Download The Two Major Structural Phosphoproteins (pp65 and ppl50) of

J. gen. Virol. (1987), 68, 1327-1337.
Printed in Great Britain
Key words: CMV, human/phosphoprotein/monospecific
1327
antisera
The Two Major Structural Phosphoproteins (pp65 and ppl50) of Human
Cytomegalovirus and Their Antigenic Properties
By G E R H A R D J A H N , * B I R G I T - C H R I S T I N E S C H O L L ,
B E R N D T R A U P E AND B E R N H A R D F L E C K E N S T E I N
Institut jfir Klinische Virologie der Universitiit Erlangen-Niirnberg, Loschgestrasse 7,
D-8520 Erlangen, F.R.G.
(Accepted 15 January 1987)
SUMMARY
Human cytomegalovirus (HCMV) purified from cell culture contains two dominant
structural phosphoproteins with apparent molecular weights of 65000 and 150000,
designated as pp65 and pp 150 respectively. The humoral immune response of infected
individuals against pp65 is relatively weak and is not always detectable by Western blot
analyses. This report shows that recent clinical isolates of HCMV do not necessarily
have pp65 as a prominent constituent, suggesting that the low immune reaction is due
to variable expression of the pp65 in natural infections. However, the HCMV strains
tested in this study produced the large structural phosphoprotein (pp 150) in about equal
amounts. The pp150 is remarkably immunogenic, if compared with all other virion
constituents; serum pools and individual sera from HCMV-infected patients
recognized this particular protein intensively in immunoblot assays. Thus, phosphoprotein pp 150 seems to be the primary polypeptide candidate for expression cloning in
order to develop reagents for novel ways of HCMV diagnosis.
INTRODUCTION
Human cytomegalovirus (HCMV) is an important pathogenic agent in newborns after
prenatal infection and in immunocompromised individuals. Serological diagnosis of HCMV
needs improvement, since the commonly used ELISA detection systems for virus-specific IgG
and IgM antibodies are based on poorly defined viral antigens. Purified HCMV particles have
at least 25 structural proteins, of which several are modified by glycosylation or phosphorylation
(Stinski, 1976; Kim et al., 1976; Pereira et al., 1982b, 1984; Gibson, 1983 ; Nowak et al., 1984a,
b; Britt, 1984; Britt & Auger, 1985, 1986; Farrar & Oram, 1984). At least four phosphorylated
structural proteins are found in HCMV (Gibson, 1983; Nowak et al., 1984a, b). One of these
proteins has an apparent molecular weight of 65000 (65K) and is designated as pp65 (Nowak
et al., 1984a) or lower matrix protein (Gibson, 1983). It is the predominant protein in
virus particles purified from lytically infected human fibroblast cultures. A second major
phosphorylated matrix protein is of 150K mol. wt. (ppl50). Two minor phosphoproteins of
HCMV have molecular weights of about 71K and 86K (Gibson, 1983; Nowak et al., 1984a).
Non-infectious enveloped particles (NIEPs) have a similar pattern of structural phosphoproteins as infectious virions (Gibson & Irmiere, 1984). However, dense bodies, a second type of
defective particles, largely consist of the major phosphoprotein pp65.
A number of studies have been done focusing on the antigenic properties of proteins from
lytically infected cells using human sera and monoclonal mouse antibodies (Schmitz et al., 1980;
Pereira et al., 1982a, 1983 ; Cremer et al., 1985; Zaia et al., 1986). Although initial studies have
also been done to determine human antibody profiles against proteins of purified virions (Kim et
al., 1983; Landini et al., 1985, 1986), far more work will be required using better defined virion
proteins. This report demonstrates that the 150K structural phosphoprotein is outstanding
among all virion constituents in eliciting a humoral immune response, whereas in contrast the
antibody reaction against the phosphoprotein pp65 is highly variable.
0000-7396 © 1987 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
1328
G. JAHN AND OTHERS
METHODS
Viruses, cell culture and virion purification. The wild-type HCMV strains were isolated from urine of infected
patients. Laboratory strain ADI69 was provided by U. Krech, St. Gallen, Switzerland. For the propagation of
viruses, human foreskin fibroblasts (HFF) were used following standard procedures. Purification of HCMV
particles from the supernatant of infected cells was done by centrifugal separation in glycerol-tartrate gradients
essentially as described by Talbot & Almeida (1977) and modified by Irmiere & Gibson (1983).
