phenotypic and genotypic analysis of helicoverpa armigera nucleopolyhedrovirus serially passaged in...

Downloaded from www.microbiologyresearch.org by

IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51

Journal of General Virology (2002), 83, 945–955. Printed in Great Britain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Phenotypic and genotypic analysis of Helicoverpa armigeranucleopolyhedrovirus serially passaged in cell culture

Linda H. L. Lua,1 Marcia R. S. Pedrini,1 Steven Reid,1 Ashley Robertson2 and David E. Tribe2

1 Department of Chemical Engineering, The University of Queensland, Queensland 4072, Australia2 Department of Microbiology and Immunology, The University of Melbourne, Victoria 3010, Australia

Rapid accumulation of few polyhedra (FP) mutants was detected during serial passaging ofHelicoverpa armigera nucleopolyhedrovirus (HaSNPV) in cell culture. 100% FP infected cells wereobserved by passage 6. The specific yield decreased from 178 polyhedra per cell at passage 2 totwo polyhedra per cell at passage 6. The polyhedra at passage 6 were not biologically active, witha 28-fold reduction in potency compared to passage 3. Electron microscopy studies revealed thatvery few polyhedra were produced in an FP infected cell (!10 polyhedra per section) and in mostcases these polyhedra contained no virions. A specific failure in the intranuclear nucleocapsidenvelopment process in the FP infected cells, leading to the accumulation of naked nucleocapsids,was observed. Genomic restriction endonuclease digestion profiles of budded virus DNA from allpassages did not indicate any large DNA insertions or deletions that are often associated with suchFP phenotypes for the extensively studied Autographa californica nucleopolyhedrovirus andGalleria mellonella nucleopolyhedrovirus. Within an HaSNPV 25K FP gene homologue, a singlebase-pair insertion (an adenine residue) within a region of repetitive sequences (seven adenineresidues) was identified in one plaque-purified HaSNPV FP mutant. Furthermore, the sequencesobtained from individual clones of the 25K FP gene PCR products of a late passage revealed pointmutations or single base-pair insertions occurring throughout the gene. The mechanism of FPmutation in HaSNPV is likely similar to that seen for Lymantria dispar nucleopolyhedrovirus,involving point mutations or small insertions/deletions of the 25K FP gene.

IntroductionSerial passaging of baculoviruses in cell culture leads to

extensive genomic changes (Miller, 1986). The most commonand rapidly accumulating mutant detected is a change from thewild-type many polyhedra (MP) per cell phenotype to the fewpolyhedra (FP) per cell phenotype. The complex FP phenotypeis generally described as a decrease in the number of polyhedraproduced per cell, few or no virions occluded within polyhedra,altered intranuclear nucleocapsid envelopment and increase innumber of infectious budded virus (BV) (Harrison & Summers,1995b). FP mutants have been reported in several baculovi-ruses, including Autographa californica multicapsid nucleopoly-

Author for correspondence: Linda Lua.

Fax 617 3365 4199. e-mail lindal!cheque.uq.edu.au

The GenBank accession numbers of the sequences reported in this

paper are AF395841 and AF395871.

hedrovirus (AcMNPV) (Hink & Vail, 1973), Trichoplusia niMNPV (TnMNPV) (Potter et al., 1976), Galleria mellonellaMNPV (GmMNPV) (Fraser et al., 1983), Lymantria disparMNPV (LdMNPV) (Slavicek et al., 1992), Orgyia pseudotsugataMNPV (OpMNPV) (Russell & Rohrmann, 1993) and Heli-coverpa armigera single-capsid NPV (HaSNPV) (Chakraborty &Reid, 1999).

Frequent mutations have been identified within a specificregion (map units 35±0–37±0) in the FP mutants of AcMNPV(Beames & Summers, 1988, 1989 ; Fraser et al., 1983) andGmMNPV (Fraser et al., 1985, 1983). This region contains the25K FP gene, which encodes a 25 kDa protein that is essentialfor virion occlusion and polyhedron formation (Beames &Summers, 1988, 1989 ; Wang et al., 1989). Most reports on FPmutations were correlated to large insertions of host DNA ordeletions of the viral genome (0±1–2±8 kb), which weredetectable by restriction endonuclease (REN) digestion analysis(Beames & Summers, 1988, 1989 ; Fraser et al., 1985, 1983 ;

0001-8201 # 2002 SGM JEF


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51

L. H. L. Lua and othersL. H. L. Lua and others

Wang et al., 1989). The FP mutations of LdMNPV wereexceptions (Bischoff & Slavicek, 1997 ; Slavicek et al., 1995).These LdMNPV FP mutants exhibit FP characteristic traits ;however, REN digestion analysis did not correlate the FPphenotypes observed to any DNA insertions or deletions ofdetectable lengths. Bischoff & Slavicek (1997) later reported1 bp insertions or small deletions (8 or 24 bp) in the 25K FPgene of LdMNPV FP mutants, suggesting that the mechanismof FP mutation in LdMNPV could be different from that ofAcMNPV.

In recent years, studies have been done to determine therole of the 25 kDa protein in the infection cycle of baculo-viruses. However, the analysis of the function of the 25 kDaprotein is not conclusive (Beniya et al., 1998 ; Braunagel et al.,1999 ; Harrison & Summers, 1995a, b). Studies showed that the25K FP protein is a structural protein in the nucleocapsids ofboth BV and occlusion derived virus (ODV) but a largefraction of the 25K FP proteins remains cytoplasmic through-out infection (Braunagel et al., 1999 ; Harrison & Summers,1995a). The relationship between the location of the 25K FPgene product and the phenotype caused by the gene mutationis still unclear. Considering the varied effects followingmutation of this gene, it is possible that the gene has more thanone function during the invasion and infection process. Studiesalso suggest that the 25K FP protein could be required forefficient protein accumulation and trafficking by regulatingtranscription, mRNA stability, translation or altered proteinstability of one or many viral proteins, which directly orindirectly affect occlusion morphogenesis (Beniya et al.,1998 ; Braunagel et al., 1999).

