Formation of nucleoprotein complexes between polyoma empty capsides and DNA.

JOURNALOF VIROLOGY, Mar. 1975,p.645-653 Copyright 01975 AmericanSocietyforMicrobiology

Vol.15,No. 3 Printed in U.SA.

Formation of

Nucleoprotein Complexes

Between

Polyoma

Empty Capsids

and

DNA

H. VASKEN APOSHIAN,* RONALD E. THAYER, ANDPRADMAN K. QASBA'

Departmentof Cell Biology andPharmacology, University of Maryland School of Medicine, Baltimore, Maryland21201

Received forpublication18October1974

Purified polyoma empty capsids and polyoma type I DNA interact in a

cell-freesystem toformnucleoprotein complexes. Complexes that consist ofone, two, three, and four empty capsids per DNA molecule have been detected.

Polyoma

virions or capsomers do not react with added DNA to form such complexes.

Productive infection

of mouse

cells

by

poly-oma virus

yields mainly

complete virions

and

empty

capsids

(3, 26)

having

densities

in

CsCl

of

about

1.32

and

1.29

g/cm3, respectively

(26).

The production

of

other

polyoma-related

parti-cles has

been summarized elsewhere (28).

The

purified

empty

capsids

are

believed

to

be devoid

of

nucleic acid (2,

3,

17). Whether

polyoma

empty

capsids

are precursors

of

complete

vi-rions,

by-products

of

virus

synthesis,

or

the

result

of

the

disruption

of mature virions

has

never

been clarified.

In a

number

of

phage

systems,

there

is increasing

evidence

for

the interactions

of

viral

DNA and

empty

capsids. For

example, the

experimental results of Pruss

et

al.

(21), using

the

phage P2-P4

system, are

compatible

with

DNA

packaging via the filling

of

preformed

empty

capsids.

Hohn and Hohn

(11) have

shown that

empty,

headlike

particles

of

phage

X are

active

in

packaging DNA

in

vitro. In the

caseof

phage T4,

at

least

some

DNA

enters

the

phage head

after

the

head has

been

constructed

(13).

Finally, experiments

by

Serwer (25)

indi-cate the formation in vivo

and

in vitro of

complexes between phage T7 capsids and T7

DNA.

The evidence of this

paper

indicates that

ina

cell-free

system

polyoma

empty

capsids and

polyoma

type

I DNA interact

to form

com-plexes.

Such

complexes

are not

formed

between

polyoma virions and added polyoma DNA.

MATERIALS AND METHODS

Media. EC medium (18) was used with primary

mouse embryo cells. Dulbecco's modifiedEagle me-diumcontaining 10%calfserumwasused with baby

mousekidney cells.

' Present address: National CancerInstitute, National In-stitutes ofHealth, Bethesda,Md.20014.

Virions and empty capsids used for binding studies. Polyoma virions and empty capsids used in the bindingexperiments wereproduced byinfecting primary mouse embryo cells, at a multiplicity of infection ofabout 1 PFU/cell, with thislaboratory's

stock of the smallplaquevariantof the Toronto strain

of polyoma virus. The method of Crawford (1) was

used for virus production and harvesting.

Toobtain radioactive virions and emptycapsids,10

,gCi

ofL-

[fH

]tryptophan

(3Ci/mmol)wereadded per

ml of medium at 18 h postinfection. The medium, after infection, contained 1% dialyzed calf serum. Virions and empty capsidslabeled with

[3H jarginine

wereproduced by adding to the cells, after the virus adsorption period, medium containing citrulline and

1%dialyzed calf serum butlackingarginine. L- ['H Jar-ginine (20.6Ci/mmol)wasadded 20 h after infection. The viruswasharvested 6 days postinfection.

The crudeextracts werepurified essentiallybythe method of Winocour(26). The peak of virions and the peak of empty capsids obtained after equilibrium centrifugation in CsCl were each subjected to two

additional CsCl equilibrium centrifugations. The finalpeak of virionsoremptycapsidswasseparately dialyzed, sedimented through the 10to40%standard

sucrose gradient, and dialyzed against 0.01 M Tris (pH7.5)-0.05 M NaCl. Virionsoremptycapsidswere

located in all of the abovegradients byradioactivity and/or hemagglutination assays. The nonradioactive polyoma empty capsids had 44 hemagglutination

units/gg

ofprotein.

Preparation of radioactive polyoma DNA. Ra-dioactive DNA was obtained by adding [4C

Ithymi-dine (56mCi/mmol) or [3H]thymidine (20Ci/mmol)

to babymousekidney cells 18 h after infection with

the large plaque strain of polyoma virus. Polyoma DNAwasextracted from the cells48h afterinfection, using the method ofHirt(9). The DNAwaspurified with phenol, precipitated with alcohol, and passed through twocyclesofcentrifugation in cesium chlo-ride-ethidium bromidegradients(23). The peak

con-taining polyoma typeI DNAwaspooled. The ethid-ium bromide was removed by four extractions with isopropanol and dialysis against0.1xstandard

saline-645

on November 10, 2019 by guest

http://jvi.asm.org/

APOSHIAN,THAYER, ANDQASBA

citrate. The DNA solution was dialyzed finally against 0.02 M Tris (pH 7.5)-0.02 M NaCl and had a 280:260absorbance ratio of 0.538.