Human and rabbit antisera. Human hyperimmunesera, pooled and concentrated from about 1000 seropositive
individuals, were provided by Biotest, Frankfurt, F.R.G. The individual HCMV antibody-positive sera were
obtained from the diagnostic laboratories of this Institute. Antisera to HCMV proteins were raised in New
Zealand white rabbits. The antigens for the injections were taken from gradient-purified virions, NIEPs, or dense
bodies. The virion proteins were separated in preparative 8"5~opolyacrylamide gels and stained with Coomassie
Brilliant Blue. Appropriate bands were excised and pulverized, and the proteins were eluted with 0.1 raNH4HCO3 pH 9.5 and 0.1 ~ SDS. Each injection was estimated to contain approximately 500 ~g of protein from
NIEPs and dense body fractions, about 200 p.g of 65K protein and approximately 20 ~g of 150K protein. The
initial doses were injected intracutaneously and subcutaneously at six to eight sites with an equal volume of
complete Freund's adjuvant (Sigma). The booster injections were given intramuscularly at monthly intervals using
Freund's incomplete adjuvant. The rabbit preimmune and immune sera were tested for recognition of HCMV
proteins by ELISA and immunoblotting.
Radioactive labelling ofviralproteins. Radiolabelling was started when viral cytopathic effect was about 60 ~ (at 4
days post-infection) and continued for another 4 days until the virus was purified from culture supernatant.
[35S]Methionine(50 ~tCi/ml)was added in methionine-free Eagle's minimal essential medium, supplemented with
5~o foetal calf serum; 32p i (50 p.Ci/ml) was added in complete medium. Radiochemicals were purchased from
Amersham. The procedure used for the in vitro phosphorylation of HCMV proteins was essentially the same as
described by Mar et al. (1981). Phosphorylation reactions were carried out at 24 °C for 60 min with 10 ~tg of the
specified protein using 7 ~tCi [32p]ATP in 200 ~tl of 50 mM-Tris-HCl buffer pH 8.0, containing 20 mM-MgCI2,
10 mM-DTT and 0.1 ~ Nonidet P40.
Protein gel electrophoresis. Proteins of purified HCMV particles were denatured in SDS buffer (2~ SDS, 10~
2-mercaptoethanol,5 ~ glycerol,0.005~ bromophenol blue, 50 mM-Tris-HClpH 7.0; heated to 100 °C for 3 rain)
and separated in denaturing 8.5~ or 10~ polyacrylamide gels essentially as described (Laemmli, 1970). In
resolving gels, the ratio of methylene bisacrylamide to acrylamide was increased to 1 :28 in order to separate the
major capsid protein from the large phosphoprotein (Irmiere & Gibson, 1983). After electrophoresis, gels
containing [35S]methioninewere fluorographed by the method of Bonner & Laskey (1974). Gels containing
unlabelled proteins were stained with Coomassie Brilliant Blue or with silver nitrate (Merril et al., 1981).Standard
proteins of known molecula- weight (Sigma) and Escherichia coli RNA polymerase (Boehringer) were run on the
same gel.
Immunoblotting. For Western blot analysis, antigens from the acrylamide gels were electrophoretically
transferred onto nitrocellulose (Towbin et al., 1979). Nitrocellulose sheets were cut into strips and blocked with
NET buffer (150 mM-NaC1,5 mM-EDTA,50 mMTris-HC1 pH 7.4, 0.25~ gelatin, 0.05~ Nonidet P40, 2 ~ bovine
serum albumin). After reaction with serum specimens (diluted from 1 : 30 to 1 :200) the strips were incubated with
horseradish peroxidase-coupled Protein A (Sigma) and staining was done with 40-chloro-l-naphthol and H202.
RESULTS
Expression o f the major structural phosphoproteins
The proteins of purified H C M V particles from culture-adapted virus strains and recent
clinical virus isolates have been found to be very similar by gel electrophoresis. Yet, a m i n i m a l
difference in the migration velocity of the glycoprotein gp58 became apparent if envelope
proteins from strains AD169 and Towne were compared by immunoprecipitation with mouse
monoclonal antibodies (Nowak et al., 1984b). Fig. 1 shows the profiles of proteins from the
prototype strain A D 169 and a wild-type virus strain designated BR84. The viruses were purified
from cell culture supernatants; structural proteins, about 0-6 to 1.5 ~tg per slot, were separated by
8-5~ (w/v) polyacrylamide gel electrophoresis and stained with silver nitrate. Corresponding
polypeptides appeared identical in electrophoretic mobility. However, a marked difference was
recognized in the relative abundance of the major matrix phosphoprotein (pp65) if the clinical
isolate BR84 and the laboratory strain were compared. Confirming numerous earlier studies
with [35S]methionine labelling, the silver staining showed that pp65 is by far the most a b u n d a n t
structural protein in strain AD169. Even more of phosphoprotein pp65 was found in the dense
body fraction (Fig. 1, lanes 3 and 4). However, the polypeptide of equivalent size appeared to be
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
1329
Cytomegalovirus phosphoproteins
2
3
4
5
6
m
m pp150
m
m pp65
Fig. 1. Profiles of HCMV proteins from the prototype strain AD169 and a wild-type isolate BR84.
