0022-538X/82/100088-10$02.00/0
Copyright ©1982,American Society for Microbiology
Herpes
Simplex
Virus Glycoprotein
gA/B:
Evidence that the
Infected Vero Cell Products Comap and Arise by Proteolysis
LENOREPEREIRA,'*DALEDONDERO,1 ANDBERNARDROIZMAN2Viral and Rickettsial Disease Laboratory, California Department of Health Services, Berkeley, California 94704,1 and TheMarjorieB. KovlerViral Oncology Laboratories, The University of Chicago, Chicago,
Illinois 606372
Received 15 March1982/Accepted4June1982
Werecently reported (Pereira et al., Proc. Natl. Acad.
Sci.
U.S.A. 78:5202-5206, 1981) that herpessimplex virus 1 and 2 glycoproteins, previously designated gA and gB, could not be differentiated by a bank of independently derivedtype-specific
and type-common monoclonal antibodies. We also reported that fromlysates of
infected Vero cells, all but one monoclonal antibody precipitated gA/Bglycoproteins which had faster electrophoretic mobility than the corresponding
infected
HEp-2 cell glycoproteins and a set of three smallpolypeptides which wedesignated g(A+B) reactive polypeptides 1, 2, and 3. Antibody H368, the single exception, failed to react with the gA/B glycoproteins or related antigens accumulating in infected Vero cells. In this paper, we report the
following
results. (i) The high-apparent-molecular-weight gA/B glycoproteins accumulating in in-fected HEp-2 cells were cleaved by a proteolytic enzyme contained in Vero cell lysates to yield more rapidly migrating proteins that were indistinguishable from authentic Vero cellgA/B
glycoproteins. Like its authentic counterpart, the cleavedgA/B glycoproteins failed
to react with H368 monoclonal antibody. In addition, the lysate cleaved HEp-2 cell gA/B glycoproteins into g(A+B) reactivepolypeptides
2and 3.(ii) The proteolytic activity contained in the uninfected celllysates was inhibited by
N-a-p-tosyl-L-lysine
chloromethyl ketone and isthere-fore
trypsin-like. (iii)
Pulse-chase experiments indicated that the cleavage ofgA/B
glycoproteins
occurredduring or soon after translation but that the accumulationofg(A+B) reactive
polypeptide
1 was a consequence ofa delayed processingevent. (iv) Analysis of herpes simplex virus 1 x herpes simplex virus 2
recombinants indicated that the determinants of
type-specific
immune reactivityand
electrophoretic mobility
ofgA/B
glycoproteins and g(A+B)polypeptides
mapneartheright terminus of herpes simplex virus 1 BamHI-G.
Herpes simplex virus 1
(HSV-1)
virions and membranes of infected cells wereinitially
re-portedtoform,
onelectrophoresis
indenaturing
polyacrylamide gels,
three bandscontaining
high-molecular-weight
glycosylated
proteins.
Theseglycoproteinsweredesignatedvirion
pro-teins VP7, VP7.5, and VP8 (4, 5, 16). Subse-quently,onthebasisofpulse-chase
studies,
theglycosylated
proteins
weredesignated
gC, gB,
and gA,
respectively
(15). However, recentre-ports from this and other laboratories indicate
that gA and gB
glycoproteins
arerelated,
inas-much asboth reactedwith
monospecific
antiseraprepared against each glycoprotein (2) or with
banks of
independently
derived monoclonal antibodies(10).Inthe courseof studieswith themonoclonal antibodies, it was notedthat
gA/B
glycoproteins
made in Vero cellsmigrated
sig-nificantly
morerapidly
thanthose made inHEp-2 cells. Furthermore, the infected Vero cell
lysates contained three relatively smallproteins which were precipitated by 23 of 24 monoclonal antibodies reactive with gA/B glycoprotein. The proteins were designated g(A+B) reactive poly-peptides. The g(A+B) polypeptides of HSV-1 (F) had apparent molecular weights of 40,000, 39,000,and29,000 and differedin electrophoret-ic mobility from those made by HSV-2 (G), whichwere 41,500,37,000, and 27,000in
appar-entmolecularweight. Theexception,
monoclo-nal antibody H368, reacted with
glycoprotein
gA/B of HSV-2made inHEp-2cells but not with
the gene product accumulating in Vero cells. These observations ledto thesuggestionthat the
differences in electrophoretic
mobility
of theglycoprotein
gA/B
made in Vero andHEp-2
cellscould reflect differencesin
glycosylation
orcleavage.
Inthispaper,wereportthat the
electrophoret-ic mobility of
glycoprotein
gA/B
verylikely
88
on November 10, 2019 by guest
http://jvi.asm.org/
reflects the cleavage of the glycoprotein gene
product in Vero cells, inasmuchas exposureof
glycoprotein gA/B made in HEp-2 cellsto
unin-fected Vero cell extracts resulted in proteolytic cleavage and yielded products indistinguishable from those accumulating in infected Vero cells.
