Physical maps for Herpes simplex virus type 1 DNA for restriction endonucleases Hind III, Hpa-1, and X. bad.

(1)

Vol.20,No. 1 Printed in U.S.A.

Physical Maps for

Herpes

Simplex

Virus

Type

1 DNA for

Restriction

Endonucleases Hind III, Hpa-1, and

X. bad

NEIL M. WILKIE

ColdSpring Harbor Laboratory, Cold Spring Harbor, New York 11724, andInstituteofVirology,

Glasgow,Gll 5JR Scotland*

Received forpublication10May 1976

Ithas beenproposed that thegenomeofherpes simplexvirustype 1 (HSV-1)

consists oftwo internal unique sequences, S and L, bounded by two sets of

redundant sequences (P. Sheldrick and N. Berthelot, 1974). In this

arrange-ment, terminal sequences (TRS and TRL) arerepeated in an internal inverted

form

(IRS

and IRL) and delimitS and L. Furthermore, abody of evidencehas

accumulated that suggests that S and L themselvesareinverted, givingriseto

four related forms of the HSVgenome. Inthisstudytheorderingofrestriction

endonuclease fragments of HSV-1 DNA for physical maps has been studied usingmolecularhybridizationtechniquesandthecleavageofisolatedrestriction

endonuclease fragmentswithfurtherrestriction endonucleases. Physicalmaps

for the fragments produced by Hind III, Hpa-1, and X. bad have been

con-structedfor the four related forms of the HSV-1genome.

TRs

and

IRS

werefound

tobe between 3.5x 106and 4.5 x 106daltons,

TR,

andIRLabout6x 106daltons, S

about 8 x 106 to 9 x 106 daltons, andL about 6.8 x 106daltons.

The genome ofherpes simplex virus type 1

(HSV-1) consists ofalinearduplex DNA

mole-cule of about100 x 10"daltons.A true terminal

redundancy of about0.5 x 106 daltons (2, 3)is

contained insomewhat largerblocks of

termi-nal sequences which Sheldrick and Berthelot

(10)found to be repeatedin

internal,

inverted forms. Two unique sequences

(S

= 107

daltons;

L =75 x 106daltons) were foundtobebounded

by these inverted repetitions. It was

proposed

that Swasboundedatthe terminus

by

TRs

and

internally by thesamesequence

inverted,

IRS.

Similarly, L was bounded by

TR,

and IRL. Sheldrick and Berthelot (10) also

proposed

that

the unique sequences S and L might become relatively inverted by recombination events

(seeFig. 10).

Subsequent

work

by

variouslaboratories

pro-videdclearevidence that thismodel, including

the relativeinversionsinS andL,was

substan-tially correct. Restriction endonuclease cleav-ageanalysis showed that submolar bands

pres-ent in

digests

ofboth HSV-1 and HSV-2 DNA

could be accounted for by the Sheldrick and Berthelot model

(1,

4, 5, 8, 11, 14). In digests

withendonucleaseswhich cleaveonlyinS and

L,two 0.5 Mterminalfragmentsaregenerated

from eachend. Four 0.25 M

fragments,

which

span the internal inverted redundant

se-quences, also arise. If it isassumedthat

TRL/IRL

and

TRS/IRS

are not identical, then enzymes that cleaveinSandL,butinonlythe

TRL/IRL

paironthe

TRs/IRs

pair, generate amolarend

from that terminus in which the redundant

sequence iscleaved. Two 0.5 Mfragmentsfrom

the otherterminusandtwointernal 0.5 M

frag-ments are also present (see Fig. 10). Direct

analysis of the terminalfragments found in

re-striction endonuclease digests of HSV-1 DNA, combinedwithsequence homology studies,

con-firmed thesehypotheses (15)and provided

fur-ther evidence that

TRL/IRL

and

TRs/IRs

were

notidentical. Finally, the relative inversionsof

S and L in thefourgenome arrangements for

HSV-1were visualized directly inthe electron

microscope via both incomplete (5) and

com-plete partial denaturation maps (Delius and

Clements, in press). In the latter study,

TRL/

IRLand

TRs/IRs

haddifferentpartial

denatura-tion profiles.

Thisreport givesthe orderofHind III,

Hpa-1, and X. bad fragments in the four genome

arrangements for HSV-1DNA.

MATERIALS AND METHODS

Enzymes. Restriction endonucleases Hind III and Hpa-1 were prepared as described previously (9, 14). X. bad, anenzyme fromXanthamonus badrii, was

the kind gift of S. Zain. DNA polymerase I was purchased from the Boehringer Mannheim Corp., andDNaseIwas fromWorthington.