Gradient-purified particles were loaded on an 8.5~ running ('high-bis' gel), 5~ stacking polyacrylamide gels, subjected to electrophoresis, and stained with silver nitrate. Lane 1, NIEP fraction of
AD169 (0.6 ~tg); lane 2, virions of strain ADI69 (0-6 ~tg);lanes 3 and 4, enriched dense bodies of strain
AD169 (1-0 ~tg); lanes 5 and 6, isolate BR84 (1.5 p.g and 0.6 ~tg)purified from the fifth passage; lane 7,
wild-type isolate (1.0 ~tg),ninth passage. Left, mol. wt. markers (Sigma) of 205K, 116K, 97K, 66K, 45K,
29K and RNA polymerase subunits (Boehringer) of 160K, 150K and 38K.
a minor constituent of the wild-type strain BR84 (Fig. 1, lanes 5 and 6). Similarly, five other
clinical isolates at low passage expressed the matrix protein pp65 to a low extent (data not
shown). Low passage wild-type viruses did not have appreciable amounts of dense body
particles. A t higher passages, pp65 of the clinical isolates did increase in quantity (Fig. 1, lane 7).
This indicates that the pp65 is not necessarily a b u n d a n t in infectious H C M V particles, and the
higher concentrations of this phosphoprotein correlate with the accumulation o f dense body
particles at higher cell culture passages.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
1330
G. J A H N A N D O T H E R S
Thus, it seems unlikely that the differences in pp65 concentration between fresh isolates and
culture-adapted laboratory strains can be explained solely by variable incorporation of this
protein into the matrix of complete virions.
Whereas the amounts of pp65 matrix protein appeared to be variable in HCMV particles
(Fig. 1), the structural proteins of approximately 150K seemed to be rather constant (Roby &
Gibson, 1986). Earlier studies had shown that one of these proteins, sometimes migrating as the
upper band, is phosphorylated (ppl50) (Nowak et al., 1984a, b). On the other hand, Gibson &
Irmiere (1984) found the phosphorylated polypeptide (from the matrix) migrated faster than the
non-phosphorylated protein (a major nucleocapsid protein) in polyacrylamide gels containing
increased concentrations of methylene bisacrylamide ('high-bis'). To resolve the discrepancy,
we analysed these labelled proteins in parallel on normal and 'high-bis' gels, combined with
autoradiography and Western blotting. The phosphorylated protein sometimes migrated more
slowly in the conventional gel system (Fig. 2, lanes 1 and 2). The two bands could not usually be
discriminated by these gel conditions. However, the phosphoprotein was always identified as
the lower band if 'high-bis' gels were used (Fig. 2, lanes 3 and 4).
Antibodies against the two major viral phosphoproteins in human sera
In order to identify immunoreactive virion polypeptides, total proteins from purified
extracellular HCMV were electrophoresed and analysed by Western blot experiments with
human sera. Fig. 3 (a) lane 1 shows the characteristic staining patterns with a pool of high titred
anti-HCMV human sera. The most intense reaction was seen with the 150K phosphoprotein
and within the area of glycoprotein gp58. On the other hand, the phosphoprotein pp65 was
rather weakly recognized, in spite of large amounts of this matrix protein on the nitrocellulose
strip. The low reactivity with pp65 was even more strikingly demonstrated with individual sera.
Fig. 3 (a) lane 2 shows a Western blot performed under identical conditions as for in lane 1, using
the serum (no. 40698) of a newborn with fatal cytomegalic inclusion disease. No antibody
response was detectable against pp65. Similarly, some sera of children or adults hardly reacted
with pp65 in Western blots, while others did under identical conditions (Fig. 4a, b). These
patterns of reactivity did not correlate with particular clinical states (Table 1); however, there
was a correlation with the titre in the complement fixation test.