Standard assay method for measuring complex formation. The assay iscarried out in a glass tube, which has been treated with Siliclad (Clay Adams Co.) 1 h before use. The standard reaction mixture (0.20 ml) contains 1.5 x 10-2 M Trisbuffer (pH 7.5), 10-2MMgCl2,0.10 mg ofcrystalline bovine albumin (Pentex), 0.61

lAg

ofpolyoma type I

[3H

]DNA (3.48 x 10' counts/min per

jg),

and designated amounts of purified polyoma empty capsid protein. (Pentex crys-talline bovine albumin was usedbecause it is free of endonuclease activity. Two different lots of Armour crystallizedbovine albumin were found to contain an endonuclease activity that converts polyoma DNA from 53S to 14 to 16S material.) After incubation at 37Cfor60min, the reaction mixture is chilledto 4C for 10 minand transferred to the top of the standard sucrosegradient at4C in aSpincoSW41 polyallomer tube. The

standard

sucrosegradient consists of10ml ofaneutral 10 to 40% sucrosegradientcontaining0.01 MTris (pH 7.5)-0.15 M NaCland 0.5 mg of crystal-line bovine albumin (Pentex) per ml. The sucrose gradient is preformedover0.30ml ofaCsClcushion having a density of 1.77 g/ml. Sedimentation is carried out in theSpincoSW41rotor at 4C for2hat

25,000 rpm. After centrifugation, the bottom ofthe tube is punctured, and 20drops/fractionarecollected directly into plastic scintillation vials. Scintillation fluid containing Triton X-100 is added, and the radioactivity in each fraction is determined with a PackardTri-Carbscintillation counter.

Gradients. In addition to the standard sucrose

gradient, described above, the following gradients were used. (i) CsCl isopycnic gradients (26) were performed at 32,000 rpm for 22 h at 25C with a Spinco SW50.1 rotor. (ii) Alkaline 5 to 20% sucrose

gradients were prepared by mixing in a Buchler gradient-forming apparatus 2.3 ml of 5% sucrose containing 0.1 NNaOH-0.5 MNaCl-0.025M EDTA and 2.3 ml of 20% sucrose containing 0.4 N NaOH-0.5 M NaCl-0.025 M EDTA. Centrifugation was for 115 min at49,000 rpm, 4C, in the Spinco SW50.1 rotor. (iii) All gradients were fractionatedby puncturing the bottomof thetubes andcollectingdrops. The radioac-tivity in each fractionwasdeterminedwithaPackard Tri-Carb scintillation counter by using a toluene-based scintillation fluid which contained 0.55% 2,5-diphenyloxazole, 0.01% 1,4-bis[2-(4-methyl-5-phen-yloxazolyl)

]-benzene,

10% water, and 33% Triton X-100(Rohm and Haas).

Other methods. Hemagglutination assays were

performed using guinea pig erythrocytes (1). Protein

wasdetermined by the method of Lowry et al. (12). Concentrationsofpolyoma DNA were determined by ultraviolet absorption, using the extinction coefficient E,,%=200. Electrophoresis was doneby the method of

Maizel (14), using 0.1% sodium dodecyl sulfate-7.5% polyacrylamide gels, 7.5 cm in length. Virions or empty capsids were incubated in a previously de-scribed,sodiumdodecyl sulfate-containing disruption bufferfor 2hat 55Cbefore electrophoresis (14).

Other materials. The following materials were obtained from the listed sources: [sH]thymidine (20 Ci/mmol), New England Nuclear; [14C ]thymidine (116mCi/mmol),Mallinckrodt,or

[14C

Jthymidine (56 mCi/mmol), New England Nuclear; Pronase (nu-cleasefree), Calbiochem; pancreatic DNase and pan-creatic RNase, Worthington Biochemicals; Pentex bovine albumin, crystallized, Miles Laboratories; L-[8HJtryptophan (3 Ci/mmol), Schwarz-Mann; L-[3H]arginine (20.6 Ci/mmol) and L-[14C ]arginine (0.313Ci/mmol), New England Nuclear.

RESULTS

Criteria of purification. The two principal criteria usedto indicate thedegree of purity of the emptycapsids were gel electrophoresis and sedimentation throughsucrose. Gel electropho-resis of purified preparations of empty capsids (Fig. 1A) indicated the presence of polyoma polypeptides P1, P2, P3, and P4 and the ab-senceof detectable amounts ofP5, P6, and P7. Thenumbering system of Roblinetal. (24) has been used to designate the polypeptides. Poly-peptides P5, P6, and P7 are "internal proteins,"

which

contain

arginine but lack tryptophan

(4, 24). They are believed to represent host cell

histones,

probably

bound to

polyoma DNA,

in the virions (4, 24). P5, P6, and P7 represent

about

15 to 20% of the

radioactivity

found in arginine-labeled virions (Fig. 1C; ref. 24). Fif-teen percent of the 12,000

counts/min

recovered in the analysis of the

[I-H]arginine-empty

cap-sids

ofFig. 1A would amount to

1,800

counts/

min. If

significant

amounts of

P5, P6,

and P7 werepresent in the [3H

]arginine-empty

capsids,

they

would be

expected

infractions 40to 46 of

Fig.