Nick translation. HSV-1 DNA was labeled with

a-32P-labeled deoxyribonucleotide triphosphates

(specific activity approximately 100 Ci/mmol; New

England Nuclear Corp.) essentially as described by 222

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

PHYSICAL MAPS FOR HSV-1 DNA 223

Maniatis etal. (7). Reaction mixtures contained, in avolume of 100 ul: 1

Ag

ofpurified HSV-1 DNA; 50

mMTris-hydrochloride, pH 7.8; 5 ,MMgCl2; 1mM

dithiothreitol; 50 ,ug of bovine serum albumin per

ml; 180 pmol of[a-32P]dTTP, [a-32P]dGTP,

[a-32P]-dCTP, and [a-32P]dATP; 2 units of DNA

polymer-ase I; 10-10 g of DNase I per ml. Incubation was at 16°C for 1 h. The DNA was then extracted with phenol and purified by gel filtration on Sephadex G-50.

The growth ofvi*usand DNA,end-labeling

stud-ies onHSV-1 DNA, restrictionendonuclease diges-tion, analysis and purification of DNAfragments,

theSouthern DNA-transfer technique (12), and hy-bridization methods were carried out as described in the previous paper (15).

RESULTS AND DISCUSSION The cleavage patterns of HSV-1 strain 17

DNA for Hind III, Hpa-1, X. bad, and an X. bad-Hpa-1 double digest are shown in Fig. 1. The molarityofeach submolarband isshown

for thesingle

digests.

The molecular

weights

of

the Hind III and Hpa-l fragments have been published previously (15), and theX.baddigest

is discussed below. The over-molar group of

fragments previously

designated

Hpa-1 o (15)

areshown belowtobe derived fromtwo separa-ble maplocations, and this group hastherefore

been subdividedintoHpa-1o and Hpa-1o'.

Highly labeled HSV-1 DNA was needed for the studies discussed below, and 32P-labeled DNA ofhigh specific activity was obtained by nick translation as described under Materials and Methods. At the end of the reaction, DNA

withaspecific

radioactivity

of3 x 107cpm per

jig

appeared to be intact (in the sense of double-strandmolecular

weight)

and could bedigested with restriction endonucleases to give normal profiles. Figure lb showsan

autoradiograph

of

nick-translated

HSV-1DNA, cleaved with

Hpa-1 and separated

by

electrophoresis

on an

aga-roseslabgel. The profile obtained appears iden-tical tothatobtained with the unlabeled DNA

showninFig. la. The

autoradiogram

wasused as a template to enable the DNAbands tobe

excised

directly

for

subsequent purification.

Figure 2 shows that when the purified frag-ments were analyzed by gel

electrophoresis

their

mobility

hadnotaltered.

Analysis of Hind

III/Hpa-1

double digests. To aid the construction of linkage groups for differentrestrictionendonucleasefragments, a

comprehensivedescriptionof the Hind

III/Hpa-1doubledigest was obtained.

32P-labeled

Hpa-1 fragments were isolated as described above and

further cleaved with Hind III. The reciprocal

experiment was also carried out. The analysis

of the Hind IIIcleavage productsof someHpa-1

fragments is shown in Fig. 2. The molecular

weights

oftheproductswereestimatedby

ref-erence to the standard cleavage products of HSV-1 DNArun ontheoutside slots of the slab

gel.

The results of several experiments are

sum-marizedinFig.3.Each fragmentpresentinthe HindIII/Hpa-1 doubledigestwasshownto de-rive from the indicated Hind III and

Hpa-1

fragments by

further cleavage. Hind III or

Hpa-1 fragmentspresent aslimitfragments in the double digest (fragments not further cleaved by the alternative enzyme) are

indi-cated. The terminal fragments in the double

digestwerelocated by end-labelingstudies

sim-ilar to those described in the previous paper

(15), and their positions in single and double

digestsare indicated.

Sequence homology between fragments.

Theassignmentofdifferentrestriction

endonu-cleasefragmentstolinkage groups canalso be deduced from sequence homology studies. 32p_ labeledHpa-1fragmentswerehybridizedto

un-labeledHind IIIfragments (and vice versa) by usingtheDNAtransferorblottechnique

devel-oped by Southern (12) and described in our

previous publication (15). Some

cross-hybridi-zation results are presented in Fig. 4. It is

evidentfromthisfigure thatfragments which

appearedtobefairlypure on the basis oftheir relative mobilities upon gel electrophoresis (Fig. 2)are, infact, contaminated bysequences

from many different regions of the genome.

This contamination can also be seen in the

further cleavage analysis of some isolated

bands shown in Fig. 2. This problem of

se-quence contamination was alsoobservedwhen

DNA labeled in vivo was used, and it seems

morelikelytobeanartifactdue-tothe

purifica-tion and separation methods used rather than

to the method oflabeling. The contaminating

sequences in the 32P-labeledHpa-1 fragments

causehybridization to most Hind IIIfragments

at levels above background. Nevertheless, the

hybridization ofthe main Hpa-1 fragment in

any reaction can readily be distinguished as

moreintensebands of

radioactivity

onthe

auto-radiographs. The data presentedinFig.5and6

summarize the data obtained in several such

experiments in which 32P-labeled Hind III

fragments were annealed to unlabeled Hpa-1

fragmentsand vice versa.