All Western blots with human sera consistently showed stronger reactivity with 150K proteins
than with any other structural protein of the virus (Fig. 3, 4 and Landini et al., 1985). To
discriminate between immunoreactivities towards the 150K nucleocapsid protein and the
matrix protein ppl50, 'high-bis' gels were used in all subsequent experiments. As shown for
instance in Fig. 3 (b) lanes 5 and 6, the strongest immune responses were always found with the
faster migrating phosphorylated matrix protein ppl50, irrespective of the stage of virus
infection in serum donors (Table 1). In general, the pooled hyperimmune serum (Fig. 3a, lane 1)
as well as individual sera consistently gave maximum reactions with pp 150. These experiments
showed that the phosphorylated matrix protein ppl50, which is reproducibly found in the
purified virus particles, is the virion protein that is always highly antigenic to the humoral
immune system in natural human infection.
Antigenicity of the two major phosphoproteins in rabbits
Raising antibodies against structural viral proteins in rabbits not only yielded monospecific
antisera for gene mapping, but also corroborated the relative immunogenic properties of the
phosphoproteins. Sera from rabbits that had been immunized with the dense body fractions of
HCMV-infected culture fluids reacted primarily with the matrix protein pp65 (Fig. 5a).
However, a concentrated NIEP fraction induced a weak immune response against pp65,
apparently due to the lower concentration of this particular protein in these defective particles.
Again, the strongest reaction was seen against phosphoprotein pp 150 (Fig. 5 b). Immunization
with gel-purified pp65 resulted in rapid seroconversion after two injections; the antibodies
reacted with pp65 exclusively in Western blots (Fig. 5 c). These experiments taken together show
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
1331
Cytomegalovirus phosphoproteins
2
3
4
CP
ppl50
150
MCP ~
••
7~:!i•i~ii!~i!i~•ii!i
•ii!i!
Fig. 2. Virion proteins of HCMV strain AD169 radiolabelled with [35S]methionine (lanes 1 and 3) or
3 2 P i (lanes 2 and 4). Labelled virions were recovered, solubilized, and subjected to electrophoresis in an
8.5~ polyacrylamide gel. The gel in the right panel was cross-linked with a high amount (1:28) of
bisacrylamide ('high-bis' gel) as described by Irmiere & Gibson (1983). Abbreviations and estimated
mol. wt. are as follows: major capsid protein (MCP), phosphoprotein 150K (ppl50), matrix
phosphoprotein 65K (pp65).
that the matrix phosphoprotein pp65 is immunogenic. The weak antibody response in the course
of natural infection is probably due to low amounts of pp65 in wild-type virions, and not the
consequence of missing antigenic epitopes.
Further immunization experiments were performed with the two structural proteins of about
150K, the phosphorylated matrix protein and the major nucleocapsid protein (MCP). In the first
series, the two proteins were injected simultaneously, and subsequently the two proteins were
cut out separately and injected into two different rabbits. If both proteins were applied
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
1332
G. JAHN A N D OTHERS
(b)
3
5
M
6
•• •: •~•~Z-••~
-- 205
_ 160
-- 150
(a)
1
. •i~i~::
•
~/:,,~,~•~
--ll6
-- 97
: "!~!U
--ppl50
:2::
-- 66
--pp65
::iX: : )?
--gp58
-- 45
--38
-- 38
}
'
%.
-- 28
Fig. 3. Immunoblot of HCMV virion proteins following electrotransfer to nitrocellulose (NC). HCMV
virion proteins of strain A D 169 were subjected to electrophoresis in a 'high-bis' polyacrylamide gel (in a
10~, in b 8-5~) electrotransferred to NC, and (a) assayed (lane I) with a pool of high-titred human
antisera (ELISA titre 1 : 50 000) and (lane 2) with a serum of a newborn with fatal cytomegalic inclusion
disease (ELISA titre 1:2560; serum 40698 in Table 1). The NC filters were incubated with Protein A
coupled to horseradish peroxidase, followed by exposure to 4-chloro-l-naphthol. Abbreviations are as
follows: 150K phosphoprotein (pp 150), 65K matrix phosphoprotein (pp65), 58K glycoprotein (gp58, as
described by Mach et aL, 1986). (b) lane 3, [3 ~S]methionine-labeUed structural proteins of strain AD169,
exposure after electrotransfer to nitrocellulose; lane 4, immune reaction of rabbit antiserum ppl50 as
shown in Fig. 5 (d) (lane 3 represents the autoradiogram of this particular immunoblot); lane 5, immune
reaction with an HCMV-positive serum (no. 40712); lane 6, immune reaction with a human serum
negative by conventional serological procedures (no. 9/193); lane 7, immune reaction with serum
43136. M, mol. wt. markers (29K is not on the figure). Information on the human sera is given in
Table 1.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
Cytomegalovirus phosphoproteins
(a)
5
1
2
3
4
1333
(b)
gt
6
7
8
9
10
-ppl50
-pp65
- gp58
Fig. 4. I m m u n e reaction of individual h u m a n sera with H C M V structural proteins separated by S D S 'high-bis' P A G E (8.5~) and electrotransferred to nitrocellulose. (a) High-titred sera were taken
(complement fixation test, C F T / > 1 : 32) with the exception of the last lane, marked by a star (serum of a
male patient with fatal acquired i m m u n e deficiency syndrome who shed the virus without titre in the
CFT). (b) Low CFT-titred sera were taken from patients with different clinical symptoms. The i m m u n e
reaction was performed as described for Fig. 3, and abbreviations are as in Fig. 3. More information
about the h u m a n sera is given in Table 1.
