Download Distinct signals in human immunodeficiency virus type 1 Pr55

Journal o f General Virology (1992), 73, 3079-3086.
3079
Printed in Great Britain
Distinct signals in human immunodeficiency virus type 1 Pr55 necessary
for RNA binding and particle formation
Jeremy B. M. Jowett, l t David J. Hockley, 2 Milan V. Nermut 2 and Ian M. Jones 1.
1NERC Institute of Virology, Mansfield Road, Oxford OX1 3SR and 2National Institute for Biological Standards and
Control, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, U.K.
The human immunodeficiency virus type 1 (HIV-1)
gag gene product Pr55 self-assembles to form virus-like
particles when expressed in Spodopterafrugiperda cells
using recombinant baculoviruses. The particles resemble immature HIV and are released from the
infected cell into the culture medium. Using this system
we have progressively truncated the gag open reading
frame from the C terminus and examined each deleted
gag protein for its particle-producing capability. We
show that deletion of Pr6 and deletions that progressively remove the distal region of the Pr7 domain,
including one Cys-His box thought to function as an
RNA capture signal, do not affect particle formation.
However deletion of two Cys-His boxes causes
production of slightly larger particles with altered
sedimentation properties. Sequence-specific North-
Western assays using an RNA probe representative of
the HIV-1 packaging signal revealed specific R N A
binding by all mutants that maintained both Cys-His
boxes. However, deletion of one Cys-His box reduced
RNA binding substantially and loss of two Cys-His
boxes abolished binding entirely. We conclude that
HIV-1 gag particle formation per se does not require
viral R N A encapsidation, but that it may act as a
cofactor in the condensation of the immature core.
Further deletion of gag sequences upstream of the CysHis boxes led to the abolition of particle-forming
ability, and we show that one boundary of the gag
sequence necessary for particle formation lies within
eight amino acids spanning one of the known protease
cleavage sites at the C terminus of Pr24.
Introduction
larger (Prl60) gag-pol fusion protein which is also
packaged into forming virions (Jacks et al., 1988). In this
way, the enzymatic functions necessary for virus infectivity (the protease, reverse transcriptase, RNase H and
integrase) are incorporated into the budding particle.
When expressed alone in a number of eukaryotic
expression systems, Pr55 produces virus-like particles
very similar to the early budding immature particles seen
in HIV-infected cells (Gheysen et al., 1989; Karacostas
et al., 1989; Overton et al., 1989; Hu et al., 1990; Luo et
al., 1990; Shioda & Shibuta, 1990; Smith et al., 1990;
Haffar et al., 1991 ; Hoshikawa et al., 1991 ; Royer et al.,
1991 ; Mergener et al., 1992). The ability of similar core
antigens from other retroviruses to self-assemble into
particle structures is now well documented (Delchambre
et al., 1989; Rasmussen et al., 1990; Morikawa et al.,
1991).
Pr55 is normally a myristylated protein and when
myristylation occurs, particles bud from the cell surface.
Lack of myristylation (by mutagenesis of the N terminus
of Pr55) does not affect particle formation, but does
prevent particle budding and release (Gheysen et al.,
1989; Overton et al., 1989).
The human immunodeficiency virus (HIV) groupspecific antigen (gag) gene product represents the major
component of the virus particle. It is first synthesized as a
precursor protein of 55K (Pr55) and forms the structural
basis of the retrovirus core. During or following virus
release, Pr55 is cleaved by the viral protease to yield
three gag-related components, Prl7 (also known as MA
protein), Pr24 (also known as CA) and Prl5 (also known
as NC protein) (for reviews see Kieny, 1990; Wills &
Craven, 1991). The Prl5 molecule undergoes further
cleavage to Pr7 (sometimes known as Pr9) and Pr6
(Veronese et al., 1987, 1988), and low frequency internal
initiation within the Pr55 open reading frame can result
in the production of a Pr41gag product of unknown
significance (Mervis et al., 1988). At a rate of about 5 %,
ribosomes translating the gag m R N A undergo a frameshift event near the Pr7/Pr6 junction, resulting in the
deletion of the Pr6 domain and the production of a much
t Present address: Department of Microbiology, UCLA School of
Medicine, Los Angeles, California 90073, U.S.A.
0001-1173 © 1992 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03
3080
J. B. M . J o w e t t and others
In contrast to Pr55, the expression of permanently
frameshifted gag-pol fusion protein does not allow
particle formation (Shioda & Shibuta, 1990; Mergener et
al., 1992), a result which is in keeping with earlier work
using other retroviruses (Felsenstein & Goff, 1988). This
might suggest a role for Pr6 in particle formation or
release because gag-pol fusions lack the gag Pr6 domain.