Construction ofa partial map. In the

pre-vious publication (15) the four terminal

frag-ments in Hind IIIdigestsof HSV-1 DNA were

identified asHind IIId, g, andmandeitherh or i (notseparated by thegel electrophoresis).

Hind III iisarbitrarilydesignatedas the

termi-nal 0.5 M band. In Hpa-1 digests, the three

terminalfragments wereidentifiedas a molar

VOL. 20, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

224 WILKIE

a

.}~~

a

F

~~~~~~is

--- k

- p

a

b

-_ c

0 -

ci

e

-

f

g o

--

----,

Ik

-~

a1

n

S

aa

I)

cl

f

ic0

0

Hindc

III

Xbaci

Hpa

0.5M

cl

g mIII

_0Q5M

accig

(Hpa I)

0.5M

abcdq

0. 25M

bceff

FIG. 1. (a)Restriction endonucleaseprofiles ofHSV-1 DNA. About1 pgofHSV-1DNAwascleaved with

the restriction endonucleases shown andseparatedbyelectrophoresison0.3%agarose.Eachfragmentwas

assignedaletterofthealphabet. Hpa-1 o',anover-molargroupoffragments,has beensubdivided into two

classes,o ando', for reasons explained in the text. Minor bands ofmolarities 0.25 or0.5 areindicated.

Terminalfragmentsareindicatedbythesolid circle(@). (b)Theprofile ofnick-translated HSV-1 DNA.32p_

labelednick-translatedDNA wascleaved withHpa-1andseparated byelectrophoresison aslabgel of0.7%

agarose.

J. VIROL.

.w ..,.',U..,7

..-r

[image:3.508.65.467.62.585.2]

,;^. iR..

.|..tr.2SX .||.

.,w: ., ;jE

iilllit

1a

.!L'

|,iF

:!Ei *.,]-';''

'.!,

r

-1111ruS

i >;

:ZE

F

B

t.

VblE

i44

q

r S

X. b

ad

on November 10, 2019 by guest

http://jvi.asm.org/

(4)

PHYSICAL MAPS FOR HSV-1 DNA

225

VOL. 20, 1976

b

R

.

b!__.

_...

,

be.,

s4.

...

___;s.

__.5S|*

s

_^S

_we M .. '0, '.i:.

... ..

+:--;.,<. j: -w','s, 27r.; ,: - :*:

:- :@ 'r+.

<,.; +'s ',.

a. :u;^s9;s,2.- ,; { 0" w:; s

*;-+ ,,@ _ , }t ...

_L

w_|-

f-

on November 10, 2019 by guest

http://jvi.asm.org/

(5)

226 WILKIE

a b cde

_4

'

r I g i

_m

Om

k

m7

FI 4h n

I. ft.f*

FIG. 2. Furthercleavage products of Hpa-1 fragments with HindIII. 32P-labeledHpa-1 fragments were

isolated as described. 2 x 103 to 2 x 104 cpm of each fragment (less than 1p) was mixed with 0.5 igof

unlabeled adenovirus-2 DNA and cleaved with Hind III. The cleaved fragments wereseparated by electropho-resis onslab gels of0.7% agarose. Undigested fragments were run inparallelslots.Ineachcasetheunlabeled adenovirus-2 DNA wasdigestedto limitproductsasdetermined byDNA-ethidium bromidefluorography.

Labeled DNA was determined by autoradiography. Standard Hpa-1 cleavageproducts of total HSV-1 DNA

areshownonthe outside slots. Hpa-1 fragments from thetermini are indicatedbythesolid circle(0).

fragment Hpa-1m andtwo 0.5 Mfragmentsin

Hpa-1 cde and hg. By a similar arbitrary deci-sion

Hpa-1

dandg aredesignatedas the

termi-nal fragments. Evidencewas alsopresentedin

the previous publication (15) to suggest that

Hind III d and i andHpa-1 m were from one

terminusof thegenome and Hind III gandm

andHpa-1candg from the other. Itcanbeseen

from Fig. 3that the terminal fragmentsin the

Hind

III/Hpa-1

double digests allarisefrom the

DNAbands of the single digests which are, or contain, ends. The data of Fig. 3 are in

com-plete agreement withtheprevious assignment (15) of end fragments to different termini of the genome.

By combining this information and the data

summarized inFig. 3, 5, and 6, apartial map

fortheHind IIIandHpa-1fragments has been

constructed. Forexample, Fig. 5 and6

demon-stratethattheuniquefragments HindIIIj and

Hpa-1

j and n are sequence related. From the

resultspresented in Fig. 2 and 3, it can be seen

that there is a Hpa- 1 site in Hind

IJIj

and a Hind

III site inHpa- 1 n, but no Hind III site in Hpa- 1

j. Furthermore,

Hpa-1

nand Hind IIIj give rise to the same double-digest fragment "o." The

molecular weights of the further

cleavage

prod-ucts (notshown)suggestthatat oneend of this linkagegrouptheremustbe Hind III and

Hpa-1 sites very close together. The only other

