Morphology of the Nucleoprotein Component of Rabies Virus

(1)

JOURNALOFVIROLOGY, OCt. 1968,p.1191-1199 Copyright (D1968 American Society forMicrobiology

Vol.2,No. 10 Prin7tediniU.S.A.

Morphology of the Nucleoprotein

Component

of Rabies

Virus

KLAUSHUMMELER, NATALE TOMASSINI, FRANTISEK SOKOL, ERNST KUWERT, AND

HILARY KOPROWSKI

Divisioni of Experimental Pathology, TheChlildreni's HospitalofPlhiladelphiai, Phliladelphia, Pennsylvania 19146,

Departmenit ofPediaitrics, School of Medicinte, Uiiiversity ofPennsylvania, antd The Wistar Ilnstituite of

Aniatomy and Biology, Philadelphia, Pennsylvania Receivedforpublication 28June1968

Theintracytoplasmic ground substance, ormatrix, associated with the develop-ment ofrabies virus and the nucleocapsid ofthe virus were investigated. The fila-ments of the matrix were identified as virus-specific by means of ferritin-labeled

antibodies. Inthinsections, the diameter was 15 nm and the strands seemed tobe incorporated into virions during morphogenesis of the virus. The nucleocapsid

was isolated frompurified viruspreparations andwas studiedin negative contrast.

The rabies nucleocapsid appeared as a single-stranded helix with a diameter of

16nmandaperiodicity of7.5nm;itslengthwasinexcessofI,im.

Themorphogenesis of rabies virus is inevitably associated with the formation of an

intracyto-plasmic

ground substance or matrix. This has beenobserved in

brain tissue

of mice (14, 23, 24, 29)aswellasin tissue cultures (3, 8).

This filamentous matrix constitutes the

ma-terial

of the

Negri

body in the infected brain of animals (25), and its development precedes the formation ofvirus particles, as studied in tissue cultures

(13).

Fluorescent antibody studies have

yielded

indirect evidence that the substance

of

the Negri body, aswell as thesamematrixininfected tissue

cultures,

is

specific

for the rabies virus

(2,

10). From the

foregoing

evidence itwassuggest.d that the strands which constitute the matrix are the

nucleoprotein

of the virus

(13).

A method has

recently

been

developed

which allows for

production

of

highly

purified

rabies virus (32). With this method, animals

injected

with the

purified

virus preparations

develop

high concentrations

of

specific

antibody

in their sera. These sera, after

conjugation

with

ferritin,

provide

direct evidence that the strands in the matrices are indeed

virus-specific

and thus give some

insight

into the role of these strands in the

morphogenesis

of rabies virus. Furthermore, isolation of the

nucleocapsid

from

purified virus,

containing

all of the viral ribonucleic acid (RNA),

permits

thedeterminationofitsstructure

and its identification with the strands of the matrix.

MATERIALS AND METHODS

Vir-us.

Thechallenge virus standard (CVS), Flury high egg passage (HEP), and Pitman-Moore (PM) fixed strains of rabies virus were used, as indicated below.

Vir-us-tissueculture systems. (i) In the ferritin

con-jugateexperiments, monolayers ofBHK-21 cells (22) were maintained in Basic Medium Eagle (BME) in Hanks' solution supplemented with 10%'- fetal calf serum. Afterremoval ofthemedium, the cell mono-layers were washed and exposed for 1 hr at 37 C to CVS-virus at a multiplicity of5. The virus was sus-pended in maintenance medium consisting of

2c%,

fetal calfserum in BME to which 50 ,ugof diethyl-aminoethyl dextran per ml had been added (15). After virus adsorption, the cell monolayers were washed, maintenance medium was added, and the cultureswereincubatedat 34C. (ii) To prepare puri-fied virusforimmunization, thePM strainwasgrown in Nil-2 cells(9) and processed as described previ-ously (32). (iii)Topreparepurifiedvirus for the isola-tion of thenucleoprotein component, theHEP-Flury strainwaspropagatedinBHK-21cells in thepresence of 3H-uridine and was purified and concentrated as before (32).

Immunle

seruim. Thepurified viruspreparation had

aproteincontentof300,pgper mlasdeterminedbythe Lowry method (19). This material was used for im-munization of adult rabbits (WhiteNewZealand).