Indeed, truncations of the Pr6 domain within H I V
proviral clones have been reported to allow particle
formation but to prevent virion release from infected
cells (Gottlinger et al., 1991). However, equivalent
deletions in a vaccinia virus expression system producing
solely gag antigen allow particle formation and release
(Hoshikawa et al., 1991). However, in this case, there is
the possibility that vacinia virus itself contributes a
function that can substitute for the missing Pr6 domain.
Pr55 is a multifunctional protein and, in addition to
driving particle formation and the co-incorporation of
gag-pol fusion proteins, it is also responsible for the
capture and incorporation into the core of the viral
genomic R N A . This function resides in the gag Pr7
domain and is associated with two Cys-His motifs
thought to function as the R N A capture signal. Point
mutations within these motifs reduce R N A incorporation and virus infectivity (Clavel & Orenstein, 1990;
Gorelick et al., 1990). Expression of gag proteins lacking
the Pr7 domain prevents particle formation (Gheysen et
al., 1989; Hoshikawa et al., 1991) suggesting that, in
addition to R N A binding, the essential signals for gaggag interaction also lie within, or very close to, Pr7. This
result has been taken to suggest a link between R N A
binding and particle formation (Gelderblom, 1991).
Intriguingly, mutations in H I V genomic R N A that lead
to poor genome incorporation also result in particles with
altered morphology (Clavel & Orenstein, 1990) and, in
other systems, a direct role for Cys-His boxes in subunit
interaction as well as nucleic acid-binding has been
demonstrated (Loeber et al., 1991). Moreover, from a
virus point of view, a link between successful R N A
capture and particle formation could be beneficial,
helping to minimize the formation of empty virus
particles.
To localize the sequences necessary for particle
formation in finer detail, and to investigate the association with R N A binding, we have expressed a detailed
series of gag truncation mutants and assessed their
ability to form and release particles, and to bind H I V
RNA.
Methods
DNA manipulation. Routine manipulations of DNA, plasmid
preparation, restriction digests and subcloning were all as described
(Sambrooket aL, 1989). Gag gene truncations Pr45, Pr44 and Pr42 were
all prepared by site-directed mutagenesis as described by Kunkel
(1985) using the followingoligonucleotides:for Pr45, 5' GCCTGTCTCTAAGTACAATCT3'; for Pr44, 5' CAACAGCCCTATTTCCTAGGG 3'; for Pr42, 5' TTGAAACACTAAACCATCTTT3'. Genes
encoding truncations Pr41.5 and Pr41 were made by the polymerase
chain reaction (PCR) using the forward primer 5' CGCGGAGCTCAG A G A T G G G T G C G A G A G C G T C 3' and reverseprimers as follows:
for Pr41.5, 5' CGCGTCTAGACTATGTATTTGTTACTTGGCTCAT 3' and for Pr41, 5' CGCGTCTAGACTACAAAACTCTTGCCTTATGGC 3'. Amplified products were eluted from a gel, cleaved
with SacI and XbaI, and cloned into the baculovirus transfer vector
pAcCL29.1 (Livingstone & Jones, 1989). All mutants and PCR
products were sequenced prior to their use for expression. Selection of
recombinant baculoviruses was done using linearized baculovirus
DNA as described (Kitts et al., 1991).
Cells and viruses. Spodopterafrugiperda (Sf9) cellswere propagated at
28 °C in TC-100 medium (Overton et al., 1989)containing 10% foetal
calf serum. AcNPV-Bgal, AcPAK6 (for transfections) and recombinant viruses were grown and assayed in confluentmonolayersof Sf9
cells as described (Summers & Smith, 1987).
RNA binding. RNA binding by each gag mutant was assayed by
North-Western blotting under the conditions described by Luban &
Goff(1991). The probe was prepared by in vitro transcription of an HIV
fragment spanning nucleotides 675 (a SacI site) to 840 (an XmnI site) of
HIV-ILAI. Non-specific probe was transcribed from the negative
strand.
Electron microscopy. For negative staining, Sf9 cells infected with
each of the truncated gag mutants were harvested 2 days post-infection,
washed once in PBS and fixed in 2.5% glutaraldehyde in 0-1 Mcacodylate bufferpH 7.2 for at least 24 h. Fixed cellswere embedded in
1% low meltingpoint agarose, and the agarose blockswere brieflyfixed
in glutaraldehyde and washed in cacodylatebufferbefore being treated
with 1% osmium tetroxide (in cacodylate buffer) for 2 h and 0.5%
uranyl acetate for a further 3 h. After dehydration in ethanol, cellswere
embedded via propylene oxide in Araldite. Sections were cut with a
diamond knife on a Reichert-Jung Ultracut E ultramicrotome, poststained with uranyl acetate and lead citrate, and examined with a
Philips CM12 electron microscope operating at 80 kV.