Reduced occlusion production and decreased virulence as aresult of FP mutations would greatly hinder the commerciali-zation of baculoviruses produced by in vitro cell culture, for useas biopesticides. As large volumes of virus inocula are neededfor large-scale production of baculoviruses (Rhodes, 1996), itmay be impossible to generate enough virus inocula at lowpassage numbers for such production, as emergence of FPmutants appears to be inevitable during the scale-up of virusinocula used for cell culture systems. Understanding the natureof FP mutations and how they are selected may help in eitherdeveloping a process for isolating virus isolates with a stableMP phenotype or to minimize the selective advantage of FPmutants. Research on FP mutation not only deserves furtherattention from the commercial perspective but also for itsintrinsic interest.

Chakraborty & Reid (1999) reported the effect of serialpassaging on HaSNPV in cell culture ; however, no insightswere given on the phenotype or genotype of HaSNPV FPmutants. The current paper reports detailed phenotypic andgenotypic studies of HaSNPV FP mutants. HaSNPV wasserially passaged in Helicoverpa zea serum-free suspensioncultures and the phenotypic changes that occurred during thetransition of an MP dominated population to a FP one wasinvestigated using transmission electron microscopy (TEM).

The ultrastructural differences between the MP and FP infectedcells during progeny virus production and assembly weredocumented. REN digestion profiles of genomic viral DNAduring serial passaging were determined, the wild-typeHaSNPV 25K FP gene was identified and sequenced, and a FPmutant that carries a mutation in the 25K FP gene was isolated.

Methods+ Cells and virus. The Helicoverpa zea cell line was obtained fromCSIRO, Division of Entomology (Canberra, Australia). It was isolated

from pupal ovarian tissue, strain BCIRL-HZ-AM1 (McIntosh & Ignoffo,1981). The uncloned cells were gradually adapted to serum-free SF900II(Life Technologies) suspension culture (Chakraborty, 1997). Cultures

were typically grown in 250 ml Erlenmeyer flasks on an orbital shakeroperated at 120 r.p.m., inside a refrigerated incubator at 28 °C. HaSNPV(uncloned) passage 1 stock was established in cells grown in SF900II

supplemented with 10% foetal bovine serum (CSL, Australia), withhaemolymph collected from infected larvae which were obtained fromthe Department of Primary Industries (Brisbane, Queensland, Australia).

The details are documented in an earlier paper (Lua & Reid, 2000).

+ Serial passaging of HaSNPV in vitro. For all passages, duplicate100 ml working volumes of Helicoverpa zea cell cultures at 5¬10&

cells}ml were infected at an m.o.i. of 0±5 p.f.u. per cell. All cultures were

seeded at 3¬10& cells}ml (50 ml volumes), and allowed to grow to1¬10' cells}ml, before being diluted back to 5¬10& cells}ml with 50 mlfresh medium (100 ml final volume). To obtain infectious BV for

subsequent passage, 50 ml of the cell suspension was harvested at 70%viability, typically 4 or 5 days post-infection (p.i.), leaving the other 50 mlcell suspension for polyhedra production. All virus stocks were stored in

1±8 ml or 4 ml aliquots at ®70 °C. The virus titre of each passage wasdetermined using a plaque assay (Lua & Reid, 2000), before use asinoculum for the next passage. Cell densities and cell viabilities of all

cultures were determined daily in triplicate using the 0±1% trypan blueexclusion cell count method (Nielsen et al., 1991). Final polyhedra yieldswere determined at 10 days p.i. Cells were first lysed with 0±5% SDS for1 h at 28 °C before triplicate counts of the polyhedra were determined

using a haemocytometer counting chamber.

+ Plaque purification of HaSNPV FP mutant. A second HaSNPVpopulation (uncloned) was established in cell culture with collected

haemolymph under similar conditions to those described above. Thevirus was serially passaged out to passage 6. BV at passage 6 was dilutedout onto culture plates as in a plaque assay and an FP plaque was picked

on the basis of its phenotypic appearance under light microscopy (Hink& Vail, 1973). The FP mutant (designated as FP8AS) was plaque purifiedthree times before being scaled up to obtain BV DNA for PCR

amplification and sequencing. The FP phenotype of the plaque purifiedmutant was confirmed using TEM.

+ TEM and percent of FP infected cells. At each passage, infected

cell pellets were harvested at 6 days p.i. for determination of percentageof FP}MP using TEM. Cell suspension (1 ml) was centrifuged at 12000r.p.m. for 10 min. The cell pellets were resuspended in 3% glutaraldehydefixative solution and kept at 4 °C. The protocol for TEM processing is

documented in Lua & Reid (2000). The% FP infected cells was determinedby assessing m infected cells under TEM and scoring each cell as eithera FP infected cell or an MP infected cell. If we assume that each

observation is independent, the number of FP infected cells observed, x,

JEG


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51

FP mutants of Helicoverpa armigera nucleopolyhedrovirusFP mutants of Helicoverpa armigera nucleopolyhedrovirus

Fig. 1. Serial passaging of HaSNPV in SF900II. The polyhedra cell specificyield (E), budded virus titre (_) and percent FP infected cells (+) ateach passage were determined.

will be binomially distributed with the probability p equal to the% FP inthe population (Taylor, 1997). The confidence interval for p is given by :

[x

x(m®x1)F(2m®2x2,2x)"−

α/#

,(x1)F(2x2,2m®2x)

"−α/#

m®x(x1)F(2x2,2m®2x)"−

α/#

]

+ Bioassay. The bioassay used to estimate the potency of thepolyhedra produced against Helicoverpa armigera larvae measured theconcentration of polyhedra required to kill a larva. The potency is usuallyexpressed as LC

&!(median lethal concentration), which is the con-

centration of polyhedra required to kill 50% of the population. Infectedcells with polyhedra at 10 days p.i. were harvested and kept frozen at®20 °C for bioassay. Polyhedra were extracted and bioassays wereperformed by Peter Christian (CSIRO Entomology, Canberra, Australia).The potency of the polyhedra to be tested was compared to a standardvirus, GemStar2 [Helicoverpa zea (Hz)SNPV produced in vivo], bysimultaneously testing both.

