Collaborative complementation study of temperature-sensitive mutants of herpes simplex virus types 1 and 2.

(1)

Vol. 27, No. 3 JOURNALOFVIROLOGY, Sept. 1978,p.490-504

0022-538X/78/0027-0490$02.00/0

Copyright i 1978 American Society forMicrobiology Printed in U.S.A.

Collaborative

Complementation

Study

of

Temperature-Sensitive

Mutants of

Herpes Simplex Virus

Types

1 and

2

PRISCILLA A. SCHAFFER,* V. CELESTE CARTER, ANDMORAG C. TIMBURY't Divisionof BasicSciences, Sidney Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts02115,andDepartment of Virology, Baylor College of Medicine, Houston, Texas 77030; and

Instituteof Virology, University of Glasgow, Glasgow G115JR,Scotland'

Received for publication 8 March1978

Twenty-three complementationgroupsof herpes simplex virustype 1(HSV-1)

and 20 of HSV-2wereidentifiedby qualitative and quantitativecomplementation

analysis from among43temperature-sensitive (ts) mutantsof HSV-1 and29ts

mutantsof HSV-2 which hadbeen isolated independently in 10laboratories.

The herpes simplex virus (HSV) genome is

large (100 x 106) and complex (5, 25, 31) and

probably encodes from 70 to80 genes. A

com-plete understanding of the structural and

func-tional organization ofthe genome will

necessi-tate theidentificationofall viral genes and their

ordering on the viral genetic map.

To date, most genetic studies of HSV have

utilized temperature-sensitive (ts) conditional

lethal mutants whichcanoccur,theoretically,in

all essential genes. Although the total number

ofessentialgenes encodedbyHSV DNA is not

known,complementation analysis of one series

of ts mutants has ledtothe identification of 15

cistrons of HSV type 1 (HSV-1) (22) and, in a

collaborative study, 18 cistrons ofHSV-2 (30).

Because it is very likely that more than this

numberofessentialcistrons is encoded bythe

HSV genome(3), a collaborative

complementa-tionstudywasundertaken toidentify additional essential cistrons from amongexistingseries of

HSV-1 and HSV-2tsmutants.

Theresultsof thisstudyestablish theidentity

of 23 cistrons among 43 ts mutants of HSV-1

and20cistronsamong29 ts mutantsofHSV-2

and demonstrate thatmostpreviously untested

t In collaboration with: S. M. Brown andJ. H. Subak-Sharpe,Institute ofVirology, University of Glasgow, Glasgow G115JR, Scotland;A.Buchan, Department of Virology, The MedicalSchool, Birmingham B15 2TJ, England; I. Hallibur-ton,Department of Microbiology, School of Medicine, Leeds LS29NL, England;R.Hughes and W. Munyon, Department ofMedical ViralOncology, Roswell Park Memorial Institute, Buffalo,NY14263; R. W. Koment and F.Rapp, Department ofMicrobiology,Milton S.Hershey Medical Center, Pennsyl-vania StateUniversity, Hershey, PA 17033;R. Manservigi, Instituto diMicrobiologia, Universita di Ferrara, 44100 Fer-rara,Italy; M. Takahashi,Research Institute for Microbial Diseases,OsakaUniversity, Yamada-Kami, Suita City, Osaka, Japan;V.Vonka,Institute of Sera andVaccines, Department ofExperimentalVirology, Praha 10, W. Pieck 108, Czechoslo-vakia; N. Zygraich and C. Huygelen, Biologics Department, RechercheetIndustrieTherapeutiques,Rixensart,Belgium.

mutants belong toalready established

comple-mentationgroups. It isclearthat the number of new complementationgroups identified in this

studyissmaller than onewould expect given the

number of independently derived mutants

tested. These findings demonstrate the

neces-sity, therefore,for(i) developingnewprocedures

for the isolation of mutants in as yet unidentified

HSV cistrons and(ii) examiningthefunctional

basis for the extensive failure to complement

observed with certain mutants.

MATERIALS AND METHODS

Celisand cellculture.Monolayercultures of

se-riallypropagatedhumanembryonic lung (HEL)

fibro-blastswere used for the preparation of virus stocks and for mixedinfectionsinthe quantitative

comple-mentation test. Vero cell monolayers were used for virusassayand for mixed infections in thequalitative complementationtest.Both HEL and Verocellswere grownat37°CinEaglemedium(Autopow,Flow

Lab-oratories,Rockville,Md.) supplementedwith 10% fetal bovine serum and containing 0.075% NaHCO3 (for

cultures in closed vessels) or 0.225% NaHCO3 (for culturesin openvesselsina5%CO2atmosphere).

Virusesand virus assays.Propertiesof theseven

seriesof HSV-1 and five series of HSV-2ts mutants

used in thisstudyare shown inTable 1. Wild-type

virus strainsand the mutagens used for the induction ofts mutants, aswellastheviralDNAphenotypesof

the mutants, are shown. Each mutant from agiven seriesincluded in thestudyhad been shownpreviously

torepresent anuniquecomplementationgroup in that series.Onlyonememberof each groupwastested.

Virusassays wereperfornedin Verocells byusing

a 2% methylcellulose overlay (7) in CO2 incubators

(Wedco,Silver Spring, Md.)with temperature varia-tionsof+0.20C.

Preparationof virus stocks. Stocks oftsmutants

werepreparedbyinfectingconfluent HELcell

mono-layersin100-mmpetriplateswith1 mlof virus

inoc-ulum at multiplicities of0.01 to 0.1 PFU/cell. After

adsorptionat37°Cfor1h,the inoculumwasdecanted, monolayerswerewashedoncewithTris bufferatpH

490

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

27,

Virus type

HSV-1

TABLE 1. Selectedproperties oftheHSV-1and HSV-2 tsmutantsused in thisstudy

Viral

Wild-type strain Mutagen' Mutant pheno- Referenceorinvestigator

typeb

KOS BUdR Al - 1,21

BUdR B2

-BUdR C4

-BUdR D9

-BUdR E6 +

NTG F18 +

BUdR G3 ±

BUdR Ill +

NTG J12 +

NTG K13 ±

NTG L14 ±

UV M19 ±

UV N20 +

UV 022 ±

AP P23 - J.Jofre, C.-T. Chu, and P. A. Schaffer,

17 BUdR

BUdR BUdR BUdR BUdR

KOS 1.1 BUdR BUdR BUdR BUdR BUdR BUdR BUdR

HFEM BUdR

BUdR BUdR NA

HFEM Spontaneous

BUdR BUdR

Spontaneous

BUdR BUdR BUdR

KOSpp60l BUdR BUdR

13 BUdR

BUdR BUdR

HSV-2 186 BUdR

BUdR BUdR BUdR BUdR BUdR BUdR UV

A +

D _

F +

G +

I +

84 478 656 661 901 822 833

Bl B5 B7 N103

LS1 LB1

LB2

LS2

LB3 LB4

LB5

3

7

D10

C4

G5

A8

B5

C2 D6 E7 F3 G4 H9

manuscript in preparation

3

14

A.Buchan +

11

NDd

ND

V.Vonka

17

+

9,20

± + + + +

4

http://jvi.asm.org/

[image:2.501.53.453.83.666.2]

(3)

[image:3.501.65.459.69.356.2]

TABLE 1-Continued

Viral

Virus type Wild-typestrain Mutagen" Mutant pDNoA Reference or investigator typeh

HSG52 BUdR 1 - 12, 13, 29

BUdR 2

-BUdR 3 +

BUdR 4 +

BUdR 5 +

BUdR 8

-BUdR 9

-BUdR 11

-BUdR 12 ±

BUdR 13 ±

IPB2 NA 1 ND 33

NA 42082 ND

333 UV 69 + 15, 32

BUdR 74 ±

UW268 BUdR 1 + 28

BUdR 5 +

BUdR 6

-BUdR 11

-BUdR 12 +

BUdR 19

-BUdR 33 +

aBUdR,5-Bromodeoxyuridine; NTG, nitrosoguanidine; AP,2-aminopurine; NA, nitrous acid.