T a b l e 1.
Listing of human sera from the patients examined with HCMV infections
Fig.
Lane
r
Initials
3 (a)
3(b)
2
5
6
7
1
2
3
4
5
6
7
8
9
10
S.R
S.B.
J.G.
L.R.
P.W.
L.G.
G.P.
S.H.
W.W.
D.S.
T.M.
G.F.
S.K.
N.K.
4(a)
4(b)
Patient
x
Age
Sex
3 days
34 years
33 years
17 years
32 years
27 years
46 years
50 years
25 years
10 years
30 years
42 years
17 years
53 years
F
F
M
M
M
M
M
F
M
F
F
M
M
F
.~
Lab. no.
Clinical status
Antibody
titre in
CFT
40698
40712
9/193
43136
50989
51710
53083
54206
45281
45850
45964
48862
46177
51 132
Cytomegalic inclusion disease
Healthy mother
Healthy scientist
Kidney transplant recipient
Kidney transplant recipient
Colitis
Reiter syndrome
Myocarditis
AIDS
Retinitis
Arthritis
Fever
Acute lymphocytic leukaemia
Pericarditis
1:8
1 : 16
1:32
1:16
1:64
1:64
1:32
1 : 16
1:4
1:8
1:16
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
Reactivity
in ELISAs
~
IgG IgM
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
1334
G. JAI-IN AND OTHERS
(a)
(b)
(c)
(d)
(e)
- ppl50
- pp65
Fig. 5. Immune reaction of rabbit antisera with HCMV proteins separated by SDS-'high-bis' PAGE
(8-5 %) and electrotransferred to nitrocellulose. For the immune reaction sera were as follows: (a) rabbit
antiserum raised against the dense body fraction of HCMV strain AD169; (b) rabbit antiserum
obtained with the NIEP fraction; (c) serum raised against gel-purified pp65 (after second booster
injection); (d) rabbit antiserum obtained with the structural proteins of about 150K (after four booster
injections, the reaction is against ppl50 exclusively); (e) rabbit antiserum raised against gel-purified
protein p150 (after six booster injections, immune reaction shows ppl50 and MCP150). The immune
reaction was performed as described for Fig. 3.
simultaneously, antibodies against ppl50 were raised significantly earlier [after four injections
(Fig. 5 d) instead of six injections]. The antisera against purified pp150 remained monospecific
and were raised after three booster injections, whereas immunization with gel-purified
nucleocapsid protein p150 resulted in an immune serum after five booster injections,
recognizing both pp 150 and the nucleocapsid polypeptide (Fig. 5 e); this was probably due to the
difficulties in cutting out ppl50 and MCP from preparative polyacrylamide gels separately
without any cross-contamination prior to immunization. In summary, the matrix protein pp150
is particularly antigenic in rabbits. This is in line with the strong immunogenic properties of
phosphoprotein ppl50 in the natural infection of humans.
DISCUSSION
Human cytomegalovirus (HCMV) particles purified from supernatants of cell cultures
contain two abundant structural phosphoproteins, designated pp65 and ppl50 according to
their apparent molecular weight in SDS-polyacrylamide gel electrophoresis. Both phosphoproteins have been described as constituents of the virion matrix (Gibson, 1983). The two proteins
are specified by the HCMV genome, as the coding genes (in strain AD169) have been identified
and their nucleotide sequence determined (Nowak et al., 1984a; ROger et al., 1987; G. Jahn, T.