+ REN digestion analysis. BV DNA from passage 2 to passage 6was purified for REN digestion analysis. To purify DNA from BV, cellswere first removed by low speed centrifugation (1000 g, 10 min). The BVwas pelleted by ultracentrifugation of the supernatant at 10000 g for45 min at 4 °C, using an ultracentrifuge with an SW28 rotor (Centrikon

Table 1. Serial passage data for HaSNPV in SF900II medium

The means and errors (95% confidence intervals, CI) are presented.

PassagePeak cell density

(105 cells/ml)Specific yield

(polyhedra per cell)BV titre

(106 p.f.u./ml)No. of FP/no.

assessed¯% FP (CI)LC50

(polyhedra/mm2)*

2 8±50³1±72 222±12³67±85 8±53(7±12–10±15)

3 7±38³0±65 177±61³17±10 5±73(4±59–7±08)

3}1083% (1–8)

0±475(0±167–0±783)

4 7±01³0±41 56±59³4±08 16±67(9±72–23±01)

67}8183% (73–90)

0±556(0±125–2±650)

5 9±71³0±69 9±54³1±00 3±73(2±82–4±85)

87}9295% (88–98)

6 5±95³0±20 2±17³0±13 0±27(0±20–0±37)

106}106100%

13±194(9±245–20±603)

* The LC&!

for the bioassay standard (GemStar2) was 0±113 (0±068–0±174)., Not determined.

T-2070, Kontron Instruments) and 38±5 ml Ultra-Clear centrifuge tubes(Beckman). Each BV pellet was resuspended with 200 µl cold PBS. DNA

was purified from the BV using the QIAamp DNA Blood Mini Kit(Qiagen) according to the protocol provided by the manufacturer. The

viral DNA was digested separately with EcoRI, HindIII and BamHI, at37 °C for 4 h. Each digestion consisted of 25 µl viral DNA, 2 µl enzymesand the appropriate buffer for specific restriction enzyme requirements.

The fragments were separated by electrophoresis on a 0±7% agarose gelat 60 W for 2 to 3 h and stained with ethidium bromide. Digital images

of the agarose gels were obtained using a Kodak EDAS 120 digitalcamera with software provided (Kodak Digital Science 1D, version 3.0.0).

+ PCR amplification and sequencing of the 25K FP gene. Thiswork was performed before the entire genome of HaSNPV was published

(Chen et al., 2001). A set of upstream and downstream degeneratedoligonucleotides was first designed from multiple alignment of the 25KFP gene of different baculoviruses and used to identify the 25K FP gene

of HaSNPV. They were oligonucleotides 5« KGYAGYGTNGARATFT-AYGG 3« (Primer 1) and 5« NATYTTNACNGGNCCRTCRTA 3«(Primer 2). The reaction buffer supplied by the manufacturer of the Taq

DNA polymerase (Boehringer Mannheim) was used with dNTPs(Pharmacia) at a final concentration of 200 µM. The amplification cycle

consisted of 3 min denaturation at 94 °C, annealing for 1 min at 45 °Cand extension at 72 °C for 1 min. The amplification cycle was repeated

30 times. Specific oligonucleotides, 5« CGGTATTCACGATAGAAA 3«(Primer 3) and 5« CGTAGTCTATGTCTAGAT 3« (Primer 4), were

designed to confirm the nucleotide sequence of the first identifiedfragment of the 25K FP gene. Primers 3 and 4 were also used for primerwalking sequencing, using the purified genomic viral DNA as template.

A 40 µl sequencing reaction consisting of 16 µl BigDye premix, 13 pMprimer and 200–300 ng viral genomic DNA was set up : 99 cycles

consisting of 30 s at 95 °C for denaturation of template, 20 s at 50 °C forannealing of primer and 4 min at 60 °C for extension of nucleotides wereoperated. The final products were cleaned up with the Centrispin-20

columns (Geneworks), according to the protocol provided by themanufacturer, before being sequenced. Oligonucleotides 5« GTACGCA-

CACATATACAC 3« (Primer 5) and 5« GCTAGTCAAATGAGTCGC 3«

JEH


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


Fig. 2. TEM images showing phenotypic characteristics of MP (passage 3) and FP (passage 4) infected cells. (A) Envelopedvirions in the ring zone of the nucleus (an MP infected cell). (B) Naked nucleocapsids (arrowheads) accumulating in the ring

JEI


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


(Primer 6) were used to amplify and sequence the entire 25K FP gene.Oligonucleotides 5« ACGGACTGGATGAGCTTC 3« (Primer 7) and 5«CGGTACTCGGTAAATCTG 3« (Primer 8) were used to amplify andsequence the entire 25K FP gene including upstream and downstreamregions of the gene. The annealing temperature used in the PCRamplification cycles for Primer 3 to Primer 8 was 50 °C. The nucleotidesequence of amplified DNA was determined directly. Sequencing wasperformed using the PRISM (Applied Biosystems) terminator chemistrymethod according to the manufacturer’s instructions : 30–90 ng DNAand 3±2 pM primer in a 16 µl reaction was prepared for sequencing, whilstthe other reagents for the reaction were supplied in the PRISM kit.Reaction conditions (including amplification cycle) and the removal ofunincorporated reaction components were performed according to themanufacturer’s recommendations. Sequence analysis were done withSEQUENCHER (Genecodes Corp.). All nucleotide sequences weredetermined on both strands from at least two independently amplifiedtemplates and consensus sequences were obtained from multipledeterminations to avoid errors arising from the use of Taq polymerase.