se-quence homology for

Hpa-1

n inHind III

di-gests is inHindIIIab (Fig. 6). The

homologous

sequences must lie in the unique

fragment

HindIII a, sincebis a 0.25 Mfragment, andits

sequences must be present in other minor

bands because of the relativeinversions ofthe uniquesequencesSandL inHSV DNA.

Hpa-1

n must therefore span the Hind IIIa-HindIIIj

junction. Analogous reductive

analysis

of the

results shown in Fig. 3, 5, and 6 indicates

thatHpa-1 b, h, i, and q must be contained

withinHindIII a.Thefollowing large linkage

group is thusconstructed:

Hind III

a

1i1

A

A~

n

(b,h, i, q) Hpa-1

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.508.69.457.82.341.2]

(6)

Hpa I fragments

bX cdxe a

'xe

f ijX c

dxe;j

kx ma de c0dxe a;edxe

0 J

(>I)p

q r

-ab;c;de;I ab;c;cle;g

hi-ox

f;ool a;c dXe;px

I

FIG. 3. Summaryoffurther cleavage experiments.

On the basisof further cleavage experiments, bands

ofDNA inHpa-1IHind III doubledigestswereshown

experimentally to have arisen from the cleavage of

HindIIIfragments with Hpa-1 andof Hpa-1

frag-mentswithHindIII. Solid circles(a) indicate frag-ments arising from termini. Crosses (X) indicate

fragments of single digestswhicharenotcleavedby

the secondenzymeand which existaslimitfragments

in the double digest. For example, double-digest

band k arisesfromtheHpa-1 cleavage ofHind III hi. SinceHpa-1 k isnotfurthercleaved with HindIII, double-digestband k andHpa-1kmustbe identical. Hind III i is a terminal fragment. Hpa-1 k must therefore be contained within either Hind III h or

Hind III i.

Space does not permita full analysisof the

construction of all the linkage groups here.

However, the data presentedinFig. 4, 5, and6

contain all the information necessary to

con-struct the partial map shown in Fig. 7. The

linkagegroup containingHind IIIa andj and

that containing Hind III h could not be

as-signedmappositionsonthe basisof the

experi-mentswithHindIII and Hpa-1describedsofar.

The limits of the map terminate at positions

wherethe Hind IlI and Hpa-1 sites are so close

together that no more overlaps could be ob-tained.

The tendency for Hind III c and de toanneal

to Hpa-1 b is notaccounted for in this partial

map. Since further mapping data below puts

Hpa-1 b firmly within Hind III a, it is thought

that this aberrant annealing data is due to

Hind IIIa sequencescontaminating Hind III c and de.

The cleavage pattern for Hpa-1 arises from

Hpa-1 sites in SandLandin either

TR,/IRs

or

in

TRL/IRL

(1, 15).Therefore, one terminal

frag-ment mustbe present in molar amounts.Hpa-1

m hasbeen identifiedas a terminal fragment (15), but was originally quantified as 0.77 M

andwasthoughttobea 0.5Mband (Hpa-11 in

reference1). Fromthe resultspresentedhere it is clear that Hpa-1 m arises from a cleavage

within TRL and must be the molar end. The

reason for the apparently low amount of this

terminalfragment indigests is notcompletely understood. We havenoted thatinmany cases

terminal fragments from bothtermini are more

spread after gel electrophoresis than other

bands of DNA (N. M. Wilkie, J. B. Clements, and R. Cortini,

unpublished

observations) as

though there was some heterogeneity intheir molecular weights. Such heterogeneity and spreading tends to reduce the peak height of the bands and in some cases may lead to a

reduction in the observed amount ofDNA in

thebands. Thereasons fortheheterogeneityin

terminal fragmentsremains a matter for

specu-lation.

The hybridization and further cleavage

ex-periments also confirm the

general

model for HSV DNAfirstproposed bySheldrick and Ber-thelot (10).Bands ofDNA containingsubmolar fragments from one enzyme digest cross-hy-bridizetotheexpected submolar fragments

pro-duced by the second enzyme. For example, Hpa-1a (0.5M) annealsto all ofthe submolar fragmentsintheHindIIIdigest (Fig.4,6). The terminal Hpa-1 fragment m hybridizes to the