Results
Expression o f truncated gag antigens
The initial series of gag deletion mutants used in this
study is shown in Fig. 1. For the first construct (Pr46) the
S a c I - B g l l I fragment spanning most of the gag O R F of
HIV-1LA~ was cloned into the baculovirus expression
vector pAcCL29.1 (Livingstone & Jones, 1989). This
fragment encodes the gag precursor protein up to and
including the frameshift signal, but the protein is deleted
downstream of amino acid 437 (the Pr6 domain). Three
further mutations truncating the upstream Pr7 domain
were prepared in the same expression vector by sitedirected mutagenesis, introducing a stop codon (TAG) at
the position shown. These three mutations resulted in the
deletion of the gag O R F downstream of (i) amino acid
427 (just downstream of the Cys-His boxes), (ii) amino
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03
HIV-1 Pr55 RNA-binding signals
/
p24 l
~
~
p6
3081
(b)
1
2
3
4
5
6
7
1 2
3
4
5
6
7
116K-84K -
58K-48K -
I
!
l
36K Pr55
Pr46
Pr45
Pr44
Pr42
(437)
(427)
(410)
(390)
Cys-His motif
Fig. 1. Construction of gag truncation mutants. The position of each
mutant made is shown in relation to the Y-terminal end of the gag
ORF. The SacI-BglII fragment of HIV'ILAI was cloned directly into a
baculovirus transfer vector to produce Pr46 which encodes a protein
deleted downstream of amino acid 437. Site-directed mutation
introduced a stop codon at codons 427, 410 and 390, producing
constructs Pr45, Pr44 and Pr41, respectively. The position of the CysHis boxes in relation to the mutations is shown.
24K -
Fig. 2. Expression of gag truncations and reactivity with anti-gag
MAbs. Recombinant viruses were grown to high titre and used to infect
Sf9 cells at a multiplicity of 10. Two days post-infection cells were
washed, lysed and fractionated on 10% SDS-polyacrylamide gels. Gels
were either stained directly with Coomassie blue (a) or transferred to
nitrocellulose and incubated with gag-specific MAbs (b). Lane 1, Pr55;
2, Pr46; 3, Pr45; 4, Pr44; 5, Pr42; 6, wild-type baculovirus-infected; 7,
mock-infected. M~s were taken from pre-stained standards run at the
same time.
Fraction no.
acid 410 (between the two Cys-His boxes) and (iii)
amino acid 390 (just before the Cys-His boxes). Based on
the sequence changes made, these mutants were predicted to encode proteins of Mr 45K, 44K and 42K and
were named accordingly (Fig. 1). All mutations were
confirmed by sequence analysis before their co-transfection into cells with viral DNA to yield recombinant
baculoviruses (Kitts et al., 1991).
Recombinant viruses producing gag protein were
screened for and identified by Western blotting with a
mix of monoclonal antibodies (MAbs) reactive with HIV
gag Prl7 and Pr24 antigens. Plaque-purified mutant
viruses from each transfection were then used to infect a
culture of Sf9 cells at a multiplicity of 10, and the cultures
were harvested 2 days post-infection and examined by
SDS-PAGE and Western blotting (Fig. 2). Cell lysates
from each truncation mutant produced stainable quantities of a new protein (a) at a mobility corresponding to
the predicted Mr of the gag protein. Each gag protein was
also stained specifically with anti-gag MAbs (b). Wildtype virus-infected (lane 6) or mock-infected cells (lane 7)
showed no evidence of any cross-reactive material. Wildtype Pr55 protein (lane 1) exhibited some breakdown, as
described previously (Gheysen et al., 1989), but the
truncation mutants showed much less breakdown suggesting that the majority of cellular proteases cleave from
the C terminus of gag.
Particle formation and release
To assess the ability of each of our truncated gag mutants
to produce HIV-like particles, Sf9 cells were infected
20%
2
(c)
3
4
:
5
~
6
~
7
8
9
60%
:S::.: :::::
Fig. 3. Sucrose gradient analysis of gag particles. Supernatants from
cultures infected at high multiplicity were harvested 2 days postinfection and layered onto preformed gradients of 20% to 60% sucrose
in PBS. The gradients were centrifuged at 36000 r.p.m, for 100 rain and
fractionated from the top. Aliquots of each fraction were electrophoresed on 10% SDS polyacrylamide gels and Western blotted with antiPr24 MAb. p46, p45, p44 and P42, (a) to (d) respectively.
with virus at high multiplicity and the supernatants were
harvested 2 days post-infection before appreciable cell
lysis had occurred. After clarification, supernatants were
applied to a sucrose density gradient as described
(Gheysen et al., 1989) and the gradient was fractionated
and analysed by Western blotting using anti-gag MAbs
(Fig. 3). We found that truncation mutants Pr46 (a), Pr45
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03
3082
J. B. M. Jowett and others
Fig. 4. Electron micrographsof truncated gag mutant-infectedSf9 cells. Cells were processed as described 2 days after infection with
Pr46 (a), Pr45 (b), Pr44 (c) or Pr42 (d). Surfacevacuoleformation was observedonlywith Pr42 and is marked V. Free Pr42 particles are
shown (FP). The bar marker represents 100 nm.