Wealso reinforce the conclusion thatgA and gB
glycoproteins and theg(A+B) reactive
polypep-tides are related by showing that they comap
withinthe samerestricted region ofHSVDNA.
MATERIALSANDMETHODS
Virusesandcells.Isolation andpropertiesof HSV-1
(mP), HSV-1 (F), and HSV-2 (G) strains and the
derivation and properties of recombinants RH1G7,
RH1G8, RH1G13, RH1G44, and RH1G48 were
de-scribed previously (1, 3, 6, 7). Human epidermoid
carcinoma 2 (HEp-2) cells and African greenmonkey
kidney (Vero) cells were grown in Earle minimum essential medium supplemented with 10%o fetal calf serum. After infection with virus, cells were main-tained in medium containing1% serum.
Monodonalantibodies to HSV.Hybridomas
produc-ing monoclonal antibodies to HSVglycoproteinswere
derivedfromfusion ofBALB/c MOPC 21 NS-1
myelo-macellswithspleencells ofBALB/cmice immunized withHSV. Procedures usedfor selection and charac-terization of hybridomas were published elsewhere
(9a, 11). Properties of H233 and H368 monoclonal
antibodies toglycoproteins gAandgBwerepreviously
described(10).
Preparation of radiolabeled infected cell extracts.
HEp-2 or Verocells, as indicated in the text, were infected with 10 PFU per cell and labeled with [355]methionine(50,uCi/ml, >400Ci/mmol;New
En-gland Nuclear Corp., Boston, Mass.) at 8 to 18 h
postinfection. For pulse-chase experiments, infected cells wereincubated in medium without methionine at 5to 6 hpostinfection. The cells were labeled for 15 min with [35S]methionine (100 ,Ci/ml) (pulse). Duplicate cultures replenished with medium containing unla-beledmethionine were incubated for an additional 3 h (chase). Lysates of infected cells were prepared by
incubating the cells on ice for 30 min in
phosphate-buffered saline containing 1% Nonidet P-40 and 1%
sodiumdeoxycholate.Theantigenlysateswere
centri-fugedfor 60 min at 25,000 rpm and4°Cin a Beckman
SW27.1 rotor.
Immunoprecipitation andpolyacrylamide gel electro-phoresis. Radiolabeled antigens were mixed with mouseascites fluids (25 to 50p,l)containing monoclo-nalantibodies. After incubation at room temperature for 60min, rabbit anti-mouseimmunoglobulin G (Miles Laboratories, Inc., Elkhart,Ind.),followedby protein
A-Sepharose (Sigma Chemical Co., St. Louis, Mo.),
were added to precipitate the antigen-antibody com-plexes. Immune precipitates adsorbed to protein A-Sepharose were washed repeatedly in cold
phosphate-bufferedsaline containing0.1% Nonidet P-40 and 1%
sodiumdeoxycholate.
Denaturedsamples weresubjected to electrophore-sis in 9.25% polyacrylamide gels cross-linked with N,N-diallyltartardiamide andcontaining sodium dode-cyl sulfate. Procedures for solubilization of the
pro-teins, electrophoresis, and autoradiography were
described previously (9). Reagents used for
poly-acrylamide gel electrophoresis were purchased from Bio-Rad Laboratories, Richmond, Calif.
Incubation ofinfected HEp-2 cell lysates with unin-fectedVerocellextracts.Uninfected Vero cell extracts were prepared from confluent cell sheets. The cells were harvested and incubated in phosphate-buffered saline containing 1% Nonidet P-40 and 1% sodium
deoxycholate at 4°C for 30 min. Extracts were centri-fuged as describedabove. Radiolabeledinfected HEp-2 cell lysates weremixed with equal volumes(usually 25,ul)of uninfected Verocell extracts and allowed to reactfor 60 min at room temperature. In some experi-ments,
10-'
MN-a-p-tosyl-L-lysine chloromethyl ke-tone (TLCK), a trypsin inhibitor (14), orL-1-tosyl-amide-2-phenylethyl chloromethyl ketone (TPCK), a
chymotrypsin inhibitor (13), were added to HEp-2 infected cell lysates before addition of the uninfected Vero cell extracts. The final concentration of either TLCK or TPCK was10'M.To obtainpartial cleav-age products ofgA/B glycoprotein, uninfected Vero cell extracts wereallowed to react with infected HEp-2 celllysates.MonoclonalantibodyH233(25 ,u)was added at different times to stop thereaction,followed
by protein A-Sepharose to precipitate the immune
complexes. TLCK and TPCK were purchased from
SigmaChemical Co.
Incubation of infected Vero cells with proteolytic inhibitors. Infected Vero cells were incubated with
106 M TLCK or TPCK in the medium at 7 h
postinfectionandlabeled with[35S]methionineat8 to
18 hpostinfection in the presenceofproteolytic inhibi-tors.