+ Cloning PCR products. Amplicons of the 25K FP gene of passage6 BV were cloned into pGEM-T Vector (Promega) and transformed intoE. coli JM109 competent cells. After transformation, the positive cloneswere picked and DNA templates for sequencing were isolated using theQiagen Plasmid Mini Kit.

ResultsSerial passaging of HaSNPV in vitro

The effect of serial passaging HaSNPV in Helicoverpa zeasuspension cultures was investigated through analysis of thecell specific yield (polyhedra produced per cell), BV productionand percentage of FP infected cells at each passage as observedusing TEM. A decrease in cell specific yield was noted as theHaSNPV was serially passaged in SF900II medium (Fig. 1). Thedrop was most pronounced from passage 3 to passage 4. In asingle passage, the cell specific yield decreased from 178polyhedra per cell to 57 polyhedra per cell, a 65–70%reduction in yield (Table 1). By the sixth passage, the cellspecific yield was less than 5 polyhedra per cell. In contrast tothe drop in polyhedra yield from the third to the fourthpassage, the BV titre increased 3-fold. However, the BV titresdeclined from passage 5 onwards. TEM examination of theinfected cells revealed an increase in the percentage of FPinfected cells with increasing passage number. A steep increasein the FP mutant population was noted from the third passageto the fourth passage. The steep slope between passage 3 andpassage 4 indicates a sharp transition from an MP dominatedpopulation to a FP one.

Bioassays were performed to determine the biologicalactivity of polyhedra from the serially passaged HaSNPV. Thebiological activity of the polyhedra decreased with passagenumber (Table 1). Passage 4 polyhedra were similar in potency

zone of a FP infected nucleus with presence of de novo intranuclear membrane profiles (arrows). (C) Virion occlusion in MPproducer. (D) FP mutant producing polyhedra without any virions occluded. (E) MP producer with many polyhedra in thenucleus and all polyhedra have virions occluded. (F) Few polyhedra were produced by the FP infected cell and the polyhedraformed have no virions occluded. Nu, nucleus ; VS, virogenic stroma; r, ring zone; v, virion ; nc, nucleocapsid ; pp, polyhedrin ;p, polyhedron ; nm, nuclear membrane; c, cytoplasm.

to passage 3 polyhedra ; however, passage 6 polyhedra were28-fold less potent compared to passage 3 polyhedra.

Nucleocapsid envelopment and virion occlusion

In a wild-type MP infected cell, nucleocapsids are producedin the virogenic stroma, and migrate to the ring zone of thenucleus to acquire their envelopes through a de novo intra-nuclear membrane synthesis process (Fig. 2A). The envelopedvirions then gather in the ring zone, near the inner nuclearmembrane, ready for occlusion into the polyhedrin proteinmatrix. Abnormalities in morphogenesis became apparent atthe fourth passage. A majority of the nucleocapsids in the ringzone of the FP infected cells did not acquire envelopes (Fig. 2B).The intranuclear membrane profiles believed to be involved inthe nucleocapsid envelopment process were clearly observedin these FP infected cells. However, only infrequently were thenucleocapsids enveloped in the FP infected cell cases, indicatingan impaired nucleocapsid envelopment process in FP infectedcells.

Examination of the occlusion process in MP infected cellssuggests that the enveloped virions were progressivelyoccluded by polymerization of polyhedrin protein (Fig. 2C).Naked, unenveloped nucleocapsids did not appear to beoccluded by deposition of polyhedrin protein on their surface.Few polyhedra were produced in FP infected cells, and usuallythese polyhedra do not have virions occluded (Fig. 2D). UnderTEM, a reduction in the number of polyhedra per section of aFP infected cell, as compared to an MP infected cell, was clearlydetected (Fig. 2E, F).

Obvious signs of infection, such as presence of virogenicstroma, nucleocapsids and intranuclear membrane profiles,were evident in the FP infected cells. Despite the signs ofinfection, some FP infected cells do not form polyhedra (Fig.3A). Another phenotypic characteristic of FP infected cells wasaccumulation of intranuclear membrane profiles in the ringzone of infected nuclei (Fig. 3B). In some FP mutant cases, theintranuclear membranes became elongated, angular and un-usual in shape, forming massive swirls in the ring zone of thenuclei. In addition, some FP infected cells also producedpolyhedra with morphologically aberrant virions occluded(Fig. 3C) or associated with the edge of an occlusion (Fig. 3D),or polyhedra with very few virions occluded (Fig. 3E).

REN digestion analysis

The genomic REN digestion patterns of passage 2 topassage 6 BV DNA were obtained with three restrictionendonucleases to determine if genomic changes had occurred

JEJ


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


Fig. 3. Appearance of various FP infected cells at passage 4 and passage 5. (A) Infected cell with many nucleocapsids andintranuclear membrane profiles (arrow) but no formation of polyhedra. (B) Massive swirls or whorls in the ring zone of nucleus.(C) Polyhedron with aberrant morphology. (D) Non-enveloped and enveloped virions failed to be occluded during thecrystallization of polyhedrin. (E) Aberrant polyhedra with few virions occluded. VS, virogenic stroma; nm, nuclear membrane;nc, nucleocapsid ; p, polyhedron ; v, virion.

(such as large insertions or deletions) during serial passagingthat could be correlated to the appearance of the FP phenotype.The REN digestion patterns of passage 2 to passage 6 virusappeared similar (Fig. 4). No large insertion or deletion wasindicated as likely from the REN digestion patterns generatedby any of the restriction endonucleases.

