Journal
of General Virology (2002), 83, 2523–2532. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
Characterization of a human immunodeficiency virus type 1
pre-integration complex in which the majority of the cDNA is
resistant to DNase I digestion
Dheeraj K. Khiytani and Nigel J. Dimmock
Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK
The human immunodeficiency virus type 1 (HIV-1) pre-integration complex (PIC) is a cytoplasmic
nucleoprotein structure derived from the core of the virion and is responsible for reverse
transcription of viral RNA to cDNA, transport to the nucleus and integration of the cDNA into the
genome of the infected target cell. Others have shown by Mu phage-mediated PCR footprinting
that only the LTRs of the cDNA of PICs isolated early in infection are protected by bound protein,
while the rest of the genome is susceptible to nuclease attack. Here, using DNase I footprinting, we
confirmed that the majority of the cDNA of PICs isolated at 8n5 h after infection with cell-free virus
was sensitive to digestion with DNase I and that only part of the LTRs (approximately 6 % of the
total cDNA) was protected. However, PICs isolated 90 min later (at 10 h post-infection) were very
different in that the majority (approximately 90 %) of cDNA was protected from nuclease
degradation. These late PICs were integration active in vitro. We conclude that HIV-1 has at least
two types of PIC, an early PIC characterized by protein bound only at the LTRs, and a late, and
possibly more mature form, in which protein is bound along the length of the cDNA.
Introduction
The human immunodeficiency virus type 1 (HIV-1) virion
is bounded by a lipid bilayer into which is inserted the
envelope protein, a homotrimer of gp120–gp41, and a variety
of host proteins. Internal to this is the matrix (MA) protein,
which surrounds the capsid (CA) protein core structure. In turn
this encloses two identical single-stranded, plus-sense viral
RNA strands, associated with nucleocapsid (NC), reverse
transcriptase (RT) and integrase (IN) proteins. Closely associated with the core are the Vif and Nef proteins. The viral
accessory protein, viral protein R (Vpr), is also found within the
virion but is most probably located outside the core. Tat may
also be located inside virions. Infection is initiated by gp120
attaching to specific target cells via a primary receptor, the
CD4 protein and a co-receptor, usually CCR5 or CXCR4 (for
a review, see Levy, 1998). As a result of these interactions the
viral lipid bilayer fuses with that of the plasma membrane.
Following fusion, it is hypothesized that the virus core
develops into an immature reverse transcription complex
Author for correspondence : Nigel Dimmock.
Fax j44 2476 523568. e-mail ndimmock!bio.warwick.ac.uk
0001-8461 # 2002 SGM
(Zennou et al., 2000), and in turn this matures into a better
defined nucleoprotein structure, termed the pre-integration
complex (PIC) (Farnet & Haseltine, 1990 ; Karageorgos et al.,
1993). These are present in the cytoplasm. PICs were first
observed in experiments with murine leukaemia virus (MLV)
(Bowerman et al., 1989 ; Farnet & Haseltine, 1991 ; Fujiwara &
Mizuuchi, 1988) when detection of viral cDNA was combined
with a functional assay for integration. HIV-1 PICs are derived
from the virion core structures and contain IN, RT, Vpr, the
cellular high mobility group protein HMG 1(Y) and a reduced
amount of MA (Bukrinsky et al., 1993b ; Ellison & Brown,
1994 ; Farnet & Bushman, 1997 ; Farnet & Haseltine, 1991 ;
Gallay et al., 1995b ; Karageorgos et al., 1993 ; Miller et al.,
1997). All these proteins are associated with the viral
RNA\cDNA. It is not clear if other internal virion proteins are
present. PICs sediment on sucrose velocity gradients between
160S and 640S after treatment with RNase (Bowerman et al.,
1989 ; Farnet & Haseltine, 1990 ; Karageorgos et al., 1993 ;
Miller et al., 1997).
The RT protein associated with the PIC reverse transcribes
the virion RNA (for a review, see Brown, 1997). PICs are also
responsible for transport of newly synthesized viral cDNA
into the nucleus and integration of the cDNA into the target
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
CFCD
D. K. Khiytani and N. J. Dimmock
cell genome (Bukrinsky et al., 1992 ; Bushman et al., 1990). Vpr
has been implicated in the targeting of PICs to the host cell
nucleus but has not been shown directly to be part of the
complex (Gallay et al., 1996 ; Heinzinger et al., 1994 ; Jenkins et
al., 1998 ; Karni et al., 1998 ; Popov et al., 1998a, b ; Vodicka et
al., 1998 ; Zhang et al., 1998). MA is also thought to contribute
to nuclear targeting of the viral cDNA (Bukrinsky et al.,
1993a ; Gallay et al., 1995b, 1996 ; Heinzinger et al., 1994 ;
Popov et al., 1998b ; von Swedler et al., 1994), although there
is also evidence that this may not be the case (Fouchier et al.,
1997). Phosphorylation of MA may be important for its
karyophilic properties (Bukrinskaya et al., 1996 ; Gallay et al.,
1995a, b). The viral IN also appears to have a role in nuclear
targeting (Gallay et al., 1997 ; Pluymers et al., 1999) and
possibly reverse transcription (Wu et al., 1999), in addition to
its major activity of integrating viral cDNA into the host cell
genome (Bushman et al., 1990 ; Ellison & Brown, 1994 ; Farnet
& Haseltine, 1990). Recently, it has been shown that a central
DNA flap (a 99 nucleotide plus-strand overlap created during
HIV-1 reverse transcription at the boundary at which left- and
right-hand segments of nascent plus-strand cDNA merge), acts
as a cis-determinant of HIV-1 DNA nuclear import (Zennou et
al., 2000).
Much has been done to determine the protein composition
and in vitro function of PICs through the use of immunoprecipitation, integration assays, nuclear import assays and, to a
certain extent, nuclease protection assays. However, less is
known about the formation and maturation of PICs especially
in regard to nuclear targeting, translocation and integration of
viral cDNA with the host genome. Although the exact
conformation of PICs is not well understood, experiments have
shown that the ends of the cDNA may be joined through
dimerization by the viral IN protein or the cell HMG 1(Y)
protein (Ellison & Brown, 1994 ; Farnet & Bushman, 1997 ;
Miller et al., 1997). Recently, the protein–DNA structure of
PICs partially purified from HIV-1-infected cells at 8n5 h after
infection has been analysed by Mu-mediated PCR footprinting
(Chen et al., 1999). The HIV-1 cDNA termini (LTRs), but not
the rest of the genome, were protected from nuclease attack by
bound protein. The termini form a unique structure, resembling
the ends of MLV PIC DNA (Wei et al., 1997), which suggests
that this may be a common feature of retrovirus PIC DNAs.
