| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, December 7, 1994; and in revised form, January
25, 1995) From the
Genomic human immunodeficiency virus type 1 (HIV-1) RNA consists
of two identical RNA molecules joined noncovalently near their 5` ends
in a region called the dimer linkage structure (DLS). Previous work has
shown that the putative DLS is localized in a 113-nucleotide domain
encompassing the 5` end of the gag gene. This region contains
conserved purine tracks that are thought to mediate dimerization
through purine quartets. However, recently, an HIV-1 All retroviruses have a common feature, namely a genome
consisting of two identical unspliced RNA molecules noncovalently
linked(1) . It has been shown by electron microscopy that this
junction is located close to their 5` ends, in a structure termed the
dimer linkage structure(2, 3, 4) . Dimerization of retroviral RNA is thought to play a crucial role in
several steps of the retroviral life cycle. RNA dimerization seems to
be closely related to the encapsidation of retroviral RNA, since cis-elements such as the dimer linkage structure (DLS) ( Several attempts have been
made to investigate factors responsible for the dimerization event.
Reports describe a crucial role for a trans-acting factor, the
HIV-1 nucleocapsid protein (NCp7) in the formation of RNA
dimers(9, 14) . It has also been demonstrated that in
the absence of any cellular or viral protein, HIV-1 RNA is able to
spontaneously dimerize in vitro under conditions of high ionic
strength(15, 16, 17, 18) .
HIV-1 However,
dimerization of viral RNA by purine quartets has recently been
discussed(20, 21, 22, 23) . First,
Berkout et al.(21) reported that the HIV-2 mutant
RNA, bereft of all PuGGAPuA sites, could dimerize in vitro.
Second, Skripkin et al.(22) recently postulated that
the linking of the two HIV-1 In
this report, we analyze the spontaneous in vitro dimerization
process of HIV-1
Plasmid pBRU2 (a gift of Dr. F. Subra) contains the HIV-1
All these oligonucleotides were loaded on an 18%
polyacrylamide gel in a buffer containing 4 M urea, 50 mM Tris-borate, pH 8.3, and 1 mM EDTA, electrophoresed, and
purified by excision from the gel by UV shadowing. DNA oligonucleotides
were 5`-
Figure 1:
Mapping of the sequences required for
HIV-1
Figure 2:
Thermal stability of the dimer RNA
224-402 as a function of temperature. Samples were analyzed by
1.5% agarose gel electrophoresis and the percent of each species
determined from the gel. The temperatures on the plot correspond to the
temperatures shown above the gel.
The same behavior observed for RNA
77-402 and 224-402 justified our choice of RNA
224-402 as the reference fragment for the following studies.
Figure 3:
RNA secondary structure predicted for
HIV-1
It is
noteworthy that our computed model is in accordance with the one
proposed by G. P. Harrison for HIV-1
The first question to be
addressed was: is the autocomplementary stem-loop I, which encompasses
nt 251-266, involved in the dimerization of HIV-1 We studied the dimerization process of RNA 224-402
when incubated with increasing concentrations of oligonucleotide 257B.
As shown in Fig. 4a, antisense oligonucleotide 257B,
which targets the autocomplementary sequence 257-262, completely
blocks dimerization. Total inhibition of dimerization is observed for
RNA 224-402 when the oligonucleotide 257B ratio was equal to 1:1.
The affinity constant can be estimated to be 0.1 to 1 µM of the antisense DNA oligonucleotide 257B for the RNA
224-402 target (for a strand concentration of 1 µM).
Autoradiogram (Fig. 4b) shows that oligonucleotide 257B
only hybridizes to RNA 224-402 in its monomeric form.
Figure 4:
Inhibition of HIV-1
Two
antisense DNA oligonucleotides were used as controls. Oligonucleotide
257 M, differing from oligonucleotide 257B by four nucleotides
in its sequence, is unable to prevent the dimer formation of RNA
224-402 (Fig. 5A). It converts 50% of the dimer
into the monomer when its concentration is 5-fold higher. The other
control oligonucleotide T (see ``Experimental Procedures'')
does not inhibit the RNA 224-402 dimerization process, even at
high concentrations (Fig. 5B). It should be noted that
oligonucleotide 257M is not bound to the RNA fragment since it migrates
as a free oligonucleotide in the gel. Consequently, it did not anneal
RNA 224-402 because of the presence of its four mutated
nucleotides. Such was not the case with oligonucleotide T which
hybridizes to the RNA target 224-402 (in Fig. 5B a difference can be observed in dimer shift mobility in the gel
when oligonucleotide T concentrations increase).