1A.

However,

fractions 40 to 46 do not represent adiscrete

peak

and

collectively

repre-sentonly 60

counts/min

or 0.5% of the recovered countsper minute.The [3H

]tryptophan-labeled

emptycapsidsalsodonotappeartobe

contami-nated with

significant

amounts of [3H ]nonviral protein (Fig. 1B).

Sedimentation of

purified

radioactiveempty capsid preparations through the standard su-crose gradient indicates that virtually all the

radioactivity

is inthemajor

peak (Fig. 2).

When

compared

to polyoma virions (240S), added as a

sedimentation

marker,

the empty

capsids

sediment at 189S. Other

investigators

have

reported

values of 140 and

134S

for

polyoma

empty

capsids

(2,

17).

Either ourempty

capsids

are

aggregated

or

they

contain

substantial

im-purities bound

tothe

capsids.

The

electrophore-sis results

(Fig.

1)

indicate that the latter

possibility

is

unlikely. On the

other

hand,

the

140S value reported

forempty

capsids

by

others

646 J.VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

POLYOMA EMPTYCAPSID-DNACOMPLEXES

30

o.1

_

I0

647

C. P2

P4

La t5

Pi P3

0 10 20 30 40 50 0 10 20 30 40 50 0 10 20 30 40 50 FRACTION FRACTION FRACTION

FIG. 1. Sodium dodecylsulfate-polyacrylamide gel electrophoresis of purified polyomaemptycapsids and virions. (A) [3H]arginine-empty capsids; (B) [3H]tryptophan-empty capsids; (C) ["4C]arginine virions.

3

2

10

E

UL

1

10 20 30 40

Fraction

FIG. 2. Sedimentationofpurified [3H]arginine

pol-yomaemptycapsids in the standardsucrosegradient.

When[14C polyomavirionsareaddedas a

sedimenta-tion marker (240S), the peak of

[14C]radioactivity

is atfraction 23.

may be due to the presence of DNA bound to

theempty capsids.

Increase in the sedimentation ofpolyoma type I DNA after incubation with polyoma empty capsids. The incubation of polyoma emptycapsids with radioactive polyomatype I

DNAcauses amarkedchangeinthe

sedimenta-tion behavior of the DNA (Fig. 3). When the DNA is incubated in the absence of theempty capsids, all of the radioactivity remains at the

top of the sucrose gradient (Fig. 3A). The

incubation of the DNA with nonradioactive polyoma empty capsids results in the

appear-ance ofanumber ofnew peaks of radioactivity

(Fig. 3B-G). At a low ratio of protein toDNA (Fig. 3B), a single new peak appears. As the

protein-to-DNA ratio increases to about 13:1, the radioactivity of peak 1 decreases, that of peak 2 increases, andathirdnewpeakappears. Eventually, as peak 3 increases, peak 2

de-creasesandtwonewpeaks,4 and5, appear(Fig. 3F andG). Further characterizations ofsomeof thesepeaks will be described later in thispaper. When radioactive empty capsidsare incubated

inthe standard reaction mixture without DNA,

aprofile identicalto that of Fig. 2 is seen.

The addition of either 7 or lOS polyoma

capsomers, intheplace ofemptycapsids,tothe standard reaction mixtureresulted in all of the radioactive DNA remaining at the top of the standard sucrose gradient. The profiles of the

gradients were identical to that of Fig. 3A. Furthermore, theaddition ofcapsomersto

reac-tion mixtures containing empty capsids and DNAdidnotinhibittheformation of theempty

capsid-DNA complexes. The 7 and lOS

capso-mers wereprepared and purified by the method

ofFriedmannand David (6),exceptthat capso-mers in thepresentexperimentswereprepared

frompurifiedemptycapsids instead of polyoma

virions.

Failure ofpolyoma virions to change the sedimentation behavior ofnon-encapsidated DNA. Figure 4A presents thesucrose gradient

profile obtained after incubation of [3lH]trypto-phan polyoma virions (fulls) with polyomatype

VOL. 15,1975

30 25 20

'.

10

05

A.

P2

Pi

p3 P4

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.504.105.403.66.227.2] [image:3.504.51.244.261.522.2]

APOSHIAN, THAYER, ANDQASBA

9

4

8 3

7

2 6

IX

5 O

x

4 E4

3 3

2 2

"I

2 2

E

cx

J.VIROL.