Genetic analysis of the 25K FP gene of wild-typeHaSNPV and a FP mutant

The phenotypic observations on the HaSNPV FP mutantssuggest mutations in the 25K FP gene as reported for otherbaculoviruses. Thus, the 25K FP gene of HaSNPV was

JFA


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


(A) (B) (C)

M P2 P3 P4 P5 P6 M M P2 P3 P4 P5 P6 M M P2 P3 P4 P5 P6 M

Fig. 4. Genomic DNA digests of HaSNPVwith BamHI (A), EcoRI (B) and HindIII(C). The lanes marked M contain DNAsize markers (1 kb DNA ladder) andlanes marked P2–P6 contain digests ofgenomic DNA obtained from buddedvirus after each serial passage.

identified and sequenced. A small fragment (250 bp) of the 25KFP gene was first amplified and sequenced with a set ofdegenerate primers that were designed based on the conservedregion of the 25K FP gene for five other baculoviruses. Theentire 25K FP gene nucleotide sequence was obtained withprimer walking sequencing directly on the genomic DNAtemplate. The 25K FP gene opening reading frame of HaSNPVis 654 bp, encoding 217 amino acids (Fig. 5A). Seven nucleotidedifferences (at nucleotide positions 45, 124, 288, 426, 561, 570and 576, as starting from the first nucleotide of the gene) weredetected between the 25K FP gene nucleotide sequencesreported in this paper and the HaSNPV genome recentlypublished by Chen et al. (2001). No predicted amino aciddifference was observed for all nucleotide differences except atnucleotide position 576, which results in a conservative residuechange (an aspartate for our isolate whereas it is a glutamatefor Chen’s group). The amino acid sequence of the 25K FP geneis conserved among baculoviruses (Fig. 5B). The only regionlacking significant conservation is the last 11–35 C-terminalamino acids.

The nucleotide sequence of the 25K FP gene was de-termined from the wild-type HaSNPV (in vivo producedODV), passage 2 BV and passage 6 BV for comparison.Contrary to expectations, no nucleotide differences within the25K FP gene were detected in any of these samples. Even theupstream and downstream regions of the gene (200–250 bpon each end) were identical for all sources of DNA. Cleanchromatographic sequence data without overlapping or hiddenpeaks were obtained from all sequencing samples. As it waspossible that passage 6 BV might be composed of a mixture ofFP and MP viruses that appeared to lack detectable DNAinsertions or deletions within this gene in the consensussequences obtained from the amplicons, further sequence workwas warranted. The 25K FP gene amplicons of passage 6 BVwere cloned : nine individual clones were picked and sequencedto measure the variability of this gene at passage 6. A threetimes plaque-purified FP mutant (FP8AS) was also isolated. As

presented in Fig. 6(A), many point mutations and}or pointinsertions were observed throughout the entire 25K FP gene.Only two out of the nine clones were found to carry the wild-type 25K FP gene sequence.

Four clones (C8, C19, C52 and C59) were identified tocontain one or two point mutations, which are predicted toresult in a single amino acid change. No amino acid change waspredicted for clone C44, which contains a single nucleotidechange in position 102 (C8 also carries this change along witha change at nucleotide position 218). Clone C48 has an extraadenine residue inserted within a string of six adenine residuesbetween nucleotide position 256 and 262, whereas an extraadenine residue was inserted within a string of seven adenineresidues (nucleotide position 418–424) in both FP8AS andclone C2. These insertions found on FP8AS, clones C2 andC48, result in a coding frameshift, leading to the introductionof stop codons (after amino acid 143 for FP8AS and clone C2,after amino acid 112 for clone C48), which would then resultin truncated proteins compared to the wild-type 25K FPprotein (Fig. 6B). Electron microscopy has also confirmed thatthe isolate FP8AS exhibited typical phenotypic characteristicsof an FP mutant, such as few polyhedra per cell and polyhedrawithout virions occluded.

DiscussionHaSNPV appears to undergo rapid mutation and selection

events when serially passaged in cell culture, leading to theformation and selection of FP mutants. The HaSNPV isolatewas predominately a population of MP virus at passage 1 inculture but transformed totally to a FP population after sixpassages. This transition was obvious between passage 3 topassage 4 as a 65–70% reduction in cell specific yield and an80% increase in FP mutants was observed within one passage(Fig. 1). Upon more careful examination, these FP mutantsexhibit characteristic traits such as low polyhedra production(on average less than 10 polyhedra per ultrathin section),

JFB


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


(A)

(B)

Fig. 5. Genetic analysis of the HaSNPV 25K FP gene. (A) Nucleotide sequences of the 25K FP gene including the upstreamand the downstream regions of the gene. The 25K FP gene is 654 bp (bold). The region that contains a string of sevenadenine residues is underlined. (B) Multiple amino acid sequence alignment of 25K FP protein from variousnucleopolyhedroviruses. The number of amino acids encoded by each gene is indicated at the end of each sequence. AcMNPV(Ayres et al., 1994); BmMNPV (Gomi et al., 1999); GmMNPV (Beames & Summers, 1989); LdMNPV (Bischoff & Slavicek,1997); OpMNPV (Ahrens et al., 1997); SeMNPV (IJkel et al., 1999).

reduced or no virions occluded, an impaired intranuclearnucleocapsid envelopment process and reduced potency (Fig.2). The phenotypic effects of FP mutations observed with

HaSNPV are similar to those previously reported for otherbaculoviruses such as AcMNPV (Harrison & Summers, 1995b),TnMNPV (Potter et al., 1976), GmMNPV (Fraser et al., 1983),

JFC


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


(B)

(A)

Fig. 6. Mutations in the HaSNPV 25K FP gene. (A) Different mutationsidentified in the 25K FP gene from the clones of the 25K FP gene PCRproducts. Point mutations were found randomly throughout the entiregene. (B) Predicted effects of the mutations on the 25K FP protein aminoacid sequences expressed by the HaSNPV FP mutants. Point mutations atnucleotide position 102 (clones C8 and C44) would not result in apredicted amino acid change; 1 bp insertions (FP8AS, clones C2 andC48) result in a coding frameshift and truncated proteins. Solid areasindicate homologous regions of the mutant compared to the wild-typeprotein and the hatched area indicates non-homologous regions.