RESULTS
Electrophoretic
mobility
ofproteinsprecipitat-edfrominfectedHEp-2and Vero cellsby
mono-clonal antibody to glycoproteins gA/B after a
pulseandaftera chase.One unresolved question is whether theeventunderlying the difference in electrophoretic mobility of gA/B glycoprotein made ininfected HEp-2 and Vero cells occurred
during
orimmediately
after translationorduring
delayed processing of
thenewly translatedpro-tein. To differentiate between thesetwo
possibil-ities,
wecompared theglycoprotein gA/Breac-tive antigens, labeled as described in Materials andMethods, in infected Vero and HEp-2 cells aftera 15-minpulse and aftera chase.
Autora-diograms
of the immuneprecipitated
polypep-tides
electrophoretically
separated indenaturing
polyacrylamide gels are shown in Fig. 1. The
results showed thefollowing.
(i) Theeventresponsible forthedifference in electrophoretic
mobility
ofgA/B
glycoproteins made in infected Vero and HEp-2 cells occursduringor soonafter translation, inasmuchasthe
gA/B glycoproteins labeled during the pulse
re-flectedthedifferenceinelectrophoretic mobility reportedpreviously (10).
(ii)Theg(A+B) reactivepolypeptides2and 3
on November 10, 2019 by guest
http://jvi.asm.org/
90 PEREIRA, DONDERO, AND ROIZMAN
HSV-" sV G
~~~t
r..P0
PA-electrophoretic
mobility
ofpolypeptide 3 afterthe chase. Therefore,theelectrophoretic
mobil-ity of
glycoproteins gA/B
and theaccumulation
ofpolypeptides 2and 3 reflecta
translational
orrapidpost-translational event.Thepresenceand
mobility of polypeptides1and 3 as well as of the
fully processed forms of
gA/B
glycoproteins
reflect slow
post-translational
events.Cleavage ofglycoprotein
gA/B
made in HEp-2cells
byproteolytic enzymes containedinextracts;HSV-1VF C
I..-".TP K ..- 6:G , - !S"I
J0
3 3
. ih :~Q_:
_
ZgAM0go !o s
.W
twlm-FIG. 1. Autoradiographic imagesof
electrophoret-ically separated [35S]methionine-labeledpolypeptides
immunoprecipitated withmonoclonalantibody H233.
HEp-2 and Vero cells wereinfected with HSV-1 (F)
and HSV-2(G)andradiolabeledduringa15-minpulse
at6hpostinfectionasdescribed in thetext.1,2,and3,
g(A+B) reactive polypeptides; gA and gB,
more-slowlymigratingforms of theglycoproteins; P, pulse;
P-C, pulse-chase; Ag, labeled antigens present in
infected-cell lysates available for reaction with anti-body.
.3
3
am -.:----
Nom
are also
produced
during or soon aftertransla-tion, inasmuch as both polypeptides 2 and 3 were present in immune
precipitates
obtainedwith
lysates
ofpulse-labeled
infected cells. Itshould be noted that, although the HSV-1
g(A+B)reactive
polypeptides
were toofainttoappear in theprints of the autoradiograms,
they
were uniformly present in the
lysates
of [image:3.496.55.242.78.427.2]pulse-labeled,
HSV-1-infected
Verocells.The HSV-1 and HSV-2g(A+B)reactive polypeptide1 was detectedduringthechasebutnotafter thepulse.
In addition, we consistently noted a change inFIG. 2. Autoradiographicimages of
electrophoret-icallyseparated
[35S]methionine-labeled
polypeptidesimmunoprecipitated with H233 monoclonal antibody
from HSV-1(F)-infectedHEp-2 cell
lysates.
Uninfect-ed Vero cellextracts weremixed withinfectedHEp-2cell
lysates
andincubated for 0, 10,or60min.TLCKorTPCKwasaddedtoinfectedHEp-2cell
lysates
to a finalconcentration of10-'
MbeforemixingwithVerocell extracts.
on November 10, 2019 by guest
http://jvi.asm.org/
[image:3.496.277.421.192.587.2]Ag
AG 0 ^0e K- .
SV-2lG'i 0 10 6a-'
-a
-S q*gB
gA *X gB9A
.9m.
extracts (Fig. 2 and3). The cleavage was
com-plete after 60 min. In addition to the more
rapidly migrating forms of glycoproteins gA/B,
the autoradiograms revealed the presence of
g(A+B) reactive polypeptides2and3butnotof
1.
To determine the nature of the proteolytic
activity responsible for the cleavage, the
mix-tures weretreated with TLCK and TPCK. The
cleavage was inhibited by TLCK but not by TPCK (Fig.2and3),suggesting thatthe proteo-lytic activity is trypsin-like withrespect to
ami-noacid specificityat the cleavagesite.
Further-more, TLCK prevented cleavage not only in
vitro but also in infected cells. Addition of TLCK1hbefore the addition of[35S]methionine
to infected Vero cells precluded cleavage of glycoprotein gA/B (Fig. 4).
