| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, June 22,
1995; and in revised form, August 9, 1995) From the
Reverse transcription of a retroviral RNA genome requires two
template jumps to generate the linear double-stranded DNA required for
integration. The RNase H activity of reverse transcriptase has several
roles during this process. We have examined RNase H cleavages that
define the maximal 3` and 5` ends of Moloney murine leukemia virus
minus strand DNA prior to the second template jump. In both the
endogenous reaction and on model substrates in vitro, RNase H
cleaves the genomic RNA template between the second and third
ribonucleotides 5` of the U5/PBS junction, but other minor cleavages
between 1 and 10 nucleotides 5` of this junction are also observed.
Similar experiments examining the specificity of RNase H for tRNA
primer removal revealed that cleavage generally leaves a ribo A residue
at the 5` end of minus strand DNA. These observations suggest that
three bases are typically duplicated on the ends of the minus strands,
leading to an intermediate following the second jump which contains
unpaired nucleotides. Model substrates mimicking the structure of this
intermediate demonstrate that reverse transcriptase has little
difficulty in utilizing such a branched structure for the initiation of
displacement synthesis.
The single-stranded plus sense RNA genome of a retrovirus is
converted into double-stranded linear DNA by the process of reverse
transcription (for review, see (1) and (2) ). At the
two ends of the viral genome, there is a direct repeat (R) sequence
that is bordered by an internal unique 5` (U5) or 3` (U3) sequence.
Immediately downstream of the 5` R-U5 sequence is the primer binding
site (PBS), ( The enzyme responsible for reverse transcription is the
viral-encoded reverse transcriptase (RT), which contains an RNA- and
DNA-dependent DNA polymerase and an RNase H
activity(1, 2) . The RNase H activity serves three
distinct roles during reverse transcription (for review, see (8) ). First, RNase H degrades the viral RNA genome in the
RNA-DNA hybrids generated during minus strand DNA synthesis (see Fig. 1). This degradation appears to be relatively nonspecific
and produces fragments ranging from 2 to 20 bases in length (9, 10, 11, 12, 13, 14, 15, 16, 17) .
While concomitant DNA synthesis is not required for RNase H
activity(18, 19) , the positioning of the RNase H
active site is strongly influenced by binding of the polymerase to a 3`
DNA primer
terminus(16, 18, 20, 21) . The
second role of RNase H is to cleave the viral template at the
polypurine tract (PPT), a purine-rich region immediately upstream of
U3, to generate the primer for plus strand DNA synthesis. Unlike genome
degradation, the PPT cleavage is sequence-specific and highly
precise(22, 23, 24, 25, 26) .
Third, RNase H removes the RNA primers used to initiate plus and minus
strand DNA synthesis. Removal of the plus strand PPT primer uniformly
occurs at the junction between the RNA primer and the first
deoxynucleotide of plus strand DNA (25, 26) . In
contrast, removal of the tRNA primer from minus strand DNA appears to
depend on the type of virus examined. For avian myeloblastosis virus,
the primer is cleaved precisely between the last ribonucleotide of the
tRNA and the first deoxynucleotide of the minus
strand(27, 28) . However, the tRNA primer of human
immunodeficiency virus type 1 (HIV-1) is cleaved between the last and
the penultimate ribonucleotides to leave a ribo A residue at the 5` end
of the minus strand DNA(18, 29, 30) .
Figure 1:
RNase H cleavage sites important for
second jump in reverse transcription. The current model of reverse
transcription is depicted in sequential steps (stages
i-v) through the second template jump, with RNA
strands as light lines and DNA strands as dark lines.
Viral sequences that serve as landmarks for replication are shown in
capital letters, and the complementary sequences are indicated with a prime. The 3` end of each strand is indicated by an arrowhead. For the second jump, RNase H cleavages are required
in the regions marked I and II. RNase H cleavage of
the genome template in region I (stage i) determines
the maximal position to which the minus strand can be extended prior to
the second jump (stage iv). RNase H cleavage in region II (stage iii) removes the tRNA primer to free the PBS
sequences for the second jump and determines the 5` end of the minus
strand (stage iv). The second jump intermediate is depicted in stage v.
The
model of reverse transcription predicts that RNase H activity is
necessary for both the first and second template jumps to occur (3) (see Fig. 1). After minus strand synthesis has
extended the tRNA primer to the end of the genome (minus strong stop
DNA; stage i), RNase H facilitates the first jump by degrading some
portion of R at the 5` end of the viral RNA. This frees minus strand R`
for transfer to R at the 3` end of the genome and continuation of
synthesis (stage ii). Consistent with this model, RNase H cleaves the
template RNA up to 14-24 nucleotides from its 5` end in DNA-RNA
hybrids, including those which contain
R(20, 31, 32, 33) . While some in vitro studies using model substrates and RNase H-deficient
RT have found that RNase H activity is necessary for a template
switch(15, 34) , others have reported that RNase H
activity is not required(32, 35) . However, viral
mutants lacking RNase H activity do not complete reverse transcription in vivo and only synthesize minus strong stop DNA in
endogenous reactions (36). For the second jump, cleavages by RNase H
are important for three different stages of reverse transcription (see Fig. 1). The first cleavage occurs on the genome 5` of the PBS
after minus strand synthesis has begun (designated ``region
I,'' stage i). The 3`-most cleavage in this region determines the
maximum point to which the minus strand DNA can be extended prior to
the second jump (stage iv). Whether this RNase H cleavage is random or
highly specific has not been previously addressed. The second RNase H
cleavage removes the tRNA primer after plus strand synthesis has
extended from the PPT primer through to the end of PBS (designated
``region II,'' stage iii). This cleavage determines the 5`
end of the minus strand and presumably frees the 3` end of plus strand
DNA for the second jump (stage iv). Finally, the PBS` sequence of the
extended minus strand DNA is exposed by RNase H degradation of the
genomic template (stage iv). Together, these RNase H cleavages permit
the PBS sequence of the plus strand to pair with the PBS` sequence of
the minus strand for the second jump (stage v). The specificity of
RNase H for the first two cleavages is an intriguing issue, as any
duplication in sequence between the 3` and 5` ends of minus strand DNA
could generate a problematic branched intermediate subsequent to the
second jump (stage v). Using Moloney murine leukemia virus (M-MuLV)
as a model system, we have examined the specificities of RNase H that
define the two termini of minus strand DNA prior to the second jump.
These results allow us to predict the predominant structure that would
result from the second jump according to the current model of reverse
transcription.
