Analysis of the Spodoptera frugiperda Nuclear Polyhedrosis Virus Genome by Restriction Endonucleases and Electron Microscopy

Analysis of the

Spodoptera

frugiperda

Nuclear

Polyhedrosis

Virus Genome by

Restriction

Endonucleases and Electron

Microscopy

LAMBERT C. LOH,1tJOHN J.HAMM,2 CLINTONKAWANISHI,3ANDENG-SHANGHUANG1*

Cancer Research Center, UniversityofNorth Carolina, Chapel Hill,North Carolina27514;1 Southern Grain

InsectsLaboratory, U.S. Department ofAgriculture, Tifton, Georgia 31793;2and U.S. Environmental

Protection Agency,Research Triangle Park, Durham,North Carolina277093

Received17May1982/Accepted 5 August 1982

Restriction endonucleaseanalysiswasused to differentiate betweenfourstrains

ofSpodopterafrugiperda nuclearpolyhedrosisvirus from differentgeographical

areas. Inaddition, partial denaturationwas

performed,

andapartialdenaturation

map was constructed for theOhio strain of this virus.

With the increasing interest in the use of

insectviruses as agents for the biological control

of insect pests, there isanurgent needto

identi-fy and characterize insect viruses and various

virus isolates. In this report, the restriction

endonucleasepatterns ofDNAfrom a strain of

Spodoptera frugiperda nuclearpolyhedrosis

vi-rus (SfNPV) for BamHI, EcoRI, and HindlIl

were determined and used to differentiate

be-tween SfNPV isolatesfrom Georgia (GA),

Mis-sissippi (MS), North Carolina (NC), and Ohio

(OH). Inaddition, apartialdenaturation map of

theOH strain of SfNPV was constructed.

Thestrains of SfNPV were originally isolated

from diseased fall armyworm larvae at Tifton, Ga.; Starkville, Miss.; Plymouth, N.C.; and

Cleveland,Ohio. Thevirionswerepurifiedfrom

thelysate of virus-infected fall armyworm larvae

by differential centrifugationand sucrose

gradi-ents aspreviously described (8). Theextraction

of DNAfromthevirions,itsdigestionby

restric-tionendonucleases, and the in vitro labeling of

DNA restriction fragments and their

visualiza-tionafter agarose gel electrophoresis were

per-formedessentially as described in ourprevious

report (8).

Thepartialdenaturation map was constructed

as follows. Purified viral DNA was partially

denatured by a modification of the method of

Inman and Schnos (3) as described by

Wads-worth et al. (11) andKilpatrick and Huang (5).

Specifically,a

10-tI

sample of DNA (6 to 10 ,ug/

ml)wasmixed with an equal volume of

denatur-ation buffer at room temperature andallowedto

reactfor 7 min. The denaturationbuffer

consist-ed of 20To (vol/vol) formaldehyde, 0.02 M

t Present address: Divisionof MedicalMicrobiology, Uni-versity of British Columbia, Vancouver, British Columbia V6T1W5,Canada.

Na2CO3, 5 mM EDTA, and enough NaOH to

bring the pH up to an appropriate value. Itwas

found empirically that a pH of 11.15 gave the

most distinct partial denaturation pattern, and

denaturation was already quite extensive at pH

11.25.Thereaction was stopped by the addition

of 80 p.1 ofice-coldspreadingsolution consisting

of 70pAlof1Mammonium acetate, 51ldof 0.2 M

acetic acid, and 5 p.lof cytochrome c (2mg/ml)

per 20 p.1 of the denatured DNA solution. The

pH of thefinal solution was about 5.2.

Theaqueous method (6) ofspreading partially

denatured DNA molecules (5, 11) was used to prepare the specimen grids. Immediately after thetermination ofpartial denaturation, 1

p,l

each of denatured and completely alkaline-denatured

4)X174RFmolecules were added to the reaction

mixture as internal length standards. A 50-pul

amount of this solution was spread over the

surfaceofan0.3 Mammoniumacetate solution

adjustedto pH 5.2. The DNA-cytochrome cfilm

was immediately transferred to

parlodion-coat-ed, 200-mesh copper grids by surface contact,

stained withuranyl acetate, dehydrated in 90%o ethanol,rotary shadowed with

platinum-palladi-um (80:20) alloy, and stabilized with a carbon

coatingtominimize distortions from the electron beam.