(b) and Pr44 (c) all produced gag-containing particles
that migrated to a position similar to Pr55 (45 ~ sucrose).
Truncation mutant Pr42 (d) also produced core-like
particles, but they sedimented in 25 to 3 0 ~ sucrose as
compared to 45~o sucrose for wild-type.
Electron microscopy
Deletion of gag C-terminal sequences was clearly not
incompatible with expression and secretion of gag
protein that could be banded by velocity gradient
centrifugation (Fig. 3). However, the exact macromolecular state of this protein could not be deduced by
gradient analysis alone. Accordingly, we examined the
culture supernatant by negative staining and also
prepared thin sections through infected Sf9 cells for
direct visualization of the particle budding process.
Typical virus-like particles were found in all samples
(Fig. 4). Particle morphology and budding for Pr55 was
similar to that described earlier and that Pr46 (a), Pr45
(b) and Pr44 (c) was similar.
Generally particle formation occurred at the plasma
membrane, producing roughly spherical particles about
120 nm in diameter. There was some evidence of intracytoplasmic gag 'ring' structures typical of the structures
formed in the absence of gag myristylation (Gheysen et
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03
HIV-1 Pr55 RNA-binding signals
al., 1989; Overton et al., 1989; Gelderblom, 1991), but
the number of these structures was small. However, Pr42
in addition to normal budding at the cell surface, showed
some budding into intracellular and cell surface vacuoles
(marked V in Fig. 4d) as well as from the plasma
membrane. Particles formed within vacuoles had essentially normal morphology but showed a greater range of
size than the surface particles (average approx. 130 nm).
The increase in particle diameter associated with Pr42
may partly explain the altered banding properties of this
mutant in velocity gradients (Fig. 3d).
We conclude from the data presented in Fig. 3 and 4
that, within the baculovirus expression system, deletion
of the gag Pr6 domain and a large proportion of the Pr7
domain including the two Cys-His boxes does not
prevent gag particle assembly or release.
A link with RNA binding?
Although particle assembly was clearly independent of
the presence of the Cys-His domain of gag, a link with
RNA binding could not be ruled out. For example, in
some retroviruses specific RNA binding has been shown
to be associated with the matrix domain of the gag
protein (Katoh & Yoshinaka, 1990), which is present in
all our deletion mutants. Therefore we examined the
ability of each of our truncated gag proteins to capture
RNA representative of the HIV genome. The exact
extent of the cis-acting sequences (the psi site) necessary
for the efficient incorporation of genomic RNA into
HIV particles is not yet clear. The region of the molecule
from the 3' end of the 5' long terminal repeat to the
beginning of the gag ORF is clearly necessary, but has
not yet been shown to be sufficient for efficient
packaging (Lever et al., 1989; Clavel & Orenstein, 1990).
To assess the truncated gag proteins for RNA binding we
used North-Western blotting with a probe encompassing
the supposed psi site and some of the 5' end of gag. This
system has been shown recently to give specific RNA
binding to Pr55 (Luban & Goff, 1991). The protein
profile observed on the blot was similar to that shown in
Fig. 2(a).
We found efficient RNA binding associated with Pr55
and with each of the Pr55 breakdown products (Fig. 5,
lane 1) suggested earlier to lack C-terminal sequences
(Fig. 2 and discussion thereof). Consistent with this we
also observed RNA binding with Pr46 and Pr45 (Fig. 5,
lanes 2 and 3). However, we observed substantially
reduced RNA binding associated with Pr44 (one CysHis box; lane 4) and none with Pr42 (no Cys-His boxes;
lane 5). Background binding to a set of cellular proteins
can be discerned in wild-type baculovirus- and mockinfected cells (lanes 6 and 7), but was poor compared to
gag-related activity and did not obscure any of the
2
1
3
4
5
6
3083
7
Fig. 5. Specific RNA binding by each gag construct. Infections were
processed and transferred as described in the legend to Fig. 2 except
that lanes were loaded with samples normalized for the amount of gag
antigen present before the transfer. North-Western blotting was done
as described in Methods. Lane 1, Pr55; 2, Pr46; 3, Pr45; 4, Pr44; 5,
Pr42; 6, wild-type baculovirus-infected cells; 7, mock-infected cells.
L
p17
]
p24
I
p7
] p6 ]
ARVLAEAMSQVTNTATIMMQRGNFRNQRKMVKCFN
!
l
!
!