Figure 5:
Analysis of HIV-1
The second question
was: Were antisense DNA oligonucleotides 320B, 327B, 365B, and 382B,
which are complementary to polypurine tracks, able to interfere with
dimer formation? Oligonucleotides 327B, 365B, and 382B are
complementary to PuGGAPuA sequences, and oligonucleotide 320B targets a
GGAGG sequence which was proposed by Sundquist and Heaphy (16) as an attractive candidate for a purine-rich region
involved in dimerization. As shown in Fig. 6A, neither
completely inhibited RNA 224-402 dimer formation, and
oligonucleotide 327B never reduced the amount of dimer by more than
50%. In every case, each [
Figure 6:
Analysis of HIV-1
Each RNA 224-402
molecule contains four polypurine sequences. We postulate that any of
these sequences are able to interact indifferently with any of those in
a second RNA 224-402 molecule. We therefore incubated the four
oligonucleotides 320B, 327B, 365B, and 382B together with the RNA
224-402 (Fig. 6B) and found that dimerization
still occurred. [
To confirm
the role of the autocomplementary G
Figure 8:
Model of HIV-1
We therefore
conclude that the 257-266 sequence is involved in the formation
of dimeric HIV-1
Figure 7:
Analysis on agarose gel electrophoresis of
HIV-1
Thus, the
mechanism of HIV-1 We present here data on the in vitro dimerization of
HIV-1 RNA transcripts (strain Lai) under conditions of low ionic
strength. As shown in Fig. 1, HIV-1 A hypothetical mechanism for RNA dimerization,
based on an extensive comparison of sequences in the leader regions of
30 retroviral genomes, has been proposed(15) . This comparative
analysis suggested that the consensus sequence PuGGAPuA participates in
the dimerization process through the formation of purine quartets
involving both adenine(s) and guanine(s). As HIV-1 Our
results suggest that a recognition mechanism between the two identical
RNA molecules is operating via a loop-loop interaction through
nucleotides G Such a mechanism for the HIV-1 The nucleotides GCGCGC are
remarkably conserved in 18 of the 21 HIV-1 strains
observed(29) . Interestingly, mutations observed in the
stem-loop of various HIV-1 strains were offset so that the structure
and the autocomplementary sequence were maintained (except for HIVHXB2
strain). A potential link between dimerization and encapsidation of
HIV-1 genomic RNA has been
proposed(7, 9, 31) . Kim et al.(35) have recently constructed several recombinant
HIV-1 * This work was supported by the Agence Nationale de la
Recherche sur le SIDA (ANRS). The costs of publication of this article
were defrayed in part by the payment of page charges. This article must
therefore by hereby marked ``advertisement'' in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Addendum-While this manuscript was reviewed, Laughrea et al.(24) have speculated that the 248-270 or
233-285 region forms a hairpin that is the core dimerization
domain of HIV-1
Volume 270,
Number 14,
Issue of April 7, 1995 pp. 8209-8216
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
RNA at Low Ionic Strength
AN AUTOCOMPLEMENTARY SEQUENCE IN THE 5` LEADER REGION IS EVIDENCED
BY AN ANTISENSE OLIGONUCLEOTIDE (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
RNA
dimerization model was proposed as the HIV-1 RNA
dimerization initiation site, involving another region upstream from
the splice donor site and possibly confined within a stem-loop. In the
present study, we have investigated the dimerization of HIV-1
RNA, using in vitro dimerization assays under conditions
of low ionic strength, predictive RNA secondary structures determined
by computer folding, and antisense DNA oligonucleotides in order to
discriminate between these two models. Our results suggest that purine
quartets are not involved in the dimer structure of HIV-1 RNA and have led to the identification of a region upstream from
the splice donor site. This region, comprising an autocomplementary
sequence in a possible stem-loop structure, is responsible for the
formation of dimeric HIV-1 RNA.
)and the encapsidation site (E) are localized within the
same region in the genome of MoMuLV
(5) ,
RSV(6) , REV(7) , BLV(8) , and
HIV-1(9) . It has also been suggested that dimerization of
retroviral RNA is associated with reverse transcription through
interstrand switching(10) , genomic
recombination(11, 12, 13) , and
down-regulated translation(6) . sequences able to support this spontaneous
dimerization have been localized between nucleotides 311 and 415
encompassing the 5` end of the gag gene(9) . Published
results indicate that (i) the HIV-1 RNA dimer is very stable, (ii)
antisense RNA cannot form a dimer, and (iii) a consensus sequence,
PuGGAPuA, is found in the putative dimerization-encapsidation region of
HIV and other retroviral genomes(15) . Marquet et al.(15) and others (16, 17, 18) have
suggested that the structural motif mediating the association of two
identical viral RNA molecules should involve purine quartets. This
model is based on the dimerization of telomeric DNA which occurs via
the formation of unusual intermolecular quadruple helical structures
that are stabilized by guanine base tetrads(19) . RNA molecules is initiated
at the level of a short RNA region, located upstream from the splice
donor site and consisting of a palindromic sequence in a stem-loop
structure. However, this process requires high ionic strength and the
presence of MgCl
in the electrophoresis gel in order to be
observed(22, 23) . Similarly, based on an RNA-RNA
recognition model, Girard et al.()identified a
possible stem-loop structure containing a palindromic sequence, which
is probably responsible for the formation of dimeric MoMuLV RNA. RNA, under conditions of low ionic
strength, in an attempt to enhance our understanding of the nature of
the interactions leading to dimerization and to distinguish between the
above described dimerization models. We have characterized the dimer
linkage structure by using HIV-1 RNA fragments of
different lengths, complementary DNA oligonucleotides, and heterodimer
formation.