0.25 MHindIIIfragments andtothetwoHind

IIIterminal fragmentsd andi, but itdoesnot

anneal significantly to the Hind III terminal fragmentsg and m. This confirms the

assign-ment ofHpa-1 m to the same terminus from

which Hind III d and i arise. It might have

been expectedthat the actual terminal

redun-dancy of about 0.5 x 106 daltons previously proposed byother groups (2, 3, 10)mighthave

caused some cross-hybridization between

ter-minal fragments fromdifferent endsof the

ge-nome. However, thenearby inverted repeat of

the terminal sequences found in HSV-1 DNA

by Hyman et al. (6) and subsequently

con-shownto bederived from

Hind III fragments

a ab

b-- ab;de

c- . c

de __ ab;g

f- dAe

g- k

h ab

___. j;eX

k hi

I - de;hi

m I n.~ nx OC₀ 0

Im

a6 I)

VOL. 20, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.508.50.242.69.420.2]

(7)

228 WILKIE

32p

_Hpa

_fragments

a

cof

I

m n

o0

d

o

ab

Cd-ire

f

-9k.- *

i-k

-=W

a)I(

U.,.

-c

-5

U

0

Ca

b

i

k

ab-

+

c-de

-f _

h

*.

h

-k

-I

_

m-n-

S

n-

0-FIG. 4. Hybridizationof isolated 32P-labeledHpa-1 fragmentstounlabeled HindIIIfragments of HSV-1

DNA.Thetechnique was that described previously (12,15) and annealing of labeledDNAto unlabeled DNA onnitrocellulose membrane strips was determined byautoradiography. The positionsofalloftheunlabeled

DNAbands on nitrocellulose membrane strips were located by annealing with fragmented, unfractionated 32P-labeled HSV-1 DNA. Terminal fragments are indicated by the solid circle.

firmed by Wadsworth et al. (13) may cause

internal reassociation (in the unlabeled DNA)

which might interfere with the reassociation

reaction.

AnalysisofX. baddigests. X. bad fragments

of HSV-1 DNAcanbeseparatedinto sixbands

by electrophoresis on 0.3% agarose (Fig. 1). Analysisof themassdistribution of the DNAin J.

VIROL.-,^- Ai 0

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.508.75.465.73.560.2]

(8)

32P-LABELED HIND III FRAGMENTS

ab c de f g

hi'

j k

+ + + + + + ++ ++ ++ + + + + + +

1 me n 0

[image:8.508.67.430.71.201.2]

+ ++ ++ ++ ++ ++ ++ (+) + ++ + + + + ++ ++

FIG. 5. Thehybridizationof 32P-labeledHind IIIfragmentsto unlabeledHpa-1 fragments.Solidcircles

(a)indicate terminalfragments.Theamountoflabel in each band(abovethat level causedby contaminating

sequences) wasassessedbyeyeforeachstripand allocated +, ++,or+ + +inorderofincreasing intensity.

E-l a b cde

Z ab + + + +

c + +

de ++ +

¢4

f + +

x g + +

,_ hi + + +

Zk

SI +

om. + +

O n + +

0v

32P-LABELED HPA-1FRAGMENTS

e f gh i j k 1 m

++ ++ + + ++

n oo' p

+ + + + + + + + + + ++ ++ + + ++ ++ ++ ++ + + + ++ + ++ ++

q r s

+ ab

+ c

+ de f

FIG. 6. Thehybridization of32P-labeledHpa-1 fragmentstounlabeledHind IIIfragments.

those bands by the techniques described in a

previous publication (1) give themass

percent-ageof thesix bandsa, b,c,de,f,gas 19, 11.2,

26, 27, 13, 3.2. The molecular weights of these

fragments were estimated from slab gels and

from the further analysis and mapping ofX.

badfragmentstobe 34x 106,22 x 106, 21 x 106,

18 x 106, 15 x 106, and 6.5 x 10-6, respectively.

From thesefigures, and assumingamolecular

weight of about 98 x 206for HSV-1 DNA, the

molarities for a, b, c, de, f, and g could be

calculated to be 0.55, 0.51, 1.2, 1.46, 0.85, and

0.48.Thissuggeststhata, b, d,ore,andgare

0.5 M fragments and thatc, d ore, andfare

molar fragments. As before, d wasdesignated

asthe 0.5Mfragment, andewasdesignatedas

the molarfragment.

Analysis of X. bad/Hind III, and X. bad!

Hpa-1 double digests (e.g., Fig. 1) clearly

indi-catethat thereare only fourcleavage sites for

X. bad inHSV-1 DNA. One site lies in HindIII

I andHpa-1 e, two in Hind IIIa (one each in

Hpa-1 b and i),and the remaining site in Hind

IIIi (and its inversion Hind III

f)

andinHpa-1

1. These X. bad sites are shown inthe partial