Kouzarides, B. Fleckenstein & B. G. Barrell, unpublished results). The matrix phosphoprotein
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
Cytomegalovirus phosphoproteins
1335
pp65 is found in large amounts in cell culture-adapted laboratory HCMV strains. However, as
shown in this study, pp65 is in low abundance in virions of recent clinical isolates gradientpurified from supernatants of infected cells. Continuous culture of clinical HCMV isolates
resulted in the gradual accumulation of pp65-containing particles. The most plausible
explanation seems to be that infectious virions have pp65 as a minor component, and the
amount of this phosphoprotein correlates with the presence of dense bodies. Theoretically it is
also possible that pp65 occurs exclusively in dense bodies, and whatever pp65 is found in the
virion fraction is due to contamination by dense bodies. However, this seems not to be very
likely, since we do not know of a proven precedent that a defined gene for a structural viral
protein codes for components of defective particles only. The human immune response against
pp65 is variable and is even undetectable in some individuals. Here we show that the
phosphoprotein pp65 can be highly immunogenic upon inoculation of the isolated protein into
rabbits. These results lead to the conclusion that the phosphoprotein pp65 is scarcely expressed
in natural infection, and the extreme accumulation of this phosphoprotein in cytomegalovirusinfected cell cultures seems rather to be a ceU culture artefact. Other known herpesviruses do not
produce defective particles of the HCMV dense body type mostly consisting of a single defined
matrix protein.
In the Towne strain of HCMV, the nucleotide sequence of the gene for a major 67K
phosphorylated viral protein has been described (Davis & Huang, 1985). A different genome
fragment coding for a weakly glycosylated abundant 64K protein was mapped by Pande et al.
(1984), indicating that there are at least two different viral proteins with apparent molecular
weights of about 65K. The genomic region of the Towne strain identified by Pande et al. (1984)
is equivalent to the HCMV AD169 gene for pp65, which was described in detail in our
laboratory (Nowak et al., 1984a; R/iger et al., 1987). Kim et al. (1983) have shown that clinical
isolates of HCMV as well as laboratory strains could be captured by monoclonal antibodies
against a structural protein of about 66K in a double antibody sandwich ELISA. These
monoclonal antibodies have not been assigned to either of the two more recently defined
proteins in this size range (Forman et al., 1985; Britt & Auger, 1985; Rfiger et al., 1987). Thus,
the experiments do not contradict the observed pronounced variability of pp65 in purified
HCMV particles. Cremer et al. (1985) described the antibody response to cytomegalovirus
polypeptides captured by different monoclonal antibodies in a solid-phase enzyme immunoassay. Other studies had shown that multiple bands of protein can be precipitated from HCMVinfected cell cultures using sera from patients with evidence of active virus infection (Pereira et
al., 1982a, 1983; Zaia et al., 1986). Though it seemed to be in contrast to our observation that
strong Western blot reactivity is limited to a small number of virion polypeptides, the
discrepancy may most easily be explained by the fact that, in contrast to others, our study was
done with gradient-purified virus particles. In a series of immunoblot experiments, Landini et
al. (1985, 1986) showed recently that serum antibodies are directed against a limited number of
individual cytomegalovirus structural polypeptides.
The high molecular weight phosphoprotein of HCMV (ppl50) which is synthesized in
equivalent amounts by wild-type and laboratory strains, is the most antigenic viral component
in humans and rabbits. Although previous studies did not discriminate the antibody responses
against pp 150 and the nucleocapsid protein of similar size (Land ini et al., 1985, 1986), we have
shown here by reproducible separation of the two proteins in 'high-bis' gels that the stronger
reactivity of all sera tested was directed against the phosphorylated matrix protein. The
immunogenic nature of ppl50 may be explained from the putative polypeptide secondary
structure and the clustering of hydrophilic amino acids (Chou & Fasman, 1974; Hopp & Woods,
1981). Other herpesviruses that have been investigated do not possess such an antigenic
phosphorylated structural protein in this size range. For instance, the strongest reactivities of
herpes simplex virus-immune sera are directed against membrane glycoproteins such as gB
(Bernstein et al., 1985; Kahlon et al., 1986) and against the major nucleocapsid protein VP5 of
about 150K (Eberle & Mou, 1983). The matrix protein pp 150 of HCMV or peptides thereof will
be useful in the development of new diagnostic reagents. Thus, for instance, it was possible to
use high-titred antisera against native and genetically engineered pp 150 for in situ detection of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
1336
G. JAHN AND OTHERS
viral antigens in tissues (B. Borisch & G. Jahn, unpublished). It may also be possible to perform
serodiagnosis with mixtures of selected viral proteins after expression in bacterial or eukaryotic
cloning systems.
We thank Michael Mach for many helpful discussions and suggestions and Jutta von Hintzenstern for
cooperation. The technical assistance of Barbara Schiitze is thankfully appreciated. The work was supported by
Deutsche Forschungsgemeinschaft and Wilhelm-Sander-Stiftung.