Oligonucleotides were 5` end-labeled in a 15-µl reaction
containing 50 mM Tris-HCl (pH 7.5), 10 mM MgCl
Figure 2:
Assays for RNase H cleavage sites. The
drawings show the different assays used to map RNase H cleavages in the
genome 5` of the PBS (A) and the RNase H cleavage event that
removes the tRNA primer (B). RNA strands are depicted as light lines, DNA strands as dark lines, and relevant
viral sequences in capital letters. The asterisks indicate
Figure 3:
Determination of the cleavage sites within
the plus strand U5-PBS region. RNA oligo Mol15R was annealed to a
template RNA containing R-U5-PBS sequences and either used directly as
a substrate for RT (Primer only) or Mol15R was extended with
RNase H
We next examined the RNase H cleavage pattern on genomic
RNA upstream of the PBS in endogenous reactions. Detergent-disrupted
M-MuLV virions were incubated with dNTPs, and products synthesized over
time were analyzed by primer extension using oligo 1 (Fig. 2A). In the absence of synthesis (zero time), the
primer extension products were similar to those observed with
substrates containing unextended RNA in the in vitro assay
with model substrates (compare Fig. 4, lane 1, with Fig. 3, lane 4). The vast majority of products mapped
to the 5` end of the genomic RNA (Fig. 4, lane 1, arrow a), with a small fraction mapping to the PBS region
itself (arrow c). Cleavages within the PBS could have resulted
from very low levels of RNase H* activity within the virions prior to
DNA synthesis. It would appear that these minor cleavage products
persisted throughout the time course, although the bands are fainter in
the 4-24 h time points (Fig. 4, lanes 2-5)
owing to a reduced recovery of products for these samples. After 4 h of
synthesis, the amount of genomic length products decreased
dramatically, and several products mapping immediately 5` of the PBS
appeared. The major product was identical to that observed using the
RNA-DNA hybrid substrate in the in vitro assay and
corresponded to RNase H cleaving the viral genome between the second
and third ribonucleotides 5` of the PBS (Fig. 4, lane
2, arrow b). Several minor cleavages were also detected,
beginning with the first ribonucleotide 5` of the PBS and extending up
to 10 ribonucleotides 5` of the PBS. The relative abundance of these
secondary cleavage sites decreased with increasing lengths of time,
reflecting further RNase H activity on the genomic RNA template as
reverse transcription progressed (Fig. 4, lanes
3-5).
Figure 4:
Cleavage sites on genomic RNA after minus
strand synthesis in the endogenous reaction. Using detergent-disrupted
M-MuLV virions, endogenous reverse transcription reactions were carried
out for 4 h (lane 2), 8 h (lane 3), 16 h (lane
4), or 24 h (lane 5), and products were recovered by
phenol extraction and ethanol precipitation. Products were similarly
prepared from detergent-disrupted virions in the absence of any
synthesis (0 h; lane 1). To map the 5` ends of genomic RNA
upstream of the PBS, primer extensions were performed with labeled
oligo 1, and samples were analyzed as described in Fig. 3. The
designations for the arrows and RNA sequence are
identical to Fig. 3.
Figure 5:
Determination of the 5` end of minus
strand DNA at the U5`-PBS` junction after primer removal. A,
RNA oligo Mol15R was annealed to single-stranded EcoRI-cut
M13mp7/PBS
Since a portion of the molecules was not cleaved by RNase H in this
experiment (Fig. 5A, lane 4, arrow
a), we tested whether longer reaction times might facilitate
removal of the ribo A. As shown in Fig. 5B, treatment
with RT through 60 min greatly reduced the amount of 5` ends containing
the full-length RNA primer (arrow a, lanes
3-7). Notably, 5` ends containing a single ribo A increased (arrow b), and in other experiments, the ribo A persisted
after incubating the substrates with RT for as long as 120 min (data
not shown). We next addressed whether primer removal by RNase H
during the endogenous reaction also leaves a ribo A at the 5` end of
minus strand DNA. The minus strand products of a 16-h endogenous
reaction were isolated and analyzed by primer extension reactions using
oligo 2 before and after alkali treatment to determine whether any
ribonucleotides remained on the 5` end of the DNA. Pretreatment with
alkali yielded the expected T4 DNA polymerase primer extension product
corresponding to the 5` end of the minus strand DNA (Fig. 6, lane 3). When the same alkali-treated sample was analyzed
using RNase H
Figure 6:
Determination of the 5` end of minus
strand DNA synthesized in the endogenous reaction. An endogenous
reaction was incubated for 16 h, and the 5` ends of the minus strand
DNAs were mapped using labeled oligo 2A and RNase H
To ask whether the recombinant RT differs from the
virion-derived RT in its ability to cleave at the RNA-DNA junction, we
directly compared the RNase H activities of these enzymes on the model
substrate described in Fig. 2B. The substrate was
incubated with either RT purified from M-MuLV virions or the
recombinant RT for 10 or 90 min, and the products were analyzed by
primer extension using labeled oligo 2 and RNase H
Figure 7:
Primer removal by RT purified from
virions. Substrates were prepared as described in Fig. 5and
treated with purified virion M-MuLV RT (Virion RT; lanes
3-6) or recombinant RT (Recomb. RT; lanes 7-10) for 10 min (lanes 3, 4, 7, and 8) or 90 min (lanes 5, 6, 9, and 10). Untreated samples are shown in lanes
1 and 2. Samples in lanes 1, 3, 5, 7, and 9 were treated with alkali prior
to the primer extension analysis by RNase H
Although retroviral RNase H activity is maximal in the
absence of KCl, typical RT conditions contain 40-50 mM KCl(31, 41) . We have varied the concentration of
KCl with the model substrates and endogenous reactions to find that,
under all conditions, the ribo A remained at the 5` end of minus sense
DNA (data not shown).
Figure 8:
Time course of RNA primer cleavage by
RNase H. 5` end-labeled RNA oligo Mol15R was annealed to
single-stranded EcoRI-cut M13mp7/PBS
Figure 9:
Removal of the RNA primer by M-MuLV versus HIV-1 RT. Substrates prepared as described in Fig. 8were treated with recombinant M-MuLV (lanes
2-4) or purified HIV-1 (lanes 6-8) RT for 0.3
min (lanes 2 and 6), 3 min (lanes 3 and 7), or 27 min (lanes 4 and 8) and analyzed
as described in Fig. 8. Substrates without treatment (0 min; lanes 1 and 5) and the labeled RNA oligonucleotide
without extension (lane 9) are also shown. Positions of size
markers are shown on the right as described in Fig. 8.
To determine whether such a structure
would affect completion of the minus strand, we used a model system
designed to compare initiation and early synthesis on templates
representing either the branched or nicked product of the second jump.