In a previous report (10), we showed that, whereas 23 of 24 independently derived
mono-clonal antibodies reacted with gA/B
glycopro-3 3
2 2
..4--4-a SP .olImd
a
...a
[image:4.496.75.219.73.454.2]...
FIG. 3. Autoradiographic images of
electrophoret-ically separated
[35S]methionine-labeled
polypeptidesimmunoprecipitated with H233 monoclonal antibody
from HSV-2 (G)-infected HEp-2 celllysatestreated as described in the legendtoFig. 2.
ofuninfected Vero cells. One
hypothesis
whichcould explain the difference in electrophoretic mobility of Vero and HEp-2 cell gA/B glycopro-teins is that the nascent polypeptide is rapidly
cleaved by a host cell enzyme. To test this
hypothesis, equal portions of infected HEp-2 cell lysates were mixed with uninfected Vero
cellextracts preparedasdescribedin Materials
andMethods. Atintervals aftermixing, thegA/
B glycoproteins were precipitated with
mono-clonalantibodytotheglycoproteinsand subject-ed to electrophoresis in denaturing polyacryl-amide gels. Limited cleavage of
gA/B
glycoprotein made in HEp-2 cellswas detected
10 min after exposure to uninfected Vero cell
Wm
_.4:
:-.,-40 ft US -., 6W
-~ id
FIG. 4. Autoradiographic images of electrophoreti-cally separated [35S]methionine-labeled polypeptides
immunoprecipitated with H233 monoclonal antibody
from HSV-1 (F)- and HSV-2 (G)-infected Vero cells treated with10-6 MTLCK or TPCK present during infection before preparation of infected Vero cell
lysates.
9 illmqB
9A
on November 10, 2019 by guest
http://jvi.asm.org/
[image:4.496.254.444.309.606.2]'
--*
Aowm
o 3
_i1
.9B
_40.
9A4 3 3
_24 2
_
FIG. 5. Autoradiographic images of electrophoretically separated
[35S]methionine-labeled
polypeptides
immunoprecipitatedwith H233 and H368 monoclonalantibodies from HSV-1(F)-and HSV-2
(G)-infected
HEp-2and Verocelllysates andinfectedHEp-2celllysates mixed withextractsofuninfected Verocells.
teins
accumulating
in both Vero andHEp-2
celllysates,one,H368, reacted with
gA/B
glycopro-teins
accumulating
in HEp-2 but not in Veroinfectedcells. GlycoproteinsgA/B madeby
ex-posing HEp-2 cell lysatestouninfectedVerocell
extractcomigrated withtheauthentic
gA/B
gly-coproteins
accumulating
in infected Vero cells(Fig. 5). Thecleavedproduct, like the authentic product, did
niot
reactwith the H368antibodies,
suggesting thatthese antibodies aredirected to an antigenic determinant site contained within
thedegraded orlostportion ofthemolecule.
FinemappingofglycoproteinsgA/Bandg(A+B)
reactivepolypeptides.On the basis of
electropho-retic mobilities of the glycoproteins specified by HSV-1 x HSV-2 recombinants, the glycopro-teins gA and gB weremappedina
region
oftheHSV genome bounded by coordinates 0.30 to
0.42 map units (12). More recent studies
nar-rowed the coordinates of the gA/B genes to
DNA BamHI fragment G (8). This fragment
maps between 0.337 and 0.388 map units. The
experiments described below and concerned
with moreprecisemappingofgA/Bgenes were
donefor two reasons. First,the narrowest map
coordinates available do not exclude the
possi-bility that the glycoproteins gA/B are specified by independent genes. The second reason
con-cerned theg(A+B) reactivepolypeptides.
Previ-ous studies have shown that g(A+B) reactive
HSV-1 polypeptides differ in electrophoretic
mobility from HSV-2 and that all monoclonal
antibodies which precipitated glycoproteins gA/
Bfrom infected Vero cell lysatesalso
precipitat-edtheg(A+B) reactive peptides(10).
Unambigu-ous demonstration that glycoprotein gA/B is a
J.VIROL.
on November 10, 2019 by guest
http://jvi.asm.org/
[image:5.496.113.400.72.430.2]I I I I
1.0
0.40 0.45
RHIG7
RHIG8 2
RHiG13 2
RHIG44 2
RHIG48 2
Sall
AU
HSV-1 BamHlI T IJI Q I U I E'lEG I V R IW IC'
I1
DH0
Kpnl N1'5 M1 5 I N ' P I V I A I X I B
Hpal I I V I B I H
Bglll G I i I 0 1 C
Hpal A I H 0D
Kpnl HI N I P J I M ITI C
FIG. 6. Restriction endonuclease maps of HSV-1 x HSV-2 recombinants RH1G7, RH1G8, RHlG13, RH1G44, and RH1G48 produced by markerrescueofHSV-1 (mP) tsHAl with HSV-2 (G).
precursor of theg(A+B) reactive polypeptides
requires evidence thatthe DNAsequences
spec-ifying the glycoprotein gA/B polypeptides and
g(A+B) reactive peptides comap. The mapping
of the DNA sequences specifying gA/B and
g(A+B) took advantage of the availability ofa
set of suitable HSV-1 x HSV-2 recombinants
produced by marker rescue of HSV-1 (mP)
tsHAl with HSV-2 (G) DNA and monoclonal
antibodies whichreactwith either HSV-2
glyco-proteins gA/B (e.g., H368) orwith both HSV-1
and HSV-2glycoproteins gA/B (e.g.,H233).
