Download An active foamy virus integrase is required for virus replication

Journal
of General Virology (1999), 80, 1445–1452. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
An active foamy virus integrase is required for virus replication
Jo$ rg Enssle,1† Astrid Moebes,1‡ Martin Heinkelein,1 Markus Panhuysen,1 Bettina Mauer,1
Matthias Schweizer,2 Dieter Neumann-Haefelin2 and Axel Rethwilm1, 3
Institut fu$ r Virologie und Immunbiologie, Universita$ t Wu$ rzburg, Germany
Abteilung Virologie, Institut fu$ r Medizinische Mikrobiologie und Hygiene, Universita$ t Freiburg, Germany
3
Institut fu$ r Virologie, Medizinische Fakulta$ t Carl Gustav Carus, Technische Universita$ t Dresden, Gerichtsstr. 5, 01069 Dresden, Germany
1
2
Foamy viruses (FVs) make use of a replication strategy which is unique among retroviruses and
shows analogies to hepadnaviruses. The presence of an integrase (IN) and obligate provirus
integration distinguish retroviruses from hepadnaviruses. To clarify whether a functional IN is
required for FV replication, a mutant in the highly conserved DD35E motif of the active centre was
analysed. This mutant was found to be able to express Gag and Pol protein precursors and cleavage
products and to generate and deliver cDNA. However, this mutant was replication-deficient. The
junctions of individual foamy proviruses with cellular DNA were sequenced. The findings suggest
that FV integration is asymmetrical, because the proviruses started with what is believed to be the
U3 end of the free linear DNA to generate the conventional TG dinucleotide, while apparently two
nucleotides from the U5 end were cleaved to create the complementary CA dinucleotide. Alignment
of known FV genome sequences indicated that this mechanism of integration is not restricted to the
two FV isolates from which integrates were studied, but appears to be a common feature of this
retrovirus subfamily. In conclusion, with respect to the necessity of a functionally active IN for virus
replication FVs behave like other retroviruses ; their mechanism of integration, however, is
probably unique.
Introduction
Foamy viruses (FVs) are retroid viruses which have a
replication strategy different from that of retroviruses and
hepadnaviruses (Weiss, 1996 ; Linial, 1999). Recent studies of
various aspects of the FV replication cycle have led to the
assumption that FVs represent a functional link between
retroviruses and hepadnaviruses (Rethwilm, 1996 ; Linial,
1999). In this respect it is most important that large amounts of
DNA have been found in FV particles (Yu et al., 1996) and that
virion DNA appears to be the functionally relevant nucleic acid
(Moebes et al., 1997 ; Yu et al., 1999). In addition, FV particle
formation and Env protein function and particle incorporation
Author for correspondence : Axel Rethwilm (at the Technische
Universita$ t Dresden). Fax j49 351 459 3530.
e-mail Axel.Rethwilm!mailbox.tu-dresden.de
† Present address : Institute for Cancer Research and Treatment,
University of Torino School of Medicine, Torino, Italy.
‡ Present address : Department of Internal Medicine II/Molecular
Biology, University Hospital Freiburg, Freiburg, Germany.
0001-6233 # 1999 SGM
are unique among retroviruses and also take some intermediate
position between the retrovirus and hepadnavirus assembly
process (Enssle et al., 1997 ; Fischer et al., 1998 ; Baldwin &
Linial, 1998 ; Pietschmann et al., 1999).
Despite these functional differences the overall structure of
FV genomes is much more similar to that of typical complex
retroviruses than that of hepadnaviruses (Rethwilm, 1995). In
its provirus form the FV coding sequences are flanked by long
terminal repeats (LTRs) (Rethwilm, 1995). Furthermore, FVs
encode an integrase (IN) protein, which has been studied in
some detail by in vitro experiments (Pahl & Flu$ gel, 1993, 1995),
and provirus integration has already been demonstrated for
several FVs or FV vectors (Rethwilm et al., 1987 ; Schweizer
et al., 1993 ; Renshaw et al., 1991 ; Russel & Miller, 1996 ;
Tobaly-Tapiero et al., 1996 ; Neves et al., 1998).
Further characteristics which distinguish the replication
strategy of retrovirus from that of hepadnavirus include the
presence of an IN protein and obligate provirus integration
(Coffin, 1996). Given the many analogies between FVs and
hepadnaviruses we wanted to determine whether an active IN
protein is essential for FV replication. In addition, we noticed
a structural peculiarity in the primary nucleotide sequence of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07
BEEF
J. Enssle and others
the polypurine tract (ppt)–U3 LTR border in FVs (see below).
Since this peculiarity may influence the process of integration,
we first set out to determine the integration junctions of foamy
proviruses with cellular DNA. We used the chimpanzee strain
of FV, CFV\hu [previously called human foamy virus (HFV)],
which was reported to be isolated from human material
(Achong et al., 1971). However, this strain is almost identical to
FV isolates from chimpanzees and evidence for natural HFV
infections is lacking (Herchenro$ der et al., 1994, 1996 ; Schweizer
et al., 1995 ; Ali et al., 1996).
Methods
Cells and viruses. Human 293T kidney cells (DuBridge et al.,
1987), human KMST-6 skin fibroblastoid cells (Namba et al., 1985), baby
hamster kidney cells (BHK-21), BHK\bel-1 (Bieniasz et al., 1997) and
BHK\LTRlacZ cells (Schmidt & Rethwilm, 1995) were cultivated in
Eagle’s minimal essential medium (MEM) or Dulbecco’s modified MEM
containing 5–10 % foetal calf serum, antibiotics and 0n5 and 1 mg\ml
G418 in the case of BHK\bel-1 and BHK\LTRlacZ cells, respectively.
Plasmid-derived viruses were generated by calcium phosphate
cotransfection of BHK-21, KMST-6 or 293T cells (Ausubel et al., 1987).