Gag p41
Gag p41.5 AcNPVgagl4myr
+
vAcGagCfr 1
Gag p42
Fig. 6. Fine endpoint mapping of sequences involved in gag particle
assembly. The amino acid sequence of the gag Pr24-Pr7 junction is
shown with each of the two potential HIV protease cleavage sites
shown as large arrow heads. The position of the two gag deletions
reported previously (vAcGagCfrl and AcNPVgagl4myr) is shown in
addition to that of the constructs described here. In each truncation the
codon for the amino acid indicated is replaced by a stop codon.
truncated gag bands. One breakdown product of Pr55
with an Mr of approximately 20K also bound probe but
was absent from the gag truncation mutants, suggesting
it may represent a Prl5-containing fragment. NorthWestern blotting with a non-specific RNA probe gave
only background binding (data not shown). This result
confirms that Cys-His boxes are essential for RNA
capture, although it does not rule out the involvement of
other gag sequences. It is also clear that particle
formation, demonstrated for each truncation mutant, is
not linked to the ability to bind RNA, as represented by
the probe used for our blotting experiments. Although we
cannot wholly rule out co-incorporation of non-specific
RNA fragments into the developing particle, the fact
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03
3084
J. B. M. Jowett and others
Fig. 7. Electronmicrographsof Sf9 cellsinfectedwith Pr41.5 (a) and Pr41 (b). The micrographsshownare typicalof manysections
examined.Grosscellsurfacedistortionis apparentin bothinfections,but no particlesare apparentin either.The bar markerrepresents
200 nm.
that non-specific probes do not bind to gag protein
suggests that this is unlikely.
Endpoint mapping of sequences involved in particle
formation
Previous work using baculovirus (Gheysen et al., 1989)
and vaccinia virus expression systems (Hoshikawa et al.,
1991) has suggested that truncation of the HIV gagcoding sequence at the distal end of the Pr24 domain
abolishes particle formation. However, one recent report
has claimed that deletions in this area still allow particles
to be formed, although the evidence presented in favour
of this was scant (Royer et al., 1991). The HIV protease
can cleave at two distinct sites at the end of the Pr24
domain, VLAE (amino acids 362 to 365) and IMMQ
(amino acids 376 to 379) (see Fig. 6). This can lead to
some confusion over the exact C terminus of Pr24. We
found that one of the two reports detailing particle
abolition had truncated the gag coding sequence from
the upstream site whereas the report detailing continued
particle formation had deleted sequences C-terminal to
the alternative downstream site. Thus, the 3' limit of gag
sequences necessary for particle formation might lie
between these two sites, both of which are upstream of
the gag construct, Pr42 (Fig. 1), with the largest deletion
which still forms particles. To confirm the boundary for
particle formation, we constructed two more gag
truncation mutants (Pr41 and Pr41.5; Fig. 6). The first is
deleted from the upper of the two Pr24 C-terminal
protease cleavage sites (amino acid 363) and the second
from amino acid 372, located between the two possible
cleavage sites. The level of expression of each mutant
was similar to that of the constructs already described,
but no antigen was found in the tissue culture superna-
tant (data not shown). When analysed by electron
microscopy, neither mutant showed evidence of particle
formation (Fig. 7). Electron-dense material typical of gag
protein was found just beneath the plasma membrane of
infected cells. The cells had abundant large surface
vacuoles, with the dense protein layer found between the
vacuole and the cell surface membranes, and these
electron-dense layers labelled specifically with an antigag MAb and gold conjugate (not shown). Similar
structures have been observed in earlier studies (Gheysen et al., 1989). Based on these results and those of Royer
et al. (1991), we deduce that the C-terminal boundary for
the sequences involved in HIV gag particle formation
lies between amino acids 372 and 379, and spans the
known downstream protease cleavage site at the end of
the Pr24 domain.
Discussion
The ability of gag gene products to self-assemble
following expression has been established for a number
of retroviruses, bovine (Rasmussen et al., 1990), feline
(Morikawa et al., 1991), human (Gheysen et al., 1989;
Karacostas et al., 1989; Overton et al., 1989; Hu et al.,
1990; Luo et al., 1990; Shioda & Shibuta, 1990; Smith et
al., 1990; Haffar et al., 1991; Hoshikawa et al., 1991;
Royer et al., 1991; Mergener et al., 1992) and simian
immunodeficiency viruses (Delchambre et al., 1989).
However, the boundaries of the gag sequences necessary
for particle formation have not been systematically
determined. Gag-pol fusion proteins expressed in
similar systems have failed to yield particles and, as the
frameshift event producing the gag-pol fusion protein
also removes the Pr6 domain of gag, a role for the Pr6
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03
H I V - 1 Pr55 R N A - b i n d i n g signals
domain in particle formation was feasible. However, we
have shown that deletion of Pr6 leads to the production
of gag particles that are indistinguishable from those
produced by Pr55. Therefore the failure of gag-pol
fusion proteins to assemble is most likely due to steric
effects imposed by the presence of the large pol domain.