Theanimals were first immunized subcutaneously withadoseofabout 0.03 pg ofprotein;2weekslater they received 100 pg of protein, emulsified in equal

parts of complete Freund's

adjuvant,

injected

intra-dermallyinto thefootpadsandat twosites in the chest region. After this, boosterdosesof100pg ofprotein

perdose, withouLt adjuvant, weregiven at 1-, 2-, and 5-week intervals bytheintramuscular route.

1191

on November 11, 2019 by guest

http://jvi.asm.org/

(2)

Nil-2cells, BHK-21 cells, and bovineserumalbumin used as a protein supplement in the tissue culture medium. Againstthehomologous virus,the absorbed serumhadaneutralizing antibodytiter of 1:60,000in the 50% plaque reduction test (40 plaque-forming units) and a complement-fixing titer of 1:2,000 as

measuredagainst 10complement-fixing antigen units of the CVS strain.

Ferritin conjugation. Serum globulin was precipi-tated from the above antiserum bymeansof sodium sulfate andwassubsequently conjugatedto recrystal-lized ferritin (1).

Stainingprocedure. Four days after infection, the cellmonolayers were washed with

PBS,

fixed

briefly

with 4% Formalin, and frozen in situ by means of dry ice. After rapid thawing, the antibody-ferritin conjugate, diluted 1:10 in PBS, was addedand the monolayerswerestained for 1 hrat roomtemperature. After repeated washing with PBS, the cells were

re-moved byscraping;thentheywerefixed in3%

glutar-aldehyde,washedagaininPBS,andpelleted.The cell pelet waspostfixed in osmic acid, dehydrated, and embeddedinepoxy resin.

Isolation of the

ribonucleoproteini

component. The purified virus preparation was mixed with an equal volume of2%sodiumdeoxycholateindistilledwater.

The mixturewaskeptat4 Cfor2min.Subsequently, 1.4mlof the mixturewaslayeredon alinear10 to30% sucrose density gradient (32). The

gradient

was

cen-trifugedinanSW 25.1 rotorina

Spinco

modelL

cen-trifuge for 1 hr at 50,000 X gat 4 C. Thenarrow

bandcontainingthelabelwascollected.

Electron microscopy. Thin sections were stained with uranylacetate and leadcitrate and wereviewed in a Siemens Elmiskop electron microscope at a

magnification of 10,500 X. For demonstration in negative contrast, a drop of the previously dialyzed gradient bandcontaining the nucleoprotein component was transferred to a carbon-covered Formvar grid bymeans ofa platinum loop. A drop of phospho-tungstic acid (pH 6.8) wasadded to this, and the ex-cessfluidwasremovedby filter paper. The preparation was transferred into the electron microscope while still moist and was viewed at a magnification of 50,000 X.

particles

and the absence of such

label

on cell

constituents demonstrated

the

specificity

of the reaction.The distribution of the ferritin on single

particles

was

of interest.

Longitudinal

sections through aparticle (Fig. 4a) or above it (Fig. 4b andc) gave the

impression of ferritin

attachment on the

spikes of

the

viruses,

following

the

honey-comb

array of the

spikes (Fig. 4c) which

has been

described

previously (13).

The

matrices

were

found

to

be labeled

heavily, tothe

exclusion of all

other

cell

constituents (Fig. 5). At a

higher magnification,

the label was

re-stricted

to the strands of the

matrix

(Fig. 6). Measurementsof these strands

varied,

depending on the

plane of sectioning

and possibly the

embedding

procedures used. In areas indicating proper longitudinal sections, the diameter of the strands was

about

15 nm.

The

relationship

of thestrands to the budding

virus particles

can be seen inFig. 7. The labeled

strands

are near the cell

surface,

some of them

seemingly reaching into

the

virus

particles form-ing on the

membrane.

This isalso evident in Fig.

8;

in thisfigure, the

matrix material

bears a close relationship to the virus particle which has been

synthesized

onthemembrane. The early budding process is shown in both figures. A thickening of the

membrane

is

evident with

labeled filaments as an

inner

lining.

The nucleoprotein component obtained from

purified

virus

preparations

is

illustrated

in Fig. 9. The

single

stranded helix had an average outside

diameter

of 15.6 nm and a periodicity of 7.5 nm. The single strand itself was about 3 nm wide. These measurements must be considered as

ap-proximations,

because the coils exhibited an

irregular

periodicity. This lability may be an inherent property of the rabies nucleocapsid or an

artifact

resulting from the procedures used. Most of the strands were tangled, making an accurate measurement of their length difficult.

on November 11, 2019 by guest

(3)

jt*

v4

ik

r'

{t+

S.