Figure 6 shows themapsofcrossoversites in
the DNAs of recombinants RH1G7, RH1G8,
RH1G13, RH1G44, and RH1G48 described in
detailbyConleyetal. (1). Themapof the DNA
of each recombinant is shown within the
con-fines ofa double line representing HSV-1 and
HSV-2 DNAs(designated1and2, respectively).
The heavy line represents the DNA of the
recombinant.Thediagonal lines linking the lines
representing HSV-1 and HSV-2 DNAsrepresent
the approximatecrossover sites. Thecrossover
sites are based on the restriction endonuclease
maps for HSV-1 (mP) and HSV-2 (G) DNAs
shown in the lowerportion of the figures.
The coordinates of the crossover sites differ
slightly from those published by Conley et al.
(1), largely because the availability of cloned
DNAfragments permitteda moreprecise
align-mentof the variousrestrictionenzymemaps.As
previously established, the HSV-2 sequences
substituting HSV-1 sequences inRH1G7 have,
astheir maximumand minimum leftboundaries,
the restriction enzyme cleavage sites HSV-1
BamHI-T-J' and HSV-2 BglII-G-J, respectively;
the maximumand minimumright boundariesare
the HSV-2 KpnI-G-D and HSV-1 HpaI-B-H
restriction enzyme cleavage sites, respectively.
The left maximum and minimum boundaries of
theHSV-2sequencesinRH1GH8aredefinedby
HSV-1KpnI-N-P and HSV-2 BglII-J-Ocleavage
sites, respectively, whereas the right maximum
andminimum boundariesare identical tothose
ofRH1G7 DNA. The maximum and minimum
leftboundaries ofHSV-2sequencesinRH1G13
are HSV-1 HpaI-I-V and -V-B cleavage sites,
respectively; the maximum and minimum right
boundaries are HSV-2 BglII-O-C and HSV-1
BamHI-R-W cleavage sites, respectively. In
recombinant RH1G44, the maximum and
mini-mum left HSV-2 boundaries are contained
be-ab
El-0
Ul
Recombinants: '.0
0.30 I 0.35I
Bglll
II InI
M
HSV-2
II9I
baac us ca
I
I
on November 10, 2019 by guest
http://jvi.asm.org/
[image:6.496.56.450.51.360.2]4. a U.
FIG. 7. Autoradiographic imagesof electrophoret-ically separated [35S]methionine-labeled polypeptides immunoprecipitated with cross-reacting H233
mono-clonal antibody fromlysates ofHEp-2and Verocells infected with recombinantsand parent strains.
tween HSV-1 KpnI-J-M cleavage sites, whereas
the right maximum and minimum boundariesare
contained between the HSV-1 BamHI-R-W and
HSV-1 KpnI-P-V cleavagesites. Inthe
recombi-nant RH1G48, the left maximum and minimum
boundariesof the HSV-2sequences areidentical
to the corresponding boundaries of RH1G8
DNA; the right maximum and minimum
bound-ariesarecontained between HSV-1 BamHI-G-V
and HSV-1 BamHI-V-R cleavage sites. Also
shown is the map location of HSV-1 Sall
frag-mentAO, whichwasshowntorescuethe tsHAl
mutation in marker rescue tests (1).
Figures 7 and 8 show the autoradiographic
images of the [35S]methionine-labeled
glycopro-teins gA/B precipitated with monoclonal
anti-bodies H233 and H368, respectively, from
ly-sates of Vero and HEp-2 cells infected with parent and recombinantvirus strains.
Monoclo-nalantibody H233, aspreviously reported,
pre-cipitated both HSV-1 and HSV-2 glycoproteins gA/B from Vero and HEp-2 cell lysates and g(A+ B) reactive polypeptides from infected Vero cell lysates. H368 monoclonal antibody
reacted onlywith the HSV-2 glycoproteins gA/B
produced in HEp-2 cells. On the basis of electro-phoretic mobility and reactivity with
monoclo-nalantibody, the glycoproteins gA/B andg(A+B)
reactive polypeptides made by RH1G8 resemble
those of the HSV-1 (mP) parent, whereas the
glycoproteins gA/B and g(A+B) reactive
[image:7.496.257.446.73.274.2]poly-peptides of all other recombinants resemble
FIG. 8. Autoradiographic images of electrophoret-ically separated [35S]methionine-labeled polypeptides immunoprecipitated with type2-specific H368
mono-clonal antibody fromlysates of HEp-2 and Vero cells infected with recombinants and parentstrains.
those ofthe HSV-2 parent. Resultsof
neutraliza-tiontestswith monoclonal antibodiesH233 and
H368 and the RHlG seriesofrecombinants are
shownin Table 1. The results are in accordance
withimmunoprecipitation reactions, inthat type
2-specific monoclonal antibody H368
neutral-ized the HSV-2 (G) parent and recombinants RH1G7, RH1G13, RH1G44, and RH1G48 but
failedtoneutralize the HSV-1parent and
recom-binant RH1G8.