Virus titrations were performed on BHK\LTRlacZ indicator cells as
described (Schmidt & Rethwilm, 1995).
Recombinant DNA. Conventional techniques were used to
generate FV recombinants (Ausubel et al., 1987). The infectious FV
plasmids pHSRV2 and pcHSRV2 and the replication-deficient FV vectors
pMH4 and pMH5, which express green fluorescent protein (GFP) under
the transcriptional control of the spleen focus-forming virus (SFFV) U3
region, have been described recently (Moebes et al., 1997 ; Fischer et al.,
1998 ; Heinkelein et al., 1998 ; Lindemann & Rethwilm, 1998). The
functional env inactivation mutant pcHSRV2\M68 was made in a similar
way to pHSRV2\M68 (Moebes et al., 1997), by Klenow enzyme fill-in of
a MroI site at the beginning of env. IN mutants of the infectious plasmids
were generated by recombinant PCR on a subgenomic 0n33 kb
EcoRI–HincII fragment (Higuchi, 1990). In pHSRV2\M70 the DYIG
motif was changed to DYTG, similar to a previous report (Pahl & Flu$ gel,
1995), and in pHSRV2\M73 and pcHSRV2\M73 the DD35E motif was
changed to DA35E. The mutations were verified by automated DNA
sequence analysis of the complete PCR fragment using AmpliTaqFS and
the ABI 310 sequence analysis system (Perkin Elmer). Introduction of the
M70 and M73 mutants into the replication-deficient FV vector pMH5
(Fischer et al., 1998 ; Heinkelein et al., 1998) was facilitated via exchange
of a 1n43 kb PacI–EcoRI fragment and resulted in pMH99 and pMH80,
respectively.
A 1n73 kb fragment covering the complete FV IN was amplified from
total DNA from HSRV2\M70-infected BHK-21 cells by PCR. Following
purification over glass milk the amplicon was subjected to automated
sequencing without further subcloning.
Sequence analysis of DNA junctions. The molecular cloning
and parts of the DNA sequence of the proviral–cellular DNA junctions
in the Vero-L cell line, which harbours a single provirus of simian FV type
3 (SFV-3), have been described previously (Schweizer et al., 1993).
293T cells were transfected with plasmid pMH4 and the cell-free
supernatant was used to transduce BHK-21 cells. Forty-eight hours
following transduction single GFP-positive BHK-21 cells were sorted in
a FACS Vantage (Beckton Dickinson), seeded into culture medium and
subcultivated. Total DNA was prepared by standard methods from
clonally expanded single cell colonies (Ausubel et al., 1987). Proviral–
cellular DNA junctions were amplified by inverse PCR, adapting
BEEG
described protocols for application to FV integrates (Triglia et al., 1988 ;
Silver & Keerikatte, 1989). In particular, prior to ligation DNA was
restriction-enzyme-digested to generate incompatible overhangs and to
specifically suppress the self-ligation and amplification of internal FV
fragments. For the amplification of genomic DNA–U3 junctions (A) total
DNA was digested with EcoNI and AluI ; for the amplification of
U5–genomic DNA junctions (B) digestion was performed with AvrII and
AluI ; the DNA was cut only with AluI to amplify both junctions (C).
DNA was self-ligated and subsequently recut with Eco91I (cases A and C)
or SspI (cases B and C) to enhance the amplification from linear substrates.
Inverse PCR was performed using the Panscript system (Boehringer
Mannheim) and FV LTR-specific primers. A and B amplicons were
purified by binding to glass milk and sequenced directly, C amplicons
were sequenced following electrophoretic gel separation and purification.
FV protein analysis. pHSRV2 or mutant plasmids were transfected
into BHK\bel-1 cells, and 4 days later a cellular lysate was prepared in
detergent buffer as described (Hahn et al., 1994). Following separation in
SDS–8 % polyacrylamide gel and semi-dry blotting to nitrocellulose
membranes (Schleicher & Schu$ ll), the viral proteins were allowed to react
with Gag-specific, and following stripping of the blot with RNaseHspecific, rabbit antisera (Hahn et al., 1994 ; Ko$ gel et al., 1995) and stained
using the ECL system (Amersham).
Southern blot analysis. pcHSRV2, pcHSRV2\M68 and
pcHSRV2\M73 were transfected into 293T cells. Four days later total
cellular DNA was prepared, run on a 0n8 % agarose gel, blotted onto a
nylon membrane (Qiagen) and hybridized to an FV cRNA probe
essentially as described previously (Moebes et al., 1997).
Analysis of FV cDNA delivery into recipient cells. 293T cells
were transfected with 15 µg pcHSRV2, pcHSRV2-M68, pcHSRV2-M73
or mock-transfected. Two days later the supernatants were collected,
passed through a 0n45 µm filter and used to infect BHK-21 cells. One day
after infection the low molecular mass DNA was extracted from the cells
by using a modified Hirt method (Hirt, 1967). Briefly, low molecular mass
DNA was separated from high molecular mass DNA by alkaline lysis of
the cells and centrifugation in high salt, as usually done for bacterial
plasmid preparation (Ausubel et al., 1987). DNA from the supernatant
was extracted using the Qiamp tissue kit (Qiagen). Viral DNA (onetwentieth of the original volume eluted from the preparation column) was
amplified by PCR with pol gene-specific primers (F196 and F197) as
described previously (Schmidt et al., 1997 b). Amplicons were separated in
ethidium bromide-containing agarose gels and photographed under UV
light.