The sample grids were examined in a Hitachi

H-500 electron microscope at 50 kV. The

elec-tronmicrographs of DNA molecules were taken

at magnifications ranging from 3,000 to 9,000.

Themicrographs were enlarged by an overhead

projector, and only intact, circular, relatively untangled DNA molecules were used for length

measurements. Aprogrammed Hewlett-Packard

9825Acalculator and digitizer was used to trace

theprojected DNA molecules, and lengths were

recorded in microns.

4OX174

RF DNA, with a

747

on November 10, 2019 by guest

http://jvi.asm.org/

AS

a

F H

G

50 100 150

12

0 50 100 150

D2

D3

200

C

JI J2

N

KL N2

0 R

S

0 50 100 I50

MIGRATION DISTANCE

200

( mm)

FIG. 1. Microdensitometer scansofautoradiogaphs ofelectrophoretically separated, end-labeled SfNPV OHDNA cleavedby (a)BamHI, (b)HindIII,or(c)EcoRI.FragmentS in the EcoRI digestistheonlyfraginent present in submolar(0.5 mol)amounts. Allcleavage pattemswerescannedata1:1 scan-to-recordratio.

known molecularweight of 3.48 x 106(10),was

used as a standard. For partially denatured molecules, the lengthsof thesingle-stranded and double-stranded regions were measured

sepa-rately. The lengths of the single-stranded regions

werethen corrected forshrinkage byafactor of

1.418, avalue obtained empirically by

compar-ing the molecular lengthsofalkaline-denatured

andintact 4X174 RF DNAmolecules cospread

withthepartiallydenatured SfNPV DNA

mole-cules.

The OH strain of SfNPV was chosen for

detailed analysis. The buoyant density ofthe

viral DNAwasfoundtobe 1.6992±0.0003g/ml

by equilibrium CsCl gradientcentrifugationina

Spinco model E analyticalultracentrifuge, with

MicrococcuslysodeikticusDNAusedas a densi-tymarker(p= 1.731g/ml). Thus, theviral DNA

0 ₂₀₀

b

N

A

on November 10, 2019 by guest

http://jvi.asm.org/

Hind III

\ ( )I \M1 (i;

_

-...'.

...oiy*#4b

BamHI

\( ()S \ri1;G

FIG. 2. Cleavagepatternsof DN)

OH, MS, and GAstrains ofSfNPV

endonucleases BamHI, HindIII, and

strictionfragmentswereend-labeledv

andelectrophoretically separated in,

gel.

should have an average guanine

(G+C) content of 40% as calc

equation derived by Schildkraut molecularweight of the viralgenc tobe 82.5 (t5.2) x106byelectro

By these parameters, it was viri

guishable from the genomeof th

SfNPV, from which the SfNPVsti

ed in most other laboratories States andEurope wereoriginalln molecular weight and density da

DNA obtained in our laboratory ablywellwith thevaluesreported 2, 4, 7).

The SfNPV OH genome wasc

15, and 25 fragments by therestr

cleasesBamHI, HindIII,and EcoI ly (Fig. 1). The molecular wei

were reported in a previous paper (8).

Viral DNA from the GA, MS, OH, and NC

QL Ms'i ' strains of SfNPV were cleaved with BamHI,

HindIII,orEcoRI, end-labeled, and

electropho-retically separated on 0.7% agarose gels. The

resulting autoradiographs are shown in Fig. 2.

The migration patterns of the HindIII digests

wereidentical for theMS,NC,and OHstrains.

- Theextrafragmentpresentin theGA strainmay

_ be duetoheterogeneity within the virus

prepara-- g~3s tion. The EcoRI digests of the GA and MS

-^ R strains had migration patterns that were easily

_-~ distinguishable from those ofthe NC and OH

strains.Heterogeneitymay accountfor the

pres-ence of some of the submolar fragments

ob-served. Loss ofEcoRI sites, possibly between

someof the linked comigrating fragments such

_ a-_ as EcoRI fragments C and D, mayalso explain

_ the appearance ofextra high-molecular-weight

restriction fragments (e.g., the EcoRI fragment

-28> above EcoRI-A in the MS digest [Fig. 1C and

X

2]).^

-^ - Virions usedfor DNA

purification

were

puri--^= fied from the lysate of virus-infected larvae

clonedin vivo. Invitroplaquepurificationof the

various virus strains was not done because of

the lackofasoundpermissivecellsystemwhich

cangenerateinfectious virus in SfNPV-infected

cellcultures.Therefore,confirmation of the loss

ofaspecific restrictionenzymesite mustawait

DNAsequencing dataand the construction ofa

complete restrictionmapoftheviralgenomefor

ks from the NC,

EcoRI.

