4 | DISCUSSION
Historically, the mechanism of fp25k mutations was reported as transposon insertion [7, 11, 12, 22]. Since AcMNPV genome does not appear to have a transposase gene, the enzymes that catalyze the insertion of host DNA to fp25k has to be provided by the host cells (Ayres et al., 1994). These host derived transposable elements have been identified as TED [23], IFP2 or PiggyBac [24], TFP3 or tagalong [25], Hitchhiker [26]. Of the 13 TTAA sites of the AcMNPV fp25k , only four TTAA sites have been reported with host DNA insertions [27, 28]. Some of these TTAA insertion sites were found in the fp25k of Galleria mellonella nuclear polyhedrosis virus (GmMNPV).
This 10th TTAA is also the site of host DNA insertion of AcMNPV FP virus M5 that was isolated from Sf9 cells [11]. Further confirmation by site directed mutagenesis demonstrated that this 10th TTAA site is responsible for host DNA insertion (Fig. 2 B). Since the host transposase has to recognize the TTAA site first before it cleaves the site for DNA insertion, different cells may have different transposases with different recognition sequences for cleavage.
The first TTAA transposition recognition site was found in thefp25k of GmMNPV IFP2 that was obtained during passage in Tn368 cells derived from T. ni ovaries [29]. The recognition site is determined to be 5’-gggtta-3’ in the fp25k of GmMNPV by mutagenesis studies [30]. In our studies, the only host DNA insertion site is at the 10th TTAA of the fp25kof AcMNPV during passage in Hi5 cells (Fig. 2). The upstream sequences of the 10th TTAA site of AcMNPV fp25k is TAC. It is highly like that the TACTTA of the 10th TTA is the recognition site of the transposase active in Hi5 cells. Event though, both the Hi5 cells and Tn368 cells are derived from T. nithat may have different transposase genes in the genome, only certain transposases are specifically expressed in different cells. This transposition site is currently under studies by the site-directed mutagenesis method.
In addition to host transposon insertions at the TTAA site of thefp25k gene of certain NPVs, insertion of a single nucleotide, base substitution, and deletion of nucleotides in the promoter regions or the coding region of fp25k [7, 8, 11, 12]. To our surprise, in our AcMNPV passage studies in three commonly used insect cell lines (Sf21, Sf9 and Hi5) only three types of mutations were found: an A nucleotide insertion at the two A7 MNRs and a 287 bp host transposable DNA insertion. It is possible for the existence of large DNA insertions in the fp25k of AcMNPV so we tried to extend the amplification time in the hope of getting PCR products larger than the 1.5 kbp product for cloning and DNA sequencing. Faint PCR products larger than 1.5 kbp and smaller than 1.2 kbp were observed but cloning and PCR sequencing failed to identify any large DNA insertions and other types of deletions. Plague purification of more FP viruses is required to pursue in more FP mutation identification.
The Hi5 cell line was originally cloned from T. ni embryos for propagation of betabaculoviruses that are often more difficult than alphabaculoviruses to have cells for infection [31]. Later, the Hi5 cells are used for high protein expression due to the larger cell size than Sf9 and Sf21 cells. As soon as AcMNPV enters Hi5 cells, the transposase in the cells recognizes the TTAA site and insert the 287 host DNA in the fp25k to produce the FP virus that results in lower polyhedrin expression and higher BV production than the wt AcMNPV (Fig. 2 B. Fig. 3). It is likely that the BV from the FP virus outcompeted that from the wt AcMNPV. This high BV budding efficiency of AcMNPV FP is likely the cause of the rapid accumulation of FP and diminishing wt AcMNPV during Hi5 cell infection (Fig. 3).
The success of obtaining AcP3-FPSDM virus that does not have the 2 A7 MNRs and the 10Th TTAA site of the fp25 is a major achievement in this project. We also demonstrated that during Hi5 cell passage, the fp25k showed no MNR mutations and host DNA insertions by PCR sequencing the fp25k locus as well as other deletion mutations. The A adding at the A7 MNRs is likely due to replication slippage as the viral DNA polymerase pauses at the MNR [32]. The isolation of AcP2, AcFP1and Acp13 in Sf21 [17] and only host DNA insertion and no MNR mutations in this paper during AcMNPV passage in Hi5 cells may suggest that host DNA replication and error repair systems are involved in AcMNPV DNA replication. This notion is supported by a report that host DNA polymerase is involved in AcMNPV replication [33]. If this is true, in some Sf21 cells, AcMNPV may use its own DNA polymerase for genome replication that leads to the MNR slippage replication errors in the fp25k , whereas in Hi5 cells, it may use Hi5 host cellular DNA polymerase for genome replication. The property of the Hi5 DNA polymerase is unknown but cellular DNA polymerases of Saccharomyces cerevisiae and human have a minimum of a ray of 8 base monopolymer for slippage replication [34, 35]. Although no fp25k mutations of AcP3-FPSDM have been detected in Hi5 cell passage, this does not eliminate other DNA mutations in the viral genome.
Comparing viral genome between AcP3-FPSDM P0 and P10 showed difference in EcoRI and HindIII RFLP (Fig. 5). However, no major DNA deletion or insertion was deduced from the RFLP comparison. It is likely that the small difference in EcoR1 and HindIII fragment mobility changes are due to base substitution that introduced more EcoRI and HindIII restriction sites into the AcP3-FPSDM P10 genome. Virion occlusion essays also supports that no fp25k mutations of AcP3-FPSDM P10 since an increased polyhedra yield was observed compared to AcP3 P10 (Fig. 4). This also supports that virion occlusion occurred in Hi5 cells infected with AcMNPV FP viruses discovered early [18]. The reduction of polyhedra yield in Hi5 cell infection by AcP3-FPSDM P10 may suggest mutations in other AcMNPV genes such as da26 and 94k for few polyhedra generation [22, 36, 37].
The hallmark of the FP phenotype is its ability to increase virus titer in cell culture [9, 13, 38-40]. In this paper, we confirmed the reports of high virus budding from these authors. In addition, we showed our AcP3-SDM virus has high virion occlusion efficiency for insect pest control. We are currently investigating the mechanism of this high virus budding mechanism.