4..'%S..e A.

.

FIG. 1. Extracellular rabies virus labeled with ferritin. X 105,000.

FIG. 2.

Idenitification

ofrabies virus in cytoplasmic vacuole. No attachment offerritin to surrounding cellular material. X105,000.

1193

on November 11, 2019 by guest

http://jvi.asm.org/

[image:3.489.62.423.25.609.2]

(4)

FIG. 3.

Long.form

ofrabiesvirusin

cvtoplasmic

vacuole,labeledspecifically. Asin

Fig.

2,1to

ferritill

label

onl

cellcontstituentts. X 105,000.

.. 1.-, :'

y

¢ ...

E L *'

.. t; t;

''t, t t

o .

Wt

FIG. 4. Distributiolnof

ferritinl attcached

to virusparticles.

(A)

Sectionz

throuiglh

a

particle;

(B) sectioni

sliglhtl/

above;

antd

(C)sectionzaboveaparticle. The

ferritiin

conjugate

seemingly

attachesottthe

spikes

ofthe virus

(A

antd

B),

antd

thesurface view

(C)

inidicatesthe

regular

oltlinie

ofthe

spikes.

X 152,000.

on November 11, 2019 by guest

[image:4.489.73.434.75.361.2] [image:4.489.56.450.408.564.2]

(5)

FIG. 5. Cytoplasmicmatrixislabeled distinctly. O0 FIG. 6. Highermagnzificationt of thematrixshIownI i

labelisconfinedto them. Measutrements ini selectedG

lim. X 105,000.