These results suggest that incorporation of the HIV gagpol precursor into forming virions is gag-driven as shown
for other retrovirus systems (Felsenstein & Goff, 1988).
Gottlinger et al. (1991) have shown that deletion of the
Pr6 domain within HIV proviral clones results in HIV
particle formation but suppression of particle release
from transfected COS-7 cells. They suggested that lack of
Pr6 might prevent the final stages of particle assembly
which, in turn, leads to prevention of particle release.
From our own findings and those recently published
(Hoshikawa et al., 1991 ; Royer et al., 1991) we suggest
that particle assembly and release are separable events.
Lack of Pr6 evidently does not affect particle assembly
per se, but does appear to prevent particle release when
all other HIV-encoded proteins are present. During the
expression of gag products only, and irrespective of the
expression system used (recombinant baculoviruses, this
work and Royer et al., 1991; recombinant vaccinia
viruses, Hoshikawa et al., 1991), gag proteins lacking the
Pr6 domain are efficiently released from infected cells.
This finding argues that as yet undefined HIV-encoded
products may be actively involved in HIV budding from
permissive cells once formation of the membraneassociated protein shell is complete, as suggested for
other retroviruses (Luftig et al., 1990). Alternatively,
vaccinia virus or baculoviruses could provide the
functions necessary for gag particle release in the
absence of the Pr6 domain.
We also investigated the role of the Pr7 domain in gag
particle assembly. Pr7 is undoubtedly the nucleocapsid
protein of HIV as mutations in the Cys-His arrays
within this domain severely depress the incorporation of
genomic R N A into forming virions (Gorelick et al.,
1990). Gheysen et al. (1989) originally showed that
complete deletion of Pr7 leads to loss of particle
assembly, a result recently confirmed by Hoshikawa et
al. (199t). Thus, in addition to the RNA-binding
activity, the C-terminal boundary of the region enabling
gag particle formation is also within the Pr7 domain. To
determine the endpoint of such sequences and to
investigate a possible link with R N A binding, we
produced a series of truncations through Pr7 up to and
including the endpoint originally described by Gheysen
et al. (1989). Deletion of both Cys-His motifs in these
constructs led to a complete loss of specific R N A
binding, with loss of the downstream motif (construct
Pr44) causing a 7 0 ~ reduction in binding. These results
are consistent with mutation studies that disrupt one or
3085
other of the Cys-His motifs (Gorelick et al., 1990; Luban
& Goff, 1991). Particle formation was apparent despite
removal of one (Pr44) or two (Pr42) Cys-His arrays,
indicating that the signals necessary for particle formation do not include either motif. However it is worth
noting that Pr42 demonstrated some cell surface vacuolation in addition to typical particle formation, suggesting
that a proportion of the expressed gag protein in this
construct fails to assemble successfully into particles.
Pr42 particles are marginally larger (about 10~) on
average than the particles produced by constructs that
retain RNA-binding capability. Therefore it is possible
that RNA incorporation acts as a condensing agent
helping the gag core to adopt a tighter conformation.
Absence of R N A might also contribute to the altered
sedimentation of Pr42 particles in sucrose gradients.
As Pr42 retains particle formation, the downstream
sequence essential for particle assembly lies N-terminal
to amino acid 390. Royer et al. (1991) have recently
proposed that deletion of gag sequences downstream of
amino acid 379 (their mutant AcNPVgagl4myr) does
not affect particle morphogenesis. However the electron
micrograph of cells infected with this truncation mutant
shows no free particles when compared to those of the
other gag truncations described, but does show evidence
of plasma membrane vacuolation, similar to our Pr42
mutant. These data suggest that both mutants encroach
upon the signals necessary for particle formation,
although they do not delete them entirely.
Two further truncations (Pr41 and Pr41.5) N-terminal
to Pr42 and AcNPVgagl4myr fail to show any evidence
of particle assembly, producing only cell surface vacuoles
with associated layers of protein. This result confirms
and extends the observations of Gheysen et al. (1989) to
locate the downstream boundary for particle formation
to between amino acids 372 (our mutant Pr41.5) and 379
(AcNPVgagl4myr; Royer et al., 1991). The identification of this region as overlapping the downstream
cleavage site for the HIV protease at the Pr24/Pr7
junction is fitting as activation of the protease at this site
during HIV budding but prior to release would result in
the abolition of particle formation. Further deletion
analysis of Pr55 should allow the definition of the
minimum sequences required for particle assembly, as
has been reported for other retroviruses (reviewed in
Wills & Craven, 1991).