Two features ofthe results shown in Fig. 6
through 8 areofsignificance. First,theg(A+B) reactivepolypeptides generatedin infected Vero
cellsbyHSV-1 (mP)differ from thosegenerated
by HSV-1 (F), reinforcing the conclusion that
TABLE 1. Plaquereductiontests with H233and H368monoclonalantibodies, intertypic
recombinants,andparental strains
PFU of virussurviving/totalPFU of
Strain virus(%)a
H233 H368
RH1G7 0/110(0) 38/110(34.5)
RH1G8 0/235(0) 226/235(96.2)
RH1G13 0/233(0) 83/233(35.6)
RH1G44 0/313(0) 83/313(26.5)
RH1G48 0/200(0) 37/200(18.5)
HSV-1 (mP) 0/121(0) 115/121 (95.0)
HSV-2(G) 0/228(0) 49/228(21.5)
aAverage results ofduplicatetests.
on November 10, 2019 by guest
http://jvi.asm.org/
the g(A+B) products are virus specific. The
secondaspect concerns the physical map
loca-tion of the sequences specifying the HSV-2
antigenic determinants in the RHlG series of
recombinants. The RHlG series of
recombi-nantswereselected by markerrescueof HSV-1
(mP) tsHAl with HSV-2 (G) DNA, and those
shown in Fig. 7 and 8 specify HSV-2-infected
cell polypeptide 8. Furthermore, the
tempera-ture-sensitive mutation was rescuedby a small
cloned fragment (SalI-AO) which selects from
infected cells mRNA capable of translating in
vitro the infected cell polypeptide 8. The
tem-perature-sensitive mutation is,therefore, closely
linkedtothegene specifying polypeptide 8. The
results shown in Fig. 6 through 8 indicate that
the sequences specifying the antigenic
determi-nant sites for gA/B and g(A+B) are in close
proximity tothe gene specifying polypeptide 8,
inasmuch asthe HSV-2 antigenic determinants
areexpressed by all of the recombinants except
RH1G8. Since RH1G8 and RH1G48overlap, the
mostprobable sequenceencodingtheantigenic
determinant sites is withinaDNAregion
bound-edby HSV-1 KpnI-N-P and the HSV-2
BglII-J-O cleavage sites. This region contains the left
crossovers of both RH1G8 and RH1G48. Since the sequences specifying the gA/B antigenic
determinantsitesaretothe leftof those
specify-ing polypeptide 8, it is likely that theleft
cross-overof RH1G8 istotheright of that ofRH1G48.
DISCUSSION
Derivation and structure of the g(A+B)
reac-tive polypeptides. The results presented in this
report indicate that polypeptides
indistinguish-able from authentic g(A+B) reactive
polypep-tides 2and 3canariseby cleavage of
glycopro-teins gA/B made in HEp-2 cells by uninfected
Vero cellextractsandsuggestthat the authentic
g(A+B) polypeptides arise bya similar
mecha-nism. The g(A+B) reactive polypeptide 1 and
themorerapidly migrating forms of polypeptide
3areproducts of processing takingplace
signifi-cantly later. All three polypeptides retain a
major portion of antigenic determinant sites of
the gA/B glycoproteins, even though they are
considerablysmallerinmolecularweight. These
conclusionsarebasedonthefollowing
observa-tions. (i) The g(A+B) reactive polypeptides
shareantigenic determinant sites with
glycopro-teins gA/B. (ii) g(A+B) reactive polypeptides
are not host gene products, inasmuch as their
electrophoreticmobilityindenaturing
polyacryl-amide gels is genetically determined by the
virus, and thephenotypic properties of g(A+B)
reactivepolypeptides and of gA/B glycoproteins
comap within a narrow region of the HSV
genome. (iii) The g(A+B) reactivepolypeptides
contain asignificant fraction of the major
anti-genic determinants of glycoproteins gA/B,even
though they represent somewhatless than 30o
ofthe apparent molecular weight of the gA/B
glycoproteins. This conclusion is based on the
observation that all g(A+B) reactive polypep-tideswereprecipitatedby23of24 independent-ly derived monoclonal antibodies to glycopro-teins gA/B. These monoclonal antibodies comprised11whichneutralized the virus and 12 which didnot.Theexceptionwas amonoclonal antibody (H368) which immunoprecipitated HSV-2 (G) glycoproteins gA/B made in HEp-2 butnotthose made ininfected Vero cells. If the frequency ofemergenceofmonoclonal antibod-ies per fractional length ofthe polypeptide
re-flected the density of antigenic determinant sites, it could be concluded thatg(A+B)reactive polypeptide 1 from both HSV-1 and HSV-2
(apparent molecular weights,29,000and27,000,
respectively) contained a major portion ofthe
type-common antigenic determinant sites of the
parentgA/B glycoprotein molecule.
