Sequence identity of the direct repeats, DR1 and DR2, contributes to the discrimination between primer translocation and in situ priming during replication of the duck hepatitis B virus

Abstract

There are two mutually exclusive pathways for plus-strand DNA synthesis in hepadnavirus reverse transcription. The predominant pathway gives rise to relaxed circular DNA, while the other pathway yields duplex linear DNA. At the completion of minus-strand DNA synthesis, the final RNase H cleavage generates the plus-strand primer at direct repeat 1 (DR1). A small fraction of viruses make duplex linear DNA after initiating plus-strand DNA synthesis from this site, a process called in situ priming. To make relaxed circular DNA, a template switch is necessary for the RNA primer generated at DR1 to initiate plus-strand DNA synthesis from the direct repeat 2 (DR2) located near the opposite end of the minus-strand DNA, a process called primer translocation. We are interested in understanding the mechanism that discriminates between these two processes. Previously, we showed that a small DNA hairpin forms at DR1 in the avihepadnaviruses and acts as an inhibitor of in situ priming. Here, using genetic approaches, we show that sequence identity between DR1 and DR2 is necessary, but not sufficient for primer translocation in the duck hepatitis B virus. The discrimination between in situ priming and primer translocation depends upon suppression of in situ priming, a process that is dependent upon both sequence identity between DR1 and DR2, and the presence of the hairpin at DR1. Finally, our analysis indicates the entire RNA primer can contribute to primer translocation and is translocated to DR2 before initiation of plus-strand DNA synthesis from that site.

Figure 1. Two mutually exclusive pathways for initiation of plus-strand DNA synthesis

Minus-strand DNA (thick black line) is generated by reverse transcription of an RNA intermediate (pregenomic RNA) by the covalently attached P protein (circle). (A) At the completion of minus-strand DNA synthesis, the final RNase H cleavage generates an 18 nt oligoribonucleotide which serves as the primer for plus-strand DNA synthesis from one of the two sites. The 3′ end of the RNA primer is coincident with the 5′ end of DR1 in the minus-strand DNA. (B) A small fraction of viruses produce a duplex linear form of the genome by extending the primer from the site it was generated, a process called in situ priming. (C) The predominant pathway requires a template switch, called primer translocation, as some portion of the RNA primer is used to initiate plus-strand DNA synthesis from DR2 located near the opposite end of the minus-strand DNA. Plus-strand synthesis from DR2 results in a relaxed circular form of the genome upon completion of a second plus-strand template switch, termed circularization, which is facilitated in part by the small terminal redundancy in the minus-strand DNA, indicated by 5′ r and 3′ r. (D and E) The placement of the three oligonucleotides used in the primer extension analyses. Panel D represents an in situ-primed molecule while panel E represents a molecule primed from DR2 and circularized. See Materials and Methods for description of use of the 3 primers.

Figure 2. Representation of variants used to examine the role of complementarity between the RNA primer and DR2 without altering the hairpin at DR1

(A) A DHBV variant was created that contained a second copy of the 12 nt direct repeat inserted adjacent to the normal copy of DR2 as shown. The remaining minus-strand DNA (thick black line) was wildtype in sequence, including the 3′ end. The small DNA hairpin overlapping the 5′ end of DR1 in the minus-strand is indicated. The capped RNA primer is shown annealed to the 3′ end of the minus-strand DNA (B). Depicted is a DHBV variant (3/12 DR1/2) that contains a 3 nt substitution in both DR1 and DR2. Mutations that affect the 3′ end of the minus-strand also affect the RNA primer as indicated by lower case letters. The 3/12 DR1 and 3/12 DR2 variants contain the 3 nt substitution at one site (DR1 or DR2, respectively) and the wildtype sequence at the other. (C) The sequence of the wildtype direct repeats is shown (minus-sense polarity). The 3 nt substitution (3/12) used in this manuscript is depicted with the altered nucleotides identified by underscores (minus-sense polarity). The 5/12 mutation described in the text and previously is shown for comparison.

Figure 3. Acceptor site selection for primer translocation is influenced by the extent of complementarity between the RNA primer and the acceptor site