On the other hand, the migration pat-withrestriction terns of the BamHI digests were quite distinct

EcoRI. The re- foreach of the four strains of SfNPV. The fact

vithka-32P]dATP that SfNPV OH had the largest number of an0.7%agarose

BamHI

sites was one reason this strain was

chosen for detailedanalysisandrestriction

map-ping inourlaboratory. From theknown

restric-tion map for BamHI (8) and the sizes of the

BamHI restriction fragments, we can deduce

plus cytosine that the NC strain might have lost the BamHI

ulated by the sitebetween BamHIfragments A and G; the MS

etal. (9). The strainhaslost BamHI fragments A and D, and

:me wasfound the GA strain has lost the BamHI site between

in

microscopy. BamHIfragmentsAand D(Fig. 1and2).Again,

tually indistin- afinal conclusion about the loss ofBamHI sites

e GA strain of in thesecases canonlybe made with thesupport

rainspropagat- of DNA sequencing data.

in the United Preliminary experiments showed that the

yderived. The nick-translated BamHI-G and D/E fragments

ita for SfNPV eluted from gels ofaBamHI digest of SfNPV

agree reason- OH DNA did hybridize to the largestBamHI

[previously

(1, restrictionfragment of the other strains,as

pre-dicted.

:leaved into 8, Toprovideameansfororienting the circular

iction endonu- viral DNA molecule, we constructed a partial

RI,respective- denaturation map for the SfNPV OH genome.

ights of these Partial denaturation ofSfNPV DNAwasinitially

0--w=*.-.o4moo

to" 4bgqo

.. .. ...406..*

on November 10, 2019 by guest

http://jvi.asm.org/

FIG. 3. Electronmicrographofapartially denatured SfNPVOH DNA molecule. Partial denaturation was

accomplishedin alkalineformaldehydeatpH 11.15 for 7 minat25°C,andthe DNA wasspread byanaqueous

method asdescribedinthe text. The small bubbles represent the denaturedsingle-stranded regions.Bar, 1 ,um.

performed at pH values ranging from 11.0 to

11.6, the reaction time being fixed at 7 minat

room temperature (25°C). Small denatured

re-gions were detectable as tiny "bubbles" along

circular molecules at pH 11.15 (Fig. 3), but

denaturation became extensive at pH 11.25. Whenthe pHwasraised above11.25,extensive

single-stranded regions were seen throughout

thecircularmolecule,anddistinctpatternswere

no longerdiscernible. Thus, itwas notfeasible

toconstruct a precise partial denaturation map

of the SfNPV genome because of the lack ofa

restrictionenzymewhich cleaves this DNA

mol-ecule at only one site. Therefore, the partial

denaturationmapof the SfNPVgenome

present-edhere wasconstructed fromdata obtained by

denaturation atpH 11.15 and supplementedby

datafrom denaturationatpH 11.25byarbitrarily

setting the majoradenineplus thymine (A+T)-richregion of the molecule asthe origin ofthe

partial denaturation map during the alignment

proceduredescribed below.

We facilitated the data analysis by

photo-graphing only relaxed circular DNA molecules

forlength measurements. Because of the slight

length variations between different DNA

spreads, it was decided that the best way to

compare the data from different experiments

was to express all single- and double-stranded

lengths as a percentage of total circular length

insteadofasabsolute length units. Itwasfound

that with few exceptions, after correction for

single-stranded shrinkage, the total circular

lengths of partially denatured molecules were

comparabletothose ofundenatured DNA

mole-cules spread under similar conditions as

con-trols. This justified the use of the shrinkage

factor of 1.418 described above.

During the alignment of the partially

dena-turedmolecules, representations of these DNA

moleculeswereplottedonstripsofgraphpaper

on ascale of10% of the total length perinch.

Themoleculeswerethenarranged formaximum

overlap between the few A+T-rich and

on November 10, 2019 by guest

http://jvi.asm.org/

1n

012 w

<r 8

z

0

z I 10

8

cn 0 6

4<4

z

0 2 d

z

O 10 20 30 40 50 60 70 80 90 IC

% TOTAL LENGTH

DO

b

0 10 20 30 40 50 60 70 80 90 100

% TOTAL LENGTH

FIG. 4. Histograms showing the positions and

fre-quencies of denatured sites for SfNPV OH DNA after

partial denaturation at(a) pH 11.15or(b) pH 11.25.