Gag-only particles of the type described have a
number of uses as 'particle carriers', for example of the
HIV env protein, to produce non-replicating vaccine
candidates (Haffar et al., 1991). We suggest that if this
technology comes into widespread use for the production
of recombinant vaccines, the use of constructs similar to
Pr42 be encouraged as they would prevent the possible
transduction of R N A to recipient ceils.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03
3086
J. B. M . Jowett and others
I.M.J. thanks Y.M. for many things. This work was supported by the
UK Medical Research Council's AIDS Directed Programme.
References
CLAVEL, F. & ORENSTEIN, J. M. (1990). A mutant of human immunodeficiency virus with reduced RNA packaging and abnormal
particle morphology. Journal of Virology 64, 5230-5234.
DELCHAMBRE, M., GHEYSEN, D., THINES, n., THIRIART, C., JACOBS,E.,
VERDIN, E., HORTH, M., BURNY, A. & BEX, F. (1989). The Gag
precursor of simian immunodeficiency virus assembles into viruslike particles. EMBO Journal 8, 2653-2660.
FELSENSTEIN, K. M. & GOFF, S. P. (1988). Expression of the Gag-Pol
fusion protein of Moloney murine leukemia virus without Gag
protein does not induce virion formation or proteolytic processing.
Journal of Virology 62, 2179-2182.
GELDERBLOM, H. R. (1991). Assembly and morphology of HIV:
potential effect of structure on viral function (editorial). AIDS 5,
617~537.
GHEYSEN, n., JACOBS,E., DE-FORESTA, F., THIRIART, C., FRANCOTTE,
M., THINES, D. & DE-WILDE, M. (1989). Assembly and release of
HIV-1 precursor Pr55gag virus-like particles from recombinant
baculovirus-infected insect cells. Cell 59, 103-112.
GORELICK, R. J., NIGIDA, S. M., JR, BESS, J. W., JR, ARTHUR, L. O.,
HENDERSON, L. E. & REIN, A. (1990). Noninfectious human
immunodeficiency virus type 1 mutants deficient in genomic RNA.
Journal of Virology 64, 3207-3211.
GOTTLINGER, H. G., DORFMAN, T., SODROSKI, J. G. & HASELTINE,
W. A. (1991). Effect of mutations affecting the Pr6 Gag protein on
human immunodeficiency virus particle release. Proceedings of the
National Academy of Sciences, U.S.A. 88, 3195-3199.
HAFFAR, O. K., SMITHGALL, M. D., MORAN, P. A., TRAVlS, B. M.,
ZARLING, J. M. ~£ HU, S. L. (1991). HIV-specific humoral and
cellular immunity in rabbits vaccinated with recombinant human
immunodeficiency virus Gag-Env particles. Virology 183, 487-495.
HOSHIKAWA, N., KOJIMA, A., YASUDA, A., TAKAYASHIKI,E., MUSAKO,
S., CHIBA, J., SATA,T. & KURATA, T. (1991). Role of the gag andpol
genes of human immunodeficiency virus in the morphogenesis and
maturation of retrovirus-like particles expressed by recombinant
vaccinia virus: an ultrastructural study. Journal of General Virology
72, 2509-2517.
Hu, S. L., TRAVIS, B. M., GARRIGUES,J., ZARLING, J. M., SRIDHAR, P.,
DYKERS, T., EICHBERG, J. W. & ALPERS, C. (1990). Processing,
assembly, and immunogenicity of human immunodeficiency virus
core antigens expressed by recombinant vaccinia virus. Virology 179,
321-329.
JACKS, T., POWER, M. D., MASlARZ,F. R., LucIw, P. A., BARR, P. J. &
VARMUS, H. E. (1988). Characterization of ribosomal frameshifting
in HIV-I Gag-Pol expression. Nature, London 331, 280-283.
KARACOSTAS, V., NAGASHIMA,K., GONDA, M. A. & MOSS, B. (1989).
Human immunodeficiency virus-like particles produced by a
vaccinia virus expression vector. Proceedings of the National
Academy of Sciences, U.S.A. 86, 8964-8967.
KATOH, I. & YOSHINAKA, Y. (1990). RNA-binding activities of bovine
leukemia virus gag proteins. Abstracts of the VIIIth International
Conference of Virology. Berlin, 26 to 31 August, 1990. Abstract
W45-005, p. 80.
KIENY, M. P. (1990). Structure and regulation of the human AIDS
virus. Journalof Acquired Immune Deficiency Syndromes 3, 395-402.
KITTS, P., AYRES, M. & POSSEE, R. (1991). Linearisation of baculovirus
DNA improves the selection of recombinants. Nucleic Acids
Research lg, 5667-5672.
KUNKEL, T. A. (1985). Rapid and efficient site-specific mutagenesis
without phenotypic selection. Proceedingsof the National Academy of
Sciences, U.S.A. 82, 488-492.
LEVER, A., GOTTLINGER, H., HASELTINE, W. & SODROSKI, S. (1989).
Identification of a sequence required for efficient packaging of
human immunodeficiency virus type 1 RNA into virions. Journalof
Virology 63, 4085-4087.