Bombyx mori MNPV (BmMNPV) (Katsuma et al., 1999), andLdMNPV (Slavicek et al., 1992).

The number of polyhedra produced by a HaSNPV FPinfected cell was significantly lower (! 10 polyhedra persection) than a HaSNPV MP infected cell ( 50 polyhedra persection) (see Table 1 and Fig. 2). Harrison et al. (1996)demonstrated that the rate of polyhedrin mRNA expression issignificantly reduced in cells infected with AcMNPV FPmutants compared to that of an MP infection. In FP infectedcells, polyhedrin localizes less efficiently to the nucleus duringthe early occlusion phase of infection (24 h post-infection), yetpolyhedrin mRNA stability is similar in MP and FP mutantinfected cells. The reduction in polyhedrin synthesis andnuclear localization could account for the reduced number ofpolyhedra observed with HaSNPV FP mutant infections andother baculoviruses by resulting in a significantly decreasedconcentration of polyhedrin available in the nucleus forocclusion assembly. Clearly, this requires further investigationof the effect of the 25K FP gene mutation on the expressionand localization of polyhedrin in the HaSNPV infected cells.

An accumulation of a large number of naked nucleocapsidsin the nuclei of FP infected cells, presumably a result of animpaired nucleocapsid envelopment, was a striking observationin this study. Although nucleocapsid assembly is apparentlynormal, there are few enveloped virions (the ODV) present in

FP mutant infected cells. The appearance of intranuclearaccumulations of short, open-ended, membrane profiles in theFP mutant infected cells also suggested that there is a specificfailure in the envelopment process of nucleocapsids, not in theproduction of envelopes themselves. Altered intranuclearnucleocapsid envelopment due to FP mutation in AcMNPVwas reported as early as 1974 (Ramoska & Hink, 1974). Nakednucleocapsids did not appear to be occluded. It is welldocumented that only enveloped virions are associated withpolyhedrin protein, leading to occlusion of the virions (Harrap,1972). The virion envelope (or proteins embedded in it) may benecessary for an interaction of the virion with polyhedrinprotein during the occlusion process (Wood, 1980). Thus, animpaired intranuclear nucleocapsid envelopment process inHaSNPV FP mutant infected cells may directly or indirectlyresult in the production of FP polyhedra that have no virionsoccluded.

The rate of FP mutant proliferation relative to parental typein different cell–virus systems is variable. In this work, morethan 80% of cells were infected by FP mutants at passage 4during serial passaging of HaSNPV, and 100% FP mutantinfected cells were obtained by passage 6. With TnMNPV, thevirus population was 90% FP only after 10 passages (Potteret al., 1976) but mutation}selection was much faster forLdMNPV, which shows a 96% FP population after passage 2(Slavicek et al., 1995). OpSNPV FP mutation is affected by hostcells (Sohi et al., 1984), but formation of FP variants ofGmMNPV is not a host-dependent phenomenon and theirdetection can be influenced by the overlay formulation usedduring plaque assays (Fraser & Hink, 1982). This suggests thatdifferences in media used to propagate cells and prepareoverlays may contribute to inconsistencies in the quantitativedetection of FP plaques. FP mutants are often distinguishedfrom MP virus on the basis of a less refractive plaquemorphology, which can be subjective. Therefore, in this study,TEM was used to score MP and FP mutant infected cells ateach passage instead of conventional plaque assays.

A trait commonly observed in most baculovirus FPmutations is an increase in the production of BV. Higher BVtitres were reported in FP mutants of AcMNPV (Harrison &Summers, 1995b ; Wood, 1980), TnMNPV (Potter et al., 1976,1978) and LdMNPV (Slavicek et al., 1995). The BV titres ofLdMNPV FP mutants increase significantly, as high as 150- to250-fold (Slavicek et al., 1995). However, an increase of BVproduction due to FP mutation was not observed for HaSNPV.A significant 3-fold increase in the BV titre was only detectedat the fourth passage of the HaSNPV isolate, but decreased inthe subsequent passages. Chakraborty & Reid (1999) reporteda steady increase in BV titres from passage 2 to passage 5during serial passaging of HaSNPV in SF900II mediumsupplemented with 10% serum but titres decreased in thesubsequent passages. Interestingly, declining BV titres duringserial passaging of HzSNPV (Yamada et al., 1982) andLdMNPV (Lynn, 1994) were also reported. A decrease in BV

JFD


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


production during serial passaging in cell culture could eitherbe a trait specific to the Helicoverpa sp. or may be dependent onthe cell–virus system. As HaSNPV FP mutations do not causean increase in BV production, one possible reason for the FPmutants to out-complete MP virus could be their ability tobind and infect cells at a higher efficiency than MP virus.

The phenotypic observations on the HaSNPV FP mutantssuggest genomic changes occurring in the 25K FP gene, whichcould be large insertions or deletions within the gene.However, the genomic digestion profiles of all passagesgenerated separately with three restriction endonucleases weresimilar, indicating that no large insertions or deletions of theviral genome occurred during serial passaging. The consensussequence data obtained from the 25K FP gene amplicon of bothpassage 2 and passage 6 material are identical. This result wasunexpected since the HaSNPV FP mutants exhibit well-documented FP phenotypes that are correlated to mutations inthe 25K FP gene of other baculoviruses. Speculations on thepossibility that passage 6 comprised a mixture of FP mutantsthat carry very small insertions or deletions as a result ofrandom mutations, thus resulting in a consensus sequence ofthe wild-type 25K FP gene, was confirmed when 25K FP genesequences obtained from individual clones of passage 6amplicons revealed a mixture of point mutations and}or pointinsertions within the gene. Two out of nine clones (C2 andC48) carried 1 bp insertions within a region of adeninerepetitive sequences. Five other clones carried point mutationson the gene. Further investigation with a plaque-purified FPmutant (FP8AS) also revealed a 1 bp insertion in the sameregion of repetitive sequences as clone C2 within the 25K FPgene. Frameshift mutations as a result of point insertions inFP8AS, clones C2 and C48 would lead to truncated 25K FPproteins. The 25K FP gene mutations observed in HaSNPV arenot specific to a particular region but rather they are randomlyfound throughout the gene. Hence, not all baculovirus FPmutants have similar genetic mutations within the 25K FPgene.