Materials
BssHII, HaeIII, RsaI, SacI, and SmaI restriction enzymes were obtained from
New England Biolabs and EcoRI, HindIII, and PstI restriction enzymes and T4 Polynucleotide Kinase from
Boehringer Mannheim. Standard procedures and sequencing were used for
restriction enzyme digestion and plasmid construction(25) . The Escherichia coli strain DH5
was used for plasmid
manipulation and preparation. All the constructions were sequenced
(United State Biochemical Corp. Kit 70770, Sequenase Version 2.0 DNA
Sequencing Kit) in order to verify the exact sequence of the
synthesized RNA fragment(26) . The constructed plasmids were
transcribed, after appropriate linearization, with T7 RNA polymerase
under conditions provided by the RiboMAX
Large Scale RNA
production System (Promega).Plasmid Construction and Digestion
Derivatives of the plasmid pGEM-3zf (Promega) were
constructed to generate viral RNA by in vitro transcription. cDNA in the pKP59 vector with XbaI-AatII as the
cloning site. This plasmid has a deletion of 2701 base pairs from
+5367 to +8061 to prevent it from being infectious.Plasmid pDM2 and HIV-1
The pBRU2 plasmid was
digested with SacI (position +224) and PstI
(position +960) to produce the viral fragment 224-960 which
was inserted into pGEM-3zf (sites SacI/PstI) and
resulted in pDM2. The pDM2 plasmid was digested either by HaeIII or RsaI and transcribed by T7 RNA polymerase
giving rise to transcripts starting from position 224 of the genomic
HIV-1 RNA Fragments from
Nucleotides 224 to 402 and 224 to 296 RNA sequence and ending at positions 402 and 296,
respectively. These fragments will be referred to as fragments
224-402 and 224-296.Plasmid pDM3 and HIV-1
The pBRU2 plasmid was
cleaved with HindIII, and the viral fragment from positions
+77 to +631 was inserted into pGEM-3zf previously digested by HindIII and resulted in pDM3. The plasmid pDM3 was digested
either by HaeIII or BssHII and transcribed by T7 RNA
polymerase giving rise to transcripts starting from position 77 of the
genomic HIV-1 RNA Fragments from
Nucleotides 77 to 402 and 77 to 256 RNA sequence and ending at positions 402
and 257, respectively. These fragments will be referred to as fragments
77-402 and 77-257.Plasmid pDM6 and HIV-1
The pDM3 plasmid was cleaved with RsaI and HindIII: the viral fragment from position
296 to 631 was isolated and inserted into pGEM-3zf previously digested
with SmaI and HindIII, and resulted in pDM6. The pDM6
plasmid was digested by HaeIII and transcribed by T7 RNA
polymerase giving rise to transcripts starting from position 296 of the
genomic HIV-1 RNA Fragment from
Nucleotides 296 to 402 RNA sequence and ending at position 402.
This fragment will be referred to as fragment 296-402.Plasmid pDM7 and HIV-1
Polymerase chain
reactions (27) were performed on pDM2 with two sets of
oligonucleotides. The first set contained the T7 universal primer and
oligonucleotide D1, corresponding to the complementary positions
+256(5`) to +236(3`) of the HIV-1 RNA Fragment
224-402 Lacking Nucleotides 257-266 genome. The
second set contained oligonucleotide D2 corresponding to positions
+263(5`) to +280(3`) of the HIV-1 genome and
the complementary sequence of the SP6 universal primer. The first
polymerase chain reaction product was digested by the EcoRI
enzyme and the second by PstI. Both DNA fragments obtained
were purified by agarose gel electrophoresis. These two fragments were
ligated at their blunt ends and inserted between EcoRI and PstI sites of pGEM-3zf(-). One clone of interest was
recovered, pDM7, which contained a viral fragment from position 224 to
960 lacking nt 257-266 of the HIV-1 genome. The
pDM7 plasmid, first digested by HaeIII, was transcribed with
T7 RNA polymerase and gave rise to a RNA fragment 224-402 bereft
of nt 257-266.In Vitro RNA Synthesis and Purification
Under the conditions stipulated by the RiboMAX Large Scale RNA Production System, 5 µg of the linearized
plasmids were transcribed with T7 RNA polymerase. The sample was
subjected to three phenol chloroform extractions after DNase treatment,
and the RNA transcripts were ethanol precipitated in the presence of
0.3 M sodium acetate, pH 5.2. The RNA precipitate was
dissolved in double-distilled sterile water (20 µl) and
microdialyzed (Millipore filters type V6, 0.025 µm) for 2 h against
sterile double-distilled water. The purity and integrity of the RNA
samples were checked by denaturing polyacrylamide gel electrophoresis.