Oligonucleotides corresponding to the 5` end (oligo U5mis) and two
possible 3` ends of the minus strand were annealed to single-stranded
DNA containing the plus sense LTR (Fig. 10A). Oligo
U5mis contained 5 unpaired 3`-terminal nucleotides to prevent its
utilization as a primer, and the two 3` upstream oligonucleotides
contained either the minus sense PBS sequence (oligo PB) or the
extended PBS sequence (oligo PBA) to generate a nicked or three
base-branched substrate, respectively. Shown in Fig. 10B is a graph depicting the time course of displacement of the 5`
end-labeled downstream oligonucleotide (oligo U5mis) as determined by
nondenaturing polyacrylamide gel electrophoresis. Of note is that RT
utilized and extended oligo PBA nearly as well as fully base-paired
oligo PB, although a slight but reproducible lag in displacement and an
apparent decrease in the total amount of displaced product were
associated with the branched substrate. In a separate experiment, it
was found that incubation of the templates with RT for 3 min prior to
the initiation of synthesis with dNTPs did not decrease the observed
lag (data not shown).
Figure 10:
Displacement synthesis on model
primer-templates comparing an unpaired with a paired primer terminus. A, schematic diagram showing the structures of the model
primer-templates constructed from a tandem pair of DNA oligonucleotides
corresponding to the putative 5` and 3` ends of the completed minus
strand annealed to plus sense LTR-containing single-stranded DNA. The arrow indicates the primer terminus that in structure I is base paired with the template and in II is unpaired
and extends beyond the 5` end of the downstream oligonucleotide by
three deoxyadenosine residues. The 3` terminus of the downstream
oligonucleotide is not complementary so it cannot be used as a primer,
and the 5` end is
We have examined the RNase H cleavages that determine the
length of minus strand DNA prior to the second jump during reverse
transcription of M-MuLV. In both the endogenous reaction and on a
hybrid RNA-DNA substrate, RNase H cleaved the genomic template in U5
very proximal to the PBS. The primary RNase H cleavage site was between
the second and third ribonucleotides 5` of the PBS (Fig. 11A), but other cleavages extending through 10
nucleotides 5` of the PBS were also apparent. These cleavages define
the 5` end of the template for minus strand synthesis and would allow
the majority of the minus strands to extend only a few nucleotides
beyond the PBS prior to the second jump (Fig. 11B).
Interestingly, Götte et al.(42) have shown that HIV-1 RT also cleaves the RNA genome
two bases upstream from the U5/PBS boundary during minus strand
synthesis. Using model substrates, we also found that RNase H cleaves
the tRNA primer between the last and penultimate ribonucleotides (Fig. 11C), leaving a single ribo A residue at the 5`
end of minus strand DNA (Fig. 11D). Similar results
were obtained from an analysis of the products of the endogenous
reaction except in this case some of the molecules were apparently
cleaved instead at the RNA-DNA junction. We cannot rule out the
possibility that a ribonuclease other than RNase H was responsible for
the cleavage at the DNA-RNA junction or that this difference results
from a coupling of the cleavage to concomitant DNA synthesis.
Subsequent to the initial cleavage, RNase H cleaved within the RNA
primer to generate smaller fragments.
Figure 11:
RNase H cleavages that generate the two
ends of the minus DNA strand and the resulting structure after the
second jump. The sequences at the U5-PBS border are indicated for a
number of structures synthesized during reverse transcription with the
RNA sequences shaded. A, the arrow above the RNA sequence
indicates the major RNase H cleavage site 5` of the PBS that determines
the 5` end of the template for minus strand DNA synthesis. B,
predicted 3` sequence for the major minus strand product assuming
synthesis proceeds to the end of the template diagrammed in A. C, the arrow indicates the site of the RNase H cleavage that
removes the tRNA primer and determines the 5` end of the minus DNA
strand shown in D. E, the predominant structure that
is predicted after the second jump. Note that the three unpaired A
residues, which extend from the site of the nick, can become paired
with the opposing plus strand to varying extents by branch migration
(see text).
The apparent 5` and 3` termini
for M-MuLV minus strand DNA suggest a detailed structure for the
intermediate after the second jump of reverse transcription (Fig. 11). Upon initiation at the polypurine tract, plus strand
DNA synthesis extends through the PBS and terminates at the first
modified base in the tRNA(43) . To continue synthesis, plus
strand strong-stop DNA transfers to the 3` terminus of minus strand DNA (Fig. 1, stage v), with the complementary sequences
(PBS/PBS`) at the ends of minus strand DNA facilitating the
jump(3) . We predict that the first three bases at the 3` end
of the minus strand (3`-AAA . . . 5`) are not initially base-paired to
the plus strand DNA (Fig. 11E). As a consequence,
branch migration is necessary to position the free 3` end of the minus
strand for use as a primer, and then displacement synthesis is required
for minus strand elongation. In contrast, the 3` end of plus strand DNA
is situated for immediate extension after the second jump. Only the
intermediate arising from the major RNase H cleavage between the second
and third nucleotides upstream of PBS is shown in Fig. 11E, but others are also predicted by our results.
These additional structures would possess unpaired 3` termini, which
are even longer, resulting in a greater requirement for branch
migration prior to the initiation of displacement synthesis. For the
intermediate structure proposed for the second jump, we inferred the
structure of the 3` end of the minus strand by assuming that synthesis
proceeds to the 5` end of the RNA template prior to the jump. Although
we cannot rigorously exclude that the second jump occurs before
completion of the minus strand, this possibility seems unlikely for the
following reasons. First, given the displacement activity of RT (44) , there is no compelling reason to suppose that the known
secondary structure in the template in the U5-PBS region (for review,
see (45) ) would present a barrier to the completion of the
minus strand. Second, it seems unlikely that strand transfer would
precede the extension of minus strand DNA through to the end of the RNA
template because degradation of the template by RNase H is required for
the first jump (36) and is likely to be required for the second
jump as well(3) . Based on footprinting and structural studies,
the DNA polymerase and RNase H active sites of RT are separated by
approximately 18 base pairs(46, 47, 48) .