Single replication cycle assay. 293T cells were transfected with
the replication-deficient FV vectors pMH5, pMH80 and pMH99 together
with the env expression vector pCenv-1, as described (Fischer et al., 1998 ;
Heinkelein et al., 1998 ; Pietschmann et al., 1999). The supernatant of the
transfected cells was used to transduce recipient BHK-21 and KMST-6
cells which were plated the day before at a density of 10% cells per 6 cm
dish. The number of GFP-positive cells was determined over time by
fluorescence-activated cell sorting (FACS) analysis on a FACScan using
the LysisII and CellQuest software packages (Becton Dickinson) as
reported (Lindemann et al., 1997 ; Heinkelein et al., 1998 ; Pietschmann
et al., 1999).
Results
Sequence determination of proviral–cellular DNA
junctions
Retrovirus integration is a highly ordered process mediated
by the IN protein. It involves the recognition and trimming of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07
Foamy virus integration
Fig. 1. (A) FV sequences at the ppt–U3 and U5–primer binding site (pbs) borders and proviral junctions to cellular DNA.
Proviral 5h U3 sequence of the pHSRV infectious molecular clones (Rethwilm et al., 1990 ; Lo$ chelt et al., 1991 ; Schmidt et al.,
1997 a) and published FV genome sequences from CFV/hu (Flu$ gel et al., 1987 ; Maurer et al., 1988), CFV (Herchenro$ der et
al., 1994), simian FV type 1 (SFV-1) (Kupiec et al., 1991), SFV-3 (Renne et al., 1992), bovine FV (BFV) (Renshaw & Casey,
1994 a, b) and feline FV (FFV) (Helps & Harbour, 1997 ; Winkler et al., 1997) were aligned. The sequences are arranged in
such a way that the LTR borders can readily be compared with the proviral–cellular DNA junctions. The Vero-L cell line
harbouring a single SFV-3 provirus and part of the junction sequences have been reported previously (Schweizer et al., 1993).
The murine leukaemia virus (MLV) and HIV-1 sequences are shown for comparison. In HIV-1 the terminal base removed from
the U5 end is the terminal A of the tRNA primer used to initiate reverse transcription (Whitcomb et al., 1990). All BHK-21 cell
clones which were used for the determination of junction sequences showed single proviral integrations as revealed by inverse
PCR analysis and ethidium bromide-containing agarose gel electrophoresis (data not shown). n, Any nucleotide ; x, duplicated
nucleotides at the MLV and HIV-1 proviral junctions ; n.a., sequence is not available. (B) Strategy to generate BHK-21 cells
harbouring single FV proviruses. pMH4 is an FV vector able to perform one round of replication (Heinkelein et al., 1998). ∆U3
is the U3 region of the HSRV2 LTR deletion variant (Schmidt et al., 1997 a).
the 3h OH ends of the unintegrated linear viral DNA and its
joining to cellular DNA, accompanied by duplication of a short
stretch of host sequences (Goff, 1992 ; Brown, 1997). During
integration two nucleotides, which form an inverted repeat, are
invariantly lost from both ends of the provirus and all
proviruses start (on the plus-strand) with the subterminal TG
dinucleotide and terminate with the subterminal CA
dinucleotide (Goff, 1992 ; Brown, 1997). The DNA sequences
of the available infectious CFV\hu molecular clones start with
TGTG (Fig. 1 A) (Maurer et al., 1988 ; Rethwilm et al., 1990 ;
Lo$ chelt et al., 1991 ; Schmidt et al., 1997 a). The 5h end of these
clones was originally derived from one provirus which was
integrated into another CFV\hu DNA copy (Rethwilm et al.,
1987). Since the U3 region of CFV\hu also starts with TGTG
(Fig. 1 A), the integration of this single provirus reflects an
error in integration or unorthodox FV integration. Specifically,
FVs could be unlike other retroviruses, because their integration does not involve removal of the two terminal
nucleotides of its U3 region.
We therefore wished to clarify this matter by sequence
determination of cellular DNA–U3 junctions of independent
single FV proviruses obtained following transduction of BHK21 cells with a replication-deficient CFV\hu vector (Fig. 1 B).
As shown in Fig. 1 (A), all sequenced CFV\hu proviruses
started with TGTG. Hence, the terminal dinucleotide of U3
was not removed during integration. In contrast, the sequence
determination of the U5–cellular DNA junctions revealed, in
all cases except clone F2 (Fig. 1 A), that the two terminal
nucleotides (AT) had been removed in order to provide the
subterminal CA for the joining reaction to host cell DNA. In
the case of provirus F2 the terminal AT dinucleotide was not
removed from the right end of the unintegrated viral DNA,
and this was probably reflected by only partial duplication of
host cell sequences (Fig. 1 A).
These findings on CFV\hu integration were corroborated
by sequence determination of the provirus–cellular DNA
junctions in Vero-L cells which harbour a single infectious SFV3 provirus (Schweizer et al., 1993). This provirus also starts
with TGTG and terminates with CA (Fig. 1 A). Taken together,
our findings indicate that removal of two nucleotides from the
U5 end of the viral DNA is common in FV integration, while
the trimming of the U3 end appears to be unusual. The
conservation of the ppt–U3 sequence in FVs (Fig. 1 A) strongly
suggests that our finding does not only apply to CFV\hu and
SFV-3, but to all FVs. Furthermore, this conservation among
the different FV isolates and the infectiousness of several fulllength molecular clones also rules out the possibility that
defective genomes may have been cloned and sequenced
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07
BEEH
J. Enssle and others
Fig. 2. Construction and properties of FV
IN mutants. (A) Genome organization of
FV and mutants in the pHSRV2 and
pcHSRV2 backbone. (B) Gag and Pol
protein expression of pHSRV2 and M70
and M73. (C) Southern blot analysis to
demonstrate the ability of the DD35E
mutant M73 to generate cDNA. The lanes
from transfected 293T cells were
exposed for 3 days, and the control lanes
(DNA from untransfected 293T and DNA
from 293T cells spiked with pcHSRV2
plasmid) were exposed for 14 days. (D)
Detection of viral DNA in recipient BHK21 cells exposed to virus produced by
transient transfection of 293T cells with
individual proviral plasmids (pcHSRV2
background).