The FP mutation of HaSNPV is likely to arise through adifferent mechanism from AcMNPV and GmMNPV (Beames& Summers, 1988 ; Fraser et al., 1983) but probably similar tothat for LdMNPV (Bischoff & Slavicek, 1997 ; Slavicek et al.,1995). HaSNPV FP mutants are not generated through largeDNA insertions or deletions that are of sufficient size to allowdetection by genomic REN digestion analysis. Five nucleotidesites consisting of 5« TTAA 3«, which are frequently associ-ated with transposon insertions within the AcMNPV andGmMNPV 25K FP genes (Beames & Summers, 1988 ; Fraser etal., 1985 ; Wang et al., 1989), were found in the HaSNPV 25KFP gene. However, 14 such sites were identified in theAcMNPV 25K FP gene (Wang et al., 1989), which could leadto a higher rate of transposon insertions for AcMNPV. One bpinsertions in repetitive sequences and point mutations ob-served for HaSNPV are more similar to the 25K FP gene mu-tations reported for LdMNPV (Bischoff & Slavicek, 1997). As

discussed by Bischoff & Slavicek (1997), 1 bp insertions withinrepetitive sequences (observed here with isolate FP8AS, clonesC2 and C48) are most likely the result of DNA polymeraseslippage on the newly synthesized strand.

The work has further confirmed the importance of the 25KFP gene during the production and assembly of progeny virus.From both a fundamental and applied science perspective, theeffects of mutation in this gene deserve further study.Characterization of HaSNPV MP and FP virus is currentlyunder way in our laboratory, in an effort to understand theselective advantage of HaSNPV FP mutants since they do notappear to gain their advantage simply through the productionof higher BV titres. In particular, the cell–virus binding anduptake characteristics of MP and FP virus are under investiga-tion.

We are grateful for the bioassays carried out by Dr Peter Christian(CSIRO) and the technical advice provided by the Centre for Microscopyand Microanalysis (UQ). We also thank Dr Lars Nielsen for his help withstatistical analysis. This research was supported by the AustralianResearch Council and the Grains Research and Development Cor-poration. Scholarships were provided by the University of Queenslandand the Brazilian Ministry of Education and Culture.

ReferencesAhrens, C. H., Russell, R. L., Funk, C. J., Evans, J. T., Harwood, S. H. &Rohrmann, G. F. (1997). The sequence of the Orgyia pseudotsugatamultinucleocapsid nuclear polyhedrosis virus genome. Virology 229,381–399.

Ayres, M. D., Howard, S. C., Kuzio, J., Lopez-Ferber, M. & Possee,R. D. (1994). The complete DNA sequence of Autographa californicanuclear polyhedrosis virus. Virology 202, 586.

Beames, B. & Summers, M. D. (1988). Comparisons of host cell DNAinsertions and altered transcription at the site of insertions in fewpolyhedra baculovirus mutants. Virology 162, 206–220.

Beames, B. & Summers, M. D. (1989). Location and nucleotide sequenceof the 25K protein missing from baculovirus few polyhedra (FP) mutants.Virology 168, 344–353.

Beniya, H., Braunagel, S. C. & Summers, M. D. (1998). Autographacalifornica nuclear polyhedrosis virus : subcellular localization and pro-tein trafficking of BV}ODV-E26 to intranuclear membranes and viralenvelopes. Virology 240, 64–75.

Bischoff, D. S. & Slavicek, J. M. (1997). Phenotypic and genotypicanalysis of Lymantria dispar nucleopolyhedrovirus few polyhedra mu-tants : mutations in 25K FP gene may be caused by DNA replicationerrors. Journal of Virology 71, 1097–1106.

Braunagel, S. C., Burks, J. K., Rosas-Acosta, G., Harrison, R. L., Ma, H.& Summers, M. D. (1999). Mutations within the Autographa californicanucleopolyhedrovirus FP25K gene decrease the accumulation of ODV-E66 and alter its intranuclear transport. Journal of Virology 73, 8559–8570.

Chakraborty, S. (1997). In vitro production of wild-type Helicoverpabaculovirus. PhD thesis, University of Queensland, Australia.

Chakraborty, S. & Reid, S. (1999). Serial passage of a Helicoverpaarmigera nucleopolyhedrovirus in Helicoverpa zea cell cultures. Journal ofInvertebrate Pathology 73, 303–308.

Chen, X., IJkel, F. W. J., Tarchini, R., Sun, X., Sandbrink, H., Wang, H.,Peters, S., Zuidema, D., Lankhorst, R. K., Vlak, J. M. & Hu, Z. (2001).

JFE


IP: 54.159.97.142

On: Sat, 23 Jul 2016 12:49:51


The sequence of the Helicoverpa armigera single nucleocapsid nucleopoly-hedrovirus genome. Journal of General Virology 82, 241–257.

Fraser, M. J. & Hink, W. F. (1982). Comparative sensitivity of severalplaque assay techniques employing TN-368 and IPLB-SF 21AE insect celllines for plaque variants of Galleria mellonella nuclear polyhedrosis virus.Journal of Invertebrate Pathology 40, 89–97.

Fraser, M. J., Smith, G. E. & Summers, M. D. (1983). Acquisition ofhost cell DNA sequences by baculoviruses : relationship between hostDNA insertions and FP mutants of Autographa californica and Galleriamellonella nuclear polyhedrosis viruses. Journal of Virology 67, 287–300.