Since the PBS of M-MuLV is 18 nucleotides in length, the RT that is
synthesizing the minus strand would be near the 5` end of the genomic
template before the RNase H activity would be positioned for cleavages
in the PBS. The properties of partial deletion and substitution mutants
in the HIV-1 PBS further suggest that minus strand synthesis may
typically traverse the entire PBS prior to the second
jump(49, 50) . Notably, Luo and Taylor (32) found that 20 but not 10 nucleotides are sufficient for
the M-MuLV strand transfer reaction in vitro. An
interesting question is whether the proposed branched intermediate
created by the second jump slows minus strand DNA synthesis
significantly (Fig. 11E). Extension of minus strands
subsequent to the second jump would require resolution of the unpaired
primer by branch migration, which in principle could be facilitated by
RT. If such branch migration were to be an impediment to extension of
the minus strand, resolution could result from displacement through the
LTR from the 5` end by extension of the plus strand DNA. Using
substrates that model the predicted second jump product, RT extended a
3` end requiring a three-base branch migration nearly as efficiently as
a completely base-paired 3` primer terminus. The slight lag in
extension associated with the unpaired primer terminus suggests that
the structure was not resolved by branch migration during the 5-min
annealing period prior to the initiation of synthesis. Additionally,
the failure of RT to reduce the lag following an incubation with the
template minus dNTPs suggests that, at least under nonsynthesis
conditions, RT does not alter the equilibrium between the base paired
and unpaired forms. Given that reverse transcription occurs over a
period of hours(1) , these results suggest that, although an
unpaired 3` terminus is not without consequence for RT, the magnitude
of the delay is small enough that it is unlikely to have a significant
impact on the overall course of retroviral replication. It is uncertain
what effect longer extensions of the minus strand 3` end would have on
minus strand synthesis following the second jump. We would anticipate
that an increase in the size of the unpaired primer terminus would lead
to an increase in the lag before synthesis and perhaps a decrease in
the overall utilization of the 3` minus strand following the second
jump. The presence of a ribo A at the 5` terminus of the majority of
M-MuLV minus strands may affect the structure of the right LTR end and
is interesting to consider with regard to integration. During
retroviral integration, two base pairs are generally lost from each end
of the linear viral DNA such that the first and last two bases of the
integrated virus are highly conserved and begin with 5`-TG . . . at the
left terminus and end with . . . CA-3` at the right terminus (for
review, see (7) ). The avian RT precisely cleaves its tRNA
primer at the junction between minus strand DNA and RNA(27) ,
which generates symmetrical ends for integration. In contrast, the tRNA
primer of HIV-1 is incompletely removed, and a ribo A remains at the 5`
end of the minus strand on model
substrates(18, 29, 30) . These biochemical
studies are supported by the observation that most HIV-1 circle
junctions contain an additional A, which is not present in the
predicted sequence (51, 52, 53) and that
viral double-stranded DNA contains the extra base pair in the right
LTR(54) . However, the presence of a ribo A at the 5` end of
the HIV-1 minus strand DNA still results in symmetrical LTRs from which
two base pairs are removed from both ends during integration. In the
case of M-MuLV, if the ribo A residue is retained on the 5` end of
minus strands, and synthesis of plus strands extends to the end of the
minus strand template to copy the ribo A, then the right end of linear
unintegrated DNA would end with the sequence 5` . . . CATTT-3`, and
three base pairs would have to be removed from the right LTR end during
integration. It seems unlikely that the additional base pair would
affect viral integration as previous studies have demonstrated that
extra sequences flanking the conserved 5`-TG . . . and . . . CA-3`
bases at the termini of M-MuLV do not interfere with
integration(55, 56) . The addition of two extra base
pairs to the right LTR resulted in a viral mutant that replicated and
retained the two base pair addition. Interestingly, two of the four
circle junction sequences analyzed for this virus contained an
additional T that might have arisen from incomplete removal of the
tRNA(56) . Alternatively, M-MuLV minus strand DNA may
undergo some additional processing that specifically removes the ribo A
prior to the completion of plus strand synthesis or integration.
Although we did not observe removal of the ribo A residue in the
reconstructed reactions in vitro, it appeared that the ribo A
was removed from some minus sense DNA ends in the endogenous reaction.
Thus we might expect a mixed population of molecules with respect to
the structure of the 5` ends of minus strands in vivo. The
sequence of the one reported M-MuLV circle junction (57) does
not contain an extra A at the position predicted by our data, but given
the small sample size, it is difficult to generalize from this
observation. Several studies examining the M-MuLV double-stranded DNA
product of reverse transcription failed to detect minus strand DNA with
a length suggesting the presence of an extra A at the 5`
end(43, 58, 59) , and one of these studies
did not find an extra T at the 3` end of plus strand DNA(43) .
While these observations might indicate that the ends of intracellular
intermediates are different from those generated in this study, it
remains equally likely that the ribo A is somehow removed from minus
strand 5` ends after the second jump but before completion of the plus
strand. Sequence analyses of additional M-MuLV circle junctions might
reveal the relative frequency of occurrence of the two predicted
structures for the 5` ends of minus strands. It remains to be
determined why the RNase H activities of both M-MuLV and HIV-1 can
remove the plus strand primer precisely at the RNA-DNA junction (25, 26) yet prefer to cleave the tRNA primer one
nucleotide away from the junction to leave a ribonucleotide on the DNA
strand. Interestingly, Huang et al.(60) showed that
the calf thymus RNase H1 invariably cleaves one base away from the
RNA-DNA junction. Through the use of two-dimensional NMR studies, the
structure of a duplex oligonucleotide that models the M-MuLV tRNA-minus
DNA junction has been determined. (
Volume 270,
Number 41,
Issue of October 13, 1995 pp. 24135-24145
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)which positions the primer tRNA used to
initiate minus strand synthesis. During reverse transcription, these
sequences are duplicated through two template jumps to generate long
terminal repeats (LTRs), which are present at both ends of the
unintegrated viral DNA and have the structure U3-R-U5 (1, 2, 3) . Upon integration into the host
cell chromosome, two base pairs are lost from the termini such that the
left LTR begins with the sequence 5`-TG . . . , and the right LTR ends
with the sequence . . . CA-3` ( (4, 5, 6) and for review, see (7) ).
Nucleic Acids
To generate single-stranded DNAs containing the R-U5-PBS
region in both the plus and minus sense orientations, the 280-bp SfaNI-XmaIII fragment (positions 66-346 in the
M-MuLV genome(37) ) was inserted into the HincII site
of M13mp7 and clones of both the plus polarity
(M13mp7/PBS) and minus polarity
(M13mp7/PBS
) were isolated. Clone pGMPBS was
constructed by introducing the 1529-bp KpnI-XhoI
fragment (positions 32-1560 in the M-MuLV genome) into the pGEM
vector (Promega Corp.). Clone pGMPBSR was created by inserting the
165-bp PvuII-SfaNI fragment of pGMPBS into the SphI site of pGMPBS, digesting with PstI to release
the 176- and 821-bp fragments in the M-MuLV sequence, and religating
the DNA. A 738-bp region of the M-MuLV genome containing the entire LTR
region was amplified by polymerase chain reaction and cloned into
pBluescript II KS(+) at the BamHI and EcoRI
restriction sites to generate a phagemid that produced single-stranded
DNA corresponding to the plus sense LTR (pBSMOLTR+). The various
recombinant M13 phages and DNAs and the M13 phagemid DNA were isolated
by established procedures(38) .