'The viral DNA phenotype of ts mutants is based upon their ability to synthesize viral DNA at the

nonpermissivetemperature comparedwiththeability ofthe wild-type virustosynthesize viral DNA at this temperatureas100%.+,>20%;±,520%;-,nodetectableviralDNAsynthesized.

'Viral DNAphenotypesof strainKOS1.1ts mutants werekindly supplied by Myron Levine, Universityof

Michigan,AnnArbor,Mich. dND, Not done.

7.4, and 7 ml ofEagle medium containing 5% fetal bovineserumand0.225%NaHCO:)wasaddedtoeach

plate.Infected cellswereincubatedat34°Cuntil

cy-topathic effects were generalized. Cells were then

scraped into the medium, and replicate suspensions

werepooled. Thesuspensionwassonically disrupted

for 45s at10kc andcentrifugedat180xg,5°Cfor10

min. The supernatant fluidwasdispensed, frozen at

-90°C,andassayedatpermissive (34°C)and nonper-missive (38°CforHSV-2 and39°Cfor HSV-1ts

mu-tants) temperatures. Virus preparations containing

significant levels of ts+ revertants were

plaque-puri-fied, and stocks were prepared from revertant-free plaques. Allvirusstocks used in thisstudy exhibited efficiencies ofplating [EOP; (PFU permilliliter,

as-sayedatnonpermissive temperature)/(PFU per mil-liliter, assayedatpermissive temperature)] of less than

orequalto10-4. Mutants withconsistentlyhigh levels ofleak and reversion (efficiency ofplating > 10-4)

werenotincluded in these studies.

Complementation tests. Two types of

comple-mentation tests were employed in this study. The

qualitativetest(4)wasused for initialscreening pur-poses, and thequantitative test (22)wasusedasthe

moredefinitive test.Thesetwo testsdifferprimarily

inthe timepermittedfor theexchangeof gene

prod-uctsandforvirusreplication duringmixed infection. Thequantitativetestinvolvesasingleround of virus

replication,whereasthequalitativetestinvolves

mul-tiple rounds ofreplication. Correlation of results ob-tained in the two types oftests hasgenerally been verygood; however, falsenegative and false positive

qualitativeresults have been described (4,30).

(i)Qualitative complementationtests.

Qualita-tivetestswereconductedasdescribedpreviously (4). Briefly,Verocellswereinfected withtwomutantsby placing filterpaperdisks saturated with mixtures

con-taining 2 x 106, 4 x 105 and 8 x 104 PFU ofeach. mutant per ml on preformed monolayers in 60-mm

petri dishes. Control disksweresaturated with suspen-sions of individual viruses at these dilutions. After incubation at the nonpermissive temperature for 5

days under 2%methylcellulose,neutral redwasadded

toeachplate.Clearingof themonolayersin theareas

ofmixed infections was comparedwith the clearing producedinareasinoculated with eachmutantalone. Ifmoremarked clearing occurred inareasof mixed infections than in areas infected with each mutant

alone,complementationwassaidtohave occurred.

(ii) Quantitative complementation tests. The

quantitative oryield-of-progeny virus test compares theyieldof virus obtained fromcells infected withtwo

492 SCHAFFER ET AL. J. VIROL.

http://jvi.asm.org/

(4)

tsmutantsand incubatedatthe nonpermissive tem-perature with theyield obtained from cells infected

with each mutantalone and incubated at this temper-ature. The test was carried outas previously described (22). Tubecultures of HEL cells were inoculated at a

multiplicityof 2.5PFU of each virus per cell (a com-bined multiplicity of 5), and singly infected control cultures were inoculated with 5 PFU of each virus alone per cell. Infectedcultureswereincubated for 18 h at the nonpermissive temperature in water baths

(Blue M, Rock Island, Ill.)with temperature variations of ±0.1°C, and virusyieldswere titrated in Verocell monolayers. Complementation indices (CI) were cal-culated from the formula: CI= ([A +B]s9c)/([A]is,c

+ [B]}.(-), where A and B are two mutants, and infected cells were incubated at the nonpermissive temperature.Virus yields were assayed at the permis-sivetemperature. As inprevioustests(22), a value of 2orgreater was taken to indicate positive complemen-tation.

Inthe present study each mutant pair was tested

bybothprocedures.If thequalitativetest was clearly

positiveand thequantitativetestyieldeda CI of 10 or greater, no additional tests were performed. If one typeoftestindicatedcomplementationand the other did not, both tests were repeated at least once. Unless otherwise stated, the result of the quantitative test wasconsidered to be definitive. Quantitative tests of mutant pairs which failed to complement in initial tests wererepeated three times.

RESULTS

Quantitative complementationtestswith

ts mutants of HSV-1. (i) Establishment of the standard set of HSV-1 ts mutants.

Rather than undertaking the analysis of all

HSV-1 ts mutants together, complementation

tests werefirstperformed betweenmutants

de-rived from strainsKOSand 17toobtain a

stan-dard set of mutants with which to compare

mutants isolated independently from other

strains. Five ts mutants of strain 17 and 15 ts

mutantsof strainKOSweretested in all possible

pairwisecombinations. The results of thesetests arepresentedinTable2.MutantstsD, F,

G,

and

Iof strain17failedtocomplementmutantsinat

least one of the strain KOS

complementation

groups. Mutants tsJ12 and L14 of strain KOS

and mutants tsD and F of strain 17 failed to complement mutants in two different

comple-mentation groups.These data suggest thattwo

or moreof thesemutants aredoublemutantsor

that thecomplementation between strain 17 ts mutants Fand I and D and F is intracistronic.

The factor which argues moststrongly against

intracistronic complementation is that if this

were the case, F, I, and D (or three of five

mutants of strain 17) would have to contain

mutations in the same gene. This situation

wouldrequireextremelystrong selectional

arti-fact andis,therefore,unlikely.Moreplausibleis

the possibility thattsJ12 and tsL14 are double

mutants.Characteristic ofapoint mutation,the

reversion frequency oftsJ12(10-4) issimilar to that of other mutants in the series, and this

mutant can be rescued by a single restriction

fragment inmarker rescueexperiments. In

con-trast, tsL14 reverts withlow frequency (<10-6)

andcannotbe rescuedbyany onefragmentin a

series of fragments representing the entire

ge-nome(unpublished data).