mapinFig. 7. Thetwo cleavage sites inHind

IIIawereconfirmedby the furtherdigestion of

isolated Hind III ab with X. bad. Three further

cleavage productswereidentified with

molecu-larweights of15 x 106, 8 x 106, and 2.8 x 106.

The15 x 106fragments had thesamemobility asX. badf.

The fourcleavagesites all lie in the L region

of the genome and would give rise to seven

fragments because of the inversion of the L

region: anenzyme which cleaves only in the L

regionshould producetwo 0.5Mterminal

frag-ments which contain the Ssequences and two

0.5 Mfragmentsfrom the L terminus. All other

fragments should be molar. This fitsverywell

with the cleavage profile and the estimates of

molecular weights and molarities obtained for

the X. bad fragments. X. bada and b arethe

0.5 M terminifromthe S region; X. bad d andg

arethe 0.5 Mtermini fromL; andX. badc, e,

and fare the remaining internal molar

frag-ments from L. This analysis of the X. bad

cleavage patternoffers further confirmation of

the partial mapshown inFig. 7.

Completion of themap. The position of the

X. bad sites onthe HSV-1 DNA mapprovides

thenecessaryoverlaps that allow completion of

themap. 32P-labeled Hind IIIfragments k ando

andHpa-l-labeled fragments b, j, and k were

cna

E- b : f

° gh

_t4

-,J k I m

_rn

O oo' p b cde f gh j +++ k I m n oo' p

+++ (+) + + hL k

m n VOL. 20, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:8.508.61.445.248.365.2]

(9)

230 WILKIE

49 k

m

InI

0)

c A d

AoA

0

pA

e f A

A 0 i

&I

OA

m

m ,|n > e o

c2 ^-

--A d

A60A

IA

g In>

1 | k ,

A a

Ao

e A A

g a a o~~~~~pf

n2 f d

g a a A0A f A e A o'p

AoA

m

ti

t4

AA

n

a

bhiq A krA

HindI1I1|

Hpa a

X. bad t

FIG. 7. Partial mapofHind III andHpa-1 fragments ofHSV-1 DNA.

annealed to unlabeledX. bad fragments using the DNAtransfer orblottechnique (12)as

be-fore(15). The resultsaresummarized inFig.8.

Hpa-1 bhybridized mainlyto X. badfand also

to d. This confirmed the previous suggestion

that X. bad f is contained within Hind IIIa.

Hpa-1j andk and HindIIIohybridized mainly

to X. bad c, whereas Hind III k hybridized

mainlytoX. bad de. SinceX. bad disa

termi-nalfragmentfromL,itssequencesmust also be

contained inonelarger0.5 Mfragment(a orb)

from the Sregion.ThehybridizationofHind III

k mainlytoX. bad de must thereforebe to the

unique fragment e. From the partial map

shown in Fig. 7 it canbe seen thatX. bad e

must lie adjacent to the Hind III b/l and c/l

junction. Since Hind III o and Hpa-1 k both

annealtoX. badc,Hind III h must lieonthe

Hind III o side of the gap in the partial map shown inFig. 7.Thus,X. badc must lieinthe

regionof the Hind IIIh/ojunction.Thisleads to

thefollowing unequivocalorderof X. bad

frag-mentsinthe four arrangements of thegenome shownin Fig. 7 and 10:

C,)

z

Ca

9: b

c

¢ de

mq

f

ZC g

04

u

32p FRAGMENTS Hind III

k b

++

Hpa-1

k

++ ++

+

FIG. 8. Thehybridization of 32P-labeledHindIII andHpa-1fragmentstounlabeled X. badfragments.

(1)aefcg (2)bc fed

(3)aefcg (4)bcfed

Sinceonlytwoarrangementsarepossiblefor L,there areonlytwopossibleX. bad maps.

SinceHpa-1 j andk bothannealed to X. bad

c,both must beadjacenttothe HindIIIoregion

inthepartialmapofFig.7.Theonlywaythese

linkagegroups canbearrangedintheLregion

k 1 d

Af A g

Ae

AAp

&o

m

AI Am

I

J.

VIROL.-

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.508.65.458.51.472.2] [image:9.508.277.458.418.533.2]

(10)

istherefore (in the order for the Hind III frag-ments): 1 kaj ho.

Hind IIIacontains twocleavage sites for X. bad, whichgiverise tothe threefurther

cleav-ageproductsat 15 x

106,

8 x

106,

and2.8 x 106 daltons. Hpa-1 bandieachcontain oneof these

X. bad sites. Theonlyinteriorarrangement of Hpa-1b, h, andi thatallows the further

cleav-age productsactually observed is:

Hind III

a

1

| h b

A A

Hpa-1 where t=X.bad; I =HindIII, and A=Hpa-1.

The results donotallow thelocation of