Location of the sequences specifying antigenic
determinant sites of
gA/B
glycoproteins andg(A+B) reactive antigens on the physical mapof
the HSV genome. Theavailability of the RHlG
series of HSV-1 x HSV-2 recombinants (1)
permitted
us to map the gene location of thesequences
specifying
the antigenic determinantsites ofglycoproteins gA/B within asmall
por-tion of the regiontowhich the gA and gBgenes werepreviouslyassigned. Specifically, the
earli-eststudies
assigned
theinfected
cellpolypeptide11 (9) and gA and gB genes (12) to0.3 to 0.42
mapunits. Inmore recentstudies, itwasshown that the HSV-1-infected cell polypeptide 11
mapped atthe right end of HSV-1 BamHI
frag-ment G, immediately to the left of the gene
specifying
polypeptide 8. Because of thevari-ability in electrophoretic patterns of glycopro-teinsgA/B reported earlier (12), itwas notclear that thepolypeptide 11 actually correspondsto
glycoproteins gA/B,even though the genes
spec-ifying
these polypeptides have consistentlymapped in the samelocation. In thisreport, we
show that the
antigenic
determinants of glyco-proteins gA/B, as identified byimmunoprecipi-tation with both type-common (H233) and
HSV-2-specific (H368) monoclonal antibodies, map
within BamHI-G to the right of the KpnI-N-P
cleavage
site buttotheleftof Sall fragment AO(Fig. 6).
Cleavageofglycoproteins
gA/B.
Thesignificant
observation presented in thisreport is that
gly-coproteins
gA/B
made in infectedHEp-2
cellswerecleavedbyuninfected Vero cell extracts to
yieldtwo setsofproducts.These were
polypep-tides which comigrated with authentic
gA/B
on November 10, 2019 by guest
http://jvi.asm.org/
glycoproteins made in infected Vero cells and
g(A+B)reactive polypeptides which comigrated
with authentic g(A+B) reactive polypeptides
seen in infected Vero but not in HEp-2 cell
lysates. Our results suggest that both products are generated by cleavage of the gA/B precursor
by a trypsin-like proteolytic enzyme. Several
aspectsofour results are of particular interest,
specifically thefollowing.
(i)Wedidnot observe complete conversion of
the gA/B glycoproteins made in infected HEp-2
cells into g(A+B) reactive polypeptides by
cleavage withthe enzyme contained in the
unin-fectedVero celllysate.Similarly, both gA/B and g(A+B) reactive polypeptides accumulate in
in-fectedVero cells even after a long chase. These
observations suggest that only a portion of the
gA/Bprecursoris cleaved to yield g(A+B)
reac-tive polypeptides, and therefore, the cleavage
sites which generate the g(A+B) reactive
poly-peptides must be inaccessible in some of the
precursormolecules. Nothing is known
regard-ing the fate ofthe peptides cleaved from gA/B precursor to generate the authentic infected
Vero cell gA/B glycoproteins and the g(A+B)
reactive polypeptides. Inasmuch as detection of
such putative peptides requires antibody and
since all but one monoclonal antibody were
found
to react with g(A+B) reactivepolypep-tides, wehave no immunological reagents with
which to determine whether the cleaved
pep-tides
areconservedordegraded ininfectedVerocells.
(ii) The cleavage of
gA/B
precursor to yield the g(A+B) reactive polypeptides 2 and 3 andthegA/Bmoleculeswhichaccumulate in
infect-ed Verocellsmust occurduringor soonafterthe
synthesis of the precursor, inasmuch as the products of the reaction were present in the
lysates
of cellsharvestedimmediately aftera15-min pulse. In contrast, the g(A+B) reactive polypeptide 1 was not detected until after a
chase. The proteolytic event which generates
thispolypeptideappears to occurlaterin the life
of
thegA/B polypeptide.
The nature of theenzymeinvolved is not known.
(iii) The demonstration that the difference in
electrophoretic mobilities is due to cleavage of the protein rather than to differences in the
extentof glycosylation ofglycoproteinsgA/B in
infectedVero and HEp-2 cellsis of considerable significance. Itindicatesthatdifferences in
elec-trophoretic
mobilities
of HSV glycoproteins commonly observed when viruses aregrownin different cell lines must be interpreted with caution, inasmuchasthey may be due topartial proteolysis or to both cleavage and differentialglycosylation.