The consensus thus far is that HIV-1 PICs contain
condensed cDNA in a complex resembling a partially dissociated viral core in which proteins are tightly associated with the
ends of the cDNA but only loosely associated, if at all, with the
intervening sequence (Chen et al., 1999 ; Farnet & Bushman,
1997). The work described in this report found an almost
identical structure at 8n5 h post-infection using the different
technique of DNase I footprinting (protease digestion followed
by DNase digestion). However, PICs isolated in exactly the
same way at 10 h post-infection were almost completely
resistant to DNase I digestion. These late PICs were active and
integrated HIV-1 cDNA into phage DNA in vitro. Thus late
CFCE
PICs have a complex protein–DNA structure, are functionally
integrative and may represent a more mature structure than
PICs present at an earlier time.
Methods
Cells, viruses and infection. C8166 (Salahuddin et al., 1983) and
H9 (Popovic et al., 1984) CD4+ T cell lines were obtained via the
Centralized Facility for AIDS Reagents, NIBSC, Potters Bar, UK, and
maintained in RPMI 1640 medium (Gibco BRL Life Technologies)
containing 10 % foetal calf serum (FCS ; Helena BioSciences) and 2 mM glutamine (Gibco). Virus was prepared by co-cultivating H9 cells
chronically infected with HIV-1 IIIB with non-infected H9 cells, at a ratio
of 1 : 4. After 2 days the tissue culture fluid was harvested, clarified and
virus concentrated by pelleting through 20 % sucrose. Infectivity was determined by syncytium formation on C8166 cells (McLain & Dimmock,
1994).
Antibodies. We used the following antibodies : mAb H12-5C to the
HIV-1 CA protein (B. Chesebro and K. Wherly) ; rabbit antiserum to Vpr
(V. Ayyavoo and D. Weiner) – both obtained from the NIH AIDS
Research and Reference Reagent Program, Rockville, USA ; mAb 4H2B1
to MA (R. B. Ferns and R. B. Tedder) ; mAb IIG10E6 to RT (D. Helland
and A. M. Szilvay) ; rabbit antiserum to IN (S. Ranjbar) – all obtained from
the Centralized Facility for AIDS Reagents.
Preparation of PICs. PICs were prepared essentially as described
previously (Farnet & Haseltine, 1990). Infection was initiated by mixing
10) syncytium-forming units of cell-free virus with 5i10( C8166 cells in
3 ml medium at 37 mC. After 5 h, 5 ml of fresh medium was added. Cells
harvested at the required time were washed twice in buffer K (20 mM
HEPES, pH 7n4, 5 mM MgCl , 150 mM KCl, 1 mM dithiothreitol and
#
20 µg\ml aprotinin) and lysed in buffer K containing 1 % Triton X-100
for 30 min at 20 mC. The Triton X-100 concentration was optimized by
cell fractionation and monitored by phase-contrast microscopy. Nuclei
and cell debris were removed by successive centrifugations at 1000 g for
3 min and 8000 g for 10 min. The resulting supernatant (cytoplasmic
extract) was treated with 20 µg\ml RNase A for 30 min at 20 mC, and
centrifuged on a 5 ml, 15–30 % sucrose gradient for 105 min at 149 000 g
(Farnet & Haseltine, 1991). Sixteen fractions were collected from the top
of the gradient. Putative PICs containing viral cDNA as judged by PCR
were located in fractions 8–10. These co-sedimented with a 160S cowpea
mosaic virus marker. To investigate viral cDNA in the infected cell nuclei,
the 1000 g nuclear pellet (see above) was washed in buffer solution and
Dounce homogenized in 300 µl buffer solution. Disruption of nuclei was
monitored by phase contrast microscopy. Lysed nuclear extracts were
proteinase K treated (1 mg\ml ; Roche Molecular Biochemicals, PCR
grade) for 1 h at 55 mC. DNA was phenol–chloroform extracted and
recovered by precipitation with ethanol overnight at k20 mC. PCR
involved preheating in a Touchdown PCR instrument (Hybaid) at 94 mC
for 3 min, denaturing at 94 mC for 45 s, annealing at 56 mC for 45 s and
an elongation step at 72 mC for 1 min. A final elongation step at 94 mC for
10 min was performed. DNA was subjected to PCR for 30 cycles. PCR
products were analysed by electrophoresis on 3 % agarose gels.
Experiments were carried out in duplicate on different lots of infected
cells.