10 20 30 40 10 20 30 40 10 20 30 40

FRACTION FRACTION FRACTION

FIG. 3. Change in the sedimentation ofpolyoma type I [8HIDNA after incubation with polyoma empty capsids and the influence of the ratio ofproteintoDNA concentrationson the reaction. Thestandardassay

method withtypeI[9H1DNAwasused. In the reaction ofA,emptycapsidswereomitted. The reactions of B, C,

D, E, F, and G had1.6, 4.8, 6.5, 8.1, 13, and16,ug,respectively. Theencircled numbersaretheprotein-DNA

ratiosineach reaction mixture. Recoveries of radioactivity from thesegradients and others in thispaper were greaterthan96%,unlessotherwise stated.Direction of sedimentationisfrom righttoleft in this and all figures.

I ["4C

]DNA.

The single peak of [3H

]radioactiv-ity that is found doesnotcontainany

[14C

]DNA radioactivity. Unlike polyoma empty capsids, full capsids do not change the sedimentation behavior of added DNA.

The addition of [3H ]tryptophan virions to a

reaction mixture containing nonradioactive empty capsids and type I [4C ]DNA did not change the multipeaked sedimentation profile that is observed when empty capsids and

["4C

]DNAare incubatedtogether. Suchmixing

experiments indicate that the failure of virions to react with addedpolyomaDNA is notdueto the presence ofsomeinhibitory factor(s) in the

virion preparation.

Characterization oftheDNA-empty capsid complex. The approximate Svalues ofpeaks2 and 3 (Fig. 3) formed in the reaction between polyoma type I [3H]DNA andunlabeled empty

capsids were determined by the method of

Martin and Ames (15). As a sedimentation

marker,

[14C

polyoma virus (240S) was added to the reaction mixture immediately prior to

sedimentation in thestandardsucrosegradient.

The approximate S value of peak 2was 144S.

Peak 3 cosedimented exactly with the 240S marker.

When the 144S complex formed between [3H ]arginine-empty capsids and

[4C

]type I DNA is isolated from the initial sucrose

gradi-ent and either diluted in or dialyzed against

0.01 M Tris (pH 7.5)-0.15 M NaCl, resedimen-tation in the standard sucrose gradient results

intherecoveryof all the 3H and 14C

radioactiv-ity in asingle peak sedimenting at 144S.

Simi-larly, dilution or dialysis of the 240S complex

results in asingle peak sedimenting at

240S.

The 144Sproduct of the reaction contains the polyoma polypeptides found in polyomaempty capsids, indicating that the reaction involves capsid and notnoncapsid protein (Fig. 5). The 14C peaks near the origin represent the

["4C ]DNA thatwas in thecomplex.

Theprotein and DNAcontentsofsomeof the

complexes formed were calculated using the

specific activity of the ["4C ]DNA and 3Hempty capsids (Table 1). The peak fraction of the144S complex is foundtocontain the leastamountof protein relative to DNA, whereas the 370S complex contains four times asmuch. Table 2

summarizes experiments using three different preparations of 3H emptycapsidstoobtain the percent DNA and protein-DNA ratio of the 144S complex.

648

12

10

8

7

6

t5

3

2

to

E

- A i _ B I

- DNAONLY

2

II

3 Ii

I

-1

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.504.119.405.69.315.2]

POLYOMAEMPTY CAPSID-DNA COMPLEXES

Examination

by electron microscopy of the

reaction

mixture, after

incubation but without

sedimentation

through

sucrose,

shows the

pres-ence ofempty

capsid-DNA

complexes consist-ing of one, two, three, and four empty capsids per

molecule

ofDNA. When the 144, 240, and

320S

complexes are isolated from the sucrose gradient, electron microscopy showed there were one, two, and three empty capsids,

respec-tively,

per

DNA molecule.

Complexes

contain-ing one, two, or three empty capsids per DNA molecule are seen in Fig. 6. No

complexes

are seen if the DNA or the empty capsids are

omitted

fromthe reaction.

The DNA in the

144

and

240S complexes

is not protected from

hydrolysis by

pancreatic DNase (Fig. 7). After complexes

between

empty

capsids and DNA

are

formed,

incubation

of

them with nuclease-free Pronase destroys them

(Fig.

8).

They

are

also dissociated

by

1M NaCl.

Determination of the ratio of polyoma type

I

DNA to types II plus III DNA after

incuba-tion with

polyoma empty

capsids. After

incu-bation in the standard

assay

reaction

mixture,

followed

by sedimentation through alkaline

su-crose,

89% of

type I DNA

and

11% of types

II

plus

III were

found

to

be

present

(Fig.

9).

Since

the

same percentages were

observed

for

the

DNA

prior to

incubation (Fig. 9B),

it

is

con-cluded

that when

these

purified

empty

capsids

are

incubated with

type I

DNA

an

endonu-cleolytic

cleavage

of type I

DNA does

not occur.

Miscellaneous

agents and their effects. The

presence of 0.50 M

NaCl

during the reaction

completely

prevents

the reaction

between

type I

DNA and

empty

capsids. No

inhibition

is

detected

in

the

presence of up to 0.25

M

NaCl.