(iv) Because the size ofthe virus yield from
infected Vero cells is as high as or higher than
that obtained from infected HEp-2 cells, the
function ofgA/B glycoproteins may not be af-fected bytheproteolysis involved in generation of the authentic Vero infected cellgA/B glyco-proteins. Whether g(A+B) reactive polypep-tidesare capable of expressinganyofthe func-tions of gA/B glycoproteins is currently unknown.
ACKNOWLEDGMENTS
WethankRichard Emmons for support of the studies done atthe California Department of Health.
The studies conducted at theUniversity ofChicagowere aidedbyPublicHealthService grants CA 19264 and CA 08494 from the National Cancer Institute.
LITERATURECITED
1. Conley,A.J.,D.M.Knipe, P. C. Jones, andB. Roizman. 1981. Molecular genetics of herpes simplex virus. VII. Characterization ofatemperature-sensitive mutant pro-duced by in vitro mutagenesis and defective in DNA synthesis and accumulation of apolypeptides. J. Virol. 37:191-206.
2. Eberle, R.,andR.J. Courtney.1980. gA andgB glycopro-teins of herpes simplex virus type1: twoforms ofasingle polypeptide. J. Virol. 36:665-675.
3. Ejercito, P.,E.Kieff, and B. Roizman. 1968. Characteriza-tion of herpes simplex virus strains differing in their effect onsocial behavior of infected cells. J. Gen.Virol. 3:357-364.
4. Heine, J. W.,R. W.Honess,E.Cassai,andB.Roizman. 1974.Proteins specified by herpes simplex virus.XII. The virion polypeptides of type 1 strains. J. Virol. 14:640-651. 5. Heine, J. W.,P.G.Spear,and B.Roizman.1972. Proteins specified by herpes simplex virus. VI. Viralproteinsin the plasma membrane. J. Virol. 9:431-439.
6. Hoggan, M., and B. Roizman. 1959. The isolation and properties ofavariant ofherpes simplex producing multi-nucleatedgiantcells inmonolayercultures in the presence ofantibody.Am.J.Hyg. 70:208-219.
7. Kieff, E. D., S.L.Bachenheimer,andB.Roizman. 1971. Size, composition,andstructureof thedeoxyribonucleic acid ofherpes simplexvirussubtypes 1 and 2. J. Virol. 8:125-132.
8. Little, S. P., J. T. Jofre, R. J. Courtney, and P. A. Schaffer.1981. A virionassociatedglycoproteinessential forinfectivity ofherpes simplex virus type 1. Virology 115:149-159.
9. Morse, L.S.,L.Pereira, B.Roizman,andP.A.Schaffer. 1978. Anatomy ofherpessimplexvirus(HSV)DNA. X. Mappingof viral genes byanalysisofpolypeptidesand functionsspecified byHSV-1 xHSV-2recombinants. J. Virol.26:389-410.
9a.Pereira,L.1982.Monoclonal antibodiestoherpessimplex viruses1and 2, p. 119-138. In J.Hurrell(ed.), Monoclo-nalhybridoma antibodies: techniques and applications. CRCPress,BocaRaton, Fla.
10. Pereira, L., D. Dondero, B. Norrild, and B. Roizman. 1981.Differentialimmunologicandelectrophoretic prop-erties ofglycoproteins gAandgBof HSV-2producedin HEp-2 and Veroacells. Proc. Natl. Acad. Sci. U.S.A. 78:5202-5206.
11. Pereira, L.,T.Klassen,andJ.R.Baringer.1980. Type-commonandtype-specificmonoclonalantibodytoherpes simplexvirus type1.Infect. Immun. 29:724-732. 12. Ruyechan, W. T.,L. S. Morse, D. M. Knipe, and B.
Roizman. 1979. Molecular genetics of herpes simplex virus.II. Mapping ofthemajorviralglycoproteinsand of thegeneticlocispecifyingthe social behavior of infected cells. J. Virol. 29:677-697.
13. Schoelimann,G.,and E.Shaw.1963.Direct evidence for
on November 10, 2019 by guest
http://jvi.asm.org/
thepresenceofhistidine in theactive centerof chymo-trypsin. Biochemistry 2:252-255.
14. Shaw, E.,M.Mares-Guia,and W.Cohen. 1965.Evidence
foranactive-center histidine intrypsin throughuseofa
specific reagent,
1-chloro-3-tosylamido-7-amino-2-hep-tanone,thechloromethyl ketone derived from N-a-tosyl-L-lysine.Biochemistry4:2219-2224.
15. Spear, P.G.1976.Membrane proteins specified by herpes simplex viruses. I. Identification of four glycoprotein
precursors andtheirproducts intype1-infected cells. J. Virol. 17:991-1008.
16. Spear,P.G., and B.Roizman. 1972.Proteinsspecified by herpessimplex virus. V.Purification and structural
pro-teins of theherpesvirion.J.Virol. 9:143-159.