Hpa-1

q withinHind IIIa.Since

Hpa-1

q issmall (1.7

x 106 daltons) this doesnot affect the relative order of

Hpa-1

b, h,

and i within Hind IIIa.

(The hybridization of 32P-labeled Hpa-1 q to

unlabeledX. bad fragments should allow the unequivocal assignment of q.) The final

orien-tationofthemaps wasobtainedby

hybridizing

32P-labeled Hpa-1 fragments to unlabeled X.

bad/HindIII

double-digest

fragments using the DNA transfer or blot

technique

(14). The

re-sultsareshownin

Fig.

9.

Hpa-1

i

hybridizes

to two

fragments,

as

expected

from thesmall

sec-tion of the map shown

immediately

above. Hpa-1n hybridizes tothe smaller of thesetwo

fragments and also to a

higher-molecular-weight fragment.

Hpa-1

j

anneals

exclusively

to this larger

fragment,

which has a

mobility

thesame asthat of HindIIIj (whichis alimit

fragmentinthe X.bad/HindIIIdouble

digest).

This confirms the Hpa-1 maporder inthis re-gion as:

h b i nj (kr)

Lqi

Taking all these results together, the

com-plete cleavage fragment maps for Hind III,

Hpa-1,

and X. bad shown in Fig. 10 can be

constructed. Theresultsdo not allow the final

orderofHpa-1o

andp,

ofHpa -1 k and r, or the

final positionofHpa-1qwithin Hind III a to be

assigned. Thehybridizationresults with Hpa-1

s did not give a unique location on the map.

However, it ispossible thatHpa-1 s is in fact two fragments. In this case, the hybridization data shown inFig. 5 can beexplainedifone of

thefragments,

Hpa-1

s,liesinthe

o'p

region of themap, andthe other

fragment,

Hpa-1 s', lies

in the linkage group kr within Hind III h. *-.

E

Cu

I

x0

=

Cu

,. 4,v

a

u

TIj"

P

Hpa

I

fragments

FIG. 9. Hybridization of32P-labeled Hpa-1

frag-mentsof HSV-1 DNA to unlabeled X. bad-Hind III double-digest fragments. T means the hybridization of fragmented, unfractionated 32P-labeled HSV-1 DNA to cold fragments.

VOL. 20, 1976

-v

on November 10, 2019 by guest

http://jvi.asm.org/

[image:10.508.259.448.69.602.2]

(11)

232 WILKIE

TR5 S IIR IR L TRL

m n b I k a h I

[ G 1 Jk

A AAd A ^ daA

c d o ops e f h b i n j I9omskr

L q

msn e 0 h I a k I d

, A A A A A A ,

c d oI skr j n b h If e ops o m

b c f (- d

g4,n,

c , I k4 a j h o

A AA, A A A A A AA A

g a o ops e f h b i n j skr

a e f c 9g

g

n,

f o h l a k I d

4_

4,4 4,

4.

4,4,4,

a A-A AA A A A

i

A AA.

[image:11.508.68.464.54.341.2]

9%~ a kr eglJ n b h _ opsom

FIG. 10. Completephysical mapsforthe HindIII, Hpa-1, and X. badfragments of HSV-1 DNA. The

redundantsequences TRL, IRL, TRs, andIRsand the unique sequencesL andS, originallyproposed by

Sheldrick and Berthelot(10), areshown.

Accordingly, these map positions have been

tentativelyassignedtoHpa-1sand Hpa-1 s'.

The molecular weights of the fragments in

the maps shown in Fig. 10 give a genome

molecularweighttotalof 98 x 106.Sappearsto beabout 8-9x 106daltons, and L about 68 x 106

daltons. These figuresaresomewhatless than

those originally published by Sheldrick and

Berthelot (3), butareveryclosetothosefound

by Delius and Clements frompartial

denatura-tion data (in press). Themapping information

also allowslimitstobeplacedonthelengths of

the redundant sequences

TRs/IRs,

IRL, and

TRL.TheX. bad endg is about 6.5x 106daltons

andarises fromacleavagepointnearthe TRL/L

junction (Fig. 10).Thus, 6.5 x 106daltons isan

upperlimitfor TRLand IRL. Datanotpresented

inthisreportshows that there isanEcoRI site

withinTR5 togiveaterminal fragment of 3.5 x

106 daltons. Since the Hind IIIterminal

frag-mentm,arising fromacleavagesiteinS,is 4.5

x 106 daltons, these give the upperand lower

limits for

TRs

and

IRS.

In the map shown in

Fig. 10, TRL/IRLand

TRs/IRs

havebeen drawn

as 6 x 106daltonsand4 x 106daltons,

respec-tively. These estimates are verysimilartothe

figures of 6 and 4% for TRL/IRL and

TRs/IRs

derived byClementsandDelius(in press) from

their complete