Mutant tsAof strain 17complemented all 15

ts mutantsofstrain KOS and wasassigned to a

separatecomplementation group.

To examine theuniformity of

complementa-tion patternsobtainedwhenmorethanone

rep-resentative of a groupwastested, two additional

mutants of the KOS series, tsA24 and tsF25,

werealsotested.tsA24and tsF25yielded results

identicaltothose of tsAl andtsF18,respectively,

in that they complemented the representative

mutantsof all five complementation groups of

strain 17efficiently (datanotshown).

The standard set of HSV-1 ts mutants was

thuscomposedofmutantsrepresenting 16

com-plementationgroups.

(ii) Complementation between mutants in the standard set and mutants in other series. Three series of ts mutants have been

derived from the KOS strain of HSV-1by

Schaf-fer et al. (22), Hughes and Munyon (14), and

Vonka (unpublisheddata). The standard set of

HSV-1 mutantswas nexttested against ts

mu-tantsin the lattertwoseries. Theresults ofthese

tests areshown inTable3.

Although tsP23 had complemented all

mu-tantsof strain17 (Table2), itfailedto

comple-mentfourofthemutantsin thetwootherKOS

series. Because of itsunusualgeneticand

biolog-ical properties which will be discussed below,

tsP23wasplacedinaseparatecomplementation

group.

In addition to the multiple negative results

obtained withtsP23,it is evident thattsP23,ts7

and ts3 did not complement any mutant well.

Mutantts3could be adouble ortriplemutant

(1-3, 1-4, and 1-14); however, the consistently

low CI obtained with this mutant makes this

conclusiontenuousin the absence ofgeneticand

physical mapping data.

With theseexceptions, allbut two other mu-tantsfell into groupsrepresented bymutantsin

the standard set. Two were shown to represent

newgroups: ts478andts7.Onlyts661exhibited

anoverlappingpatternofcomplementationnot

involving tsP23. ts661isprobablyadouble

mu-tant because the two mutantsthat it doesnot

complementareseparated by additional cistrons

(23). ts901 is asingle mutant in the tsB2 (1-2)

http://jvi.asm.org/

(5)

[image:5.501.59.455.92.416.2]

494 SCHAFFER ET AL.

TABLE 2. Complementation between ts mutants of HSV-1 strains KOS and 17

Mutants of Mutants of strain 17 Complemen- Potential

mem-strain KOS A D F G tationgroup bers ofgroup

Al + + + + 1-1 Al

1,704b 315 52 2,311 244

B2 4,459 1.1 8.8 7,671 18,000 1-2 B2, D

C4

~+

+ + + + 13 C

C4 3,667 66 756 1,467 1,216 1-3 C4

D9

~+

+ + + + 14 D

D9 8,565 9,105 60,067 2,439 2,039 1-4 D9

E6

~+

+ + + + 15 E

E6 4,958 626 16,040 189 1,504 1-5 E6

F18

~

+ + + +

+16

F8

F18 233 97 81 21,100 3,620 1-6 F18

G3 1,771 2,417 126 599 685 1-7 G3

+ + + + + il

12,000 6,294 26,690 191,000 297 1-8

Ill

J12 + +

-+

1- J12,F

444 658 1.0 5,374 1.3 1-9 (J12),' F, I

K13

~

+ + + + + 11 1

K13 124 20 29 269 38 1-10 K13

L14 23 0.2 0.7 1,100 181 1-11 (L14),' D, F

M19

~

+ + + + + 11 1

M19 10,364 500 5,763 436,700

3,473

1-12 M19

N20

ND"

ND ND ND

ND113

N0G

576 423 706 0.7 4,150 1-13 N20,G

022

~

+ + + + + 11 2

022 877 93 219 2.0 1,192 1-14 022

P23

ND

ND ND ND ND 1-15 P23

2,100 380 280 30 11

A 1-16 A

+and-areresults ofqualitative complementationtests.For furtherexplanation,see text.

bNumbersareCIobtained in quantitative complementation tests. Pairs with indices of 10 or greater were not

retested. Pairs which failed tocomplement, yieldedCI of less than 10, oryielded disparate results in the two typesof test wererepeatedtwo tofive times. In these cases valuesrepresent the average ofalltests; results of

quantitativetests wereregardedasdefinitive.Numbers initalics indicatenegativeindices.

'Potentialdouble mutantsareshown inparentheses.

dND, Not done.

cistron. BecausetsL14 also containsadefectin

cistron 1-2, itwouldnotbeexpectedto

comple-mentts901.

Theresults ofcomplementationtests

compar-ing mutants in the standard set with mutants

derived from strains HFEM and13 areshown in

Table 4. Again, tsP23 yielded negative or low,

positive indices with nearly half the mutants

tested. Excluding results with tsP23, mutants

tsBl, B7, LB4, LB5, and D10 were shown to

constitute new groups, whereas all other

mu-tants wereassignedtoexisting groups. Overlap-pingpatterns ofcomplementationnotinvolving

tsP23wereevidentinthe case of tsLB2 andLS2.

Because tsL14 is probably a double mutant, tsLB2wasassignedtogroup 1-2withtsB2. The

failure of tsLS2 to complement two mutants

whichliesomedistanceapart on theKOS

link-age map (23) suggests that tsLS2 is a double

mutant. The mutation in tsLB3 is probably in

the F cistron of strain 17 (i.e., cistron 1-11),

otherwise it wouldnothavecomplemented tsB2.

Theoverlapping patterns of complementation

noted in Tables2, 3, and4 suggest thattsL14

and tsB2 shouldnothavecomplementedin

pre-vious studies (22). By present standards,

how-ever, they did complement, although

ineffi-ciently (CI =3.6). Such inefficient

complemen-tation in retrospect may have been

intracis-tronic.

(iii)Complementationtestsamong ts

mu-tantsof HSV-1 representing putativenew

complementation groups. Eight of the 43

HSV-1tsmutants werefoundtocomplementall

mutants (except tsP23) in the standard set of test mutants and were, therefore, assigned to

J. VIROL.

http://jvi.asm.org/

(6)

COMPLEMENTATION WITH ts MUTANTS OF HSV 495

TABLz 3. ComplementationbetweenHSV-1ts mutants in the standard set and other tsmutantsof strain KOSa

Mutants Mutantsofstrains:

Mutan Comple- Potential

mem-84 478

| . KOS1.1 KOSpp601 mentation1

ber

ofgroup

set 84 478

1

656

1

661

1

901 822 833 3 7

Al 27,000 5,294 1.1 9,600 343 17 42,727 39 13 1 Al65

B2 + + + + - + + + - 12 B,0

B2

1,210 3,111 622 653 1.9 30 18,235 25 12

1-2

B2,

901 -4

+ + + + - + - A 61,3,3

C4

40 63 34,400 0.8 980 225 0.4 0.6 6.8

1-3

C4,(661),833,(3)