(Rethwilm et al., 1990 ; Lo$ chelt et al., 1991 ; Renshaw et al.,
1991 ; Herchenro$ der et al., 1996 ; Tobaly-Tapiero et al., 1996 ;
Schmidt et al., 1997 a ; Mergia & Wu, 1998).
As also shown in Fig. 1 (A), FV integration apparently
resulted in the duplication of four nucleotides of host cell
DNA, as has been observed for feline and murine leukaemia
virus integration (Brown, 1997). In the limited cases analysed
GjC-rich sequences appeared to be the preferred target for
FV integration, which may resemble previous observations
with avian retrovirus IN protein (Kitamura et al., 1992). These
results corroborate findings published recently (Neves et al.,
1998).
Construction and characterization of FV IN mutants
Provirus integration is required for retroviral gene expression and replication (Brown, 1997). To determine whether
an active IN protein is essential for FV replication, two FV IN
BEEI
mutants were analysed (Fig. 2 A). The mutant M70 contains an
isoleucine to threonine substitution (DYIG DYTG) around
the first aspartic acid of the DD35E motif. This mutant was
described previously as having a reduced endonuclease, a
slightly elevated disintegrase and a largely impaired IN activity
(reduction to less than 3 % of wild-type activity) in in vitro
assays (Pahl & Flu$ gel, 1995). The mutant M73 contains an
aspartic acid to alanine substitution (DD35E DA35E) in the
highly conserved catalytic centre present in all known IN
proteins (Brown, 1997). Mutations introduced into this motif
have previously been demonstrated to disable the integration
and replication activity of a variety of retroviruses (Brown,
1997).
The expression of viral Gag and Pol proteins was analysed
following transient transfection of proviral (pHSRV2 background) constructs into BHK\bel-1 cells which stably express
the CFV\hu transcriptional trans-activator protein (Tas). As
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07
Foamy virus integration
Table 1. Development of cell-free virus titres/ml following transfection of BHK-21 and
KMST-6 cells with pHSRV2 or mutants
Plasmid DNA (5 µg) was transfected into the cells and cell-free virus titres were determined on BHK\LTRlacZ
cells. The results represent the mean of two experiments with less than 20 % variation.
BHK-21
KMST-6
Days posttransfection
pHSRV2
M70
M73
pHSRV2
M70
3
5
7
9
11
3n65i10#
6n45i10$
7n20i10%
2n90i10&
2n40i10'
60
1n50i10$
1n50i10%
6n30i10%
3n10i10&
0
0
0
0
0
7
1n55i10#
2n25i10#
7n00i10#
3n40i10%
0
9n75
32
1n90i10#
1n10i10$
shown in Fig. 2 (B), wild-type virus and the mutants expressed
the p70\74 kDa Gag protein doublet, the p127 kDa Pol
precursor protein and the p80 kDa RT. This demonstrated that
gene expression from the proviral plasmids and the function of
the pol-encoded protease in cleaving Gag and Pol precursor
molecules (Konvalinka et al., 1995) were not affected by the
mutations introduced.
We also investigated whether IN mutant M73 is able to
generate cDNA. It has been demonstrated previously that FV
reverse transcribes its RNA pregenome late in the replication
cycle (Moebes et al., 1997). The generation of the DNA
required an active polymerase and a correctly assembled capsid
(Enssle et al., 1997 ; Moebes et al., 1997). To analyse this for
M73, 293T cells were transfected with pcHSRV2, pcHSRV2\
M68, in which env was inactivated, and pcHSRV2\M73. DNA
from the transfected cells was probed for the presence of linear
HSRV2 DNA (Moebes et al., 1997). As shown in Fig. 2 (C),
M73 was found as able to synthesize cDNA as the unmutated
virus. This result indicated that the replication of M73 up to
this step was normal.
The delivery of viral DNA derived from M73 transfection
into recipient cells was analysed by PCR. As shown in Fig.
2 (D), we detected M73 DNA in the Hirt fraction of recipient
BHK-21 cells, which indicated that IN mutant M73 was able to
deliver virus DNA to fresh cells.
Analysis of the replication competence of FV IN
mutants
The replication of the proviral mutants over time was
analysed by transfection of BHK-21 and KMST-6 cells and
determination of the virus titres in the cell-free supernatants.
Wild-type virus replicated an average 5–10 times better
compared to the M70 mutant (Table 1). In addition, cell-free
infectious virus could be continuously transmitted to fresh cells
from cultures infected with M70 (data not shown). It has been
reported that even a replication-deficient human immunodeficiency virus (HIV) IN mutant reverted at a second site to
U3
GFP
Fig. 3. Analysis of FV IN mutants in a single replication cycle. (A) Genome
structure of the wild-type and mutant FV vectors. Transient expression of
the proviral sequence is directed by the human cytomegalovirus
immediate early enhancer/promoter and expression of the internal GFP
cassette is directed by the heterologous SFFV U3 region (Heinkelein et al.,
1998). The plasmids were transfected into 293T cells along with the Env
expression vector pCenv-1 and the supernatant was used to transduce
recipient cells (Heinkelein et al., 1998). (B) Transduction rates of BHK-21
and KMST-6 cells with the pMH vectors. Compared with wild-type pMH5
vector, pMH99 (DYIG DYTG) led to a reduction of approximately
20–50 % in transducing KMST-6 and BHK-21 cells, respectively. No
significant transduction of recipient cells was detected with the pMH5
vector, which was transfected into 293T cells in the absence of the Env
expression vector, and no or only a very low transduction was observed
with pMH80 (DD35E DA35E). The mean results of two experiments
are shown. d.p.t., days post-transduction.
regenerate a replication-competent virus (Taddeo et al., 1996).