The

appearance of

peaks

2, 3, 4,

and

5of

Fig.

3 is prevented

by

either pretreating the type I

[3H

]DNA with pancreatic DNase

or

pretreating

the

purified

empty

capsids with nuclease-free

Pronase. Purified

empty

capsids treated with

RNase and

DNase

and then

repurified

still

react

with

polyoma

DNA.

Mercaptoethanol

(0.5

x 10-3

M),

EDTA

(0.04

M),

or

preincubation

of type I

[3H]DNA with

pancreatic RNase did

not

affect the

appearance of

peaks

2,

3,

4, or 5

of

Fig.

3.

When

["C

]dTMP

was

used

in

the standard

reaction mixture in place of radioactive

poly-oma type I DNA,

all

of

the radioactivity

re-mained at the very top of the sucrose

gradient

and there was no

detectable

binding

of the

[14C

]dTMP

tothe

purified

empty

capsids.

DISCUSSION

These

experiments

demonstrate that

poly-omaDNA and

polyoma

empty

capsids

interact

5

14

N

6

3

x

I'

14

02

3

£'

I0

toI

5

3 10

2

II A 3H-FULLS

-14C-DA-DNA

B 3H-FULLS

i

C-'4C-DNA

i

- 1 .

10 20 30

FRACTION

40

FIG. 4. Incubation of polyoma type I DNA with polyoma virions does not change its sedimentation behavior. The standard assay method was used, except emptycapsidswereomitted. Inthe reaction of

A, [3H]tryptophanvirions("fulls"),8.1ug,andtypeI

["CIDNA (1.18 x 10' counts/minperug), 0.62 Mg,

wereused; B, ["4C]DNAwasomitted;C, virionswere omitted. Thespecific activity ofthevirions was1.63

x 102counts/minperAgof proteinand 74

hemaggluti-nationunits/agof protein.

to form complexes. At a low ratio of empty

capsidproteintoDNA,a1:1complexis formed.

As the ratio increases, two capsids per DNA molecule and three capsids perDNA molecule

arefound.There isaspecificitytothereaction;

VOL. 15, 1975

649

5

0

3 3

x

5

2 i,,

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.504.272.420.65.507.2]

APOSHIAN, THAYER, AND QASBA

E

0.

u

0 20 30 40 50

FRACTION

FiG. 5. Polyacrylamide gel electrophoresis of the 144S complex formed between 'Hemptycapsids andtypeI ["4CJDNA. (Left panel) Standard assay method except that the reaction mixture contained 7.2 Mig of

['HJarginine-empty capsids (2.25 x 10' counts/minper Mg) and 0.74 Mg of type I ["4CJDNA (1.04 x 10'

counts/minper Mg). The 144S peakwaspooled, ["4C]arginine polyoma virionswereaddedasamarker, and the

proteinswereprecipitated with10%otrichloroacetic acid. The precipitatewaswashedwithacetoneandanalyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. (Right panel) As above,exceptreaction mixture contained 10 Mg of [3H]tryptophan-empty capsids (3.14 x 10' counts/minper MAg) and 0.74 Ag oftype I

[image:6.504.121.407.60.258.2]

[14C DNA.

TABLE 1. Protein and DNA analysis of complexes formedbetween empty capsidsand type I DNAa

ompex

D~PrtenA

Estimated

Complmx'

DNAc

DNA amt of

capsiDs

(%oftotal) raid poen perDNA

molecule

144S 10.7 8.3 (1) 1

240S 6.3 14.9 1.8 2

320S 3.7 25.9 3.1 3

370S 3.0 32.8 4.0 4

aThestandard assay methodwasused, except that

each collected fraction of the sucrose gradient

con-tained10drops.The reaction contained

['H

Jarginine-emptycapsids,7.2ug(1.85 x 10'

counts/min

per

jg

of protein), and type I [4C

]DNA,

0.74 ,ug (1.01 x 104

counts/minperjAg).Recoveries of radioactivitieswere

90% for "ICand 88% for IH.The values in thetables

are for the peak fraction of each of the complexes collected from the sucrosegradient.

bS values were determined insucrose.

c(Micrograms ofDNA/microgramsofDNA +

mi-crograms ofprotein) x 100.

dMicrogramsofprotein/microgramsofDNA.

that is, the added

DNA

combines

with empty

capsids but

not with

complete

virus

particles.

We do

not

wish

to

imply

a

specificity

for the type of

DNA, since

polyoma

types I or II

plus

III

DNA,

as

well

as mouse

DNA,

bind the empty

TABLE 2. Relative content ofproteinand DNAin the DNA-empty capsid144Scomplexa

Empty capsids 1 DNA Protein-DNA Expt labeled with (%)' ratioc

1 [3H]tryptophan 9.4 9.6

2 [SH ]tryptophan 10.6 9.4

3 [SH

]arginine

10.2 8.7

aThe standard assay method was used. In each

experiment, 0.74 ug of polyoma type I

["4C

]DNAwas

used. The amounts of empty

'H

capsids used in experiments 1, 2, and 3 were 6.2, 10.0, and 7.2

Ag,

respectively.

b,C Calculatedasshown inlegendtoTable1.These values are basedon all thefractions obtained in the 144S peak of the gradients.

capsids (Aposhian

and Thayer,

unpublished

data). However,

it

should

be remembered that

pseudovirions,

which contain mouse DNA frag-ments,

also

are

produced

during productive

infection

by polyoma

virus (16, 22, 27).