LIVINGSTONE, C. & JONES, I. M. (1989). Baculovirus expression vectors
with single-strand capability. Nucleic Acids Research 17, 2366.
LOEBER, G., STENGER, J. E., RAY, S., PARSONS,R. E., ANDERSON, M. E.
& TEGTMEYER, P. (1991). The zinc finger region of simian virus 40
large T antigen is needed for hexamer assembly and origin melting.
Journal of Virology 65, 3167-3174.
LUBAN, J. & GOFF, S. P. (1991). Binding of human immunodeficiency
virus type 1 (HIV-1) RNA to recombinant HIV-I Gag polyprotein.
Journal of Virology 65, 3203-3212.
LUFTIG, R., IKUTA, K., BU, i . & CALKINS, P. (1990). Terminal stages
of retroviral morphogenesis. In Retroviral Proteases : Maturation and
Morphogenesis, pp. 141-148. Edited by L. H. Pearl. London:
Macmillan.
Luo, L., LI, Y. & KANG, Y. (1990). Expression of Gag precursor protein
and secretion of virus-like particles of HIV-2 from recombinant
baculovirus infected insect cells. Virology 179, 874-880.
MERGENER, K., FACKE, M., WELKER, R., BRINKMAN, V., GELDERBLOM, H. R. & KRAUSSLICH, H. G. (1992). Analysis of HIV particle
formation using transient expression of subviral constructs in
mammalian cells. Virology 186, 25 39.
MERVIS, R. J., AHMAD, N., LILLEHOJ,E. P., RAUM,M. G., SALAZAR,
F. H., CHAN, H. W. & VENKATESAN,S. (1988). The gag gene products
of human immunodeficiency virus type 1 : alignment within the gag
open reading frame, identification of post-translational modifications, and evidence for alternative Gag precursors. Journal of
Virology 62, 3993-4002.
MORIKAWA, S., BOOTH, T. F. & BISHOP, D. H. (1991). Analyses of the
requirements for the synthesis of virus-like particles by feline
immunodeficiency virus Gag using baculovirus vectors. Virology
183, 288-297.
OVERTON, H. A., FUJII, Y., PRICE, I. g. & JONES, I. M. (1989). The
protease and gag gene products of the human immunodeficiency
virus: authentic cleavage and post-translational modification in an
insect cell expression system. Virology 170, 107-116.
RASMUSSEN, L., BATTLES, J. K., ENNIS, W. H., NAGASHIMA, K. &
GONDA, M. A. (1990). Characterization of virus-like particles
produced by a recombinant baculovirus containing the gag gene of
the bovine immunodeficiency-like virus. Virology 178, 435-451.
ROYER, i . , CERUTTI, M., GAY, B., HONG, S. S., DEVAUCHELLE,G. &
BOULANGER, P. (1991). Functional domains of HIV-1 Gag-polyprotein expressed in baculovirus-infected insect cells. Virology 184,
417-422.
SAMBROOK, J., FRITSCH, E. F. & MANIATIS, T. (1989). Molecular
Cloning: A Laboratory Manual, 2nd edn. New York: Cold Spring
Harbor Laboratory.
SHIODA, T. & SHmUTA, H. (1990). Production of human immunodeficiency virus (HIV)-like particles from cells infected with recombinant vaccinia viruses carrying the gag gene of HIV. Virology
175, 139-148.
SMITH, A. J., CHO, M. I., HAMMARSKJOLD,M. L. & REKOSH, D. (1990).
Human immunodeficiency virus type 1 Pr55gag and Prl60Gag-Pol
expressed from a simian virus 40 late replacement vector are
efficiently processed and assembled into virus like particles. Journal
of Virology 64, 2743-2750.
SUMMERS, M. D. & SMITH, G. E. (1987). A Manual of Methods for
Baculovirus Vectors and Insect Cell Culture Procedures. Texas
Agricultural Experimental Research Station. Bulletin no. 1555.
VERONESE, F. D., RAHMAN, R., COPELAND, T. D., OROSZLAN, S.,
GALLO, R. C. • SARNGADHARAN,M. G. (1987). Immunological and
chemical analysis of Pr6, the carboxy terminal fragment of HIV
Prl5. AIDS Research and Human Retroviruses 3, 253-264.
VERONESE, F. D., COPELAND, T. n., OROSZLAN, S., GALLO, R. C. &
SARNGADHARAN, M. G. (1988). Biochemical and immunological
analysis of human immunodeficiency virus gag gene products Prl7
and Pr24. Journal of Virology 62, 795-801.
WILLS, J. W. & CRAVEN, R. C. (1991). Form, function, and use of
retroviral Gag proteins (editorial). AIDS 5, 639-654.
(Received 24 June 1992; Accepted 4 August 1992)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 23:40:03