REFERENCES
BERNSTEIN, D. I., GARRA'I-rY,E., LOVETT, M. A. & BRYSON, Y. J. (1985). Comparison of Western blot analysis to
microneutralization for the detection of type-specific herpes simplex virus antibodies. Journal of Medical
Virology 15, 223 230.
BONNER, w. M. & LASKEY,R. A. (1974). A film detection method for tritium-labeled proteins and nucleic acids in
polyacrylamide gels. European Journal of Biochemistry 46, 83-88.
BRITT, W. J. (1984). Neutralizing antibodies detect a disulfide-linked glycoprotein complex within the envelope of
human cytomegalovirus. Virology 135, 369-378.
BRIT'r, W. J. & AUGER, D. (1985). Identification of a 65000 dalton virion envelope protein of human
cytomegalovirus. Virus Research 4, 31-36.
BRWr, w. I. & AUGER, D. (1986). Synthesis and processing of the envelope gp55-116 complex of human
cytomegalovirus. Journal of Virology 58, 185-191.
CHOU, P. Y. & FASMAN,G. D. (1974). Prediction of protein conformation. Biochemistry 13, 222-245.
CREMER,N. E., COSSEN,C. K., SHELL,G. R. & PEREIRA,L. (1985). Antibody response to cytomegalovirus polypeptides
captured by monoclonal antibodies on the solid phase in enzyme immunoassays. Journal of Clinical
Microbiology 21, 517-521.
DAVIS, M. G. & HUANG, E.-S. (1985). Nucleotide sequence of a human cytomegalovirus D N A fragment encoding a
67-kilodalton phosphorylated viral protein. Journal of Virology 56, 7-11.
EBERLE, R. & MOU, S.-W. (1983). Relative titers of antibodies to individual polypeptide antigens of herpes simplex
virus type 1 in human sera. Journal of Infectious Diseases 148, 436-443.
FARRAR, G. H. & ORAM, J. D. (1984). Characterization of the human cytomegalovirus envelope glycoproteins.
Journal of General Virology 65, 1991-2001.
FORMAN,S. J., ZAIA,J. A., CLARK,B. R., WRIGHT, C. L., MILLS,B. J., POTTATHIL,R., RACKLIN,B. C., GALLAGHER,M. T.,
WELTE, K. & BLUME,K. G. (1985). A 64,000 dalton matrix protein of human cytomegalovirus induces in vitro
immune responses similar to those of whole virM antigen. Journal of Immunology 134, 3391-3395.
GIBSON, w. (1983). Protein counterparts of human and simian cytomegaloviruses. Virology 128, 391~406.
GIBSON, W. & IRMIERE, A. (1984). Selection of particles and proteins for use as human cytomegalovirus subunit
vaccines. In Birth Defects (Original Articles Series, vol. 20), pp. 305-324. Edited by S. A. Plotkin, S. Michelson,
J. S. Pagano & F. Rapp. New York: Alan R. Liss.
HOPP, T. P. & WOODS, K. R. (1981). Prediction of protein antigenic determinants from amino acid sequences.
Proceedings of the National Academy of Sciences, U.S.A. 78, 3824-3828.
IRMIERE,A. & GIBSON,w. (1983). Isolation and characterization of a noninfectious virion-like particle released from
cells infected with human strains of cytomegalovirus. Virology 130, 118-133.
KAHLON,J., LAKEMAN,F. D., ACKERMANN,M. & WHITLEY, R. J. (1986). Human antibody response to herpes simplex
virus-specific polypeptides after primary and recurrent infection. JournalofClinical Microbiology 23, 725-730.
KIM, K. S., SAPIENZA,V. J., CARP, R. I. & MOON, H. M. (1976). Analysis of structural polypeptides of purified human
cytomegalovirus. Journal of Virology 20, 604-611.
KIM, K. S., SAPIENZA, V. J., CHEN, C-M. & WISNIEWSKI, H. (1983). Production and characterization of monoclonal
antibodies specific for a glycosylated polypeptide of human cytomegalovirus. Journal of Clinical Microbiology
18, 331-343.
LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature, London 227, 680-685.
LANDINI, M. P., RE, M. C., MIROLO, G., BALDASSARRI,B. & LA PLACA, M. (1985). Human immune response to
cytomegalovirus structural polypeptides studied by immunoblotting. Journal of Medical Virology 17, 303 311.