We, therefore, amplified a 1n73 kb fragment spanning the
complete IN gene from BHK-21 cells infected with HSRV2\
M70 for 3 weeks. The DNA sequence of the IN gene was
determined directly without further subcloning steps and
revealed an identical nucleotide sequence as present in the
original M70 mutant.
In contrast to M70 the M73 IN mutant did not replicate at
all in BHK-21 cells (Table 1). A similar result was obtained
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07
BEEJ
J. Enssle and others
when BHK\bel-1 cells were followed for a period of 4 weeks
following transfection with pHSRV2\M73 (data not shown).
However, when pcHSRV2 and pcHSRV2\M73 were transfected into 293T cells and the amounts of infectious virus were
determined after transient production for 48 h, we obtained
from five experiments mean cell-free titres of 3n1i10%\ml and
26\ml for pcHSRV2 and pcHSRV2\M73, respectively.
Analysis of FV IN mutants in a single replication assay
To determine the effect of the IN mutants in a single
replication assay, we took advantage of indicator-geneexpressing FV vectors. Similar analyses have been performed
previously using HIV-1 IN mutants (Masuda et al., 1995 ;
Wiskerchen & Muesing, 1995 ; Liu et al., 1997). To do this we
used vectors pMH99 and pMH80, shown in Fig. 3. BHK-21
and KMST-6 cells were transduced with cell-free FV vectors,
which were produced by transient co-transfection of vector
DNA and an Env expression plasmid, and the percentage of
GFP-positive cells was monitored over time. Wild-type pMH5
vector transduced recipient cells to a similar extent, as
described recently (Heinkelein et al., 1998), while only very
low, if any, transduction was obtained with pMH80 (Fig. 3).
Transduction following transfection of the pMH99 construct
showed the GFP expression level to be reduced by 20–50 %
compared with the pMH5 wild-type vector (Fig. 3), which was
in good agreement with the result of the infection experiment
with this mutant (Table 1).
Discussion
We demonstrate in this study that FV, which is highly
divergent from all other retroviruses, essentially requires an
active IN for replication. Episomal replication of FVs, which
could have been assumed following the discovery of a variety
of analogies between FVs and hepadnaviruses appears to be
negligible or non-existent. The few blue cells observed in our
most sensitive assay (infection of BHK\LTRlacZ indicator cells
with transiently produced virus in 293T cells) indicated that
either there exists a very weak residual activity that expresses
FV tas from unintegrated DNA or that the FV DNA may
express genes from occasional and unspecific integrations into
the host cell genome. A very weak replication of IN-mutated
retroviruses and unspecific integration of such mutants has
been reported previously (Panganiban & Temin, 1984 ;
Schwartzberg et al., 1984 ; Hagino-Yamagashi et al., 1987 ;
Wiskerchen & Muesing, 1995 ; Gaur & Leavitt, 1998).
With respect to the M70 mutation (DYIG DYTG) our
results indicate that this mutant, in contrast to what has been
expected from in vitro integration assays (Pahl & Flu$ gel, 1995),
behaves in a very similar way to unmutated virus. Except for
FVs the threonine (in HIV-1 IN at position 66) is highly
conserved in retroviral IN proteins (Engelman & Craigie,
1992). Substitution of the HIV-1 IN threonine at position 66
by alanine, while having moderate effects on in vitro IN
activities, had no effect on virus replication in an established TBEFA
cell line (Engelman & Craigie, 1992 ; Taddeo et al., 1994). We
show here that mutation of the isoleucine, which is in the
similar position in FVs compared with the threonine in other
retroviral IN proteins (Engelman & Craigie, 1992), also had
only a moderate effect on virus replication, provided that it
was mutated to threonine.
Another part of our study highlighted a further difference
between FVs and the other retroviruses. Our sequencing of
individual proviruses revealed that their left ends were identical
to what is believed to be the start of the U3 region (Flu$ gel et
al., 1987), in contrast to other retroviruses, in which two bases
from U3 are lost during integration (Fig. 1 ; Brown, 1997). This
finding suggests that either the left end of the unintegrated FV
DNA remains unprocessed during integration or that the
processing is unusual and involves cleavage of RNA–DNA
hybrid sequences originating from the ppt and remaining
attached to the U3 end of the linear viral DNA. Such untidy
ends have been found previously in other retroviral systems
and in particular in the yeast retrotransposons (Miller et al.,
1997 ; Mules et al., 1998). Remarkably, the latter preferentially
integrate blunt-end substrates (Brown, 1997).
A high percentage of uncleaved proviral right and left ends
has been recently claimed for visna virus integrates (List &
Haase, 1997). However, the reported sequences were not
derived from single proviruses, which makes interpretation of
the findings difficult. Furthermore, as deduced from the primary
sequences, visna virus allows and probably requires cleavage
of the terminal nucleotides from both ends of linear DNA for
efficient integration (Sonigo et al., 1985).
There is a major discrepancy between our results and a
previous study on the in vitro activity of purified recombinant
FV IN. Pahl & Flu$ gel (1993) reported that CFV\hu IN was able
to cleave the end of what they called a ‘ wild-type ’ U3
oligonucleotide. However, to obtain this result, two nontemplated nucleotides, which were complementary to the
terminal U5 dinucleotide, were added to the U3 substrate (Pahl
& Flu$ gel, 1993). In addition, their introduction would destroy
the bet reading frame overlapping the ppt–U3 border in all FVs
(Flu$ gel et al., 1987 ; Kupiec et al., 1991 ; Renne et al., 1992 ;
Baunach et al., 1993 ; Hahn et al., 1994 ; Renshaw & Casey,
1994 b ; Herchenro$ der et al., 1994 ; Helps & Harbour, 1997 ;
Winkler et al., 1997). Therefore, we suggest that the report of
Pahl & Flu$ gel (1993) showed that FV IN is able to cleave some
kind of elongated U3 end, which may arise through the
generation of untidy ends during reverse transcription.