S

6,'o', @ b 44+v} '*

w

'A

~~~~~~~~~~I

i.. IF'

thei cellcomponelntsarefreefromferritint. X 52,500.

if the previousfiguire. Thestrandsarevisible,andthe fe, ireas, indicatedby circles, yielded a diameter of aboi

1195

?rritin tIt IS

on November 11, 2019 by guest

http://jvi.asm.org/

[image:5.489.69.427.32.621.2]

(6)

4 * 1

ti

FIG. 7and 8. Outeredge ofadisintegratedcell. Thevirusparticlesand matrixstrandsnearthem are

specifi-callylabeled. Thzeconnection ofthestrandstoemerging virusparticles can be seen. Theearly buddingprocess

(arrow

a)and thecompletevirusleavingthecelllimits(arrow b) indicateapossibleroleofthematrixstrandsin themorphogenesis ofrabies virus. X 105,000.

1196

on November 11, 2019 by guest

[image:6.489.66.426.23.610.2]

(7)

NUCLEOPROTEIN COMPONENT OF RABIES VIRUS

FIG. 9. Pcart of theniiieleocaipsidofiabies irus. Asiiiglestrciudlecdhelical structure is evident. X 300,000.

It can be stated, nevertheless, that they were at least 1 ,um long, onthe basis ofmeasurements in areaswhere measurement was feasible.

DISCUSSION

The identification of complete rabies virions by means of

ferritin-conjugated

antibodies has been described by Atanasiu et al. (4). The main interest of the present investigation centered on the nature of the fibrils which constitute the

characteristic

matrixand their roleinthe morpho-genesis of rabies virus. Morphogenesis of the investigated strain of fixed rabies virus parallels that ofcertain myxoviruses. The assembly of the viral components is similar to that of some paramyxoviruses, suchas parainfluenza type 2 or SV 5, although the nucleoprotein component is narrower (6, 12). The

intracytoplasmic

ground substance consists of strands about 15 nm in

diameter,

asmeasured ontheir sections. Synthesis ofthenucleoprotein substance occursinthe cyto-plasm, with subsequent virus assembly on mem-branes. A

possible

mechanism of incorporation of the

ribonucleoprotein

into viral particles dur-ing their maturation onmembraneswasobserved only

infrequently

when labeled strands could be seen to be associated with the

budding

particle. In most

instances,

the nucleoprotein fibrils were

only found near the budding site or they were absent in the plane of the section.

Intracytoplasmicinclusionsconsisting ofhelices have been described for another member of the rabies group of

viruses,

namely thevirus ofviral hemorrhagic

septicemia

of the rainbow trout. Similar to the

paramyxoviruses,

these helical strands havea diameterof 18 nminsections

(34).

Theabilitytogrowrabiesvirus intissuecultures

to relatively high titers and then to purify it made itpossible to investigate the structure ofthe ribonucleoprotein in negative contrast.

Thefollowing evidence indicates that thehelices in Fig. 9 represent the viral nucleoprotein: (i) the bandobtainedaftercentrifugationina sucrose density gradient of deoxycholate-disrupted

puri-fied virus contained essentially all of the RNA present in the original viruspreparation, and (ii) this material reacted strongly in complement fixation tests withspecific antiserum but did not yieldanyhemagglutinin. Theincorporationofthe matrix strands into capsids during the morpho-genesis of the rabies virus and their diameter in sections make it obvious that they are identical with the nucleocapsid isolated from virions, as seenin negative contrast.

The data obtained are consistent with the data of Pinteric and Fenje (28), who foundhelical structures 15 to 16 nm in diameter after dis-ruptionofpartially purified virus; however, these structures were not shown to contain the viral RNA.

Similarly,

vesicular stomatitis virus, another member of thisgroup,wasfoundtorelease helices with a diameter of 15 nm after disruption (7, 20, 31).

The morphology of the nucleocapsid of the investigated rabies strain exhibited a close rela-tionship with the nucleocapsids of some myxo-virusesofsubgroup II,such as SV (5),Newcastle diseas2 virus (16), and HVJ (11). Biophysical data have emphasized this similarity (Sokol et al., in preparationz). It would be prema-ture, nevertheless, to classify rabies virus in the myxovirus group. The striking morphology of the virion sets it apart from myxoviruses.

VOL.2, 1968 ₁₁₉₇

on November 11, 2019 by guest

http://jvi.asm.org/

[image:7.489.80.403.64.245.2]

(8)

in the matrices without obvious involvement of pre-existing cell membranes. This was not ob-served in the present experiment in which a

different strain of fixed rabies virus was used.

Differences in the pathogenesis of rabies virus strains have been described. Miyamoto and Matsumoto (26) found that, whereasstreet virus infections resulted in the production of large Negri bodies in mouse brain, fixed virus strains induced only small amounts of matrices in

neurons but exhibited marked cytopathogenicity. These differences in pathogenesis may also

ex-tend to morphogenesis.

No qualitative differences in antigenicity

be-tween fixed strains was evident in the present

study. Although immune serum was prepared

against PM virus and the experimental agent used was the CVSstrain, both thenucleoprotein

strandsand the viruscoatwereeffectivelylabeled,

and the nucleocapsid obtained from HEP-Flury strains reacted strongly with anti-PM serum in

