virol. 31:463-471.

4.DeLeo,A.B.,G.Jay,E.Apella,G. C.Dubois,L. W.Law, andL.J.Old.1979.Detection ofatransformation-related antigen inchemicallyinducedsarcomasand other

http://jvi.asm.org/

[image:4.488.52.449.58.316.2]

(5)

formedcellsof the mouse. Proc. Natl. Acad.Sci. U.S.A.

76:2420-2424.

5. Dippold, W. G., G. Jay, A. B. DeLeo,G. Khoury, and L. J. Old. 1981. p53 transformation-relatedprotein: detection by monoclonalantibody in mouse and human cells. Proc. NatI. Acad. Sci. U.S.A. 78:1695-1699.

6. Ey, P. L., S. J. Prowse, and C. R. Jenkin. 1978.Isolation of pure IgG1, IgG2, and IgG,b immunoglobulins from mouse serumusing proteinA-Sepharose. Immunochemis-try 15:429-436.

7. Greenspan, D.S.,andR. B.Carroll. 1981. Complexof

simianvirus 49 largetumorantigenand48,000-dalton host tumorantigen. Proc. Natl. Acad. Sci. U.S.A. 78:105-109. 8. Gurney, E. G., R. 0. Harrison, and J. Fenno. 1980. Monoclonal antibodiesagainst simian virus 40 T antigens: evidencefor distinct subclasses of large T antigen and for similarities among nonviral Tantigens.J. Virol. 34:752-763.

9. Harlow, E., D. C. Pim, and L. V. Crawford. 1981.

Complex of simian virus 40 large-T antigen and host

53,000-molecular-weight proteininmonkeycells. J. Virol. 37:564-573.

10. Kilton, L. J., M. Bradley, C. Mehta, andD.M.Livingston. 1981.Rapid and sensitive quantitative immunoassay for the large simian virus 40 Tantigen.J.Virol. 38:612-620. 11.Kress, M., E. May, R. Cassingena, and P. May. 1979. Simian virus 40-transformed cells express newspeciesof proteinsprecipitable byanti-simian virus 40 tumor serum. J.Virol. 31:472-483.

12. Lane, D. P., and L. V. Crawford. 1979.T-antigenis bound to a host protein in SV40 transformed cells. Nature

(London)278:261-263.

13. Linzer, D. I. H., W. Maltzman, andA.J. Levine. 1979. TheSV40 A geneproductisrequired for the production of a54,000 MW cellulartumorantigen. Virology 98:308-318. 14. Martin, R. G., and B. N. Ames. 1961. A method for

determiningthesedimentation behavior of enzymes:

ap-plication toprotein mixtures. J. Biol. Chem. 236:1372-1379.

15. McCormick, F., R. Clark, E. Harlow, and R. Tjian. 1981.

SV40 Tantigen bindsspecifically to a cellular 53K protein in vitro. Nature(London)292:63-65.

16. McCormick, F., and E. Harlow. 1980. Association of a murine53,000-dalton phosphoprotein with simian virus 40 large-T antigen in transformed cells. J. Virol. 34:213-224. 17. Melero, J. A., D. T.Stitt, W. F. Mangel, and R. B. Carroll. 1979.Identification of newpolypeptidespecies (48-55K) immunoprecipitable by antiserum to purified large T anti-gen and present in SV40-infected and -trans-formed cells.Virology93:466-480.

18. Myers, R., R.Williams,and R. Tjian. 1981. Oligomeric structureof SV40 T antigen in free form and as bound to SV40 DNA. J. Mol.Biol. 148:347-353.

19. Reich, N. C., and A. J. Levine.1982. Specific interactions of the SV40 T antigen-cellular p53 protein complex with SV40 DNA.Virology 117:286-290.

20. Rotter, V.,0. W. Witte, R. Coffman, and D. Baltimore. 1980. Abelson murine leukemia virus-induced tumors elicit antibodies against a host cell protein,p50.Virology

36:547-555.

21. Sarnow, P., Y. S. Ho, J. Williams, andA.J. Levine. 1982. Adenovirus Elb-58kd tumorantigen and SV40 large tu-morantigenarephysically associated with the same 54kd cellularproteinintransformed cells. Cell 28:387-394. 22. Simmons, D. T. 1980. Characterization ofTau antigens

isolated from uninfected and simian virus 40-infected monkey cells and papovirus-transformed cells. J. Virol. 36:519-525.

23. Simmons, D. T., M. A.Martin,P.T. Mora, and C.Chang. 1980. Relationship among Tau antigens isolated from various lines ofsimian virus 40-transformed cells. J. Virol. 34:650-657.

24. Smith,A.E., R. Smith, and E. Paucha. 1979. Character-ization of different tumorantigenspresent in cells trans-formedby simian virus 40. Cell 18:335-346.

25. Voller,A., D.Bidwell, and A.Bartlett. 1976.Microplate enzymeimmunoassaysfor theimmunodiagnosis of virus

infections,p. 506-512. In N. Roseand H. Friedman(ed.), Manual of clinical immunology. American Society for

Microbiology, Washington, D.C.

VOL. 1983

http://jvi.asm.org/