D9 + + + + + - - + + 14 D984 3

D9

|

1.9

113 34,800

4.2

564

100

13

0.6

17

4 D9,8,(3)

-6

+ + + + + + + - - E

E6

4.0 469 1,520 12 1,494 230 19,166 35

7.3 1-5

F18

+

1-6 F18

46 465 661 2,333 376 72 14,285 32 21

G3

+ + + + + + + + 1-7 G3

38,000 28 2,890 424 183 45 4,181 4.2 3.0

I1l

+ + + + + + + + + 1-8 ill

340,000 944 24,400 2,000 565 59 3,181 30 6.8

J12 + + + + + - + + + 1-9 (J12),822

13,000 1,062 17,600 1,273 240 0.6 15,000 14 6.4

K13 + + + + + + + + + 1-10 K13

786 344 76,000 154 238 51 1,791 30 4.0

L14 18 1,000 3,080 35 0.4 2.8 9,523 19 2.8 1-11 (L14),901

M19

+ + + + + + + + +

1-12 M19

8,000

17,500 144,000

520

518

15 15,384

8.4

14 N20

+ + + + + + + + +

1-13 N20

682

421

376

169

197

22 9,393

33

21

02 + + + + + + + + + 11 2,(6) 3

022

21

44 1,289

1.3

352

34

435

1.2

8.2 1-14

022,1(61),(3)

P23

0.9

4.6

35

5.1

230

11

1.0

0.02

0.7

1-15 P23

A

+ + + + + + + + +

A

24,000

1,062

1,880 8,800

1,877

26 7,000

18

7.6 1-16

478 1-17 478

1

1 1 1

1

1 I

1 ~~~~7 1-18 7

aSeefootnotes toTable2forexplanationofresults.

putative

new

complementation

groups.To con- Mixed infections

resulting

in

equivocal

firmthe

uniqueness

oftheir

functions, they

were and anomalous patterns of

quantitative

tested among themselves

by

the

quantitative

complementation.

CI in

positive quantitative

complementation

test.Theresults of thesetests tests

ranged

from 2.0

(e.g.,

Table 4, tsK13 +

are shown in Table 5.

Complementation

was

tsLB3)

to440,000

(e.g.,

Table4, tsMl9+

tsG5);

demonstratedto occur between mutantsin all 72% ofpositive indicesweregreaterthan50.Of

pairwise

combinations.

Therefore,

eachof these the 123

positive

indiceswithvalues of lessthan

mutantswasshowntorepresenta new

comple-

50,

nearly

halfwere

produced

inmixed infections

mentation group. Mutantpairs ts7 + tsBl, ts7 with5 mutants (e.g.,tsP23, tsLB2, tsC4 [strain

+tsLB5,ts7+tsD10 andtsLB4 + tsLB5yielded 13],

ts3,

and

ts7).

Negativeindicesrangedfrom

low,

but

positive

indices. Similar results were 0.02(e.g.,Table3,tsP23+

ts3)

to1.9(e.g., Table

obtained inthree separatetestsofthesemutant 3, tsD9 +ts84).

pairs. The low indices obtained with ts7 are Althoughthe results of mostquantitative tests

consistent withpreviousfindings(Table 3). were unequivocal, certain mutants exhibited

Quantitative

complementationtestshave thus equivocaloranomalous behavior inquantitative

resulted in theidentificationof23complemen- tests. A careful survey of results

indicates

that

tationgroups from among43ts mutantsofHSV- such behavior isaconsequence of the properties

1

(Tables

2 to5). ofindividualmutantsratherthan theproperties

http://jvi.asm.org/

(7)

J. VIROL.

cq m

O-"Z N

q m I" LO CO 000 C I" Lo co

+ ^+ + N+ + + + O+ + + + +N +

+t- + P ++P- ++ + >+;+ -4+ t-+.+ 4+ +q++ 4 + 04 + Lq

+ ~ ~t~ ~~' + r+° +° + +C,,+ +_+ _+

+_+O+U+ + +°++ +U+ s|>+ +: + 8 + +

C+ + c ++ + +m+ + +< +N+° :+s+ O+N+

+0 + + +

'o

+ +

CS+

+ +

8

+_+ +uN+°+ Ci+

+~~~ ~N-+st- ult+ + N+ + +-5 + +t-4 +O + ul

m-

t;SXx c: _ ; X

.9 9_co

1.

I

i

.4.1

1

:a

e

I

1 11-1.

O

-S a

A

_:3

e

12

it

9:

-S

t

d

Po PA m

49

E-,

4)

r.

t

0

d

Iz

LO

0

m

1.4 u

0

V-4

0

LO

5

9

u

2 cli

5

1. m

C9

m

0 1

2

-4

5

P-4

2

a

P-4

z

t--a m

2

m 0

m

1-4

m

C)1.6-ill

-10 &

-SI

ET AL. 496 SCHAFFER

.8

I

l

@.

;R

http://jvi.asm.org/

(8)

of the testitself.

Toillustrate the anomalous behavior of these

mutants,the derivation ofCI for five

represent-E8

ative mutant

pairs

is shown in Table 6. Data

froma

typical

positive

test areillustrated

by

the

pair

tsI

+ tsB2. In this test, the yield from the

mixedinfection was clearly greater than the sum

ofyieldsincontrol, single infections. Such

effi-cient complementation is in sharp contrast to

the typical negative test illustratedin Table 6

eo

No

eq eq eq by

tsN103

+tsAl in whichtheyieldof the mixed

infection was

slightly

less than the sumof the

yields

from control infections.

ull<

Despite

the fact that all mutant stocks used

in these tests exhibited low levels of leak and

reversioninplaqueassays,severalmutants,

no-tably ts3, 7, LB2, and C4 (strain 13), yielded

higher levels ofleakand/orreversion than other

mutants afterincubation for 18 h at39°C which

oo complicated the interpretation of results. As S

a shown in Table6 (ts3+ tsG3),thebackground

levels of leak and reversion produced by ts3

Lo

mresulted

in a reduction in the CI in this test.

Whether the index would have been higher if

leak were suppressed by incubation at higher

temperature

is not

known.

It

was not

surprising,

therefore,

thatmutantswith the

highest

levels ofleak andreversionyielded consistently lower CI than mutants which did not leak or revert

significantly. Consequently,

these mutants

yielded

a

greater

number of

equivocal

and

neg-C4

.lX

>ative

results.

Anadditionalcauseof

equivocal

and

negative

e results in quantitative tests is illustrated in

Ta-ble 6 by pairs tsP23 + ts84 andtsP23 + ts3. In

these teststhe yieldsof mixed infectionswere

significantly lower than the sum of the yields of

controlinfections. InthecaseoftsP23+ts3, this reduction was nearly 1.5 logs. tsP23 also

sup-pressedthe

replication

ofts+ revertantsof ts3 in

this test. In contrast to other mutants which

consistently yielded low CI, tsP23 did not exhibit

unusually highlevelsofleak and/or reversion.

0 u i3 These data indicateastrong

tendency

by

tsP23

to suppress the productionof ts virusthrough

<:-~complementation mechanisms and ts+

revert-ants andrecombinants. Theability of tsP23 to

suppress the growth of the wild-type virus by SE > >-= greater than 90% when equal

multiplicities

of

tsP23 and wild-type virus are used to infect cells has been noted by J. Jofre (personal

communi-°X

XM4

cation). In a total of 28 quantitative tests in the

present study,tsP23failedtocomplementeight

mutants and produced indices of less than 10

with four othermutants. Itis noteworthy that

five ofeight of the negative quantitative tests

.2 6^3 8 X z

Swere

positive

in

qualitative

tests,whichinvolve

multiple rounds of replication.