Fraser, M. J., Brusca, J. S., Smith, G. E. & Summers, M. D. (1985).Transposon-mediated mutagenesis of a baculovirus. Virology 145,356–361.

Gomi, S., Majima, K. & Maeda, S. (1999). Sequence analysis of thegenome of Bombyx mori nucleopolyhedrovirus. Journal of General Virology80, 1323–1337.

Harrap, K. A. (1972). The structure of nuclear polyhedrosis viruses. III.Virus assembly. Virology 50, 133–139.

Harrison, R. L. & Summers, M. D. (1995a). Biosynthesis and localizationof the Autographa californica nuclear polyhedrosis virus 25K gene product.Virology 208, 279–288.

Harrison, R. L. & Summers, M. D. (1995b). Mutations in the Autographacalifornica multinucleocapsid nuclear polyhedrosis virus 25 kDa proteingene result in reduced virion occlusion, altered intranuclear envelopmentand enhanced virus production. Journal of General Virology 76, 1451–1459.

Harrison, R. L., Jarvis, D. L. & Summers, M. D. (1996). The role of theAcMNPV 25K gene, ‘FP25, ’ in baculovirus polh and p10 expression.Virology 226, 34–46.

Hink, W. F. & Vail, P. V. (1973). A plaque assay for titration of alfalfalooper nuclear polyhedrosis virus in a cabbage looper (TN-368) cell line.Journal of Invertebrate Pathology 22, 168–174.

IJkel, W. F. J., Van Strien, E. A., Heldens, J. G. M., Broer, R., Zuidema,D., Goldbach, R. W. & Vlak, J. M. (1999). Sequence and organization ofthe Spodoptera exigua multicapsid nucleopolyhedrovirus genome. Journalof General Virology 80, 3289–3304.

Katsuma, S., Noguchi, Y., Zhou, C. L., Kobayashi, M. & Maeda, S.(1999). Characterization of the 25K FP gene of the baculovirus Bombyxmori nucleopolyhedrovirus : implications for post-mortem host degra-dation. Journal General Virology 80, 783–91.

Lua, L. H. L. & Reid, S. R. (2000). Virus morphogenesis of Helicoverpaarmigera nucleopolyhedrovirus in Helicoverpa zea serum-free suspensionculture. Journal of General Virology 81, 2531–2543.

Lynn, D. E. (1994). Enhanced infectivity of occluded virions of thegypsy moth nuclear polyhedrosis virus for cell culture. Journal ofInvertebrate Pathology 63, 268–274.

McIntosh, A. H. & Ignoffo, C. M. (1981). Replication and infectivity ofthe single-embedded nuclear polyhedrosis virus, baculovirus Heliothis, inhomologous cell line. Journal of Invertebrate Pathology 37, 258–264.

Miller, L. K. (1986). The genetics of baculoviruses. In The Biology ofBaculoviruses, pp. 217–238. Edited by R. R. Granados & B. A. Federici.Florida : CRC Press.

Nielsen, L. K., Smyth, G. K. & Greenfield, P. F. (1991). Hemocytometercell count distributions : implications of non-Poisson behavior. Biotech-nology Progress 7, 560–563.

Potter, K. N., Faulkner, P. & MacKinnon, E. A. (1976). Strain selectionduring serial passage of Trichoplusia ni nuclear polyhedrosis virus. Journalof Virology 18, 1040–1050.

Potter, K. N., Jaques, R. P. & Faulkner, P. (1978). Modification ofTrichoplusia ni nuclear polyhedrosis virus passaged in vivo. Intervirology 9,76–85.

Ramoska, W. A. & Hink, W. F. (1974). Electron microscope examinationof two plaque variants from a nuclear polyhedrosis virus of the alfalfalooper, Autographa californica. Journal of Invertebrate Pathology 23,197–201.

Rhodes, D. J. (1996). Economics of baculovirus–insect cell productionsystems. Cytotechnology 20, 291–297.

Russell, R. L. Q. & Rohrmann, G. F. (1993). A 25 kDa protein isassociated with the envelopes of occluded baculovirus virions. Journal ofGeneral Virology 195, 532–540.

Slavicek, J. M., Podgwaite, J. & Lanner-Herrera, C. (1992). Propertiesof two Lymantria dispar nuclear polyhedrosis virus isolates obtainedfrom microbial pesticide Gypchek. Journal of Invertebrate Pathology 59,142–148.

Slavicek, J. M., Hayes-Plazolles, N. & Kelly, M. E. (1995). Rapidformation of few polyhedra mutants of Lymantria dispar multinucleo-capsid nuclear polyhedrosis virus during serial passage in cell culture.Biological Control 5, 251–261.

Sohi, S. S., Percy, J., Arif, B. M. & Cunningham, J. C. (1984). Replicationand serial passage of a singly enveloped baculovirus of Orgyia leucostigmain homologous cell lines. Intervirology 21, 50–60.

Taylor, J. R. (1997). The binomial distribution. In An Introduction to ErrorAnalysis : The Study of Uncertainties in Physical Measurements, 2nd edn,pp. 227–244. California : University Science Books.

Wang, H. H., Fraser, M. J. & Cary, L. C. (1989). Transposon mutagenesisof baculoviruses : analysis of TFP3 lepidopteran transposon insertions atthe FP locus of nuclear polyhedrosis virus. Gene 81, 97–108.

Wood, H. A. (1980). Isolation and replication of an occlusion body-deficient mutant of the Autographa californica nuclear polyhedrosis virus.Virology 105, 338–344.

Yamada, K., Sherman, K. E. & Maramorosch, K. (1982). Serial passageof Heliothis zea singly embedded nuclear polyhedrosis virus in ahomologous cell line. Journal of Invertebrate Pathology 39, 185–191.

Received 6 November 2001; Accepted 20 December 2001

JFF

phenotypic and genotypic analysis of helicoverpa armigera nucleopolyhedrovirus serially passaged in...

Documents