The

results of these

experiments

suggest that

the

polyoma

empty

capsids

contain somesite(s)

which has

an affinity for DNA. These sites do not appear to exist on the outer surface of

complete

virions

and, therefore,

probably

may not

be

on

the

outer surface ofthe emptycapsids.

Rather,

the

site(s)

may

be

attached to orpart of

650

J.VIROL.

P2'

0

9'+

84

7'1

6'4

5

,I-4,

21

I P3 P4P5-7

&O'

10

9

8 7

cl-6

6

5E

0.

4.._Cs, 3 2 0

FRACTION

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.504.63.255.353.467.2] [image:6.504.265.458.354.437.2]

POLYOMA EMPTY CAPSID-DNA COMPLEXES

the inner surface of the

empty

capsids

or,

alternatively,

at a

possible point

of

incomplete-ness

of

some

of

the

empty

capsids.

However,

we do not know of any

published

evidence in

support of

such

a

postulated

incompleteness.

The

possibility

that such

sites are present on

the

outer surface of

the

complete virions,

but

are

occupied

already

by

some

nucleic

acid,

appears

unlikely.

The

virions

have

been

par-tially purified

by

at

least three

CsCl

equilib-rium

centrifugations. The

DNA-empty

capsid

complex is dissociated

in

such concentrations of

.CsCl

(Aposhian,

unpublished data).

It

seems

reasonable

to assume

that

any

complex

formed

between

unencapsidated

nucleic acid and the

possible sites, if

they did exist

on

the

outer

surface

of

virions, also would be dissociated and

separated

in

CsCl.

Also,

empty

capsids

purified

from

cells

that

were

labeled before

or after

infection with radioactive

thymidine

were

found

to

be free

of

radioactivity.

Although

we

have

been

able

to

detect,

by

the

present

methods,

only

polypeptides P1,

P2,

P3,

and P4

in

the

purified

preparations of empty

capsids

and

in

the

DNA-empty

capsid

complex,

we areaware,

of course,

that

as yet no one

has

been able

to

purify

polyoma

virions

absolutely

free

of

traces

of

contaminating proteins.

Ho et

al.

(10)

have

reported

that

polyoma

empty

capsids

do

not

appear

to react

with

mouse

[3H

]DNA,

as

judged

by

the failure

ofthe

DNA

to

reduce the

hemagglutination

activity

of

the

empty

capsids and

by the failure

to

precipi-tate

radioactivity after

mixing [3H

]DNA,

empty

capsids, and

polyoma

virusantiserum. Hoetal. (10) used very small ratios of

protein and DNA

(0.02:1 or

less).

Complex

formation

would

not

be expected

to occur at

such

small ratios

(Fig.

3).

Friedmann

(5)

has

disrupted polyoma

virions to

capsomers and DNA

by

using carbonate

buffer

(pH 10.6)

and dithiothreitol.

Attempts

(5) at

reassembly of the disrupted virus

by

overnight

dialysis of the preparations resulted

in

the formation

of a

100S DNA-protein

com-plex consisting

mostly of linear

aggregates of

capsomers,

indicating

to

Friedmann

that

poly-oma

capsomers

bind polyoma DNA. Our

experi-ments

demonstrate that

purified,

naturally

oc-curring

empty

polyoma capsids also bind

poly-oma

DNA.

We do

not

know

at present

whether the

formation

of

these

empty

capsid-DNA

com-plexes will

be useful

as a

model

system for

the

study

of

the

assembly

of

polyoma virions.

How-ever,

experiments

are in progress to

determine

whether the 55S polyoma nucleoprotein

com-plex, isolated

by

Green

et

al.

(8) and

Goldstein

et

al.

(7), also

binds

polyoma empty

capsids.

There

is

evidence supporting

aprecursor role for empty

capsids

in

the

maturation of another papova

virus,

simian virus

40,

as has

been

2

;

,'

> 4, ' 4

J4 4 r ..

X,>t

¢,X@A:U^i¢8~4

Wat7S~~~~~~4

.EL ,4.C, 'I

;,*

; tb+X

.. lb.

;=,;,,; t.v.*,-.x !

-F | t > ^ * ^ ' Ja | ' ; \tC^i;>-<-~~~4'

, -,x -ieP '9>|

a' 5;r * s

444.~~~~~~r

`5jf 'tS e t' '|aX,'

'it~~~~~~~~~~~~~i

4.