The Y axisrepresents the number ofdenatured sites

per 2% of the total length (a) or the number of

denatured sitesper1%of the totallength (b).

rich regions. The frequencyofoccurrenceof the

denaturation sites along the DNA of 14

mole-cules examined wasthen calculatedtogive the

tentativepartialdenaturation map (Fig. 4).

Beginningfrom theoriginof themap(Fig. 4a),

therewas amajorrelative A+T-richzonewhich

extended for about 15% of the total length.

There followedaregion of lower A+Tcontent

that stretched for about 30% of themolecule;at pH 11.25 (Fig. 4b) most of this region was

denatured, but at pH 11.15 there were a few

small, relatively G+C-rich sites scattered

aroundthisregion.The first G+C-richzone was

found to be located immediately next to the central portion of the map and could only be

recognizedatpH11.25. Another G+C-richzone

spannedtheterminal15% of themap.Thesetwo

regionsweretheonlyonesthatremained

unde-natured atpH 11.25. Theregion between them

wasmarkedbytworelatively A+T-rich sites.

Insummary, wecharacterized four geographi-cally different strains of SfNPV by restriction

denaturationmapof the SfNPV OH genome was constructed. There was no indication of the

presence of long stretches ofhigh G+C or high

A+T regions or of highly repetitive genome

sequences, as was the case with certain

herpes-viruses. However, the asymmetrical pattern of

the twoG+C-rich regions shown in the

denatur-ation map at pH 11.25 might provide a means for orienting the circular viral DNA molecule.

This work wassupported byU.S.Environmental Protection Agency grant 806210.

LITERATURECITED

1. Bud, H. M., and D. C.Kelly.1977. The DNAcontained by nuclear polyhedrosisviruses isolated from four Spodop-tera spp. (Lepidoptera, Noctuidae): genome size and configuration assessed by electronmicroscopy. J.Gen. Virol.37:135-143.

2. Harrap, K. A., C. C.Payne,and J.S. Robertson.1977. Thepropertiesof three bacuoloviruses from closely relat-ed hosts. Virology 79:14-31.

3. Inman, R. B., and M. Schnos. 1970. Partialdenaturation of thymine- and5-bromouracil-containing A DNA inalkali.J. Mol.Biol.49:93-98.

4. KelBy,D.C. 1977. The DNA containedby nuclear

polyhe-drosis viruses isolated from fourSpodoptera Sp. (Lepi-doptera,Noctuidae):genomesize and homology assessed byDNAreassociationkinetics. Virology 76:468-471. 5. Klpatrick, B. A., and E.-S.Huang. 1977. Human

cyto-megalovirusgenome:partial denaturationmap and organi-zationofgenome sequences.J. Virol.24:261-276. 6. Kenschmndt, A.K.1968.Monolayer techniques in

elec-tronmicroscopy of nucleic acid molecules.Methods En-zymol.22:361-377.

7. Kndson,D.L., and T. W.Tinsley.1978.Replication ofa

nuclear polyhedrosis virus in acontinuous cell lineof Spodopterafrugiperda: partial characterization ofviral DNA,comparative DNA-DNA hybridization andpattern of DNAsynthesis. Virology87:42-57.

8. Loh,L.C., J. J. Hamm, and E.-S. Huang.1981. Spodop-terafrugiperda nuclear polyhedrosisvirus genome: physi-cal maps for restriction endonucleases BamHI and HindIII.J.Virol. 38:922-931.

9. Schildkraut,C.L., J. Marmur,andP.Doty. 1962.

Deter-mination of the basecomposition of deoxyribonucleic acid from itsbuoyantdensity in CsCl.J.Mol. Biol. 4:430-443.

10. Sharp, P. A., M. T. Hus, E. Obtsubo, and N. Davidson.

1972. Electron microscope heteroduplex studies of se-quencerelationsamongplasmids ofE. coli. I.Structure of

F-primefactors.J. Mol.Biol.71:471-497.

11. Wadsworth, S., R. J. Jacob, and B. Rolzman. 1975. Anatomyofherpessimplex virusDNA. II.Size, composi-tion,and arrangementof inverted terminalrepetitions.J. Virol.15:1487-1497.

on November 10, 2019 by guest

http://jvi.asm.org/