Enzymes
Recombinant wild-type M-MuLV RT, T7 RNA polymerase, and
Sequenase version 2.0 were purchased from U. S. Biochemical Corp.,
RNase H RTs (Superscript and Superscript II) were
obtained from Life Technologies, Inc., and M-MuLV virion RT was
purified as described previously(22) . Recombinant
heterodimeric HIV-1 RT was kindly provided by Lawrence Loeb (University
of Washington). Restriction enzymes and T4 DNA polymerase were
purchased from New England Biolabs, Inc. T4 polynucleotide kinase was
purchased from both U. S. Biochemical Corp. and Life Technologies, Inc.
Oligonucleotides and 5` End Labeling
DNA oligo 1 (5`-TGGCCAGCTTACCTCCCGG-3`) was purchased from
Oligos Etc. DNA oligo 2 is represented by two oligonucleotides sharing
the same 3` end but differing at the 5` end; oligo 2A
(5`-TGGTCTCGCTGTTCCTTGGG-3`) was synthesized with a Biosearch model
8600 DNA synthesizer, and oligo 2B (5`-TGTGGTCTCGCTGTTCCTTGGG-3`) as
well as the other DNA oligonucleotides described below were purchased
from DNA Express. Results were identical using oligo 2A or oligo 2B.
RNA oligo Mol15R (5`-CCGGACGAGCCCCCA-3`), which matches the 15 3`-most
nucleotides of the tRNA primer, was kindly provided by
Brian Reid (University of Washington). The first 30 bases of DNA oligo
U5mis (5`-AAATGAAAGACCCCCGCTGACGGGTAGTCAGAGTG-3`) are complementary to
the M-MuLV sequence from positions 146 to 117 with the five 3`-most
nucleotides deliberately chosen to prevent base pairing with the
genomic sequence. DNA oligo PB (5`-ATCCCGGACGAGCCCCC-3`) is
complementary to the M-MuLV sequence from positions 163 to 147, which
corresponds to the PBS without the 3`-terminal A residue. Oligo PBA is
identical to oligo PB with the addition of three A residues on the 3`
end.
, 5 mM dithiothreitol (DTT), 50 µg/ml
of bovine serum albumin, 50 µCi of
[-
P]ATP (DuPont NEN), 17 pmol of
oligonucleotide, and 2-30 units of T4 polynucleotide kinase. The
reaction was incubated at 37 °C for 60 min and stopped with
10-50 mM EDTA. In most cases, the labeled
oligonucleotides were separated from labeled ATP using a spin column
containing Sephadex G-25 (Pharmacia Biotech Inc.)(38) .
M-MuLV Endogenous Reaction
M-MuLV virions were isolated as described previously (25) (final concentration 2 µg/ml of protein) and were
incubated in a 250-µl reaction containing 50 mM Tris-Cl
(pH 8.0), 50 mM KCl, 10 mM MgCl, 0.4
mM DTT, 0.01% Nonidet P-40, and 0.25 mM deoxynucleoside triphosphates (dNTPs) at 40 °C for the times
indicated. After termination with 15 mM EDTA, SDS and
proteinase K were added to final concentrations of 0.25% and 100
µg/ml, respectively, and the reactions were incubated at 37 °C
for 60 min. Samples were extracted twice with phenol and twice with
chloroform and precipitated with 20 µg/ml of glycogen (Boehringer
Mannheim) in the presence of 0.3 M sodium acetate (pH 5.2)
with two volumes of ethanol. After resuspension in 20 µl of TE (10
mM Tris-HCl (pH 8.0), 1 mM EDTA), 1-4 µl of
each sample was treated with 0.3 M NaOH at 65 °C for 20
min, neutralized with acetic acid, and ethanol precipitated prior to
analysis by primer extension. For the zero time control, the sample was
prepared identically except that virions were added to a 250-µl
reaction that lacked any dNTPs, and the sample was directly purified
without incubation.Analysis of Plus Strand RNA Cleavages in U5 Proximal to
the PBS
Production of RNA Template
To prepare RNA
containing plus sense R-U5-PBS sequences, 2.5 µg of HindIII-linearized pGMPBSR was transcribed in a 100-µl
reaction containing 40 mM Tris-HCl (pH 7.5), 10 mM
MgCl, 5 mM DTT, 50 µg/ml of bovine serum
albumin, 0.4 mM ribonucleoside triphosphates, 100 units of
RNasin (Promega), and 40 units of T7 RNA polymerase at 37 °C for 90
min, and the DNA template was digested with 2.5 units of DNase I
(Promega) at 37 °C for 15 min. The reaction was brought to a final
volume of 200 µl with TE containing 0.1% SDS, and extracted once
with phenol and twice with chloroform. The RNA was precipitated in the
presence of 2.5 M ammonium acetate with 2.5 volumes of ethanol
and resuspended in 15 µl of TE.Preparation of Hybrid Substrates
A 50-µl
reaction containing 120 ng of RNA oligo Mol15R and 2 µg of RNA
template in 50 mM Tris-HCl (pH 8.0), 40 mM KCl, 6
mM MgCl, and 1 mM DTT was heated at 65
°C for 3 min and cooled at room temperature for 3 min. dNTPs were
added to a final concentration of 0.5 mM each followed by 200
units of RNase H RT (Superscript II). The reaction
was incubated at 37 °C for 30 min and terminated by the addition of
EDTA to 10 mM. In addition to the desired extension of the RNA
oligonucleotide, some snap-back self-priming resulted in the formation
of an RNA-DNA hybrid region 3` of the PBS(39) . This led to
some cleavage of the RNA in this region and reduced the overall signal
in the primer extension assay but otherwise had no affect on the
results. Hybrids were ethanol precipitated twice in the presence of 2.5 M ammonium acetate and resuspended in 10 µl of TE.
Substrates in which the RNA primer was not extended were prepared by
annealing the RNA primer to the RNA template followed by direct
precipitation.
Treatment with RT
1.5 µl of the hybrids
prepared above were treated with 250 units of recombinant M-MuLV RT in
a 10-µl reaction containing 1 
RT buffer (50 mM
Tris-HCl (pH 8.0), 10 mM MgCl, and 0.8 mM DTT) at 37 °C for 15 min and terminated by addition of EDTA to
a final concentration of 10 mM. Sodium acetate was added to
0.3 M and the samples were ethanol precipitated. The pellet
was resuspended in 7.5 µl of TE and analyzed by primer extension
using oligo 1 (see Fig. 2A).
P-labeled 5` ends of RNA or DNA
oligonucleotides as specified in the text, and the 3` end of each
strand is indicated by an arrowhead. Depending on the
experiment, the RNA paired with the PBS (PBS`) is either the
RNA oligo Mol15R or the primer tRNA. A, primer extension from
labeled oligo 1 maps cleavages on the plus strand genome RNA. B, extension from labeled oligo 2 was used to map the 5`-end
of the minus strand DNA after primer removal and 5`-end-labeled Mol15R
was used to determine the fate of the RNA
primer.