Furthermore,

the

http://jvi.asm.org/

(9)

[image:9.501.61.459.89.192.2]

TABLE 5. Complementation tests with HSV-I ts mutants representing putative new complementation groups

CIfrommixed infections with ts mutant: Mutant

478 7 Bi B7 LB4 LB5 D1O

478 16 166 174 1,322 4,167 253

7 5.4 64 18 5.7 3.2

Bi 159 34,286 13,000 1,287

B7 62 30 47

LB4 2.6 86

LB5 210

[image:9.501.59.458.212.456.2]

D10

TABLE 6.

Quantitative

complementationtests

exhibiting

typical positive, negative, equivocal, and anomalousresults

tsMutantsin test Virusyield(PFU/ml)

Mixed infections Controls

(A+B) ci" Interpretation

Mutant Mutant

A B A B

34"C

390C _340C

390C 34°C 390C

I B2 1.8x10' 1.2x 104b <101 <101 1x 101 <10o 18,000' Positive N103 Al 1.1 x 102 <10' 0.5 x101 <101 1.2 x102 <10' 0.9" Negative 3 G3 1.2xlOr1 2.7 x 104b1.2x1043.3x 10'e4.0 x10' <10' 99d Low positive;

leak and re-versionby

mu-tantA

P23 84 1.0X 102 <10' 4.3x 102 <10' 5.0x 10' <10' 0.2" Negative;

inter-P23 3 5.1x1022.3x 102b4.3 x 102 <10' 1.2x 1043.3x 1o3e 0.04d ference by

mutantA

"CI= ([A+

B]e,'c

)/([A]39,c

+

[B]:j'sc)

wheninfected cellswereincubatedat

390C

and

yields

were

assayed

at34°C.

'Yields ofts+recombinants and revertants in progeny of mixed infections assayed at39°C.

'This value appears inTable2.

" This table contains datafromonlyone ofthe two to five tests with these mutant pairs; indices were averaged toproduce the resultsshown inTables3and 4.

'Yields ofts+revertants in progeny of control, single infections assayed at390C.

reasons for the efficient

complementation

ob-served between tsP23 and mutants

tsA,

D,

F

(Table 2),901(Table 3),and B5 and LB4(Table 4),areunclear.

Ofequalsignificanceinthe

interpretation

of the results ofquantitative testsis the

identifi-cationofmutantswithoverlappingpatterns of

complementation, i.e., those with multiple ts

defectsasdescribed above.

Qualitative complementation tests with

tsmutantsof HSV-1.Althoughthe

quantita-tivetest wasconsideredtobe definitive in these

studies, excellent agreement was observed

be-tweentheresults ofqualitativeandquantitative complementation tests. With ts mutants of

strainsKOS and 17

(Table

2),nofalse

positive

qualitativetests(testsyielding positive

qualita-tive and negative

quantitative

results)

andtwo

falsenegativetests(testsyielding negative

qual-itative and positive quantitative results) were

observed.

Intestsbetween mutants in the standard set

and other ts mutants ofHSV-1 (Tables3and 4),

agreement between the two types of tests was

alsogood,albeitlessso. Thirteenfalse negative

qualitative results (e.g., Table 3, tsC4 [strain

KOS] +ts84) and 12 false positive results (e.g.,

Table3, tsD9 +ts84)wereobtained. Of noteis

the observation that 5 of the 12 false

posi-tive results occurred in mixed infection with

tsP23. Thus, in atotal of433qualitative tests,

15 false negative (3.4%) and 12 false positive

tests(2.8%)wereobserved.

Quantitative complementationtestswith

ts mutants of HSV-2: the standard set of

HSV-2tsmutants.Inacollaborative

comple-mentation study involving 20 ts mutants of

HSV-2 strains HGS52 and 186,Timbury etal.

(30) identified18complementationgroups.

Rep-resentatives ofthese 18 groups were therefore

used in thisstudyasthestandardsetofHSV-2

tsmutants.

498

SCHAFFER ET AL. J. VIROL.

http://jvi.asm.org/

(10)

Complementation between the standard

setand other HSV-2tsmutants.Theresults

of complementationtestsbetweentsmutantsin

the standard set and those derived from

wild-typestrainsIPB2, 333,andUW268areshownin

Table 7. Themoststrikingfeature oftestswith

HSV-2mutantswasthe fact that CI were

con-sistently 10-foldlower thanindices obtained in

tests with HSV-1 ts mutants. Positive indices

ranged from 2.0to9,090; 39%wereless than 10.

Intestswhichyielded indices between2and5,

resultswereconsistent fromtest totest,i.e.,they

werealwayslowbutpositive.Althoughmutants

which exhibited high levels of leak were

ex-cluded from the study, HSV-2mutants

consist-ently yielded higherlevels of leak than HSV-1

mutants.Leakinesswasthusasignificantfactor

inthe generation oflowcomplementation

indi-ceswith thesemutants.Clearly,such inefficient

complementation created difficulties in

assign-ingmutantstocistrons.

The failureofts2tocomplement 10mutants

also contributedtodifficulties inassignment of mutants to cistrons. Failure of ts2 to

comple-mentwasnotduetoexcessive leak. It ispossible

that ts2may possesssimilarinterfering

proper-tiestothose of tsP23 of HSV-1 describedabove,

i.e.,whetherts2is abletosuppressreplicationof

wild-type virus atthe nonpermissive

tempera-ture is not known. To determine the relative

distance ofts2fromother mutants,allmutants

in Table 7 except tsl (strain UW268) were

screened for theirabilitytorecombine withts2

at the permissive temperature. Recombination

frequencieswereadditive for themostpart and

ranged from3to 18%, indicatingthatts2could

recombineefficiently, demonstratingthat itwas

notcloselylinkedtomutantswith which itfailed

tocomplement.

Of the 11 HSV-2tsmutantstestedagainstthe standard set, andexcluding theresults oftests

with ts2, all but two, tsll and tsl2 (strain

UW268),failed tocomplement one ortwo

mu-tantsin the standardset.In mixedinfection,tsll

+ tsl2 yieldedaCI of 35 and werethus shown

tobeunique.Thesetwomutantswere,therefore,

placedinnewcomplementationgroups2-19 and

2-20. Four of the 11mutantsagainstwhichthe standard set was tested exhibited overlapping

patterns ofcomplementation (exclusiveoftests

withts2).Because ofthisoverlapand theoverall

low indicesobtained,thesemutantscouldnotbe assigned unequivocally to complementation

groups. Inall,20complementationgroupswere

identified by quantitative complementation

testswithtsmutantsof HSV-2.

Qualitative complementation tests with

tsmutantsofHSV-2. Of198qualitativetests,

9falsenegativetests(4.5%) and12falsepositive

tests(6.1%)wereobserved. Eight of the 12 false

positive tests occurred with the

noncomple-menting mutantts2.Thus,although false

nega-tive andfalse positive tests were obtained with

bothHSV-1 andHSV-2tsmutants,the

propor-tion ofanomalousqualitativeresults was higher

in theanalysisofHSV-2ts mutants.