If the purity of the RNA was insufficient, the sample in water was
loaded on 1.5% agarose gel and electrophoresed as described above. The
corresponding RNA fragment was excised from the agarose gel and
purified using a spin-X column (Costar). Ethidium bromide was extracted
from the sample with one volume of butanol-1, and RNA was
ethanol-precipitated. The concentration was determined by UV
spectroscopic measurement at 260 nm.RNA in Vitro Dimerization
RNA dimerization assays were performed on 0.5 µg of RNA
generated in vitro, in a buffer containing 50 mM Tris-HCl, pH 7.5, and 100 mM NaCl, to a final volume of
10 µl, for 45-60 min at 37 °C. RNA in water was heated
for 2 min at 90 °C, chilled on ice for 2 min before adding the
buffer. Monomeric and dimeric RNA fragments were analyzed by 1.5%
agarose gel electrophoresis at 5V/cm in a buffer containing 50 mM Tris-borate, pH 8.3, and 1 mM EDTA, at 4 °C. Ethidium
bromide (0.2 µg/ml) was added to the buffer.Tm Determination of Dimer Dissociation
After denaturation at 95 °C for 2 min in nuclease-free
water, the RNA was incubated in a buffer containing 50 mM Tris-HCl, pH 7.5, and 100 mM NaCl, at a concentration of
1 µM (molecules), at 55 °C for 50 min for optimal
dimerization. The resulting dimer was then microdialyzed (Millipore
filters type V6, 0.025 µm) for 2 h at 4 °C against a buffer
composed of 40 mM Tris-HCl, pH 7.5, 10 mM NaCl, and 1
mM EDTA. 10-µl aliquots of each sample were then incubated
for 5 min at varying temperatures ranging from 20 to 70 °C and
electrophoresed as described above. After fluorescent scanning of the
gels, the percentage of dimer and monomer was estimated. The percentage
of the dimer was defined as the area of the dimer peak divided by the
sum of the areas of monomer and dimer peaks. The melting temperature
(Tm) was estimated from the plot of the amount of the dimer as a
function of temperature.Antisense and Control DNA Oligonucleotides
Synthetic DNA oligonucleotides complementary to positions
251-267 (oligonucleotide 257B), 312-326 (oligonucleotide
320B), 322-336 (oligonucleotide 327B), 336-350
(oligonucleotide T), 360-374 (oligonucleotide 365B), and
377-391 (oligonucleotide 382B) of the HIV-1 sequence were made using an Applied Biosystem Synthesizer.
Another oligonucleotide, 257 M, complementary to positions
268-284 of the HIV-1 sequence(29) , was
used as a control oligonucleotide. It differs from 257B by four
nucleotides.
P-labeled with [P]ATP
(Amersham, United Kingdom) and T4 polynucleotide kinase. The specific
radioactivity was about 10
cpm/µg DNA oligonucleotide.
RNA 224-402 was heat denatured in the presence of the
[P]DNA oligonucleotide before starting in
vitro dimerization process. [P]DNA
oligonucleotide
RNA complexes were analyzed by agarose gel
electrophoresis. The level of hybridization of the
[P]DNA oligonucleotide to monomer and dimer RNA
224-402 was detected by autoradiography of the corresponding gel. Determination of RNA Secondary Structure by Computer
Folding
Free energy minimization predictions were made using PCFOLD
software (version 4.0) written by M. Zucker in accordance with
parameters for the prediction of RNA structures described by Turner et al.(28) . The 5` sequence of HIV-1 has been reviewed (29) and, in an attempt to determine
the secondary structure of our RNA fragments, sequences corresponding
to these fragments were folded.
Spontaneous Dimerization of HIV-1
To obtain a more detailed characterization of the DLS and
the RNA dimerization process of HIV-1 RNA in
Vitro, a RNA fragment
spanning nucleotides 77-402, which contains the leader region
(without the TAR domain) and the 5` gag sequence (Fig. 1a), was prepared by in vitro transcription and analyzed by agarose gel electrophoresis. After
heat denaturation, the RNA is in a monomeric form (Fig. 1c, lane 1), while upon incubation at 37
°C in a buffer composed of 50 mM Tris-HCl, pH 7.5, and 100
mM NaCl, dimeric RNA appears (Fig. 1c, lane 2). As reported(30) , this spontaneous
dimerization is specific to retroviral RNA. Similarly, we produced two
more RNA fragments: RNA 77-257 and RNA 224-402 (Fig. 1b, RNAs 2 and 3). RNA
77-257, shortened at the 3` end, lost its ability to dimerize (Fig. 1c, lanes 3 and 4), while RNA
224-402, shortened at the 5` end, dimerized efficiently (lanes 5 and 6). We obtained up to 80% of the dimer
under conditions of low ionic strength. We therefore used the latter
RNA as the reference fragment for the remaining experiments.
RNA dimerization by deletion mutagenesis. a, representation of the 5` end of HIV-1 DNA.
Numbering is relative to the genomic RNA cap site (+1).The
restriction sites of interest are indicated: HindIII
(+77), SacI (+224), and RsaI (+296). PBS and ATG indicate, respectively, the primer
tRNA
-binding site and initiation codon for Pr55gag synthesis. b, HIV-1 RNAs generated in
vitro and used in this study. RNAs were generated in vitro by transcription of pDM2, pDM3, pDM6, and pDM7 linearized with the
appropriate enzyme. RNA 1 and 2 begin at position
+77, RNAs 3-5 at position +224, and RNA 6 at position +296. The deletion in RNA 5, between nt
257-266 is represented by the broken line. The column of symbols, headed RNA Dimer, indicates the
level of dimeric RNA. -, 0-5%; ±, 10-15%;
+++, 70-90% (means value from at least three
experiments). c, agarose gel electrophoresis of HIV-1 RNAs. For each sample, the RNAs are numbered as in b.
Heat-denatured RNAs (M) are shown in lanes 1, 3, 5, 7, 9, and 11, and
RNAs (D) in dimerization conditions are shown in lanes
2, 4, 6, 8, 10, and 12. Lane MK, 0.16-1.77-kb RNA ladder (Life
Technologies, Inc.).
Thermal Stability of the HIV-1
Thermal stability of the HIV-1 RNA
Dimer 70 S genomic
RNA dimer has been previously determined: the melting temperature of
HIV-1 genomic RNA purified from HIV-1 virions was around 50
°C(9, 31) . For comparison, we studied the thermal
denaturation of the RNA 224-402 under the same experimental
conditions. Once formed at 55 °C in a buffer containing 50 mM Tris-HCl, pH 7.5, and 100 mM NaCl, the dimer was
subjected to thermal denaturation in a buffer containing 40 mM Tris-HCl, pH 7.5, 10 mM NaCl, and 1 mM EDTA(9) . The results obtained after thermal denaturation
of the RNA 224-402 dimer are shown on agarose gel (Fig. 2)
together with the denaturation curve representing the relative
percentage of the dimer as a function of temperature. Thermal
transition from the dimer to the monomer occurred at 53 °C. The
same Tm value was obtained for dimer RNA 77-402 (data not shown).
This Tm value is close to that of entire HIV-1 viral
RNA(9, 31) . The correspondence between these two Tm
values suggests that the nature of the interactions between the in
vitro generated HIV-1 RNA fragments is comparable to
that of the 70 S natural dimer.
Computer-assisted Analysis of HIV-1
To further define the precise sequence
involved in HIV-1 RNA
224-402 Folding dimerization, we applied free energy
minimization computer analysis to examine the RNA 224-402
secondary structure. As shown in Fig. 3, the predicted structure
appears to be in the form of three stem-loops branched together.
Stem-loops II and III, consisting of nt 296-402, correspond to
the putative DLS 311-415 of HIV-1 previously
described by Darlix et al.(9) , and contain three
purine tracks. The remaining region, spanning nucleotides
224-296, exhibits an autocomplementary stem-loop (loop I in Fig. 3). Based on the HIV-1 dimerization
initiation site recently described by Skripkin et
al.(22) , we speculate that the presence of this
autocomplementary G
CGCGC
sequence in loop I
could be responsible for the recognition of two identical RNA molecules
through a loop-loop interaction. Only one autocomplementary loop was
found in the HIV-1 RNA region investigated.
RNA 224-402 by computer-assisted energy
minimization analysis. The full lines represent the antisense
DNA oligonucleotides, used in this work, which are complementary to the
RNA sequences covered.
(32) which
comprises this stem-loop with the same primary sequence.Mapping of the Sequence Involved in the Formation of the
HIV-1
To test the validity of this assumption,
suggested by computer-assisted analysis of RNA 224-402 folding,
we hybridized a set of synthetic DNA oligonucleotides to RNA
224-402. The DNA oligonucleotides chosen were complementary to
small sequences of RNA 224-402 (Fig. 3). The selected
oligonucleotides (see ``Experimental Procedures'') were
5`- RNA Dimer using Complementary DNA
OligonucleotidesPlabeled and heat denatured with RNA 224-402
before initiating the dimerization process. RNA?
RNA
224-402 dimer formation in the presence of increasing
oligonucleotide 257B concentrations. Fixed concentrations of RNA
224-402 (0.5 µg of heat denatured RNA/assay) were incubated
in a buffer with 50 mM Tris-HCl, pH 7.5, and 100 mM NaCl, for 45-60 min at 37 °C, in the presence of (lanes 2-11) 0, 0.05, 0.075, 0.1, 0.5, 0.75, 1, 2, 5,
and 7 molar equivalent(s) of the [P]DNA oligomer
257B, complementary to nucleotides 256-336. Lane 1 shows
the monomeric form of HIV-1 RNA 224-402 (heat
denatured for 5 min at 95 °C), lane MK, as in Fig. 1. m and d indicate monomeric and dimeric
RNAs, respectively. a, 1.5% agarose gel electrophoresis. The
samples are visualized by ethidium bromide staining. b,
autoradiogram of a.
RNA
224-402 dimer formation in the presence of oligonucleotides 257 M (A) and T (B) as controls. As described in Fig. 4, fixed concentrations of RNA 224-402 were incubated
in the presence of (lanes 2-8) 0, 0.05, 0.1, 0.5, 1, 2,
and 5 molar equivalent(s) of the DNA oligonucleotide 257M or T. Lane 1, monomeric form of HIV-1 RNA
224-402. Lane MK, as in Fig. 1.
P]DNA oligonucleotide
annealed to both monomer and dimer RNA 224-402 (see
autoradiographies, Fig. 6A).
RNA
224-402 dimer formation in the presence of each oligonucleotide
320B, 327B, 365B, and 382B (A) or the four together (B). As described in Fig. 4, fixed concentrations of
RNA 224-402 were incubated in the presence of (lanes
1-6) 0, 0.1, 0.5, 1, 2, and 5 molar equivalent of each
[P]DNA oligonucleotide (A) or in the
presence of the four together, each of them at a molar equivalent of
RNA (B). Oligonucleotides 320B, 327B, 365B, and 382B which
correspond, respectively, to nucleotides 312-326, 322-336,
360-374, and 377-391 of the HIV-1 sequence,
are indicated in A. Lane M shows monomeric form of
HIV-1 RNA 224-402. Monomer (m), dimer (d) of RNA 224-402, and free DNA oligonucleotides are
indicated. The results are visualized on 1.5% agarose gel
electrophoresis by ethidium bromide staining (gels above) and
autoradiograms (views below).
P]DNA oligonucleotides marked
monomeric and dimeric forms of the RNA 224-402 equally (see
autoradiogram, Fig. 6B).A New Cis-acting Element Required for HIV-1
Based on the results of DNA
oligonucleotide mapping, we synthetized three new shorter HIV-1 RNA Dimerization in Vitro RNA fragments, derived from RNA 224-402: RNA 224-296,
which only contains the autocomplementary stem-loop I, RNA
296-402, which contains the four polypurine tracks mentioned
above, and RNA 224-402DEL bereft of the
GCGCGCACGG
sequence (Fig. 1b, RNAs 4-6). We analyzed their
ability to dimerize spontaneously in vitro (Fig. 1c). We found that RNA 224-296
dimerized very efficiently (Fig. 1c, lanes 7 and 8), while RNA 296-402 was unable to dimerize (lanes 11 and 12) under the salt conditions used to
study the dimerization process(30) .CGCGC sequence in HIV-1 dimerization, nucleotides
257-266 were deleted from RNA 224-402. This deleted RNA was
analyzed by agarose gel electrophoresis to estimate the degree of
dimerization (Fig. 1b, RNA 5). RNA
224-402DEL lost the capacity to form dimeric RNA (Fig. 1c, lanes 9 and 10) while RNA
224-402 formed up to 80% of the dimer under the same experimental
conditions (Fig. 1c, lane 6). The
10-nucleotide deletion did not alter the overall predicted secondary
structure of the molecule in spite of a modification in the primary
sequence of the stem-loop which was no longer autocomplementary (Fig. 8c). Furthermore, it is noteworthy that RNA
77-257 corresponds to an RNA molecule spliced in loop I: it had
lost the capacity to dimerize efficiently (Fig. 1, b and c, lanes 3 and 4).
dimerization
process. a, predicted structure of stem-loop I, encompassing
nucleotides 257-262, implicated in the recognition of the two
HIV-1 RNA molecules. This structure was determined on
RNAs 77-402, 224-402, and 224-296 using PCFOLD
software(28) . b, formation of a double-stranded helix
by the opening of the predicted stem-loop structure. Nucleotides
spanning 240 through 280 are indicated. c, predicted structure
of the stem-loop lacking nucleotides 257-266 which
contain the GCGCGC sequence.
RNA.Analysis of Heterodimer Formation of HIV-1
Finally, to determine whether dimerization of
HIV-1 RNAs RNA 77-402 and RNAs 224-402 and
224-296 occurred via the same mechanism, we tried to form
heterodimers containing one HIV-1 RNA 77-402
molecule and one molecule of the shorter HIV-1 RNAs. When
an in vitro dimerization reaction was performed with RNA
77-402 and RNA 224-402, or RNA 224-296, we were able
to detect heterodimers. The new band between the monomeric and dimeric
forms of RNA 77-402 was clearly visible (Fig. 7, a and b, lanes HD). By contrast, no intermediate
band could be detected with RNA 296-402 (Fig. 7c, lane HD) or RNA 224-402DEL (Fig. 7d, lane HD): indeed no heterodimer was formed.
RNA heterodimers between RNA 77-402 and RNAs
224-402 (a), 224-296 (b),
224-402DEL lacking nt 257-266 (c) and
296-402 (d). Lanes M show heat-denatured RNA
77-402 and lanes D show RNA 77-402 in dimerization
conditions. Coincubations of the same amount of each of the different
HIV-1 RNAs and RNA 77-402 in dimerization
conditions are shown in lanes HD. Monomer (m) and
dimer (d) of the different RNA fragments are indicated.