FIG. 6. Complexesof polyomaemptycapsidsandpolyomaDNAformedin the standard reactionmixtureas seen byelectronmicroscopy.Reaction mixture contained type IDNA,0.80Ag, and12.0lAg ofemptycapsids. Protein monolayer technique was used.Magnification, 72,000.

VOL.15, 1975

₆₅₁

on November 10, 2019 by guest

http://jvi.asm.org/

APOSHIAN, THAYER, AND QASBA

In

0.

E

0

E

cq

6

Ei

E

0

k-)

4In

ion

6

06

FRACTION

FIG. 7. Sensitivity ofthe DNA in theDNA-empty capsid complex topancreatic DNase. (A) The

stan-dard assay method containing 11.2 ug of polyoma

empty capsid protein and 0.68 Mlgof polyoma typeI [8HIDNA wasused. Thefractions from thestandard

sucrose gradient were collected inplastic minivials

and 50gliters ofeach fractionwasplaced on

What-man GF/A glass filterpaper. Thefilters were dried

andplacedinplastic countingvials. Toluene scintilla-tion fluidwas added and thesamples werecounted.

(B and C) A 50-Mliter amount of fraction 37ofthe abovegradientwasplaced in each oftwo test tubes.

To one ofthe test tubes (C) were addedMgCI2, to giveafinal concentration of5x 10-3M, and pancre-atic DNase (1

gg/ml).

MgCI2 and DNase were not

added to the other tube (B). The tubes were

incu-batedfor15minat37C andthenchilledat4 C. To

each tubewereadded 0.22 ml of2.7 x 10-2MNaK

10 20 30 4

E5 4

3

2

0

10 20 30 40

FRACTION

FIG. 8. Digestionof the capsid-DNA complexes by Pronase. The standard assay method was used, except thatserum albuminwasomitted from the incubation mixture and the gradient. The reaction mixture contained 11.2 gg of polyoma empty capsid protein and0.68ugof polyoma type I[3HJDNA.After 60 min at 37C, the reaction mixture was divided into ali-quotsof 0.10ml. To onealiquot nuclease-free Pronase (100ug/ml)was added. Each aliquot was incubated at

37Cforanadditional30minand thenplacedonthe standard sucrose gradient. Symbols: With Pronase, (0); without Pronase, (a).

pointed

out

by Ozer and Tegtmeyer

(19,

20).

The

observation that polyoma

empty

capsids

form

complexes

with

polyoma

DNA

raises

a

number

of

other

interesting

questions. Do the empty

capsids

protect

the DNA

in

the

complex

from

the action

of

bacterial

restriction

en-zymes?

Are

the

empty

capsids attached

to

specific

segments of

polyoma DNA? These

questions

are

being investigated.

ACKNOWLEDGMENTS

Wearegratefuland indebted toThomas J. Kelly, Jr., for the electron microscopy. We thank Maybritt Doelp for excellent technical assistance.

This workwassupportedbyPublic Health Service grants GM-19209 from the National Institute of General Medical Sciences and CA-10497 from the NationalCancerInstitute, andbyagrantfromthe JohnA.HartfordFoundation, Inc.

EDTA and 0.03 ml of 1 NNaOH. The contents of

each tube were quantitatively transferred to the top

ofanalkaline 5 to 20%sucrosegradientand centri-fuged. Fractions (7 drops) werecollected directly in

plasticscintillation vials and counted in scintillation

fluid containing Triton 100-X. (DandE) SameasB

and C, except that 50,Lliters of fraction 30, instead

offraction 37,wasused.(FandG)SameasB andC except that 50 ,uliters of fraction 23, insteadof

frac-tion37,wasused.

FRACTION

652 J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:8.504.64.258.52.483.2] [image:8.504.269.459.60.237.2]

POLYOMA EMPTYCAPSID-DNA COMPLEXES

20 30 40 10

FRACTION

FIG. 9. Determination of theratio of polyomatype I DNA to typesII plus III DNA after incubation with polyoma emptycapsids. The standardassaymethod

was used. In each experiment, 0.61 ug of polyoma [3H]DNA typeI(3.43 x 104counts/minpergg)and 6.5 ugof polyoma empty capsid protein were used.

Afterincubation for1hat37C, the reaction mixture

was chilled to 0C and 0.02 ml of0.2MEDTA and 0.025 mlof1N NaOHwereadded withgentle mixing.

Then 0.20 ml of the mixture was placed on a

pre-formed alkaline 5 to 20%sucrosegradient and

cen-trifuged. Each fraction contained 7drops. (A) after incubation; (B) before incubation.

LITERATURE CITED

1. Crawford, L. V. 1962. The adsorption of polyoma virus. Virology18:177-181.

2. Crawford, L. V., and E. M. Crawford. 1963. A

compara-tivestudy of polyoma and papillomaviruses.Virology

21:258-263.

3. Crawford, L. V., E. M. Crawford, and D. H. Watson.

1962. Thephysical characteristics of polyoma virus.I. Two typesofparticle. Virology 18:170-176.

4. Frearson, P.M., andL. V.Crawford. 1972.Polyomavirus

basicproteins. J. Gen. Virol. 14:141-155.