LAND1NI, M. P., MIROLO, G., COPPOLECCH1A,P., RE, M. C. & LA PLACA, M. (1986). Serum antibodies to individual
cytomegalovirus structural polypeptides in renal transplant recipients during viral infection. Microbiology and
Immunology 30, 683-695.
MACH,M., UTZ, U. & FLECKENSTEIN,B. (1986). Mapping of the major glycoprotein gene for human cytomegalovirus.
Journal of General Virology 67, 1461-1467.
MAR, E. C., PATEL, P. C. & HUANG, E. S. (1981). Human cytomegalovirus-associated D N A polymerase and protein
kinase activities. Journal of General Virology 57, 149-156.
MERRIL, C. R., GOLDMANN, D., SEDMAN, S. A. & EBERT, M. H. (1981). Llltrasensitive stain for proteins in
polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211, 1437-1438.
NOWAK, B., GMEtNER, A., SARNOW, P., LEVINE, A. J. & FLECKENSTEIN, B. (1984a). Physical mapping of human
cytomegalovirus genes: identification of D N A sequences coding for a virion phosphoprotein of 71 kDa and a
viral 65-kDa polypeptide. Virology 134, 91-102.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23
Cytomegalovirus phosphoproteins
1337
NOWAY,.,B., SULLIVAN, C., SARNOW, P., THOMAS, R., BRICOUT, F., NICOLAS, J. C., FLECKENSTEIN, B. & LEVINE, A. J.
(1984b). Characterization of monoclonal antibodies and polyclonal immune sera directed against human
cytomegalovirus virion proteins. Virology 132, 325-338.
PANDE, H., BAAK,S. W., RIGGS, A. D., CLARK,B. R., SHIVELY, J. E. & ZAIA, J. A. (1984). Cloning and physical m a p p i n g
ofa gene fragment coding for a 64-kilodalton major late antigen of human cytomegalovirus. Proceedingsof the
National Academy of Sciences, U.S.A. 81, 4965-4969.
PEREIRA, L., HOFFMAN, i . & CREMER, N. (1982a). Electrophoretic analysis of polypeptides i m m u n e precipitated
from cytomegalovirus-infected cell extracts by human sera. Infection and Immunity 36, 933-942.
PEREIRA, L., HOFFMAN, M., GALLO, D. & CREMER, N. (1982b). Monoclonal antibodies to h u m a n cytomegalovirus:
three surface membrane proteins with unique immunological and electrophoretic properties specify crossreactive determinants. Infection and Immunity 36, 924-932.
PEREIRA, L., STAGNO, S., HOFFMAN, i . & VOLANAKIS, J. E. (1983). Cytomegalovirus-infected cell polypeptides
immune-precipitated by sera from children with congenital and perinatal infections. Infection and Immunity
39, 100-108.
PEREIRA, L., HOFFMAN, M., TATSUNO, M. & DONDERO, D. (1984). Polymorphism of h u m a n cytomegalovirus
glycoproteins characterized by monoclonal antibodies. Virology 139, 73-86.
ROBY, C. & GIBSON, W. (1986). Characterization of phosphoproteins and protein kinase activity of virions,
noninfectious enveloped particles, and dense bodies of human cytomegalovirus. Journal of Virology 59, 714727.
ROGER, B., KLAGES,S., WALLA, B., ALBRECHT,J., FLECKENSTEIN, B., TOMLINSON, P. & BARRELL, B. (1987). Primary
structure and transcription of the genes coding for the two virion phosphoproteins pp65 and pp71 of human
cytomegalovirus. Journal of Virology (in press).
SCHMITZ,H., MULLER-LANTZSCH,N. & PETELER, G. (1980). H u m a n i m m u n e response to proteins of cytomegalovirus.
Intervirology 13, 154-161.
STINSKI,i . F. (I 976). Human cytomegalovirus: glycoproteins associated with virions and dense bodies. Journal of
Virology 19, 594-609.
TALBOT,P. & ALMEIDA,J. D. (1977). Human cytomegalovirus: purification of enveloped virions and dense bodies.
Journal of General Virology 36, 345-349.
TOWBIN,n., STAEHELIN,T. & GORDON,J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to
nitrocellulose sheets: procedure and some applications. Proceedings of the National Academy of Sciences,
U.S.A. 76, 4350-4354.
ZAIA, A., FORMAN,J., TING, Y-P., ORBINA, E. & BLUME, G. (1986). Polypeptide-specific antibody response to h u m a n
cytomegalovirus after infection in bone marrow transplant recipients. Journal of Infectious Diseases 153, 780787.
(Received 13 August 1986)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 09 Aug 2017 20:20:23