the complement fixation test. The antigenic composition and thediscrepancy in development of theseviruses require further investigation.

ACKNOWLEDGMENTS

Thisinvestigation was supported by PublicHealth Service grants AI-04911, Al-02954, and AI-02405 from the National Institutes ofAllergy andInfectious Diseases, and by a grant from the World Health Organization. K. H.wastherecipient of Public Health

Service award K3-HD-22708. LITERATURE CITED

1. Andrew, G. A., K. C.Hsu,andB.C. Seegal. 1966.

Immunoferritin technique for the identification

ofantigens by electron microscopyp. 527-570.

InD. M.Weir (ed.), Handbook of experimental

immunology. F.A.DavisCo.,Philadelphia,Pa.

2. Atanasiu, P., P. Lepine, and P. Dragonas. 1963.

Etude cinetique du virus rabiqueenculturede

tissusaI'aidedes anticorps fluorescents et des

study of moderate and virulent virus-cell in-teractions of the parainfluenza virus SV5. Virology 30:411-426.

7. David-West, T. S., and N. A. Labzoffsky. 1968. Electromicroscopic studieson thedevelopment of vesicular stomatitis virus. Arch. Ges. Virus-forsch. 23:105-125.

8. Davies, M. D., M. E. Englert, G. R. Sharpless, and V. J. Cabasso. 1963. The electron mi-croscopy of rabies virus in cultures of chicken embryo tissues. Virology 21:642-651. 9. Diamond, L. 1967. Two spontaneously

trans-formed cell lines derived from thesamehamster embryo culture. Intern. J.Cancer. 2:143-152. 10. Goldwasser, R. A., and R. E. Kissling. 1958.

Fluorescent antibody staining of street and fixed rabies virus antigens. Proc. Soc. Exptl. Biol. Med. 98:219-223.

11. Hosaka, Y., H. Kitaro, and S. Ikeguchi. 1966. Studies on the pleomorphism of HVJ virions. Virology 29:205-221.

12. Howe, C., C. Morgan, C. de VauxSt.Cyr, K. C. Hsu,andH. M.Rose. 1967. Morphogenesisof type 2 parainfluenza virus examined by

light

and electronmicroscopy.J.Virol. 1:215-237. 13. Hummeler, K., H.Koprowski, andT. J. Wiktor.

1967. Structureanddevelopmentof rabies virus in tissue culture. J. Virol. 1:152-170. 14. Johnson, R. T., and E. H. Mercer. 1964. The

de-velopment of fixedrabies virus inmousebrain. AustralianJ.Exptl.Biol. Med. Sci. 42:449-456. 15. Kaplan, M. M., T. J. Wiktor, R. F. Maes, J. B. Campbell, and H. Koprowski. 1967. Effecton

polyionsontheinfectivityofrabies virus in tis-sue culture: construction of a single-cycle growth curve. J. Virol. 1:145-151.

16. Kingsbury, D. W., and R. W. Darlington. 1968. Isolation and properties of Newcastle disease virusnucleocapsid. J. Virol. 2:248-255. 17. Kissling, R. E., R. Q.Robinson, F. A. Murphy,

and S. G. Whitfield. 1968. Agent of disease contracted from green monkeys. Science 160: 888-890.

18. Kitajima, E. W., and A. S. Costa. 1966. Morphol-ogy and developmental stages of Gomphrena virus. Virology 29:523-539.

19. Lowry, 0. H., N. J. Rosebrough, A. L. Farr,

on November 11, 2019 by guest

(9)

NUCLEOPROTEIN COMPONENT OF RABIES VIRUS

and R.J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.

20. McCombs,R. M., M.Benyesh-Melnick, andJ. P. Brunschwig. 1966. Biophysical studies of vesicu-larstomatitis virus. J.Bacteriol. 91:803-812. 21. MacLeod, R., L. M. Black, and F. H. Moyer.

1966. The fine structure and intracellular localization of potato yellow dwarf virus. Virology 29:540-552.

22. Macpherson, I., and M. Stoker. 1962. Polyoma transformation of hamster cell clones-an in-vestigation of genetic factors affecting cell com-petence. Virology 16:147-151.

23. Matsumoto, S. 1962. Electronmicroscopy ofnerve cells infected with streetrabies virus. Virology 17:198-202.

24. Matsumoto,S. 1963. Electron microscopestudies of rabies virus inmouse brain.J.Cell Biol.19: 565-591.

25. Matsumoto, S., and K. Miyamoto. 1966. Elec-tron-microscopic studies on rabies virus multi-plication and thenatureof the Negri body. In Symp. Ser. Immunobiol. Standard. Karger, Basel1:45-54.

26. Miyamoto, K., and S. Matsumoto. 1967.

Com-parative studies betweenpathogenesis ofstreet

and fixed rabies infection. J. Exptl. Med. 125: 447-456.

27. Murphy, F. A., P. H.Coleman, and S. G.

Whit-field. 1966. Electron microscopic observations of Flanders virus. Virology 30:314-317. 28. Pinteric, L., and P. Fenje. 1966. Electron

micro-scopic observations of the rabies virus. hI Symp. Ser. Immunobiol. Standard. Karger, Basel 1:9-25.

29. Roots, E., andI.M. Schultze. 1963. Neuere elek-tronenmikroskopische Befunde an Gehirnen nach Infektion mit Tollwutvirus. Zentr. Bak-teriol. Parasitenk. Abt.IOrig. 188:159-173. 30. Siegert, R., H.-L. Shu, W. Slenczka, D. Peters,

and G. MUller. 1967. Zur Atiologie einer un-bekannten, von Affen ausgegangenen men-schlichen Infektionskrankheit. Deut. Med. Wochschr. 92:2341-2343.

31. Simpson, R. W.,and R. E. Hauser. 1966. Struc-turalcomponents of vesicular stomatitis virus. Virology 29:654-667.

32. Sokol,F., E. Kuwert, T. J.Wiktor,K.Hummeler, and H. Koprowski. 1968. Purification of rabies virus grown in tissue culture. J. Virol. 2:836-849.

33. Zlotnik, I., D. 1. H. Simpson, and D. M. R. Howard. 1968. Structure of thevervet monkey disease agent. Lancet 2:26-28.

34. Zwillenberg,L. O., M. H. Jensen, and H. H. L. Zwillenberg. 1965. Electron microscopyofthe virus of viral hemorrhagic septicemia of rain-bow trout (Egtved virus). Arch. Ges. Virus-forsch. 17:1-19.

VOL. 2, 1968