(A) The tDR2 variant contains two identical copies of DR2 inserted in a head to tail orientation as shown in the minus-strand DNA (thick black line) with the P protein (circle) attached to the 5′ end. The two copies are labeled, A and B. The 3/12@A and 3/12@B variants contain the 3/12 mutation (Fig. 2C) in either copy A or copy B, respectively. The sequence adjacent to copy A is substituted by 6 nts that increase the complementarity between copy A and the RNA primer from 12 to 18 nts in the ext@A variant, where copy B retains the normal 12 nts of complementarity. (B) Southern blotting of viral DNA from LMH cells isolated 3 days post-transfection. The positions of the prominent DNA replication intermediates (RC, DL, and SS) are indicated. The blot was hybridized with a genomic-length, minus-strand specific RNA probe. Lanes (1, wildtype reference; 2 and 3, tDR2; 4 and 5, 3/12@B; 6 and 7, 3/12@A; 8, ext@A). (C) Primer extension with primer 1 can distinguish between 5′ ends of plus-strand DNA initiating from either copy of DR2. 5′ ends are detected primarily at acceptor sites wildtype in sequence or containing extended complementarity with the RNA primer. Ends mapping to copy A or copy B are denoted next to the gel image. Using this assay, it is normal to detect bands at 4 consecutive positions where the fastest migrating species corresponds to the position just outside of DR2. The two fastest migrating bands represent authentic 5′ ends of viral DNA (Lein 1986). The third and fourth band are thought to be derived from the two fastest migrating bands as a consequence of non-templated, 3′-end additions during the primer extension reaction. Sequencing ladder was generated using primer 1 and a wildtype DHBV template.

Figure 4. Complementarity between the RNA primer and DR2 is necessary, but not sufficient for primer translocation

(A) Description of the 3/12 variants with the location of the 3/12 mutation (Fig. 2C) indicated in each case. DR1 and DR2 are indicated as boxes on the schematic of minus-strand DNA (thick black line) with P protein (circle) attached. (B) Southern blot of viral DNA isolated from LMH cells 3 days post-transfection. Lanes (1, wildtype reference; 2, 3/12 DR1; 3, 3/12 DR2; 4, 3/12 DR1/2). The positions of the prominent DNA replication intermediates (RC, DL, and SS) are indicated to the left. The blot was hybridized with a genomic-length, minus-strand specific, RNA probe. (C) Primer extension with primer 3 measured the 5′ termini of minus-strand DNA {(−) DNA} and the internal standard DNA (i.s.). Primer extension with primer 1 measured the amount of plus-strand DNA initiated from DR2 and elongated at least to the 5′ end of minus-strand DNA {(+) DR2} and the internal standard DNA (i.s.). Lanes (1, wildtype reference; 2 and 3, 3/12 DR1; 4 and 5, 3/12 DR2; 6 and 7, 3/12 DR1/2). Sequencing ladders were generated using their respective primers and a wildtype DHBV template. (D) The percentage of RC DNA synthesis (Southern blot) and primer translocation (primer extension) were calculated as described in the Materials and Methods. For each virus, the mean value was presented with error bars indicating one standard deviation. Each virus was analyzed multiple times from independent transfections.

Figure 5. Reduced complementarity between the RNA primer and DR2 affect both primer translocation and in situ priming levels

(A) In situ priming of the 3/12 DR set of variants was measured using both Southern blot (black bars) and primer extension (gray bars) and calculated as described in Materials and Methods. (B) Description of variants containing 1 or 3 nt mutations in the DR2 sequence. The wildtype sequence (minus-sense polarity) is provided as a reference. (C) Primer translocation was calculated by primer extension using primers 1 and 3 as described in Fig. 4C and D (gels not shown). (D) In situ priming was calculated using both Southern blot (black bars) and primer extension (gray bars) as described in the Materials and Methods. Samples in (A,C, and D) were normalized to a wildtype reference (primer translocation = 100, in situ priming = 1) and presented as the mean with error bars to indicate the standard deviation. Each virus was analyzed multiple times from independent transfections.

Figure 6. Complete suppression of in situ priming requires the hairpin at DR1 and complementarity between the RNA primer and DR2

(A) A schematic of the wildtype minus-strand DNA (thick black line) indicates the sequence of the direct repeats, DR1 and DR2, and the hairpin that resides overlapping DR1. The four arcs indicate the base pairing partners in the stem of the predicted hairpin. The four variants containing only DR1 (DR1-x; x = 10, 11, 12, s4) mutations reduce base pairing in the hairpin and introduce mismatches between the RNA primer and DR2. The four variants containing mutations in both DR1 and DR2 (DR1/2-x; x = 10, 11, 12, s4) reduce base pairing in the hairpin, but retain complementarity between the RNA primer and DR2. The mutations that leave the hairpin intact, but disrupt complementarity between the RNA primer and DR2 were shown in Fig. 5B (DR2-x; x = 10, 11, 12, s3). (B) In situ priming was calculated from Southern blots as described in the Materials and Methods. Samples were normalized to a wildtype reference (set to 1) and presented as the mean with error bars to indicate the standard deviation. Each virus was analyzed multiple times from independent transfections. Values for DR2-x variants (Fig. 5D) are reproduced here for clarity.

Figure 7

Two general models for primer translocation. See Discussion for thorough description of these models