Immunoprecipitation of PICs from sucrose velocity gradient
fractions. Fifty µl of sucrose gradient fractions containing detectable
160S cDNA were incubated overnight at 4 mC with 10 µl of monoclonal
antibodies against RT, MA, CA at 10 µg\ml or 10 µl of polyclonal
antibodies (1\300) against IN or Vpr. Next antibody–antigen complexes
were collected for 2 h at room temperature using either protein A or G
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
Structure of a late PIC
Table 1. Summary of the 23 regions of viral cDNA amplified and sizes of PCR products
Region and
primer pair
1. LTR1
2. LTR2
3. LTR\GAG
4. GAG1
5. GAG2
6. GAG3
7. GAG4
8. GAG\POL
9. POL1
10. POL2
11. POL3
12. POL4
13. POL5
14. POL\VIF
15. POL\VPR
16. VPR\VPU
17. VPU\ENV
18. ENV1
19. ENV2
20. ENV3
21. ENV4
22. ENV\NEF
23. NEF\LTR
Sequence
Sense (5h CACACAAGGCTACTTCCCTGA 3h)
Antisense (5h GATCTCTAGTTACCAGAGTCAC 3h)
Sense (5h CGAGAGCTGCATCCGGAGTA 3h)
Antisense (5h GCTTCAGCAAGCCGAGTCCT 3h)
Sense (5h TTGACTAGCGGAGGCTAGAA 3h)
Antisense (5h TGATTGCTGTGTCCTGTGTC 3h)
Sense (5h GCAGTTAATCCTGGCCTGTT 3h)
Antisense (5h CCATTCTGCAGCTTCCTCAT 3h)
Sense (5h GGTACATCAGGCCATATCACC 3h)
Antisense (5h ACCGGTCTACATAGTCTC 3h)
Sense (5h CCTACCAGCATTCTGGACAT 3h)
Antisense (5h GTCTATCGGCTCCTGCTTCT 3h)
Sense (5h ACTCTAAGAGCCGAGCAAGC 3h)
Antisense (5h CTTGTCTATCGGCTCCTGCT 3h)
Sense (5h GAGCCGATAGACAAGGAACT 3h)
Antisense (5h TGACAGGTGTAGGTCCTACT 3h)
Sense (5h GGACCTACACCTGTCAACAT 3h)
Antisense (5h CCACCTCAACAGATGTTGTC 3h)
Sense (5h AGGAGCTGAGACAACATCTG 3h)
Antisense (5h GCCAATACTCTGTCCACCAT 3h)
Sense (5h AGAGTATTGGCAAGCCACCT 3h)
Antisense (5h TCCTGATTCCAGCACTGACT 3h)
Sense (5h CATGGGTACCAGCACACAAAGG 3h)
Antisense (5h TCTACTTGTCCATGCATGGCTTC 3h)
Sense (5h TAACCTGCCACCTGTAGTAG 3h)
Antisense (5h GTTCAGCCTGATCTCTTACC 3h)
Sense (5h AGCAGGAAGATGGCCAGTAA 3h)
Antisense (5h CAATCATCACCTGCCATCTG 3h)
Sense (5h CAGATGGCAGGTGATGATTG 3h)
Antisense (5h ACGCCTATTCTGCTATGTCG 3h)
Sense (5h GCAGGAGTGGAAGCCATAAT 3h)
Antisense (5h TGCCACTGTCTTCTGCTCTT 3h)
Sense (5h GAAGACAGTGGCAATGAGAG 3h)
Antisense (5h AGGCCTGTGTAATGACTGAG 3h)
Sense (5h GAGGATATAATCAGTTTATGG 3h)
Antisense (5h AATTCCATGTGTACATTGTACTG 3h)
Sense (5h TGGCAGTCTAGCAGAAGAAG 3h)
Antisense (5h CCTCCAGGTCTGAAGATCTC 3h)
Sense (5h AGTCCGAGATCTTCAGACCT 3h)
Antisense (5h AACCTACCAAGCCTCCTACT 3h)
Sense (5h ATAGTAGGAGGCTTAGTAGG 3h)
Antisense (5h GCGAATAGCTCTACAAGCTC 3h)
Sense (5h AGAGCTATTCGCCACATACC 3h)
Antisense (5h GGTGTAACAAGCTGGTGTTC 3h)
Sense (5h CCTGGCTAGAAGCACAAGAG 3h)
Antisense (5h CTTGTAGCAAGCTCGATGTC 3h)
Sequence
amplified*
No. of bases
amplified
59–610
551
297–712
415
762–1169
407
922–1425
503
1215–1687
472
1624–2229
605
1696–2232
536
2218–2499
281
2484–3185
701
3157–3768
611
3758–4221
463
4150–4393
243
4307–4740
433
4541–5068
527
5049–5807
758
5721–6224
503
6212–6837
625
6540–6977
437
7010–7639
629
7615–8307
692
8286–8747
461
8736–9295
559
8963–9435
472
* From HxB2.
Sepharose beads (Sigma–Aldrich) in their respective binding buffer
(50 mM Tris, 150 mM NaCl, pH 8n0 or 0n01 M NaH PO , 0n15 M NaCl,
# %
0n01 M EDTA, pH 7n0). Beads with bound complexes were then washed
three times in wash buffer (10 mM Tris–HCl, pH 7n4, 150 mM NaCl and
1 % Triton X-100). These were then digested with proteinase K at
1 mg\ml for 1 h at 55 mC to release immunoprecipitated complexes from
the Sepharose beads and the beads were removed by centrifugation. The
supernatant containing any viral cDNA was phenol–chloroform extracted, ethanol precipitated and amplified by PCR as described above using
the GAG2 primer pair (see below and Table 1) and 30 cycles of
amplification.
Integration assay. Cytoplasmic extract (300 µl), purified PICs or
deproteinized PICs were incubated with 500 ng of linearized φX174
phage DNA (Sigma) for 45 min at 37 mC. Reactions were stopped by
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
CFCF
D. K. Khiytani and N. J. Dimmock
treatment with proteinase K (1 mg\ml) for 1 h at 55 mC, and DNA
extracted and recovered by precipitation with ethanol. Integration was
assessed by PCR (Pryciak & Varmus, 1992) with an HIV-1 cDNA-specific
forward primer and a reverse primer that recognized phage DNA. The
advantage of this assay is that there can be no PCR product until HIV-1
cDNA has integrated. Various primer pairs were tried with 30 cycles of
PCR, and one using a forward primer from HxB2, a molecular clone of
HIV-1 (8953 GCTGCTTGTGCCTGGCTAGA 8972) and a reverse
primer from φX174 (144 AGCTGCGCAAGGATAGGTCG 163), gave a
product of approximately 900 bp. This does not argue for a single
integration site, but rather that there is a bias to certain integration sites,
as others have shown (Bo et al., 1996). In the integration reaction the viral
integrase cuts the host DNA in an exchange reaction that covalently links
each viral strand to phage DNA at one end, but not the other, where
there is a gap (Whitcomb & Hughes, 1992). Thus repair of the gap is not
needed for a successful PCR reaction. PCR products were analysed as
above and partially sequenced. Experiments were done in duplicate with
different lots of infected cells.
DNase I footprinting. PICs were treated with DNase I (RNase
free ; Sigma–Aldrich) at 0n1, 1, 10 and 30 µg\ml for 30 min at 37 mC, and
then reacted with proteinase K (1 mg\ml) at 55 mC for 1 h to remove
protein. DNA was phenol–chloroform extracted and precipitated with
ethanol overnight at k20 mC. PCR was then carried out for 24, 26, 28
and 30 cycles using 23 primer pairs designed by Primer Designer for
Windows (Version 3.0 ; Scientific and Educational Software) from the
sequence of HxB2. These covered the entire HIV-1 genome in
approximately 500 bp overlapping fragments (Table 1). PCR conditions
were optimized for each primer pair so that each had specific MgCl and
#
primer concentrations in the PCR mix. Positive controls were amplified
following proteinase K treatment (1 mg\ml) but in the absence of DNase I
and negative controls (k) were amplified following proteinase K and
then DNase I treatment at 0n1 µg\ml. Footprinting was carried out twice
on PIC preparations from different batches of infected cells.
Results
Time-course of synthesis of 160S HIV-1 PIC cDNA,
accumulation of viral cDNA in the cell nucleus and
virion proteins associated with PIC cDNA
Viral cDNA was not detected in 160S PIC cDNA from
cytoplasmic extracts at 2n5 and 5 h post-infection but was
clearly present at 7n5, 8n5 and 10 h post-infection (Fig. 1a).
However, the 12n5 h sample contained little amplifiable
material. In nuclei, cDNA was first detected as a weak band at
10 h post-infection and there was a very strong band present
at 12n5 h (Fig. 1b). The experiment was repeated with a new
batch of infected cells with very similar results (data not
shown). Absence of cDNA in the nucleus at 8n5 h suggests that
there was no significant contamination of nuclei with cytoplasm and the absence of cytoplasmic cDNA at 12n5 h
suggested there was little nuclear contamination of cytoplasm.
No cDNA was detected in any of the nuclear washes (data not
shown). The co-precipitation of cDNA by antibodies to various
virion proteins incubated with 160S PICs isolated at 10 h postinfection is shown in Fig. 1(c). There were PCR products after
immunoprecipitation with antibodies specific for RT, IN, MA
and Vpr proteins but not with antibodies specific for CA or the
influenza NP protein or in the absence of antibody. Similar data
CFCG
Fig. 1. Time-course and cellular location of HIV-1 cDNA isolated from
infected C8166 cells and the detection of PIC-associated virion proteins.
(a) Detection of cDNA present in gradient-purified 160S PICs.
(b) Detection of cDNA present in nuclear extracts from the same
cells. Samples were harvested at the times shown post-infection (h).
(c) Detection of cDNA complexed with virion proteins in 160S gradientpurified PICs from 10 h cytoplasmic extracts after immunoprecipitation of
PICs with antibodies to RT, IN, MA, Vpr and CA. In all panels cDNA was
detected using PCR, the GAG2 primer pair, and 30 cycles of amplification.
PCR products were analysed by agarose gel electrophoresis and visualized
by ethidium bromide staining. The arrow represents a DNA marker of 506
bp. In panel (c) immunoprecipitated complexes were captured and then
eluted from protein G– or protein A–Sepharose beads. DNA was then
extracted with phenol–chloroform, ethanol precipitated, and amplified by
PCR. C8+, C8−, unfractionated cytoplasmic extracts from HIV-1-infected
and uninfected C8166 cells respectively ; Mo, Ra, immunoprecipitation mix
with normal mouse or rabbit antiserum replacing virus-specific mouse or
rabbit antibody respectively ; IgG1, immunoprecipitation mix with influenza
NP-specific IgG1 ; PG, PA, immunoprecipitation mix with protein G– or
protein A–Sepharose respectively but no antibody ; PCR, PCR mix only.
Each experiment was repeated at least twice.
were obtained also with 8n5 h PICs (data not shown). Although
there are slight differences in kinetics, the above data indicate
that our PICs have properties similar to those described in the
literature by others (Li & Burrell, 1992).
Integration activity of 160S PICs
We next determined the integration activity of 160S
purified PICs isolated at 2n5, 5, 7n5, 8n5 and 10 h post-infection
into φX174 phage DNA. Preliminary experiments suggested
that there were multiple integration sites, but that one
particular site was favoured (see Methods). Thus we could use
PCR with specific primers for HIV-1 DNA and for φX174
phage DNA to demonstrate that integration had taken place,
as there could be no product until the cDNA had integrated.
This approach takes less time than Southern blotting and was
consistently reproducible. Fig. 2 shows that there was no PCR
product at 2n5 and 5 h post-infection, and that a single band,
approximately 900 bp, was first detected at 7n5 h postinfection. Much stronger PCR bands were evident at 8n5 and
10 h post-infection (Fig. 2). There was no PCR product using
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
Structure of a late PIC
Fig. 2. In vitro integration activity of PICs. PICs were purified on sucrose
velocity gradients from cytoplasmic extracts of HIV-1-infected C8166 cells
at the times shown post-infection (h). Purified 160S PICs or 160S PICs
deproteinized with proteinase K were incubated with 500 ng of linearized
φX174 phage DNA for 45 min at 37 mC. After DNA extraction, samples
were subjected to PCR for 30 cycles using a primer that recognized HIV-1
and another that recognized φX174 DNA, so that no product was
obtained without integration. The arrow indicates a DNA marker of 1018
bp. PCR products were analysed by electrophoresis on agarose gels.
Similar data were obtained from a second experiment using a different
batch of infected cells.
deproteinized PICs, or when PICs and phage DNA were held
together at 4 mC. The 900 bp PCR product was excised from
the gel and partially sequenced using the HIV-1 and φX174
PCR primers. These gave approximately 250 bp of sequence
with 93 % similarity with the HxB2 sequence and a similar
length of sequence with 97 % phage similarity respectively
(data not shown). This confirmed that HIV-1 cDNA had
integrated into the phage DNA and thus that purified PICs
isolated between 7n5 and 10 h post-infection were integration
active. This activity corresponded well with the appearance of
PIC cDNA described in Fig. 1(a).
DNase footprinting of PICs harvested at 8n5 h postinfection shows that only part of the LTR is protected
The extent of protein–DNA interactions within PICs was
determined by DNase footprinting across the whole HIV-1
genome, using PCR to indicate which regions were protected.
Since PCR can be variable, we minimized primer-to-primer
variation and stringency by optimizing PCR conditions with
viral cDNA, so that all 23 primer pairs gave an amplicon of
approximately equal band intensity after 30 cycles (data not
shown). In addition, carrying out a complete set of PCRs on the
same infected cell preparation further minimized variation.
PICs were harvested from C8166 cells infected with cellfree virus and harvested at 8n5 h post-infection. Cytoplasms
were prepared and fractionated by sucrose velocity gradient
centrifugation. PICs sedimenting at 160S were subjected to
DNase footprinting using four concentrations of DNase I (0n1,
1, 10 and 30 µg\ml). Digestion with proteinase K was then
carried out and protected DNA amplified for 24, 26, 28 and 30
cycles with 23 primer pairs that covered the entire HIV-1
genome. After analysis on agarose gels it was clear that PCR
products were only seen with primer pair LTR1 and with 10,
1 and 0n1 µg\ml DNase I (Fig. 3). No PCR product was seen
using any of the other 22 primer pairs, even with 30 cycles of
PCR, or at any DNase concentration. The negative result
shown with GAG4 primers was representative of data with the
other 21 primer pairs (Fig. 3). PCR was continued up to 36
cycles in combination with the lowest DNase I concentration,
Fig. 3. Nuclease protection assay of cDNA from HIV-1 PICs harvested at
8n5 h post-infection using 23 primers pairs to cDNA regions 1–23 (Table
1). PICs were purified by sucrose gradient centrifugation and the 160S
PIC cDNA digested with DNase I at 30, 10, 1 and 0n1 µg/ml as indicated
by the triangle. The positive control (j) was treated with proteinase K but
not DNase I. After proteinase K treatment for 1 h at 55 mC, DNA was
extracted with phenol–chloroform, and precipitated with ethanol overnight.
PCR was done for 24, 26, 28 and 30 cycles. Products were
electrophoresed on 3 % agarose gels. The negative control (k) was
treated with proteinase K and then 0n1 µg/ml DNase I before PCR. PCR
products were obtained using LTR1 primers (region 1 ; top panel), but
with none of the other primers. Only the 30 cycle PCR data are shown.
The lower panel shows the negative result with the GAG4 (region 7)
primer pair that is representative of all primer pairs other than LTR1. The
arrows indicate a DNA marker of 506 bp. This experiment is
representative of two independent experiments each carried out on
different batches of infected cells.
Fig. 4. Titration and PCR of cDNA from 160S HIV-1 PICs harvested at 8n5
and 10 h post-infection, as described in Fig. 1, using the primers pairs
shown and 30 cycles of PCR. The arrows indicate a DNA marker of
506 bp.
but still no product was seen (data not shown). Positive
controls (j) were successfully amplified after protease treatment (1 mg\ml) but in the absence of DNase I (Fig. 3), and no
amplicon was seen in negative controls (k) following
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
CFCH
D. K. Khiytani and N. J. Dimmock
Fig. 5. Nuclease protection assay of the entire cDNA genome isolated from 160S HIV-1 PICs harvested at 10 h post-infection.
Two repeat experiments using different batches of infected cells are shown to demonstrate reproducibility. Twenty-three primer
pairs (numbered 1–23 and described according to the relevant region of the genome – see also Table 1) were used. The first
lane (k) is the negative control, the triangle indicates digestion with DNase I at concentrations of 30, 10, 1, 0n1 µg/ml, and
(j) is the positive control (all as described in Fig. 3). PCRs were done using 24, 26, 28 and 30 cycles but only the 30 cycle
data are shown. The arrowheads indicate a DNA marker of 506 bp.
proteinase K and DNase I treatment (0n1 µg\ml). These data
showing protection of only part of the LTR(s) are entirely
consistent with Mu-mediated PCR footprinting on PICs
harvested at a similar time (Chen et al., 1999).
DNase footprinting of PICs harvested at 10 h postinfection shows that the majority of cDNA is protected
In the next part of this study we carried out DNase I
footprinting exactly as described above, except that 160S PICs
CFCI
were harvested at 10 h post-infection. A DNA dilution series
using three different primer pairs showed that approximately
the same amount of DNA was present in both 8n5 and 10 h
samples (Fig. 4).
DNase I protection with primer pairs 1–23 amplifying the
entire genome from PICs isolated at 10 h post-infection is
shown in Fig. 5. Data from two independent experiments are
shown. PCR products were analysed after 24, 26, 28 and 30
cycles and with four concentrations of DNase I. Only the 30
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
Structure of a late PIC
cycle PCR data are shown. Positive controls, amplified in the
absence of DNase I, were uniform in appearance and intensity
for the most part, suggesting that the PICs amplified were
representative of the 10 h PIC population as a whole. No PCR
product was obtained in any of the negative controls. The first
line of Fig. 5 shows protection of the genome from the LTR to
the gag–pol junction. All three LTR regions (detected with
primer pairs LTR1, LTR2 and LTR\GAG) were well protected.
This is in marked contrast with 8n5 h PICs where only region
1 was protected. Regions of gag were also well protected. For
example in region 4, a PCR product was detected with primer
pair GAG1 after 26 cycles and 0n1 µg\ml DNase (not shown).
However, further along the genome, region 7 (GAG4) gave
PCR products only after 30 cycles and 1 µg\ml DNase.
Nonetheless the adjacent region 8 (GAG\POL) was well
amplified after 26 cycles and 1 µg\ml DNase (not shown).
The second line of Fig. 5 shows amplification of regions
9–16 from pol to the vpr–vpu junction. Regions 9–11 and
13–16 were quite well protected, with clear PCR products after
26 or 28 cycles and 0n1 or 1 µg\ml DNase (not shown). Region
12 (in the centre of pol) showed weak protection (upper data
set) or no protection (lower data set). However, this was a
relatively poor primer set, with positive control bands apparent
only after 28 or 30 cycles of PCR. The third line of Fig. 5 shows
amplification of regions 17–23 from vpu to nef. Overall, there
was less amplification than with the upstream regions, although
there was clear protection of regions 17–19, 21 and 22.
However, none of the regions gave a product with 24 and 26
cycles of PCR, even at the lowest DNase concentration (not
shown). PCR products were, however, obtained at 28 cycles
for region 17 (VPU\ENV) and region 21 (ENV4), but only at
DNase concentrations of 0n1 and 1 µg\ml. A PCR product was
seen for region 18 (ENV1) after 30 cycles and 0n1 µg\ml
DNase and for region 19 (ENV2) and 22 (ENV\NEF) only after
28 cycles at 0n1 µg\ml DNase. There was no protection of
region 20 (ENV3) even though the positive control gave
products after 26 cycles of PCR in both experiments. There
was a weak product in region 23 (NEF\LTR) after 30 cycles of
PCR.
Discussion
Recent work has provided insights into the composition,
structure and function of HIV-1 PICs (see Introduction for
references). The aim here was to elucidate PIC structure in
more detail in order to begin to understand how PICs function
in vivo. We initially characterized the time-course of appearance
of viral cDNA in gradient-purified PICs in the cytoplasm and
the accumulation of viral cDNA in the cell nucleus. 160S PIC
cDNA was detected in the cell cytoplasm at 7n5, 8n5 and 10 h
post-infection, but not at 12n5 h post-infection (Fig. 1a). Viral
cDNA was first detected in the nucleus as a weak band at 10 h
post-infection and then as a strong band at 12n5 h postinfection (Fig. 1b). These observations are consistent with
synchronous transport of PICs to the nucleus between
10–12n5 h post-infection. In agreement with others (see
Introduction for references) we found virion proteins RT, IN,
MA and Vpr, but not CA, associated with PIC cDNA (Fig. 1c).
The appearance of integration activity in 160S PICs (Fig. 2)
closely followed their detection shown in Fig. 1(a), suggesting
that the majority of the cDNA-containing PIC population was
integration-competent (Fig. 2).
The main thrust of this report is the investigation of
DNA–protein interactions over the entire HIV-1 genome
using DNase I footprinting combined with PCR. In total, 23
overlapping sequences from the 5h LTR to the 3h LTR were
probed and we examined PICs isolated at 8n5 and 10 h after
infection of C8166 cells with cell-free virus. In the 8n5 h PICs,
we found that only part of the LTR (region 1, amplified with
primer pair LTR1) was protected from DNase I (Fig. 3). This
amounted to approximately 6 % of the cDNA, and is entirely
consistent with earlier data (Chen et al., 1999). Together these
data suggest that very limited amounts of protein are tightly
bound to the cDNA of 8n5 h PICs. However, when we
analysed PICs isolated at 10 h post-infection in exactly the
same way, a completely different picture emerged. Here nearly
all (approximately 90 %) of the viral cDNA genome was
protected by bound protein from digestion by DNase I.
The fact that the one protected region 1 (LTR1) in 8n5 h PICs
(Fig. 3) was amplified to a similar extent in the 10 h sample (Fig.
5) underlines the significance of areas of higher nuclease
resistance seen elsewhere in the genome. However, not all the
10 h PIC cDNA genome was protected from DNase digestion
and region 20 was completely sensitive (Fig. 5). Regions of
high protection did not correspond with either of the two
genomic polypurine tracts (Charneau et al., 1994).
As HIV-1 infection of cells in culture comprises a series of
multiple non-synchronous infection events, we cannot fully
discount the fact that the 10 h PICs may comprise mixtures of
structures at various states of maturity that have been derived
from virions entering the cell asynchronously. However, data
in Fig. 1(a, b) suggest that there is synchronous movement of
PIC cDNA from cytoplasm to nucleus. In addition, the 8n5 and
10 h post-infection patterns of nuclease protection differ
radically as shown above, and in two separate experiments the
nuclease protection findings were reproducible even to the
extent of nuclease concentration sensitivity and number of
PCR cycles required. Such data suggest that the characteristics
of infection did not vary significantly from infection to
infection. The positive controls shown in Fig. 5 suggest that
the PICs isolated at 10 h post-infection represent the PIC
population as a whole, and thus they may be an intermediate
on the pathway into the nucleus. Further, PICs described here
are unlikely to be proviral contaminants, as in our system, as
discussed above, there was no detectable cross-contamination
of cytoplasm and nucleus. In addition, our 10 h PICs were well
characterized as originating from a cytoplasmic fraction that
sedimented at 160S (Bowerman et al., 1989 ; Farnet & Haseltine,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
CFCJ
D. K. Khiytani and N. J. Dimmock
Fig. 6. Diagram showing two conventional views of the structure of 8n5 h PICs (Chen et al., 1999 ; Farnet & Bushman, 1997)
and possible structures of 10 h PICs. All are drawn with exactly the same types and amounts of virion proteins. (a, b) 8n5 h
PICs where all proteins are tightly bound to and protect only the LTRs from nuclease digestion (a), or some proteins are tightly
bound to and protect the LTRs and the others are loosely associated with and do not protect the rest of the viral genome (b).
(c, d) 10 h PICs where all associated proteins are tightly bound to and protect most of the genome (c), or which resemble
model (a), but have recruited cellular proteins which are tightly bound and protect from nuclease digestion (d).
1990 ; Karageorgos et al., 1993 ; Miller et al., 1997) and being
associated with the viral proteins RT, IN, Vpr and MA, but not
CA, as others have also found (see Introduction for references).
It is interesting that there was no difference in the types of
virion protein detected in our 8n5 and 10 h PICs. Clearly, the
increased nuclease resistance at 10 h could only be explained if
the distribution or association of PIC-associated proteins had
changed as hypothesized in Fig. 6(a–c), or more (possibly cell)
proteins had been recruited (Fig. 6d).
It is postulated that a number of cytoplasmic viral
intermediates are formed following HIV-1–cell fusion. These
are (a) an immature reverse transcription complex (Zennou et
al., 2000), (b) an early (8n5 h) PIC described by (Chen et al.,
1999) and by us above, in which only the LTRs are tightly
associated with protein, (c) the late (10 h) PIC described above
in which approximately 90 % of the genome is tightly
associated with protein and (d) a PIC structure with a central
DNA flap, a 99 nucleotide sequence, located approximately
halfway along the HIV (Zennou et al., 2000) and feline
immunodeficiency virus genomes (Whitwam & Poeschla,
2001). The central DNA flap is hypothesized to act as a
determinant in translocating HIV-1 DNA through the nuclear
pore (Zennou et al., 2000). Interestingly, the position of the
DNA flap corresponds with regions 13 and 14 of our 10 h PICs
where there was reasonably good protection (Fig. 5). It remains
to be determined how late PICs harvested 10 h post-infection
and the PICs with the central DNA flap are related structurally
and in terms of the sequence of events on the PIC pathway.
However, our 10 h PICs were integration-competent (Fig. 2)
and appear to be nuclear translocation-competent (Fig. 1a, b).
CFDA
However, because all virus preparations have a high particle :
infectivity ratio, it cannot be proved that late PICs are destined
to become integrated provirus.
In conclusion, we have described a new form of nucleaseresistant PIC present at 10 h post-infection that appears late in
the cytoplasm. This may be a mature and integrationcompetent intermediate that appears just prior to nuclear
membrane docking and translocation. More work is needed to
support this view, and we are currently engaged in identifying
proteins that are bound near the central DNA flap of the late
PIC, and investigating if and how they might aid nuclear
translocation.
Cells were kindly provided by Dr H. Holmes at the Centralized
Facility for AIDS Reagents (NIBSC, Potters Bar, UK). Antibodies were
also provided by the Centralized Facility for AIDS Reagents : Dr S.
Ranjbar (IN antibodies), Drs D. Helland and A. M. Szilvay (RT
antibodies), Drs R. B. Ferns and R. S. Tedder (MA antibodies) ; and by the
NIH AIDS Research and Reference Reagent Program (Rockville, USA) :
Dr J. Kopp (Vpr antibodies) and Drs B. Chesebro and K. Wehrly (CA
antibodies). We also thank Axis Genetics plc for purified cowpea mosaic
virus and Dr Steve Busby (University of Birmingham, UK) for helpful
discussions on nuclease footprinting. Work in the NJD laboratory was
supported by grants from the National Heart, Lung, and Blood Institute,
NIH (5R01HL59726) and The WPH Charitable Trust.
References
Bo, Y.-C., Miller, M. D., Bushman, F. D. & Orgel, L. E. (1996). Target-
sequence preferences of HIV-1 integration complexes in vitro. Virology
222, 283–288.
Bowerman, B., Bishop, P. O., Bishop, J. O. & Varmus, H. E. (1989). A
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
Structure of a late PIC
nucleoprotein complex mediates the integration of retroviral DNA. Genes
& Development 3, 469–478.
Brown, P. O. (1997). Integration. In Retroviruses, pp. 161–203. Edited by
J. Coffin, S. H. Hughes & H. E. Varmus. Cold Spring Harbor, NY : Cold
Spring Harbor Laboratory.
Bukrinskaya, A. G., Ghorpade, A., Heinzinger, N. K., Smithgall, T. E.,
Lewis, R. E. & Stevenson, M. (1996). Phosphorylation-dependent
Gallay, P., Stitt, V., Munday, M., Oettinger, M. & Trono, D. (1996). Role
of the karyopherin pathway in human immunodeficiency virus type 1
nuclear import. Journal of Virology 70, 1027–1032.
Gallay, P., Hope, T., Chin, D. & Trono, D. (1997). HIV-1 infection of
nondividing cells through the recognition of integrase by the importin\
karopherin pathway. Proceedings of the National Academy of Sciences, USA
94, 9825–9830.
human immunodeficiency virus type 1 infection and nuclear targeting of
viral DNA. Proceedings of the National Academy of Sciences, USA 93,
367–371.
Heinzinger, N. K., Bukrinsky, M. I., Haggerty, S., Ragland, A. M.,
Kewelramani, V., Lee, M.-A., Gendelman, H. E., Ratner, L., Stevenson,
M. & Emerman, M. (1994). The Vpr protein of human immunodeficiency
Bukrinsky, M. I., Sharova, N., Dempsey, M. P., Stanwick, T. L.,
Bukrinskaya, A. G., Haggerty, S. & Stevenson, M. (1992). Active
virus type 1 influences nuclear localization of viral nucleic acids in
nondividing host cells. Proceedings of the National Academy of Sciences,
USA 91, 7311–7315.
Jenkins, Y., McEntee, M., Weis, K. & Greene, W. C. (1998). Characterization of HIV-1 Vpr nuclear import : analysis of signals and pathways.
Journal of Cell Biology 143, 875–885.
Karageorgos, L., Li, P. & Burrell, C. J. (1993). Characterization of HIV
replication complexes early after cell-to-cell infection. AIDS Research and
Human Retroviruses 9, 817–823.
Karni, O., Friedler, A., Zakai, N. & Loyter, A. (1998). A peptide derived
from the N-terminal end of HIV-1 Vpr promotes nuclear import in
permeabilized cells : elucidation of the NLS region of the Vpr. FEBS Letters
429, 421–425.
Levy, J. A. (1998). HIV and the Pathogenesis of AIDS, 2nd edn. Herndon,
VA : ASM Press.
Li, P. & Burrell, C. J. (1992). Synthesis of human immunodeficiency
virus DNA in a cell-to-cell transmission model. AIDS Research and Human
Retroviruses 8, 253–259.
McLain, L. & Dimmock, N. J. (1994). Single- and multi-hit kinetics of
immunoglobulin G neutralization of human immunodeficiency virus type
1 by monoclonal antibodies. Journal of General Virology 75, 1457–1460.
Miller, M. D., Farnet, C. M. & Bushman, F. D. (1997). Human
immunodeficiency virus type 1 preintegration complexes : studies of
organization and composition. Journal of Virology 71, 5382–5390.
nuclear import of human immunodeficiency virus type 1 preintegration
complexes. Proceedings of the National Academy of Sciences, USA 89,
6580–6584.
Bukrinsky, M. I., Haggerty, S., Dempsey, M. P., Sharova, N., Adzhubel,
A., Spitz, L., Lewis, P., Goldfarb, D., Emerman, M. & Stevenson, M.
(1993a). A nuclear localization signal within HIV-1 matrix protein that
governs infection of non-dividing cells. Nature 365, 666–669.
Bukrinsky, M. I., Sharova, N., McDonald, T. L., Pushkarskaya, T.,
Tarpley, W. G. & Stevenson, M. (1993b). Association of integrase,
matrix and reverse transcriptase antigens of human immunodeficiency
virus type 1 with viral nucleic acids following acute infection. Proceedings
of the National Academy of Sciences, USA 90, 6125–6129.
Bushman, F. D., Fujiwara, T. & Craigie, R. (1990). Retroviral integration
directed by HIV integration protein in vitro. Science 249, 1555–1558.
Charneau, P., Mirambeau, G., Roux, P., Paulous, S., Buc, H. & Clavel,
F. (1994). HIV-1 reverse transcription : a termination step at the centre
of the genome. Journal of Molecular Biology 241, 651–662.
Chen, H., Wei, S.-Q. & Engelman, A. (1999). Multiple integrase
functions are required to form the native structure of the human
immunodeficiency virus type 1 intasome. Journal of Biological Chemistry
274, 17358–17364.
Ellison, V. & Brown, P. O. (1994). A stable complex between integrase
and viral DNA ends mediates human immunodeficiency virus integration
in vitro. Proceedings of the National Academy of Sciences, USA 91,
7316–7320.
Farnet, C. M. & Haseltine, W. A. (1990). Integration of human
immunodeficiency virus type 1 DNA in vitro. Proceedings of the National
Academy of Sciences, USA 87, 4164–4168.
Farnet, C. M. & Haseltine, W. A. (1991). Determination of viral proteins
present in the human immunodeficiency virus type 1 preintegration
complex. Proceedings of the National Academy of Sciences, USA 65,
1910–1915.
Farnet, C. M. & Bushman, F. D. (1997). HIV-1 cDNA integration :
requirement of HMG I(Y) protein for function of preintegration
complexes in vitro. Cell 88, 483–492.
Fouchier, R. A. M., Meyer, B. E., Simon, J. H. M., Fischer, U. & Malim,
M. H. (1997). HIV-1 infection of non-dividing cells : evidence that the
amino-terminal basic region of the viral matrix protein is important for
Gag processing but not for post-entry nuclear import. EMBO Journal 16,
4531–4539.
Fujiwara, T. & Mizuuchi, K. (1988). Retroviral DNA integration :
structure of an integration intermediate. Cell 54, 497–504.
Gallay, P., Swingler, S., Aiken, C. & Trono, D. (1995a). HIV-1 infection
of non-dividing cells : C-terminal tyrosine phosphorylation of the viral
matrix protein is a key regulator. Cell 80, 379–388.
Pluymers, W., Cherepanov, P., Schols, D., de Clercq, E. & Debyser, Z.
(1999). Nuclear localization of human immunodeficiency virus type 1
integrase expressed as a fusion protein with green fluorescence protein.
Virology 258, 327–332.
Popov, S., Rexach, M., Ratner, L., Blobel, G. & Bukrinsky, M. (1998a).
Viral protein R regulates docking of the HIV-1 preintegration complex to
the nuclear pore complex. Journal of Biological Chemistry 273, 13347–
13352.
Popov, S., Rexach, M., Zybarth, G., Reiling, N., Lee, N.-A., Ratner, L.,
Lane, C. M., Moore, M. S., Blobel, G. & Bukrinsky, M. (1998b). Virus
protein R regulates nuclear import of the HIV-1 pre-integration complex.
EMBO Journal 17, 909–917.
Popovic, M., Sarngadharan, M. G., Read, E. & Gallo, R. C. (1984).
Detection, isolation, and continuous production of cytopathic retroviruses (HTLV-III) from patients with AIDS and pre-AIDS. Science 224,
497–500.
Pryciak, P. M. & Varmus, H. E. (1992). Nucleosomes, DNA-binding
proteins, and DNA sequence modulate retroviral integration target site
selection. Cell 69, 769–780.
Salahuddin, S. Z., Markham, P. D., Wong-Staal, F., Franchini, G.,
Kalyanaraman, V. S. & Gallo, R. C. (1983). Restricted expression of
human T-cell leukemia-lymphoma virus (HTLV) in transformed human
umbilical cord blood lymphocytes. Virology 129, 51–64.
Gallay, P., Swingler, S., Song, J., Bushman, F. & Trono, D. (1995b).
Vodicka, M. A., Koepp, D. M., Silver, P. A. & Emerman, M. (1998).
HIV nuclear import is governed by the phosphotyrosine-mediated
binding of matrix to the core domain of integrase. Cell 83, 569–576.
HIV-1 Vpr interacts with the nuclear transport pathway to promote
macrophage infection. Genes & Development 12, 175–185.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30
CFDB
D. K. Khiytani and N. J. Dimmock
von Swedler, U., Jornbluth, R. S. & Trono, D. (1994). The nuclear
localization signal of the matrix protein of human immunodeficiency
virus type 1 allows the establishment of infection in macrophages and
quiescent T lymphocytes. Proceedings of the National Academy of Sciences,
USA 91, 6992–6996.
Wei, S.-Q., Mizuuchi, K. & Craigie, R. (1997). A large nucleoprotein
complex assembly at the ends of the viral DNA mediates retroviral
integration. EMBO Journal 16, 7511–7520.
Whitcomb, J. M. & Hughes, S. H. (1992). Retroviral reverse transcription and integration : progress and problems. Annual Review of Cell
Biology 8, 275–306.
Whitwam, T. & Poeschla, E. M. (2001). Identification of central DNA
flap in feline immunodeficiency virus. Journal of Virology 75, 9407–9414.
CFDC
Wu, X., Liu, H., Xiao, H., Conway, J. A., Hehl, E., Kalpana, G. V., Prasad,
V. & Kappes, J. C. (1999). Human immunodeficiency virus type 1
integrase protein promotes reverse transcription through specific interactions with the nucleoprotein reverse transcription complex. Journal of
Virology 73, 2126–2135.
Zennou, V., Petit, C., Guetard, D., Nehbass, U., Montagnier, L. &
Charneau, P. (2000). Genome nuclear import is mediated by a central
DNA flap. Cell 101, 173–185.
Zhang, S., Pointer, D., Singer, G., Feng, Y., Park, K. & Zhao, L.-J.
(1998). Direct binding to nucleic acids by Vpr of human immuno-
deficiency virus type 1. Gene 212, 157–166.
Received 18 March 2002 ; Accepted 25 June 2002
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:16:30