paltial-denaturation

maps of

HSV-1 DNA. Since there issomeindication (15)

that thetrueterminalredundancy of 0.5 x 106

daltons (2, 3, 10)mayberepeatedin the

inter-nal inversions of

TRs

and TRL, this sequence

has beenindicated in themapsshown in Fig. 10

as a separate division within each redundant

region.

Therestrictionenzyme mapsof HSV-1 DNA

for Hind III,Hpa-1, andX. bad shown inFig.

10shouldform the basis for furtherrestriction

endonuclease maps. Additional mapping work

should also eliminatetheremainingminor

un-certaintiesintheHpa-1maps.

ACKNOWLEDGMENTS

Thisworkwas carriedout at theCold Spring Harbor Laboratory duringtemporary leave of absence from the Institute ofVirologyinGlasgow. IamgratefultoJ.Watson formaking availablethefacilities at the ColdSpring

Har-borLaboratory andtoJ.Sambrook formanyhelpful discus-sions. JanetArrandand RitaCortiniprovidedinvaluable assistance.

LITERATURE CITED

1. Clements, J. B., R. Cortini, and N. M. Wilkie. 1976.

AnalysisofherpesvirusDNAsubstructurebymeans

ofrestriction endonucleases. J. Gen. Virol. 30:243-256.

2. Grafstrom,R.H.,J. C.Alwine,W. L.Steinhart,and C. W. Hill.1974.Terminalrepetitionsinherpessimplex J. VIROL.

2

3

HindIII *

Hpa A

Xbad

4 on November 10, 2019 by guest

http://jvi.asm.org/

(12)

PHYSICAL MAPS FOR HSV-1 DNA

233

virustype 1DNA. ColdSpring Harbor Symp. Quant. Biol. 39:679-681.

3. Grafstrom,R.W.,J.C.Alwine,W. L.Steinhart,C. W. Hill,and R. W. Hyman. 1975. The terminal repeti-tion of herpessimplex virus DNA. Virology 67:144-157.

4. Hayward, G. S., N. Frenkel, and B. Roizman. 1975. Anatomy of herpes simplex virus DNA: strain differ-encesandheterogeneityinthelocation of restriction endonuclease cleavage sites. Proc. Natl. Acad. Sci. U.S.A. 72:1768-1772.

5. Hayward, G. S., R. J. Jacob, S. C.Wadsworth,and B. Roizman. 1975. Anatomy of herpes simplex virus DNA:evidence for four populations ofmoleculesthat differ in the relativeorientationsof theirlongand short components. Proc. Natl. Acad. Sci. U.S.A. 72:4243-4247.

6. Hyman, R. W., S. Burke, and L. Kudler. 1975. A nearby inverted repeat of the terminal sequence of herpessimplex virus DNA. Biochem. Biophys. Res. Commun. 68:609-615.

7. Maniatas, T., A. Jeffrey, and D.G.Kleid.1975. Nucleo-tide sequences of the rightward operator ofphage lambda. Proc. Natl. Acad. Sci. U.S.A. 72:1184-1188. 8. Roizman, B., M. Kozak, R. W. Honess, and G. Hay-ward.1974. Regulation ofherpesvirus macromolecu-lar synthesis:evidence for multilevel regulation of herpes simplex 1 RNAand protein synthesis. Cold Spring HarborSymp. Quant.Biol. 39:687-701.

9. Sharpe,P.A., B.Sugden,and J.Sambrook.1975. De-tectionoftwo restrictionendonuclease activities in Haemophilus parainfluenzae using analytical aga-rose-ethidium bromide electrophoresis. Biochemistry 12:3055.

10. Sheldrick, P.and N. Berthelot. 1974. Inverted repeti-tions in thechromosome of herpes simplex virus. Cold Spring Harbor Symp. Quant. Biol. 39:667-678. 11. Skare,J., W. P. Summers, and W. C. Summers. 1975.

Structure and function of herpesvirus genomes. I. Comparisonoffive HSV-1 and two HSV-2 strains by cleavage of their DNA with EcoRI restriction endonu-clease. J. Virol. 15:726-732.

12. Southern,E. M. 1975. Detection of specific sequences among DNAfragments separated by gel electropho-resis. J. Mol.Biol. 98:503-518.

13. Wadsworth,S., G. S. Hayward, and B. Roizman. 1976. Anatomy of herpes simplex DNA. V. Terminally re-petitive sequences. J.Virol. 17:503-512.

14. Wilkie, N. M., J. B. Clements, J. C. M. Macnab, and J. H.Subak-Sharpe. 1974. The structure and biological properties of herpes simplex virus DNA. Cold Spring Harbor Symp. Quant. Biol. 39:657-666.

15. Wilkie,N.M.,and R. Cortini. 1976. Sequence arrange-ment inherpes simplex type 1DNA: identification of terminal fragments in restriction endonuclease di-gestsand evidence for inversions inredundant and unique sequences. J. Virol. 20:211-221.

VOL. 20, 1976