However, such a substrate, which remains to be identified,
would probably be different from the one used in the study in
question (Pahl & Flu$ gel, 1993).
We thank Anke Marxer and Dirk Lindemann for invaluable help and
Rolf Flu$ gel for providing reagents.
This work was supported by grants from Bayerische Forschungsstiftung, BMBF (01KV9817\0), DFG (SFB 165 and Re 627\6-1) and EU
(BMH4-CT97-2010).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07
Foamy virus integration
References
Achong, B. G., Mansell, P. W. A., Epstein, M. A. & Clifford, P. (1971).
An unusual virus in cultures from a human nasopharyngeal carcinoma.
Journal of the National Cancer Institute 46, 299–307.
Ali, M., Taylor, G. P., Pitman, R. J., Parker, D., Rethwilm, A.,
Cheingsong-Popov, R., Weber, J. N., Bieniasz, P. D., Bradley, J. &
McClure, M. O. (1996). No evidence of antibody to human foamy virus
sequence in the pol region required to transfer human foamy virus
vectors. Journal of Virology 72, 6307–6314.
Helps, C. R. & Harbour, D. A. (1997). Comparison of the complete
sequence of feline spumavirus with those of the primate spumaviruses
reveals a shorter gag gene. Journal of General Virology 78, 2549–2564.
Herchenro$ der, O., Renne, R., Loncar, D., Cobb, E. K., Murthy, K. K.,
Schneider, J., Mergia, A. & Luciw, P. A. (1994). Isolation, cloning, and
in widespread human populations. AIDS Research & Human Retroviruses
12, 1473–1483.
sequencing of simian foamy viruses from chimpanzees (SFVcpz) : high
homology to human foamy virus (HFV). Virology 201, 187–199.
Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D., Seidman, J. G.,
Smith, J. A. & Struhl, K. (1987). Current Protocols in Molecular Biology.
Herchenro$ der, O., Turek, R., Neumann-Haefelin, D., Rethwilm, A. &
Schneider, J. (1996). Infectious proviral molecular clones of foamy virus
New York : John Wiley.
Baldwin, D. N. & Linial, M. L. (1998). The roles of pol and env in the
assembly pathway of human foamy virus. Journal of Virology 72,
3658–3665.
Baunach, G., Maurer, B., Hahn, H., Kranz, M. & Rethwilm, A. (1993).
Functional analysis of human foamy virus accessory reading frames.
Journal of Virology 67, 5411–5418.
Bieniasz, P. D., Erlwein, O., Aguzzi, A., Rethwilm, A. & McClure, M. O.
(1997). Gene transfer using replication-defective human foamy virus
vectors. Virology 235, 65–72.
Brown, P. O. (1997). Integration. In Retroviruses, pp. 161–203. Edited by
J. M. Coffin, S. H. Hughes & H. E. Varmus. Cold Spring Harbor, NY :
Cold Spring Harbor Laboratory Press.
Coffin, J. M. (1996). Retroviridae : the viruses and their replication. In
Virology, 3rd edn, pp. 1767–1847. Edited by B. N. Fields, D. M. Knipe &
P. M. Howley. Philadelphia : Lippincott–Raven.
DuBridge, R. B., Tang, P., Hsia, H. C., Leong, P.-M., Miller, J. H. &
Calos, M. P. (1987). Analysis of mutation in human cells by using
Epstein–Barr virus shuttle system. Molecular and Cellular Biology 7,
379–387.
Engelman, A. & Craigie, R. (1992). Identification of conserved amino
acid residues critical for human immunodeficiency virus type 1 integrase
function in vitro. Journal of Virology 66, 6361–6369.
Enssle, J., Fischer, N., Moebes, A., Mauer, B., Smola, U. & Rethwilm, A.
(1997). Carboxy-terminal cleavage of the human foamy virus gag
precursor molecule is an essential step in the viral life cycle. Journal of
Virology 71, 7312–7317.
Fischer, N., Heinkelein, M., Lindemann, D., Enssle, J., Baum, C.,
Werder, E., Zentgraf, H., Mu$ ller, J. G. & Rethwilm, A. (1998). Foamy
virus particle formation. Journal of Virology 72, 1610–1615.
Flu$ gel, R. M., Rethwilm, A., Maurer, B. & Darai, G. (1987). Nucleotide
sequence analysis of the env gene and its flanking regions of the human
spumaretrovirus reveals two novel genes. EMBO Journal 6, 2077–2084.
Gaur, M. & Leavitt, A. D. (1998). Mutations in the human immunodeficiency virus type 1 integrase D,D(35)E motif do not eliminate
provirus formation. Journal of Virology 72, 4678–4685.
Goff, S. P. (1992). Genetics of retroviral integration. Annual Review in
Genetics 26, 527–544.
Hagino-Yamagashi, K., Donehower, L. A. & Varmus, H. E. (1987).
Retroviral DNA integrated during infection by an integration-deficient
mutant of murine leukemia virus is oligomeric. Journal of Virology 61,
1964–1971.
Hahn, H., Baunach, G., Bra$ utigam, S., Mergia, A., Neumann-Haefelin,
D., Daniel, M. D., McClure, M. O. & Rethwilm, A. (1994). Reactivity of
primate sera to foamy virus Gag and Bet proteins. Journal of General
Virology 75, 2635–2644.
Heinkelein, M., Schmidt, M., Fischer, N., Moebes, A., Lindemann, D.,
Enssle, J. & Rethwilm, A. (1998). Characterization of a cis-acting
from chimpanzee (SFVcpz) generated by ‘ long PCR ’ provide evidence
for functional relationship with human foamy virus (HFV). Virology 214,
685–689.