Analysis of the 5` End of Minus Sense DNA
The 322-base insert of M13mp7/PBS was
released from the single-stranded recombinant using EcoRI as
described previously(22) . The resulting mixture of
single-stranded insert and vector DNA is referred to as
``EcoRI-cut M13mp7/PBS
DNA.''
Preparation of Substrates
490 ng of RNA oligo Mol15R was
annealed to 3.5 µg of EcoRI-cut M13mp7/PBS DNA, and the RNA was extended with Sequenase essentially as
described previously(29) .
Treatment with RT
One-fourth of the substrates
prepared above were treated with 100-250 units of recombinant RT
or 40 units of purified virion RT in a 10-µl reaction containing 1
RT buffer at 37 °C for the indicated times. After
termination with 10 mM EDTA, one-half of each reaction was
treated with alkali as described above, brought to a final volume of 20
µl, and ethanol precipitated. The remaining half of each reaction
was diluted to 20 µl with TE and ethanol precipitated in the
presence of 0.3 M sodium acetate. All precipitates were
resuspended in 7.5 µl of TE and analyzed by primer extension using
oligo 2A or 2B as indicated in the figure legends (see Fig. 2B).Primer Extension Analyses
One-half of each RT-treated sample and 0.4 pmol of the
specified 5` end-labeled DNA oligonucleotide were used in primer
extension reactions with RNase H RT (Superscript)
essentially as described (29) or with 1 unit of T4 DNA
polymerase in 10 mM Tris-HCl (pH 8.0), 50 mM NaCl, 10
mM MgCl
, 1 mM DTT, 0.2 mM dNTPs
and analyzed in an 8% polyacrylamide gel containing 8 M urea.Analysis of RNA Primer Fate
Preparation of Substrates
25 ng of 5`
end-labeled RNA oligo Mol15R was annealed to 21 µg of EcoRI-cut M13mp7/PBS DNA in 1
Sequenase buffer and extended with Sequenase as described above. The
reaction was stopped by the addition of 25 µl of 5 DNA
sample buffer (25% Ficoll, 100 mM EDTA, 5% xylene cyanole, 5%
bromphenol blue) and subjected to electrophoresis in a 5%
polyacrylamide gel in a Tris borate/EDTA buffer(38) . The
extended products were eluted from a gel slice in 0.5 M ammonium acetate, 1 mM EDTA (pH 7.9) for a total of 16 h,
precipitated with 2 volumes of ethanol, and then reprecipitated in the
presence of 0.3 M sodium acetate with two volumes of ethanol.
The products were resuspended in a small volume of TE, heated at 60
°C for 5 min, and purified a second time by 5% polyacrylamide gel
electrophoresis. Alternatively, reactions containing the extended
products were terminated by adding an equal volume of formamide stop
buffer and heating at 85 °C for 3 min, and the single-stranded
extension products were isolated from a 5% polyacrylamide gel as
described above. The recovered extension products were annealed to 2.4
µg of cut M13mp7/PBS DNA in 20 mM Tris-HCl (pH 7.5), 0.3 M NaCl, 1 mM EDTA at 65
°C for 10 min, cooled to room temperature for 20 min, and
precipitated with ethanol. To generate substrates in which the RNA
primer was not extended, RNA oligo Mol15R was annealed to EcoRI-cut M13mp7/PBS
DNA and isolated
directly from a 5% polyacrylamide gel as described above. In all cases
the products were resuspended in TE after precipitation.
Treatment with RT
One-fourth to one-half of the
purified substrates was treated with 10 units of recombinant M-MuLV or
HIV-1 RT in a 10-µl reaction containing 1 RT buffer at 37
°C for the indicated times. Samples were mixed with an equal volume
of formamide stop buffer, denatured at 90 °C for 5 min, and
analyzed by electrophoresis in a 20% polyacrylamide gel containing 8 M urea.Size Markers
Sequencing ladders were prepared using a Sequenase kit
essentially as described by the manufacturer (U. S. Biochemical Corp.).
The plus sense ladder was generated using oligo 2A or 2B and
M13mp7/PBS DNA, while the minus sense ladder was
generated using oligo 1 and M13mp7/PBS
DNA. A size
ladder of the labeled RNA oligo Mol15R was produced by partial
digestion with 0.2 µg/ml of nuclease P1 (P-L Biochemicals) in a
10-µl reaction containing 50 mM sodium acetate (pH 5.5) at
37 °C for 30 min.
Assay for Capacity of RT to Initiate Displacement
Synthesis with a Mismatched Primer Terminus
5` end-labeled oligo U5mis (2 pmol) was annealed to 2.7 pmol
of pBSMOLTR+ single-stranded circular template DNA in 30 µl of
100 mM Tris-HCl (pH 8.3), 100 mM KCl, and 12
mM MgCl. Control experiments showed that the five
mismatched nucleotides at the 3` end of oligo U5 mis were sufficient to
prevent its usage as a primer and that no displacement by RT occurred
in the absence of added upstream primer (data not shown). The annealing
reaction was divided in half and annealed with 1 pmol of unlabeled
oligo U5mis and 7.5 pmol of oligo PB or PBA in 100 mM Tris-HCl
(pH 8.3), 100 mM KCl, and 12 mM MgCl (final volume 20 µl). To minimize the potential for
resolution of the unpaired oligo PBA 3`-terminus by branch migration,
the second annealing step was carried out for only 5 min at 37 °C.
The subsequent extension reactions (20 µl final volume) contained 9
µl of the annealed mixture in 50 mM Tris-Cl (pH 8.3), 50
mM KCl, 6 mM MgCl, 5 mM DTT, and
0.2 mM dNTPs. A sample was removed for analysis as the zero
time point, synthesis was initiated with 120 units of recombinant RT,
and 3.9-µl aliquots were terminated at 30, 60, 90, 120, and 150 s
time points by the addition to a final concentration of 13% Ficoll, 20
mM EDTA, and 0.05% SDS. The products were treated with 40
µg/µl of proteinase K for 30 min at 37 °C before
electrophoresis in a 15% nondenaturing polyacrylamide gel. Quantitation
of the reaction products was carried out after exposure of the wet gel
to a PhosphorImager screen at 4 °C. The amounts of free (displaced)
and total oligo U5mis were determined by the area integration function
of ImageQuant software (Molecular Dynamics).