Numbering corresponds to that given for RNA fragments in Fig. 1b.
RNA dimerization is most probably
common to RNAs of different sizes, only if they contain the
257-266 sequence. The heterodimers have a dimerization region
similar to that of the homodimers.
RNA
77-402, a 5` leader RNA fragment, can efficiently dimerize in
vitro. This RNA does not contain the TAR domain but, according to
Berkout et al.(21) , who has implicated TAR inverted
sequences in the dimerization process, this domain should not play any
role. Likewise, shorter RNAs 224-402 and 224-296 can
dimerize up to 80-90% in low ionic strength buffer whereas RNAs
77-257 and 296-402 cannot. RNA 224-402 was chosen as
the reference fragment for the study. Computer folding analysis of RNA
224-402 showed that an autocomplementary sequence could be a part
of a stem-loop structure (loop I in Fig. 3). When this
sequence is deleted from this region, the RNA (224-402DEL) is
unable to dimerize (Fig. 1) although the predicted stem-loop
structure may be conserved (Fig. 8c). Furthermore, and
consistent with this finding, a DNA oligonucleotide, complementary to
this region, totally inhibits the HIV-1 dimerization
process (Fig. 4), whereas control oligonucleotides 257M and T do
not (Fig. 5). RNA
224-402 contains three such sites as well as the GGAGG sequence
described by Sundquist and Heaphy (16) , we tested the role of
these elements in the dimerization process. Surprisingly, RNA
224-296, derived from RNA 224-402 but totally bereft of
purine consensus sites, was able to dimerize spontaneously in vitro with the same efficiency (Fig. 1). Furthermore, this RNA
224-296 was able to form a heterodimer with the leader RNA
77-402. RNA 224-296 therefore acts as an antisense RNA
since we observed more heterodimer than the homodimer 77-402 (Fig. 7b). The remaining RNA region 296-402,
which corresponds to the DLS 311-415 of HIV-1,
previously shown to support the necessary sequence for HIV-1
dimerization(9) , was neither able to dimerize in vitro under conditions of low ionic strength (Fig. 1), nor to
form a heterodimer with RNA 77-402 (Fig. 7d). We
therefore failed to inhibit the formation of RNA dimer 224-402 in
the presence of the antisense DNA oligonucleotides 320B, 327B, 365B,
and 382B (Fig. 6) which target the purine tracks. We observed
that the oligonucleotide 327B reduced the amount of the dimer by only
50% (Fig. 6A). The partial inhibition of dimerization
by the oligonucleotide 327B probably implicates another region that may
participate, to a certain extent, in the stabilization of the final
structure of the dimer. However, the total inhibition observed with
oligonucleotide 257B indicates that this stabilization, in itself, is
not sufficient enough to lead to the formation of the dimer.CGCGC in loop I (Fig. 3). We postulate that a transient complex is formed
between complementary nucleotides in loops. The opening of both
stem-loops during the interaction could lead to a double-stranded
region via Watson-Crick base pairing, as shown in Fig. 8, a and b. This mechanism is proposed in the light of two
similar mechanisms described for (i) the formation of a duplex between
RNA I and RNA II from plasmid ColE1 (33) and (ii) the antisense
RNA CopA and its target RNA CopT in the replication of plasmid
R1(34) . DLS
signal is in good agreement with recent analytical studies on
HIV-1 RNA dimerization(22) , except for a
difference observed by us in dimer stability. This difference could be
attributed to a sequence modification when it passes from HIV-1 to HIV-1 RNA (5` . . .
G
AAGCGCGCACGG . . . 3` to 5` . . .
G
AGGUGCACACAG
. . . 3`). The HIV-1 RNA 224-402 dimer is strongly stabilized through the GC
residues present in the dimerization sequence (Fig. 8). It
dissociates into the monomer at 53 °C (Fig. 2). This Tm
value is in good agreement with that found by Darlix et al.(9) with the full-length HIV-1 viral RNA
isolated from wild-type virus particles. proviral DNA clones. One of them, lacking 13 bases
upstream from the splice donor site, produced a virus with less
efficient packaging of its genomic RNA than a virus bearing mutations
between the 5` splice donor site and the gag gene. This
deletion encompasses nucleotides 241-253 which are a part of
stem-loop I shown in Fig. 3and Fig. 8a.
Understanding the structures present in dimeric retroviral RNAs is
therefore an essential prerequisite for antiviral strategies. We
therefore propose stem-loop I as a potential target in attempts to
interfere with HIV-1 replication. Since one of the current antisense
strategies is based on targeting oligonucleotides to complementary
sequences of an RNA molecule, oligonucleotide 257B is proposed for this
function. HIV-1 RNA is totally inhibited in
vitro, under conditions akin to physiological ones, by this
antisense DNA oligonucleotide complementary to the putative stem-loop I
which contains the autocomplementary GCGCGC sequence.
) or or ,
human immunodeficiency virus type 1 (strains Lai or Mal or NL43);
MoMuLV, Moloney murine leukemia virus; RSV, Rous sarcoma virus; BLV,
bovine leukemia virus; REV, reticuloendotheliosis virus; TAR,
trans-activation response element; nt, nucleotide.)
We thank
Frédéric Subra (IGR,
Villejuif) for the gift of plasmid pBRU2, Philippe
Fossé (IGR, Villejuif) for helpful discussions,
and Filippo Rusconi di Lugano (CNRS, Gif-sur-Yvette) and Lorna St-Ange
(IGR, Villejuif) for critical reading of the manuscript. RNA. Our results and theirs are
complementary and mutually supportive: we reached the same postulated
dimerization model of HIV-1 RNA.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. L. Jones, S. Sonza, and J. Mak Primary T-lymphocytes rescue the replication of HIV-1 DIS RNA mutants in part by facilitating reverse transcription Nucleic Acids Res., March 1, 2008; 36(5): 1578 - 1588. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bernacchi, S. Freisz, C. Maechling, B. Spiess, R. Marquet, P. Dumas, and E. Ennifar Aminoglycoside binding to the HIV-1 RNA dimerization initiation site: thermodynamics and effect on the kissing-loop to duplex conversion Nucleic Acids Res., December 18, 2007; 35(21): 7128 - 7139. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Sinck, D. Richer, J. Howard, M. Alexander, D. F.J. Purcell, R. Marquet, and J.-C. Paillart In vitro dimerization of human immunodeficiency virus type 1 (HIV-1) spliced RNAs RNA, December 1, 2007; 13(12): 2141 - 2150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Reblova, E. Fadrna, J. Sarzynska, T. Kulinski, P. Kulhanek, E. Ennifar, J. Koca, and J. Sponer Conformations of Flanking Bases in HIV-1 RNA DIS Kissing Complexes Studied by Molecular Dynamics Biophys. J., December 1, 2007; 93(11): 3932 - 3949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. T. Baig, J.-M. Lanchy, and J. S. Lodmell HIV-2 RNA dimerization is regulated by intramolecular interactions in vitro RNA, August 1, 2007; 13(8): 1341 - 1354. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Kieken, F. Paquet, F. Brule, J. Paoletti, and G. Lancelot A new NMR solution structure of the SL1 HIV-1Lai loop-loop dimer Nucleic Acids Res., January 12, 2006; 34(1): 343 - 352. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Ennifar, J.-C. Paillart, A. Bodlenner, P. Walter, J.-M. Weibel, A.-M. Aubertin, P. Pale, P. Dumas, and R. Marquet Targeting the dimerization initiation site of HIV-1 RNA with aminoglycosides: from crystal to cell. Nucleic Acids Res., January 1, 2006; 34(8): 2328 - 2339. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Baba, K.-i. Takahashi, S. Noguchi, H. Takaku, Y. Koyanagi, N. Yamamoto, and G. Kawai Solution RNA Structures of the HIV-1 Dimerization Initiation Site in the Kissing-Loop and Extended-Duplex Dimers J. Biochem., November 1, 2005; 138(5): 583 - 592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. S. Chin, T. D. Rhodes, J. Chen, W. Fu, and W.-S. Hu Identification of a major restriction in HIV-1 intersubtype recombination PNAS, June 21, 2005; 102(25): 9002 - 9007. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Buxton, G. Tachedjian, and J. Mak Analysis of the Contribution of Reverse Transcriptase and Integrase Proteins to Retroviral RNA Dimer Conformation J. Virol., May 15, 2005; 79(10): 6338 - 6348. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-C. Paillart, M. Dettenhofer, X.-f. Yu, C. Ehresmann, B. Ehresmann, and R. Marquet First Snapshots of the HIV-1 RNA Structure in Infected Cells and in Virions J. Biol. Chem., November 12, 2004; 279(46): 48397 - 48403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Huthoff, A. T. Das, M. Vink, B. Klaver, F. Zorgdrager, M. Cornelissen, and B. Berkhout A Human Immunodeficiency Virus Type 1-Infected Individual with Low Viral Load Harbors a Virus Variant That Exhibits an In Vitro RNA Dimerization Defect J. Virol., May 1, 2004; 78(9): 4907 - 4913. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-M. LANCHY, J. D. IVANOVITCH, and J. S. LODMELL A structural linkage between the dimerization and encapsidation signals in HIV-2 leader RNA RNA, August 1, 2003; 9(8): 1007 - 1018. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Balakrishnan, B. P. Roques, P. J. Fay, and R. A. Bambara Template Dimerization Promotes an Acceptor Invasion-Induced Transfer Mechanism during Human Immunodeficiency Virus Type 1 Minus-Strand Synthesis J. Virol., April 15, 2003; 77(8): 4710 - 4721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-I. Sakuragi, S. Ueda, A. Iwamoto, and T. Shioda Possible Role of Dimerization in Human Immunodeficiency Virus Type 1 Genome RNA Packaging J. Virol., April 1, 2003; 77(7): 4060 - 4069. [Abstract] [Full Text] [PDF]
|
|
|
|
|
< |