5.Friedmann, T. 1971. In vitro reassembly of shell-like particles from disrupted polyoma virus. Proc. Nat.

Acad.Sci. U.S.A. 68:2574-2578.

6.Friedmann, T., and D. David. 1972.Structural roles of

polyomavirusproteins.J.Virol. 10:776-782. 7.Goldstein, D. A., M. R. Hall, and W. Meinke. 1973.

Propertiesofnucleoprotein complexes containing repli-cating polyomaDNA. J.Virol.12:887-900.

8. Green, M. H., H. I. Miller, and S. Hendler. 1971. Isolation of a polyoma-nucleoprotein complex from

infected mouse-cell cultures. Proc. Nat. Acad. Sci.

U.S.A. 68:1032-1036.

9. Hirt,B.1967. Selective extraction ofpolyomaDNA from

infectedmousecell cultures.J. Mol.Biol. 26:365-369.

10. Ho, J. K., G. Y. Chan, and J. C. Chan. 1970.Lackof

interaction between polyoma virus capsid andmouse

cell DNA in vitro. Life Sciences 9:1255-1260.

11. Hohn, B., and T. Hohn. 1974. Activity ofempty,headlike particles for packaging of DNA of bacteriophagesAin

vitro.Proc.Nat. Acad. Sci. U.S.A. 71:2372-2376.

12. Lowry, 0. H., N. J.Rosebrough, A. L. Farr, and R. J.

Randall. 1951. Protein measurement with the Folin phenolreagent.J. Biol. Chem. 193:265-275.

13. Luftig, R. B., W. B. Wood, and R. Okinaka. 1971. Bacteriophage T4 head morphogenesis. On thenature

ofgene49-defective heads and their roleas

intermedi-ates.J.Mol. Biol. 57:555-573.

14. Maizel, J. V., Jr. 1971.Polyacrylamide gel electrophoresis ofviralproteins, p. 179-246.In K. Maramoroschand

H. Koprowski (ed.), Methods in virology, vol. V.

AcademicPress Inc., New York.

15. Martin, R. G., and B. N. Ames. 1961. A method for determining the sedimentation behavior ofenzymes:

application to protein mixtures. J. Biol. Chem. 236:1372-1379.

16. Michel, M. R., B. Hirt, and R. Weil. 1967. Mouse cellular DNA enclosed in polyoma viral capsids (pseudovir-ions). Proc. Nat. Acad. Sci. U.S.A. 58:1381-1388.

17. Murakami, W. T., R. Fine, M. R. Harrington, and Z. Ben Sassan.1968.Properties and amino acid compositionof

polyomaviruspurified by zonal ultracentrifugation. J. Mol. Biol. 36:153-166.

18. Osterman, J. V., A. Waddell,and H. V.Aposhian. 1970. DNA andgenetherapy: uncoating of polyoma pseudo-virus in mouseembryo cells. Proc. Nat. Acad. Sci.

U.S.A.67:37-40.

19. Ozer, H. L.1972.Synthesis and assembly ofsimian virus 40. I.Differentialsynthesis of intact virions andempty

shells. J. Virol. 9:41-51.

20. Ozer, H. L., and P. Tegtmeyer. 1972. Synthesis and assembly of simian virus 40. II. Synthesis of the major capsid protein and its incorporation into viral particles. J. Virol. 9:52-60.

21. Pruss, G., R. N. Goldstein, and R. Calendar. 1974. In vitropackaging of satellite phage P4 DNA. Proc. Nat. Acad. Sci.U.S.A. 71:2367-2371.

22. Qasba, P. K., and H. V.Aposhian. 1971. DNA andgene

therapy: transfer ofmouseDNAtohuman andmouse

embryonic cells by polyoma pseudovirion. Proc. Nat. Acad. Sci.U.S.A.68:2345-2349.

23. Radloff, R., W. Bauer, and J. Vinograd. 1967. A dye-buoyant-density method for the detection and isolation ofclosedcircular duplex DNA: the closed circular DNA in HeLa cells. Proc. Nat. Acad. Sci. U.S.A. 57:1514-1521.

24. Roblin, R., E. Harle, and R. Dulbecco. 1971.Polyoma virus proteins. I. Multiple virionscomponents.Virology 45:555-566.

25. Serwer, P. 1974.Complexes between bacteriophage T7 capsids and T7 DNA. Virology 59:89-107.

26. Winocour, E. 1963.Purification ofpolyoma virus.

Virol-ogy19:158-168.

27. Winocour, E. 1968. Further studiesontheincorporation

ofcell DNA into polyoma-related particles. Virology

34:571-582.

28. Yelton, D. B., and H. V. Aposhian. 1972. Polyoma pseudovirions.I.Sequenceofeventsinprimarymouse

embryo cells leading to pseudovirus production. J.

Virol. 10:340-346.

VOL.15, 1975

653

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.504.53.247.60.235.2]