Viral DNA phenotypes of members of complementation groups. Summaries of the complementation groups and the viral DNA

phenotypes ofmutantsin each group are shown inTables8and 9.

Excluding mutants which exhibited

overlap-ping patterns of complementation, the viral

DNA phenotypes of mutants in the 10 HSV-1 complementation groups containing two or more

members are inexcellentagreement (Table 8).

Six of the 23 groups contain DNA- mutants, 3

contain mutants which synthesize20% or less of

wild-type levels of viral DNA, and 11 contain

DNA'mutants.

Ten of the20HSV-2 complementationgroups

maycontainmorethan one member; 4 ofthese

10 groups(2-2, 2-3, 2-5, and 2-15)clearlycontain

two or more mutants. The viral DNA

pheno-typesof mutants inmultimemberHSV-2 groups

were not ingoodagreement. Thebasis for this

disparity maylie in the inavailabilityof defini-tivecomplementationdata orperhapsto

differ-ences in the conditions used for determining

viral DNA phenotypes of mutants in different

laboratories. In any event,availabledataclearly

indicate that 7ofthe 20groups contain

DNA-mutants(groups 1, 2, 3, 4, 7, 8, and

2-19), 2contain DNA'mutants(groups 2-13 and

2-14), and8contain DNA+ mutants (groups

2-10, 2-11,2-12, 2-15, 2-16, 2-17, 2-18, 2-20).

Despitethe fact that theresultsof some

com-plementationtests wereequivocal and that the

viral DNA phenotypes of mutants in certain

multimember groups were dissimilar, 6 of 23

HSV-1 groups and 7 of 20 HSV-2 groups were

shown to contain DNA- mutants. Thus, these

investigationsconfirmthe previous observation

thatalargenumber of cistronsareessential for

viralDNAsynthesis (21, 27).

DISCUSSION

Basedprimarilyonthe results of quantitative

complementation tests,23cistrons of HSV-1and

20 cistrons ofHSV-2 have been identified.

Al-though overlapping patterns of complementa-tion complicated the assignment of some

mu-tants tocistrons,they didnotdiminish the total

number of cistrons identified. Onthe contrary,

overlappingpatterns ofcomplementation have facilitated the identification of potential double

mutants.

The results of quantitative complementation

http://jvi.asm.org/

(11)

R ET AL.

m

J. VIROL.

ko

^e

6

CD

_C4

l

^~~

eq -Meq e Xtwo___~__I"- LO

eq eq eq eq eq eq eq eq eq eq eq eq ce eq eq eq

+

L3+O

+

N;+ cs+

+

D+ _+ co+ l£+

+ + L + + +

24

+ lq+

+s

~~~

|

+0e>t+

>

~

e~~q

~

++t.+o+.+.+.+e,

* e

+.+3

+I+`e+ +++g + + ++ m+ X+ + +a

- ' + -++Lo

-++ + + C>+& +c + +D++ + ++ +cQ +ll +ao I 'O + +e

+g +0 +s

+oI

+LO++uVt

*+E

+°,l+"q

N+10 so+t-+t+e+eq+

-"' N +.+ +

`@

'tI

+

o

+'- + -! +

S.+ +.|

+In + +

LO m -4mm e cq w w -w m

eq O -4 m e

+ +Lq+)en + + + + L +-!+ Iq+

m V-4_ eq X Ur3e cq

SCHAFFE]

0

.2

eq

500

.4

a

P.

.s

.S

r11

0

.5

ct

O

2

t1

8

P. 4 laE

http://jvi.asm.org/

(12)

tests were considered to be definitive in these

o £ studies. A CI of 2 was selected as the value

x8

AQsignifying

complementation when

complemen-tation studies with HSV ts mutantswere

initi-2 l, | ated (22). The levelsofcomplementationwhich

representinter- and intracistronic

complemen-v-4 cq tation were notknown at that time, and are not

known now, due to the absence of fine structural

mapping data. Thus, the number of cistrons

AA

S c; X o g

Sidentified

in this study by the quantitative test

7c.s R , , , @ may be somewhat greaterorless than thetrue

numberofcistrons

represented by

themutants

studied. On theotherhand, problems alsoexist

+88 o with the

qualitative

test.

By

using

tsmutantsof

HSV-2,

the

qualitative

test

yielded

resultswhich weresimilartothose obtained ininfectious

cen-ters tests

(30).

Both the

qualitative

and infec-o + tious centers tests involve

multiple

rounds of

replication,and both arecomplicated bythe fact .2 that false positive results couldbe produced

if

recombinants can be

generated

in the absence

ofcomplementation. The classical definition of

the cistron is basedupon virusyields produced

X

~

during one round of replication (2, 8) and

3_ + E _ g whether recombinationcanoccurinthe absence

of

complementation

isnot

presently known;

con-.! ; sequently, the quantitativetestwas considered

to be more definitive.

I CD + <

.<

Because the limitsofeithertestarenot

pres-ently known, it is essential that the cistron

as-signments ofall mutants beconfirmedby

recom-+

*t g bination and markerrescue tests. Genetic and

ba<+e- physical mapping data would be particularly

2E useful in the case of mutants which appeared to

bedoublemutantsand those which

yielded

mul-- +

tiple

low and

negative

results

(tsP23

ofHSV-1

and ts2 of HSV-2) in complementation tests.

Thefeasibilityofmappingviral genes and

poly-o+ .~ peptidesbyanalysisofHSV-1 x HSV-2 recom-binants has recently been demonstrated (18, 19;

N. M.

Wilkie,

N. D.

Stow,

H. S.

Marsden,

V.

Brown, R. Cortini, M. C.

Timbury,

and J. H.

8 + @

.i

Subak-Sharpe, Abstracts of the Third

Interna-tionalSymposium On Oncogenesisand

Herpes-viruses,

Cambridge,

Mass., 1977). It is

antici-> o pated that

complementation,

recombination and

31+*_ | markerrescue testscombined with analysesof

HSV-1 x HSV-2 recombinants

will

ultimately

lead to the identification of most essential viral

_ M = f cistrons and to thepositioningofthese cistrons

- + _ EZ a onthe physical maps of HSV DNA.

The identification ofadditional viralcistrons

, X

~

!.

5 should continuetoconstitutea

primary goal

of

genetic studies of HSV. Brown et al. (3)

con-g8eQ E g @

~~cluded

that the HSV genome encodesa

mini-La

]$R¢ <^, == mum of 30 essential functions. Given the large

-nX>e theoretical coding capacity of the genome, this

value isprobablyaconsiderableunderestimate.

27,

http://jvi.asm.org/

(13)

[image:13.501.59.455.96.359.2]

J. VIROL. 502 SCHAFFER ET AL.

TABLE 8. HSV-I complementation groups

tsMutants derived from strains: Predominant

Complemen- viralDNA

tationgroup KOS 17 KOS 1.1 KOSpp HFEM" HFEMh 13 phenotype

601

1-1 A1' 656- N103-

-1-2 B2- D- 901- LB2:

-1-3 C4 (661j)d,833- (3NI))e

1-4 D9- 84- (3ND))

1-5 E6+ (LS2+) +

1-6 F18+ LS1+, (LS2+) +

1-7 G3: LB1+ C4+ +

1-8 I11+ +

1-9 (J12+) I+ 822+ B5+ G5+ +

1-10 K135 i

1-11 (L145k) F+ LB3+ +

1-12 M195 +

1-13 N20+ G+ +

1-14 022± (661 ) (3N) ±

1-15

P23-1-16 A+ +

1-17 478+ +

1-18 7ND) ND"

1-19

Br-1-20

B7-1-21 LB4+ +

1-22 LB5+ +

1-23 D10-

-"Mutants

derivedfromstrain HFEMbyA.Buchan.

bMutants derived from strain HFEMbyI.

Halliburton.

' +, ±, -,and ND are viral DNAphenotypesfromTable 1.

dMutantsinparenthesesareputativedoubleortriplemutants.

'ND, Notdone.

TABLE 9. HSV-2complementation groups

Complemen- ts Mutants derived fromstrains: Predominant viralDNA

tationgroup KOS HSG52 IPB2 333 UW268 phenotype

2-1 1-a

2-2 H9- 1,(42082)b 19-

-2-3 B5- 6-

-2-4 9

-2-5 C25 10- 1+ Mixedphenotype

2-6 11- (74+) (5+) Mixedphenotype

2-7 A8- (42082)

2-8

2-2-9 8- (69+)(74k) Mixedphenotype

2-10 3+ +

2-11 4+ +

2-12 5+ (69+) +

2-13 12: ±

2-14 13- (42082) ±

2-15 D6+ (5+) +

2-16 E7+ +

2-17 F3+ 33+ +

2-18 G4+ +

2-19 11-

-2-20 12+ +

"+,±, and-areviral DNAphenotypes fromTable 1. If phenotype isnotgiven, it hasnotbeendetermined. bMutants inparentheses are putative multiple mutants.

http://jvi.asm.org/

[image:13.501.60.455.436.648.2]

(14)

HSV 503

The results of the present study demonstrate

thatmostinvestigatorshave isolatedtsmutants

inthe same genesand that the numberof

essen-tial cistrons which remain to be identified by

techniquesalready employedisprobablysmall.

Anexplanation for the failure toidentify

ad-ditional cistrons isreadilyapparentfrom Table

1. Of the 72 ts mutantstested, 57wereinduced

with BUdR. Amongthe 23HSV-1

complemen-tation groups, 3 containasinglemutantinduced

by a mutagen other than BUdR (1-10,NTG;

1-12,NTG;and 1-15,AP).Specialmentionshould

bemade of LS1 and LS2-theonlytwo mutants

isolated in the absence of mutagen. Although

difficulttoisolate,spontaneousmutantsarefree ofmutationalartifact.Ofthe20HSV-2

comple-mentation groups, each contains one or more

BUdR-inducedmutants,reflectingthefactthat

26 of the 29 HSV-2 ts mutants tested were

BUdR-induced.

Thatmutational hot-spottingmay occur with HSV DNA and BUdR must be considered in

evaluating the number of cistrons identified in thisstudy.Mutational artifact isa common fea-tureofgeneticsystems inwhich mutagens with

highly specificmodes of actionare

employed

(6,

16). Because nucleotidesare notrandomly

dis-tributedinHSVDNA,largeT-rich geneswould bepreferentially mutagenized bytheT-specific

mutagen BUdR. In thefutureit willbe of

inter-est to compare the base compositions of the

sequences known to encode individual viral

geneswith the mutagens usedtoinduce

muta-tions in those sequences. For the present, it is

clear that further attemptstoisolate HSV

mu-tantsshould utilize mutagens whosespecificities

differ from that of BUdR. Nitrousacid, which

canproducetransitions and transversions in all fourbases, wouldseem aprimecandidate.

With regard to mutagenesis with NTG, it

should be noted thattwo of the four

NTG-in-ducedmutantsofstrain KOS usedtoestablish

the standard set of HSV-1 mutants may be

doublemutants. Because NTG isnotorious for

inducing multiple closelylinked mutations (10,

26), this mutagen should be used advisedly. In

additiontotheuseof different mutagens, future

effortsmightalso be directed towards the

isola-tion of other types ofconditional lethalmutants.

Cold-sensitivemutants,forexample, have been

shown tobeclustered inregions of the genome

notcontainingtsmutations(24).

Thepresentstudywas undertakenin partto

identify additional essential HSV cistrons from

amongts mutants ofHSV-1and HSV-2 isolated

in different laboratories. New cistrons both of

HSV-1 and HSV-2have,indeed, been identified.

However, having demonstrated that most

inves-tigators have isolated mutants with defects in

the same series ofcistrons, the need to identify

additionalcistronsbynewmethods isclear. Fur-thermore, recombination analysis, marker

res-cuestudies, and phenotypic characterization of

mutantsassignedtothe same cistron are

essen-tial toverify the results of the complementation

testsreported herein.

The second purpose of thisstudywas to

facil-itatefutureinvestigationswith HSV ts mutants

by (i) introducing a uniformsystem of mutant

nomenclature and (ii)

identifying

mutants in different series which are defective in the same

viralgenes toavoid duplicationofeffort in

phe-notypiccharacterization of mutants. In addition,

it is theintent of thecollaboratorsinthis study

tomake available series of ts mutants

represent-ingeach ofthe known cistrons to facilitate the

identification of additional essential viral genes. Auniformsystemofnomenclatureforts

mu-tants of HSV-1 and HSV-2 as suggested by

Timbury et al. (30) has been introduced. Al-thoughthe systempermits individual

laborato-ries to retaintheir own mutant designations, it

ishopedthatpublicationsinvolving studies with

HSVts mutantswillinclude the cistron

assign-ment of mutants by reference to the common

system suggested here. Numbers rather than

letters have been used toreflectthe large coding

capacity of the virus, and the prefix "1, or "2"

has been used todifferentiateHSV-1and

HSV-2 complementation groups, respectively. The

proposedsystem isnotintended to be rigid but

toform the basis for asystemof nomenclature

which is boundtoevolve asknowledge advances

in thefield.

ACKNOWLEDGMENTS

We thank M. L.Hendricks forexcellenttechnical assistance andM.J. Tevethia forhelpfulcommentsandsuggestions.

Thisinvestigationwasconductedin part inthe Department ofVirologyandEpidemiology, Baylor College ofMedicine, Houston, Tex., supported byPublic HealthServiceresearch contractN01 CP53 526 within theVirus CancerProgram of the NationalCancerInstituteand research grant CA 10,893 fromthe NationalCancerInstitute. The workconductedin Boston wassupported by research grant CA 20,260 and Pro-gram Project grant CA 21,082 from the National Cancer Institute. M.C.T. was therecipient ofanEleanor Roosevelt International Cancer Fellowshipfrom theAmericanCancer Society and aBellahoustonTraveling Fellowship from the ListerCentenary Appeal Fund, UniversityofGlasgow.

LITERATURE CITED

1. Aron,G.M.,D. J. M. Purifoy,and P. A. Schaffer. 1975.DNAsynthesis and DNA polymerase activity of herpes simplex virus type 1 temperature-sensitive mu-tants.J.Virol.16:498-507.

2. Benzer, S. 1957. The elementaryunits ofheredity, p. 70-93. In W. D. McElroy and B. Glass (ed.), The chemicalbasis ofheredity. The Johns Hopkins Press, Baltimore.

http://jvi.asm.org/

(15)

504 SCHAFFER ET AL.

3. Brown,S.M.,D. A.Ritchie,and J. H.Subak-Sharpe. 1973.Geneticstudies withherpessimplexvirus type 1: theisolation oftemperature-sensitive mutants, their arrangement intocomplementationgroups and recom-binationanalysisleadingtoalinkagemap. J. Gen. Virol. 18:329-346.

4. Chu,C.-T.,and P.A. Schaffer. 1975.Qualitative com-plementationtestfortemperature-sensitivemutantsof herpessimplexvirus. J. Virol. 16:1131-1136.

5. Delius, H.,and J. B. Clements. 1976. Apartial denatur-ationmapofherpessimplex virustype1DNA:evidence for inversions of the twouniqueDNAregions. J. Gen. Virol. 33:125-133.

6. Drake, J.W. 1963.Propertiesofultraviolet-induced rII mutantsofbacteriophage T4. J. Mol. Biol. 6:268-283. 7. Dreesman, G. R., and M. Benyesh-Melnick. 1967. Spectrum of humancytomegalovirus complementfixing antigens. J. Immunol.99:1106-1114.

8. Edgar,R.S., G.H.Denhardt,andR. H.Epstein.1964. Acomparativegeneticstudyofconditionallethal mu-tations ofbacteriophage T4D.Genetics49:635-648. 9. Esparza, J., D. J. M.Purifoy,P.A.Schaffer,andM.

Benyesh-Melnick. 1974. Isolation, complementation andpreliminaryphenotypic characterization oftemper-ature-sensitivemutantsofherpessimplexvirus type 2. Virology57:554-565.

10.Guerola, N., J. L Ingraham,and E.Cerda-Olmedo. 1971.Introductionofclosely-linkedmultiplemutations bynitrosoguanidine. Nature (London) 230:122-124. 11. Halliburton,L.W., R. E. Randall, R. A.Kfllington,

and D. H. Watson.1977.Some propertiesofintertypic recombinants betweentype 1andtype 2 herpes simplex virus.J.Gen. Virol.36:471-484.

12. Halliburton,L. W., and M. C. Timbury. 1973. Charac-terizationoftemperature-sensitivemutants of herpes simplex virustype 2.GrowthandDNAsynthesis. Vi-rology54:60-68.

13. Halliburton,I.W., andM.C.Timbury.1976. Temper-ature-sensitivemutants of herpes simplex virus type 2: description ofthree newcomplementationgroups and studiesontheinhibitionsofhostcellDNA synthesis. J. Gen. Virol.30:207-221.

14. Hughes, R. G., Jr., and W.H.Munyon.1975. Temper-ature-sensitivemutantsofherpes simplex virus type 1 defectiveinlysis but not intransformation. J.Virol. 16:275-283.

15. Koment,R.W.,andF.Rapp. 1975. Temperature-sen-sitive hostrange mutants of herpes simplex virus type 2.J. Virol.15:812-819.

16. Krieg,D.R. 1963.Specificityofchemicalmutagenesis. Prog.NucleicAcidRes.2:125-168.

17. Manservigi,R. 1974.Methodforisolationandselection oftemperature-sensitivemutants of herpes simplex vi-rus.Appl. Microbiol.27:1034-1040.

18.Morse,LS., T. G.Buchman,B.Roizman, andP.A. Schaffer.1977.Anatomyofherpes simplex virus DNA. IX.Apparentexclusionofsome parental DNA

arrange-mentsin the generation ofintertypic (HSV-1 xHSV-2) recombinants. J. Virol.24:231-248.

19.Morse,LS.,LPereira,B.Roixman,and P. A. Schaf-fer. 1978. Anatomy ofherpes simplex virus (HSV) DNA. X.Mappingof viral genes by analysis of polypep-tides and functionsspecifiedbyHSV-1 x HSV-2 recom-binants. J. Virol. 26:389-410.

20.Purifoy, D. J. M., and M. Benyesh-Melnick. 1975. DNA polymerase induction by DNA-negative temper-ature-sensitive mutants ofherpessimplex virus type 2. Virology68:374-386.

21. Schaffer, P. A. 1975.Temperature-sensitive mutants of herpesviruses. Curr. Top. Microbiol. Immunol. 70:51-100.

22.Schaffer, P. A., G. M. Aron, N. Biswal, and M. Ben-yesh-Melnick.1973. Temperature-sensitive mutants of herpes simplex virus type 1:isolation,complementation and partial characterization.Virology 52:57-71. 23. Schaffer, P. A., M. J. Tevethia, and M.

Benyesh-Melnick. 1974.Recombinationbetween temperature-sensitive mutants ofherpessimplexvirustype1. Virol-ogy58:219-228.

24. Scotti, P. D. 1968. A new class of temperature conditional lethal mutants of bacteriophage T4D. Mutat. Res. 6:1-14.

25. Sheldrick, P., and N. Berthelot. 1974. Inverted repeti-tions in the chromosome of herpes simplex virus. Cold SpringHarborSymp. Quant.Biol. 39:667-681. 26. Siccardi, A. G., F. A.Ferrari,G.Mazza,and A.

Gal-izzi.1976.Identificationofcoreplicating chromosomal sectorsin Bacillus subtilisby nitrosoguanidine-induced comutation. J. Bacteriol. 125:755-761.

27. Subak-Sharpe,J.H., and M. C.Timbury. 1977. Ge-netics ofherpesviruses,p. 89-131. In H. Fraenkel-Con-rat and R. R.Wagner(ed.),Comprehensive virology, vol.9.PlenumPublishing Corp.,New York.

28. Takahashi,M., and K. Yamanishi.1974. Transforma-tionof hamsterembryo and human embryo cells by temperature-sensitivemutantsofherpessimplexvirus type 2.Virology61:306-311.

29.Timbury,M. C. 1971.Temperature-sensitivemutantsof herpessimplexvirus type 2. J.Gen. Virol. 13:373-376. 30.Timbury,M.C.,M.L.Hendricks,andP.A.Schaffer. 1976.Collaborativestudyoftemperature-sensitive mu-tants of herpes simplex virus type 2. J. Virol. 18:1139-1142.

31. Wadsworth, S.,R. J.Jacob,and B.Roizman. 1975. AnatomyofherpessimplexvirusDNA.II.Size, com-position andarrangement ofinvertedterminal repeti-tions. J. Virol. 15:1487-1497.

32.Westmoreland, D., and F. Rapp. 1976. Host range temperature-sensitivemutantsofherpessimplexvirus type 2. J. Virol. 18:92-102.

33. Zygraich, N.,and C.Huygelen.1973.Invivobehavior of atemperature-sensitive (ts)mutantsofherpesvirus hominis type 2. Arch.Gesamte Virusforsch.43:103-111. J. VIROL.

http://jvi.asm.org/