Higuchi, R. (1990). Recombinant PCR. In PCR Protocols : A Guide to
Methods and Applications. Edited by M. A. Innis, D. H. Gelfand & T. J.
White. San Diego, CA : Academic Press.
Hirt, B. (1967). Selective extraction of polyoma DNA from infected
mouse cell cultures. Journal of Molecular Biology 26, 365–369.
Kitamura, Y., Lee, Y. M. H. & Coffin, J. M. (1992). Nonrandom
integration of retroviral DNA in vitro : effect of CpG methylation.
Proceedings of the National Academy of Sciences, USA 89, 5532–5536.
Ko$ gel, D., Aboud, M. & Flu$ gel, R. M. (1995). Molecular biological
characterization of the human foamy virus reverse transcriptase and
ribonuclease H domains. Virology 213, 97–108.
Konvalinka, J., Lo$ chelt, M., Zentgraf, H., Flu$ gel, R. M. & Kra$ usslich,
H.-G. (1995). Active foamy virus proteinase is essential for virus
infectivity but not for formation of a pol polyprotein. Journal of Virology
69, 7264–7268.
Kupiec, J. J., Kay, A., Hayat, M., Ravier, R., Peries, J. & Galibert, F.
(1991). Sequence analysis of the simian foamy virus type 1 genome.
Gene 101, 185–194.
Lindemann, D. & Rethwilm, A. (1998). Characterization of a human
foamy virus 170 kD env–bet fusion protein generated by alternative
splicing. Journal of Virology 72, 4088–4094.
Lindemann, D., Bock, M., Schweizer, M. & Rethwilm, A. (1997).
Efficient pseudotyping of murine leukemia virus particles with chimeric
human foamy virus envelope proteins. Journal of Virology 71, 4815–4820.
Linial, M. L. (1999). Foamy viruses are nonconventional retroviruses.
Journal of Virology 73 (in press).
List, J. & Haase, A. T. (1997). Integration of visna virus DNA occurs and
may be necessary for productive infection. Virology 237, 189–197.
Liu, H., Wu, X., Xiao, H., Conway, J. A. & Kappes, J. C. (1997).
Incorporation of functional human immunodeficiency virus type 1
integrase into virions independent of the gag–pol precursor protein.
Journal of Virology 71, 7704–7710.
Lo$ chelt, M., Zentgraf, H. & Flu$ gel, R. M. (1991). Construction of an
infectious DNA clone of the full-length human spumaretrovirus genome
and mutagenesis of the bel 1 gene. Virology 184, 43–54.
Masuda, T., Planelles, V., Krogstad, P. & Chen, I. S. Y. (1995). Genetic
analysis of human immunodeficiency virus type 1 integrase and the U3
att site : unusual phenotype of mutants in the zinc finger-like domain.
Journal of Virology 69, 6687–6696.
Maurer, B., Bannert, H., Darai, G. & Flu$ gel, R. M. (1988). Analysis of
the primary structure of the long terminal repeat and the gag and pol
genes of the human spumaretrovirus. Journal of Virology 62, 1590–1597.
Mergia, A. & Wu, M. (1998). Characterization of provirus clones of
simian foamy virus type 1. Journal of Virology 72, 817–822.
Miller, M. D., Farnet, C. M. & Bushman, F. D. (1997). Human
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07
BEFB
J. Enssle and others
immunodeficiency virus type 1 preintegration complexes : studies of
organization and composition. Journal of Virology 71, 5382–5390.
Long terminal repeat U3 length polymorphism of human foamy virus.
Virology 230, 167–178.
Moebes, A., Enssle, J., Bieniasz, P. D., Heinkelein, M., Lindemann, D.,
Bock, M., McClure, M. O. & Rethwilm, A. (1997). Human foamy virus
Schmidt, M., Niewiesk, S., Heeney, J., Aguzzi, A. & Rethwilm, A.
(1997 b). Mouse model to study the replication of primate foamy
reverse transcription that occurs late in the viral replication cycle. Journal
of Virology 71, 7305–7311.
Mules, E. H., Uzun, O. & Gabriel, A. (1998). In vivo Ty1 reverse
transcription can generate replication intermediates with untidy ends.
Journal of Virology 2, 6490–6503.
viruses. Journal of General Virology 78, 1929–1933.
Namba, M., Nishitani, K., Hyodoh, F., Fukushima, F. & Kimoto, T.
(1985). Neoplastic transformation of human diploid fibroblasts (KMST-
6) by treatment with '!Co gamma rays. International Journal of Cancer 35,
275–280.
Neves, M., Pe! rie' s, J. & Saib, A. (1998). Study of human foamy virus
proviral integration in chronically infected murine cells. Research in
Virology 149, 393–401.
Pahl, A. & Flu$ gel, R. M. (1993). Endonucleolytic cleavages and DNAjoining activities of the integration protein of human foamy virus. Journal
of Virology 67, 5426–5434.
Pahl, A. & Flu$ gel, R. M. (1995). Characterization of the human spuma
retrovirus integrase by site-directed mutagenesis, by complementation
analysis, and by swapping the zinc finger domain of HIV-1. Journal of
Biological Chemistry 270, 2957–2966.
Panganiban, A. T. & Temin, H. M. (1984). The retrovirus pol gene
encodes a product required for DNA integration : identification of a
retrovirus int locus. Proceedings of the National Academy of Sciences, USA
81, 7885–7889.
Pietschmann, T., Heinkelein, M., Zentgraf, H., Rethwilm, A. &
Lindemann, D. (1999). Foamy virus capsids require the cognate
envelope protein for particle export. Journal of Virology 73, 2613–2621.