Sites of RNase H Cleavage 5` of the PBS after
Initiation of Minus Strand DNA Synthesis
To define the template
that determines the 3` end of minus strand DNA prior to the second jump (Fig. 1), we examined RNase H specificity using a model
substrate in vitro. Mol15R, a 15-mer RNA oligonucleotide that
corresponds to the last 15 bases of the tRNA primer for M-MuLV, was
annealed to RNA containing plus sense R-U5-PBS sequences and extended
using RNase H RT (Superscript II) (Fig. 2A). The resulting RNA-DNA hybrids were incubated
with recombinant wild-type RT alongside controls in which the RNA
primer had not been extended. To identify RNase H cleavage sites that
occurred 5` of the PBS sequences on the plus strand RNA (Fig. 1, stage i, region I), the products were denatured, and
a labeled DNA oligonucleotide complementary to the template RNA was
used in a primer extension assay (Fig. 2A, oligo
1). As expected in control reactions without RT, a major primer
extension product corresponding to the 5` end of the plus strand RNA
template was observed (Fig. 3, lanes 1 and 3, arrow a). Products slightly smaller than full-length were also
apparent; these species likely resulted from premature termination
during extension on the full-length RNA. Treatment of the hybrid
substrates with RT resulted in the loss of all full-length and near
full-length products, indicating that most RNA molecules had been
cleaved somewhere along their length (Fig. 3, lane 2).
These cleavages did not result from nonspecific ribonuclease activity
associated with the recombinant RT, as no degradation was observed when
template RNA was treated with RT in the absence of RNA primer Mol15R
(data not shown). A major primer extension product was detected (Fig. 3, lane 2, arrow b) corresponding to
RNase H cleavage between the second and third ribonucleotides
immediately 5` of the PBS. Cleavages one base upstream and both one and
two bases downstream of this primary cleavage site were less abundant
but clearly detectable (Fig. 3, lane 2). Longer
exposures revealed minor products corresponding to cleavages at every
position up to 10 nucleotides 5` of the PBS (data not shown). As
expected, no cleavages were detected upstream of the U5/PBS boundary
when the control substrate containing unextended RNA primer was
examined (Fig. 3, lane 4). However, some cleavages were
observed within the RNA-RNA duplex region at the PBS site (Fig. 3. compare lanes 3 and 4, arrow
c). Since cleavage of RNA-RNA duplexes has been reported for RNase
H(40) , these species might result from a low level of RNase H*
activity.
RT (Superscript II) prior to use as a
substrate (Extended primer) (see Fig. 2A). The
5` ends of the template RNAs were mapped either before (lanes 1 and 3) or after treatment with recombinant M-MuLV RT for
15 min (lanes 2 and 4) by primer extension using
labeled oligo 1. Primer extension products were analyzed in an 8%
polyacrylamide, 8 M urea gel adjacent to a dideoxy-sequencing
ladder generated using the same labeled oligonucleotide (lanes
T, G, C, and A). The lower signal in lane 2 resulted from cleavages in the RNA 3` of the PBS due to
the presence of some snap-back priming during the preparation of the
substrate (see ``Materials and Methods''). Arrow a indicates the mobility of products that extend to the 5` end of
the RNA template; arrow b marks the predominant extension
product that results from the RNase H cleavage site within the hybrid
region just 5` of the PBS; and arrow c indicates the position
of products resulting from cleavage within the PBS. The RNA sequence of
the U5-PBS region is shown with the PBS boxed at the left.
tRNA Primer Removal by RNase H Leaves a Ribo A on the 5`
End of Minus Strand DNA
The 5` end of the minus strand is
defined by the RNase H cleavage that removes the tRNA primer in region
II after the PPT primer has been extended through U5-PBS (Fig. 1, stage iii). To initially address the
specificity of tRNA primer removal, we used an in vitro assay
with a model substrate. The RNA oligonucleotide Mol15R was annealed to
single-stranded DNA containing R-U5-PBS sequences and extended with
Sequenase to generate a DNA-DNA duplex containing the RNA primer (Fig. 2B). These substrates were incubated with
recombinant RT and analyzed by a primer extension assay using RNase
H RT and labeled oligo 2. As shown in Fig. 5A, some of the starting substrate remained, as
reflected by the presence of a primer extension product, which extended
through to the end of the RNA primer and co-migrated with the single
band present in the untreated material (arrow a, lanes 2 and 4). The major primer extension product migrated
slower by the equivalent of one base (arrow b, lane
4) than the extension product in the control samples that had been
treated with alkali prior to the assay (arrow c, lanes 1 and 3). Furthermore, this species co-migrated with the
band in the sequencing ladder consistent with a single ribo A remaining
on the 5` end of the minus DNA strand. Primer extension analyses on the
same samples using T4 DNA polymerase gave similar results (Fig. 5A, lanes 5-8) except that the T4
enzyme is much less efficient at utilizing the RNA portion of the
molecules as a template as compared with RNase H
RT.
DNA and extended with Sequenase (see Fig. 2B). The resulting substrate was treated with
recombinant M-MuLV RT for 10 min (lanes 3, 4, 7, and 8) or not treated (lanes 1, 2, 5, and 6), and the 5` ends of the minus
strand DNAs were mapped by primer extension using labeled oligo 2A and
RNase H
RT (Superscript) (lanes 1-4)
or T4 DNA polymerase (lanes 5-8). To remove any
remaining RNA oligonucleotide from the 5` ends of the minus strands,
the products in the indicated lanes were treated with alkali before
carrying out the primer extension reactions. Primer extension products
were analyzed adjacent to a dideoxy sequencing ladder generated using
labeled oligo 2A (lanes T, G, C, and A) as described in Fig. 3. B, same as for A except that the substrate was not treated (0 min, lanes
1 and 2) or treated with recombinant RT for 1 min (lane 3), 3 min (lane 4), 10 min (lane 5),
30 min (lane 6), or 60 min (lane 7), and primer
extensions were performed only with RNase H
RT. For
each panel, the sequence of the RNA oligonucleotide is indicated at the left. Arrow a marks the position of extension
products corresponding to the 5` end of the RNA primer; arrow b indicates the position of the major cleavage product containing
one ribonucleotide A on the 5` end of the minus strand; and arrow c indicates the position of the 5` end of the
DNA.