Renne, R., Friedl, E., Schweizer, M., Fleps, U., Turek, R. & NeumannHaefelin, D. (1992). Genomic organisation and expression of simian
foamy virus type 3 (SFV-3). Virology 18, 597–608.
Renshaw, R. W. & Casey, J. W. (1994 a). Analysis of the 5h long terminal
repeat of bovine syncytial virus. Gene 141, 221–224.
Renshaw, R. W. & Casey, J. W. (1994 b). Transcriptional mapping of the
3h end of the bovine syncytial virus genome. Journal of Virology 68,
1021–1028.
Renshaw, W. R., Gonda, M. A. & Casey, J. W. (1991). Structure and
transcriptional status of bovine syncytial virus in cytopathic infections.
Gene 105, 179–184.
Rethwilm, A. (1995). Regulation of foamy virus gene expression. Current
Topics in Microbiology and Immunology 193, 1–23.
Rethwilm, A. (1996). Unexpected replication pathways of foamy viruses.
Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology
13 (Suppl. 1), S248–S253.
Rethwilm, A., Darai, G., Ro$ sen, A. & Flu$ gel, R. M. (1987). Molecular
cloning of the genome of human spumaretrovirus. Gene 59, 19–28.
Rethwilm, A., Baunach, G., Netzer, K. O., Maurer, B., Borisch, B. &
Meulen, V. T. (1990). Infectious DNA of the human spumaretrovirus.
Nucleic Acids Research 18, 733–738.
Russel, D. W. & Miller, A. D. (1996). Foamy virus vectors. Journal of
Virology 70, 217–222.
Schmidt, M. & Rethwilm, A. (1995). Replicating foamy virus-based
vectors directing high level expression of foreign genes. Virology 210,
167–178.
Schmidt, M., Herchenro$ der, O., Heeney, J. & Rethwilm, A. (1997 a).
BEFC
Schwartzberg, P., Colicelli, J. & Goff, S. P. (1984). Construction and
analysis of deletion mutations in the pol gene of Moloney murine
leukemia virus : a new viral function required for productive infection. Cell
37, 1043–1052.
Schweizer, M., Fleps, U., Ja$ ckle, A., Renne, R., Turek, R. & NeumannHaefelin, D. (1993). Simian foamy virus type 3 (SFV-3) in latently
infected Vero cells : reactivation by demethylation of proviral DNA.
Virology 192, 663–666.
Schweizer, M., Turek, R., Hahn, H., Schliephake, A., Netzer, K.-O.,
Eder, G., Reinhardt, M., Rethwilm, A. & Neumann-Haefelin, D. (1995).
Markers of foamy virus (FV) infections in monkeys, apes, and accidentally
infected humans : appropriate testing fails to confirm suspected FV
prevalence in man. AIDS Research and Human Retroviruses 11, 161–170.
Silver, J. & Keerikatte, V. (1989). Novel use of polymerase chain
reaction to amplify cellular DNA adjacent to an integrated provirus.
Journal of Virology 63, 1924–1928.
Sonigo, P., Alizon, M., Staskus, K., Klatzmann, D., Cole, S., Danos, O.,
Retzel, E., Tiollais, P., Haase, A. & Wain-Hobson, S. (1985). Nucleotide
sequence of the visna lentivirus : relationship to the AIDS virus. Cell 42,
369–382.
Taddeo, B., Haseltine, W. A. & Farnet, C. M. (1994). Integrase mutants
of human immunodeficiency virus type 1 with a specific defect in
integration. Journal of Virology 68, 8401–8405.
Taddeo, B., Carlini, F., Verani, P. & Engelman, A. (1996). Reversion of
a human immunodeficiency virus type 1 integrase mutant at a second site
restores enzyme function and virus infectivity. Journal of Virology 70,
8277–8284.
Tobaly-Tapiero, J., Celis-Kosmas, J. D., Bittoun, P., Lasneret, J.,
Poorters, A. M., Eladari, M. E. & Emanoil-Ravier, R. (1996). Isolation
and characterization of infectious full-length DNA clones of chimpanzee
foamy viruses SFV6 and SFV7 : evidence for a taf-dependent internal
promoter. Research in Virology 147, 17–27.
Triglia, T., Peterson, M. G. & Kemp, D. J. (1988). A procedure for in
vitro amplification of DNA segments that lie outside the boundaries of
known sequences. Nucleic Acids Research 16, 8186.
Weiss, R. A. (1996). Foamy viruses bubble on. Nature 380, 201.
Whitcomb, J. M., Kumar, R. & Hughes, S. H. (1990). Sequence of the
circle junction of human immunodeficiency virus type 1 : implications for
reverse transcription and integration. Journal of Virology 64, 4903–4906.
Winkler, I., Bodem, J., Haas, L., Zemba, M., Delius, H., Flower, R.,
Flu$ gel, R. M. & Lo$ chelt, M. (1997). Characterization of the genome of
feline foamy virus and its proteins shows features different from those of
primate spumaviruses. Journal of Virology 71, 6727–6741.
Wiskerchen, M. & Muesing, M. A. (1995). Human immunodeficiency
virus type 1 integrase : effects of mutations on viral ability to integrate,
direct viral gene expression from unintegrated viral DNA templates, and
sustain viral propagation in primary cells. Journal of Virology 69, 376–386.
Yu, S. F., Baldwin, D. N., Gwynn, S. R., Yendapalli, S. & Linial, M. L.
(1996). Human foamy virus replication : a pathway distinct from that of
retroviruses and hepadnaviruses. Science 271, 1579–1582.
Yu, S. F., Sullivan, M. D. & Linial, M. L. (1999). Evidence that the
human foamy virus genome is DNA. Journal of Virology 73 (in press).
Received 9 February 1999 ; Accepted 19 February 1999
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 05 May 2017 05:49:07