RT as the primer-extension polymerase,
a second band one nucleotide shorter than the expected product was
observed (Fig. 6, lane 2). Since only a single product
was observed with T4 DNA polymerase, we conclude that the shorter
primer-extension product observed in lane 2 resulted from
RNase H
RT pausing one nucleotide short of the 5` end
of the template. It is unclear why pausing at this site was
consistently more pronounced with the products of the endogenous
reaction as compared with those of the reconstituted reaction (see Fig. 5A, lane 4). The primer extension
analysis on the sample that had not been treated with alkali revealed
three bands in addition to the expected full-length product resulting
from complete extension through the RNA primer (Fig. 6, lane
1). The upper of these three bands corresponds to a ribo A on the
end of a significant fraction of the minus strands initiated in the
endogenous reaction since this product disappears after alkali
treatment (compare lanes 1 and 2). The middle band
corresponds to cleavage precisely at the RNA-DNA junction, while the
fastest migrating species results from premature termination by RNase
H
RT during the primer extension assay. Because
premature termination could, in principle, occur on molecules either
with or without the ribo A residue, it is not possible to accurately
quantify the proportion of molecules in each class from the endogenous
reaction. Since T4 DNA polymerase did not efficiently copy
ribonucleotides in primer extensions carried out on the model
substrates (see Fig. 5A, lanes 6 and 8), this enzyme could not be used to confirm that the
extension product, which is two nucleotides shorter than ribo A species
in Fig. 6, lane 1, indeed resulted from premature
termination.
RT (Superscript) (lanes 1 and 2) or T4 DNA
polymerase (lane 3) as described in Fig. 5. Samples
shown in lanes 2 and 3 were treated with alkali. The
sequence of the tRNA primer is indicated at the left.
RT. For both enzymes, a ribo A remained at the 5` end of minus
strand DNA (Fig. 7, lanes 4, 6, 8,
and 10). Treatment of identical samples with alkali prior to
the primer extension analysis confirmed the presence of the ribo A on
the 5` ends of the DNAs (Fig. 7, lanes 3, 5, 7, and 9). Based on these results, it would appear
that the virion-derived RT is indistinguishable from the recombinant
enzyme.
RT using
oligo 2B. The sequence of the RNA oligonucleotide is indicated at the left.
Fate of the RNA Primer
To address the fate of the
RNA primer in the cleavage reaction, 5` end-labeled RNA oligonucleotide
Mol15R was annealed to single-stranded DNA containing R-U5-PBS
sequences and extended with Sequenase to generate the model substrate
described above (Fig. 2B). When analyzed by denaturing
polyacrylamide gel electrophoresis, the labeled extended RNA migrated
much more slowly than the original 15-mer RNA oligonucleotide (Fig. 8, compare lanes 2 and 1, arrows a and b). As early as 20 s after incubation with RT, the
major product was clearly shorter than the starting 15-mer RNA
oligonucleotide and by comparison with a sizing ladder migrated as a
species 14 nucleotides in length (Fig. 8, lane 3).
After 3 min, the amount of starting material decreased dramatically,
but the major RNase H cleavage product was still a 14-mer RNA (Fig. 8, lane 5). With increasing incubation times, the
14-mer RNA was cleaved further into smaller products (Fig. 8, lanes 6 and 7). In these experiments, a 15-mer RNA
corresponding to cleavage at the RNA-DNA junction was not observed.
When the labeled RNA primer was annealed to the single-stranded DNA but
not extended and subsequently incubated with RT for as long as 27 min,
the primer was not cleaved by RNase H at all (data not shown). These
data support the conclusion that the first RNase H cleavage on the RNA
primer occurs between the last and penultimate ribonucleotides, and
show that the primer is subsequently cleaved into smaller fragments.
DNA and
extended with Sequenase (see Fig. 2B). The resulting
substrate was gel-purified, treated with recombinant RT for 0.3 min (lane 3), 1 min (lane 4), 3 min (lane 5), 9
min (lane 6), or 27 min (lane 7) and analyzed in a
20% polyacrylamide, 8 M urea gel. As controls, the untreated
substrate (lane 2, arrow a) and 15-mer RNA oligo
Mol15R without extension (lane 1, arrow b) are shown.
Positions of size markers generated by nuclease P1 digestion of labeled
Mol15R are indicated at the right.
Removal of M-MuLV RNA Primer by HIV-1 and Avian
Myeloblastosis Virus RTs
Previous studies that examined RNA
primer removal by HIV-1 RT using the homologous tRNA substrate showed
that, similar to the case for M-MuLV described above, a ribo A remained
at the 5` end of minus sense DNA and that the RNA primer is initially
removed intact (18, 29, 30) . With the same
HIV-1 substrate, both the M-MuLV and avian myeloblastosis virus RTs
also left a ribo A at the 5` end of the minus sense DNA(29) .
To test whether the RNase H activity of HIV-1 RT might similarly leave
a ribo A on the 5` end of the M-MuLV substrate, a model substrate (Fig. 2B) containing the labeled Mol15R primer was
incubated with either the HIV-1 or M-MuLV RTs for increasing lengths of
time, and the products were examined as described above. In contrast to
the studies with the M-MuLV enzyme, the HIV-1 RT cleaved the RNA primer
to release a small amount of intact 15-mer as well as the 14-mer (Fig. 9, compare lanes 1-4 with lanes
5-8, respectively). Like the M-MuLV enzyme, HIV-1 RT cleaved
the RNA primer into smaller fragments with longer incubation times (Fig. 9, compare lanes 3 and 4 with lanes
7 and 8). Primer extension analyses to map the 5` ends of
the minus strand DNA after cleavage confirmed that the HIV-1 enzyme
cleaved the M-MuLV substrate both at the DNA-RNA primer junction and
after the first ribo A in the RNA primer (data not shown). Similar
experiments carried out with avian myeloblastosis virus RT on the
M-MuLV substrate also yielded heterogeneous cleavage products in which
the RNA primer was either removed completely or as a 14-mer (data not
shown).
Initiation of Displacement Synthesis from the Primer
Terminus Generated by the Second Jump
Based upon the foregoing
results, which map the 5` and 3` limits of the minus strand, it is
possible to predict the structure of the replicative intermediate
generated by the second jump. Because the minus DNA strand defined in
the above experiments is typically longer than genome length by three
nucleotides (one 5` rA and two 3` dAs), the second jump would be
expected to create a branched rather than a nicked replication
intermediate (3) (Fig. 1, stage v; see
``Discussion'').
P-labeled (indicated with an asterisk). B, The release of the labeled
oligonucleotide by RT-mediated displacement synthesis is shown as a
function of time for the two primer-template
combinations.
)Structural anomalies at
the RNA-DNA junction, which include distortions in the base pairs, a
bend in the helix axis, and unusual sugar conformations could influence
the cleavage specificity of RNase H. Recently, it was suggested that
the width of the minor groove could be an important feature recognized
by RNase H(61) . Based on the NMR analysis of the Moloney
structure, it is possible that narrowing effects of the junctional DNA
base pair on the minor groove could preclude cleavage by RNase H at the
junction. The determination of which, if any, of these structural
features prevents removal of the ribo A by the viral RNase H awaits a
structural analysis of a substrate that is cleaved precisely at the
RNA-DNA junction.
))
We thank Knut Madden, Leon Parker, Caterina Randolph,
and Lance Stewart for helpful discussions during the course of these
experiments.
©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:
