Vlak J.M., de Gooijer C.D., Tramper J., Miltenburger H.G. (Eds.) Insect Cell Cultures: Fundamental and Applied Aspects

Подождите немного. Документ загружается.

122

Production of libraries of very high complexity will

require further advances in baculovirus recombinant

DNA technology. For instance, it may be possible to

achieve efficient conversion of plasmid into virus and

a high yield of recombinants by combining the in vit-

ro site-specific recombination system with the Ganci-

clovir, p35

or ORF1629 selections for recombinant

viruses.

Acknowledgments

I am grateful to Kristen Mayo for reviewing this man-

uscript.

References

Abremski K, Hoess R & Sternberg N (1983) Studies on the proper-

ties of P1 site-specific recombination: evidence for topologically

unlinked products following recombination. Cell 32: 1301–1311.

Ayres MD, Howard SC, Kuzio J, Lopez-Ferber M & Possee RD

(1994) The complete DNA sequence of Autographa californica

nuclear polyhedrosis virus. Virology 202: 586–605.

Bonning BC & Hammock BD (1992) Development and potential

of genetically engineered viral insecticides. In: Tombs MP(ed.)

Biotechnology and genetic engineering reviews, Vol. 10(pp.455–

489) Intercept, Andover.

Clem RJ, Fechheimer M & Miller LK (1991) Prevention of apoptosis

by a baculovirus gene during infection of insect cells. Science

254: 1388–1390.

Copeland WC & Wang TS-F (1993) Mutational analysis of the

human DNA polymerase a. J. Biol. Chem. 268: 11028–11040.

Day A, Wright T, Sewall A, Price-Laface M, Srivastava N & Fin-

layson M (1995) Rapid procedures for the isolation and PCR

analysis of recombinant baculovirus. In: Richardson (ed.) Meth-

ods in Molecular Biology, Vol. 39: Baculovirus Expression Pro-

tocols (pp. 143–159) Humana Press, Totowa, NJ.

Davies AH (1994) Current methods for manipulating baculoviruses.

Bio/Technology 12: 47–50.

Ernst WJ, Grabherr RM & Katinger HWD (1994) Direct cloning

into the Autographa californica nuclear polyhedrosis virus for

generation of recombinant baculoviruses. Nucleic Acids Res. 22:

2855–2856.

Fung M-C, Chiu KYM, Weber T, Chang T-W & Chang NT (1988)

Detection and purification of a recombinant human B lym-

photropic virus hhv–6 in the baculovirus expression system by

limiting dilution and DNA dot-blot hybridization. J. Virol. Meth.

19: 33–42.

Godeau F, Saucier C & Kourilsky P (1992) Replication inhibition

by nucleoside analogues of a recombinant Autographa californi-

ca multicapsid nuclear polyhedrosis virus harboring the herpes

thymidine kinase gene driven by the IE–1(0) promoter: a new

way to select recombinant baculoviruses. Nucleic Acids Res. 20:

6239–6246.

Granados RR & Federici BA (1986) The biology of baculoviruses,

Vol. II. CRC Press, Boca Raton, Florida.

Grosse F & Manns A (1995) Fast isolation of recombinant bac-

ulovirus by antibody screening. In: Richardson (ed.) Methods in

Molecular Biology, Vol. 39: Baculovirus Expression Protocols

(pp. 179–185) Humana Press, Totowa, NJ.

Hartig PC & Cardon MC (1992) Rapid efficient production of bac-

ulovirus expression vectors. J. Virological Methods 38: 61–70.

Hershberger PA, Dickson JA & Friesen PD (1992) Site-specific

mutagenesis of the 35-kilodalton protein gene encoded by Auto-

grapha californica nuclear polyhedrosis virus: cell line-specific

effects on virus replication. J. Virol. 66: 5525–5533.

Hoess R & Abremski K (1984) Interaction of the bacteriophage

P1 recombinase Cre with the recombining site loxP. Proc. Natl.

Acad. Sci. USA 81: 1026–1029.

Kitts PA (1995) Production of recombinant baculoviruses using lin-

earized viral DNA. In: Richardson (ed.) Methods in Molecular

Biology, Vol. 39: Baculovirus Expression Protocols (pp. 129–

142) Humana Press, Totowa, NJ.

Kitts PA, Ayres MD & Possee RD (1990) Linearization of bac-

ulovirus DNA enhances the recovery of recombinant virus expres-

sion vectors. Nucleic Acids Res. 18: 5667–5672.

Kitts PA & Possee RD (1993) A method for producing recombinant

baculovirus expression vectors at high frequency. BioTechniques

14:810–817.

Lalumière M & Richardson CD (1995) Production of recombinant

baculoviruses using rapid screening vectors that contain the gene

for In: Richardson (ed.) Methods in Molecular

Biology, Vol. 39: Baculovirus Expression Protocols (pp. 161-

177) Humana Press, Totowa, NJ.

Lerch RA & Friesen PD (1993) The 35-kilodalton protein gene (p35)

of Autographa californica nuclear polyhedrosis virus and the

neomycin resistance gene provide dominant selection of recom-

binant baculoviruses. Nucleic Acids Res. 21: 1753–1760.

López-Ferber M, Sisk WP & Possee RD (1995) Baculovirus trans-

fer vectors. In: Richardson (ed.) Methods in Molecular Biology,

Vol. 39: Baculovirus Expression Protocols (pp. 25–63) Humana

Press, Totowa, NJ.

Luckow VA, Lee SC, Barry GF & Olins PO (1993) Efficient gen-

eration of infectious recombinant baculoviruses by site-specific

transposon-mediated insertion of foreign genes into a baculovirus

genome propagated in Escherichia coli. J. Virol. 67: 4566–4579.

Manns A & Grosse F (1991) Fast isolation of recombinant bac-

ulovirus by antibody screening. BioTechniques 10: 154–158.

Martens JWM, van Oers MM, van de Bilt BD, Oudshoorn P & Vlak

JM (1995) Development of a baculovirus vector that facilitates

the generation of pl0-based recombinants. J. Virological Meths.

52: 15–19.

Patel G, Nasmyth K & Jones N (1992) A new method for the isolation

of recombinant baculovirus. Nucleic Acids Res. 20: 97–104.

Peakman T, Page M & Gewert D (1989) Increased recombinational

efficiency in insect cells irradiated with short wavelength ultra-

violet light. Nucleic Acids Res. 17: 5403.

Peakman TC, Harris RA & Gewert DR (1992) Highly efficient gen-

eration of recombinant baculoviruses by enzymatically mediated

site-specific in vitro recombination. Nucleic Acids Res. 20: 495–

500.

Pen J, Welling GW & Welling-Wester S (1989) An efficient proce-

dure for the isolation of recombinant baculovirus. Nucleic Acids

Res. 17:451.

Peng S, Sommerfelt MA, Berta G, Berry AK, Kirk KL, Hunter E &

Sorscher EJ (1993) Rapid purification of recombinant baculovirus

using fluorescence-activated cell sorting. BioTechniques 14: 274–

277.

Pennock GD, Shoemaker C & Miller LK (1984) Strong and regulated

expression of Escherichia coli in insect cells with

a baculovirus vector. Mol. Cell. Biol. 4: 399–406.

123

Pham DQ-D, Hice RH, Sivasubramanian & Federici BA (1993)

The 1629-bp open reading frame of the Autographa californica

multinucleocapsid nuclear polyhedrosis virus encodes a virion

structural protein. Gene 137: 275–280.

Possee RD, Sun T-P, Howard SC, Ayres MD, Hill-Perkins M &

Gearing KL (1991) Nucleotide sequence

f the Autographa cali-

fornica nuclear polyhedrosis 9.4 kbp EcoRI-I and -R (polyhedrin

gene) region. Virology 185: 229–241.

Richardson C, Attia J, Dunn R, Gupta S, O’Connor M, Semeniuk D,

Tam J, Hamel M, Lambert G, Dennis M, Jacobs F, Martin L, Iorio

C & Vialard J (1992a) Engineering glycoproteins for secretion

using the baculovirus expression system. In: Vlak JM, Schlaeger

E-J & Bernard AR (eds) Baculoviruses and recombinant protein

production processes (pp. 67–74) Roche, Basel.

Richardson CD, Banville M, Lalumière M, Vialard J & Meighen

EA (1992b) Bacterial luciferase produced with rapid screening

baculovirus vectors is a sensitive reporter for infection of insect

cells and larvae. Intervirology 34: 213–227.

Smith GE, Fraser MJ & Summers MD (1983) Molecular engineering

of the Autographa californica nuclear polyhedrosis virus genome:

deletion mutations within the polyhedrin gene. J. Virol. 46: 584–

593.

Sternberg N & Hoess R (1983) The molecular genetics of bacterio-

phage P1. Ann. Rev. Genet. 17: 123–154.

Summers MD & Smith GE (1987) A manual of methods for bac-

ulovirus vectors and insect cell culture procedures. Texas Agri-

cultural Experiment Station Bulletin No. 1555.

Vialard J, Lalumière M, Vernet T, Briedis D, Alkhatib G, Henning

D, Levin D & Richardson C (1990) Synthesis of the membrane

fusion and hemagglutinin proteins of measles virus, using a novel

baculovirus vector containing the gene. J. Virol.

64: 37–50.

Vialard JE & Richardson CD (1993) The 1,629-nucleotide open

reading frame located downstream of the Autographa cali-

fornica nuclear polyhedrosis virus polyhedrin gene encodes a

nucleocapsid-associated phosphoprotein. J. Virol. 67: 5859–

5866.

Vlak JM, Schouten A, Usmany M, Belsham GJ, Klinge-Roode EC,

Maule AJ, van Lent JWM & Zuidema D (1990) Expression of

cauliflower mosaic virus gene I using a baculovirus vector based

upon the p10 gene and a novel selection method. Virology 179:

312-320.

Weyer U, Knight S & Possee RD (1990) Analysis of very late gene

expression by Autographa californica nuclear polyhedrosis virus

and the further development of multiple expression vectors. J.

Gen. Virol. 71: 1525–1534.

Wood HA & Granados RR (1991) Genetically engineered bac-

uloviruses as agents for pest control. Ann. Rev. Microbiol. 45:

69–87.

Zuidema D, Schouten A, Usmany M, Maule AJ, Belsham GJ,

Roosien J, Klinge-Roode EC, van Lent JWM & Vlak JM (1990)

Expression of cauliflower mosaic virus gene I in insect cells using

a novel polyhedrin-based baculovirus expression vector. J. gen.

Virol. 71:2201–2209.

Address for correspondence: Paul A. Kitts CLONTECH Laborato-

ries, Inc., 1020 East Meadow Circle, Palo Alto, CA 94303–4230,

U.S.A.

This page intentionally left blank.

Cytotechnology 20

125–137, 1996.

125

Passage effect of virus infection in insect cells

Peter J. Krell

Department of Microbiology, University of Guelph, Guelph, Ontario, N1G 2W1 Canada

Key words: baculovirus, genomic alteration, passage, bioreactors, viral persistence, viral fitness

Introduction

Genomic plasticity is the hallmark of all biological

processes and for viruses is more apparent because

of their short generation times and high number of

“progeny” produced during each replication cycle or

passage. Despite the inherent fidelity of DNA replica-

tion, DNA genomes, including those of baculovirus-

es, continuously undergo variations in their nucleotide

sequences. Due to recombination and errors in DNA

replication there is a certain probability of error at

each passage which for the baculovirus genome leads

to point mutations, deletions, rearrangements, inver-

sions, reiterations and acquisition of host cell DNA.

It is in this background that baculoviruses replicate

either in cells of insect larvae or of insect tissue cul-

tures. Whether the baculovirus replicates in its insect

host as a “natural” pathogen, or in cultured cells for the

production of microbial insecticides and foreign pro-

teins of economic or scientific interest, the potential

for such genomic alterations is omnipresent.

Faulkner (1981) first suggested that the production

of variants, atypical morphogenesis and lower viru-

lence of baculovirus preparations occurred as a con-

sequence of repeated serial passage, herein referred to

as the “passage effect”. The result of prolonged (undi-

luted) passage of baculovirus in vitro

was first shown

by MacKinnon et al. (1974) for Trichoplusia ni NPV

(TnNPV) derived from “late passage” (up to 50) in

T. ni TN–368 cells. At late passage fewer polyhedra

per cell and less virulent polyhedra were produced.

Moreover polyhedral morphology mutants and aber-

rant and shorter nucleocapsids resulted and the number

of normal occluded virions per polyhedral inclusion

body (PIB; the form of baculoviruses responsible for

insect to insect horizontal transmission) also decreased.

Payne (1988) noted the importance of passage effect

in the production of effective viral pathogens for insect

control. Tramper & Vlak (1986) first recognized the

importance of the passage effect in bioreactor produc-

tion of baculoviruses and foreign protein in the context

of decreased productivity.

The accumulation of genomic alterations through

several replication cycles either through individual

insects or tissue culture flasks or through the sever-

al cycles of replication in bioreactors can be referred

to as the passage effect. Some of these new genotypic

variants could replicate at the expense of the origi-

nal parental virus and could, with sufficient numbers

of passages become dominant. The major problems

with the passage effect on baculovirus replication and

expression is the loss of virulence of PIB for virus

grown in tissue culture and, for viruses used as expres-

sion vectors, a reduced ability to produce the foreign

protein of interest. For growth of PIB in tissue culture

there is no selection pressure for virus capable of pro-

duction of infectious polyhedra. Similarly for foreign

genes in recombinant baculoviruses there is often no

selection to maintain and express those genes.

Baculovirus genomic alterations

That genomic alterations occur frequently among bac-

uloviruses was evident from the many genotypic vari-

ants (McIntosh et al., 1987) initially detected by

restriction fragment length polymorphism of many nat-

ural baculovirus isolates including Autographa cali-

fornica nuclear polyhedrosis virus (AcMNPV; Lee &

Miller, 1978; Miller, 1984), Spodoptera frugiperda

126

NPV (SfNPV; Knell & Summers, 1981), S. exempta

MNPV (Brown et al., 1985), S. litura NPV (Maeda et

al., 1990) and Panolis flammea MNPV (Weitzman et

al., 1992). That baculovirus genomes can recombine

was initially demonstrated by Summers et al. (1980)

for AcMNPV and Rachiplusia ou NPV and also by

recombination between mutant strains of AcMNPV

(Carstens et al., 1987). Moreover the development of

baculoviruses as expression vectors is predicated on

homologous recombination between baculovirus and

transfer vector DNAs.

Of the many genomic changes that can occur some

are silent and provide no selective advantage, some

are lethal and, except under unusual circumstances,

would be eliminated with passage. Certain changes

could enhance replicative advantage so that the mutant

virus could outcompete the original genotype during

successive passages. A single nucleotide change which

alters a codon but not the corresponding amino acid

could, except for factors such as codon usage (Ayres

et al., 1994), have no effect. An alteration in a codon

leading to substitution of a different amino acid could

lead to proteins such as polyhedrin with altered struc-

ture (e.g., mutant M29 and M934, Carstens et al., 1987;

1992) or influence the activity of viral enzymes criti-

cal for viral replication such as the AcMNPV helicase

(Lu & Carstens, 1991). Other changes such as dele-

tions or insertions of DNA in an essential gene, could

alter the reading frame. If deletions are large enough,

entire genes could be eliminated. Another alteration

which could influence gene expression is reiteration of

viral DNA such as reported by Lu & Carstens (1990).

Those genotypic changes which enhance the replica-

tion ability in host cells, or insects will be amplified at

each passage and will eventually dominate the original

genotype. The outcome depends on the virus species

and genotype, the cell (species and tissue source of

cells in culture or in insects), the nature of the genotypic

changes (genes affected, source and location of DNA

insertions or deletions), the method of virus culture

(passage in insects, cells in suspension, monolayers or

bioreactors, frequency of transfer) and multiplicity of

infection (moi).

For AcMNPV and other baculoviruses there are

numerous isolates, strains, variants or mutants which

have been extensively analyzed to determine the nature

of the genotypic variations (Eraser, 1987; Kool et al.,

1991; Kool&Vlak, 1993; Xiong et al., 1991; Table 1).

Genotypic variation in the AcMNPV and other bac-

ulovirus genomes, include point mutations, both small

and large deletions and insertions, recombination and

reiterations. There may be hot spots for certain genom-

ic alterations such as insertions due to transposable ele-

ments or the deletions in the hypervariable DA26 gene

region (O’Reilly et al., 1990), but all regions of the

genome appear capable of variation. Some variations

such as transposon-mediated insertions occur readily

even under low passage number and low moi (Fraser

et al., 1985). Other variations occur under conditions

of genomic stress during prolonged passage (Kumar

& Miller, 1987; O’Reilly & Miller, 1990; Miller &

Miller, 1982; Friesen et al., 1986) or under high mul-

tiplicity passage (HMP) in flasks or bioreactors (Kool

et al., 1991).

Deletions of viral DNA occurs readily. For

example, the genomes of the majority of the 667

AcMNPVLl-X plaques analyzed by Kumar & Miller

(1987), after only 10 passages at an moi of 0.1, had

deletions of 0.7 kb in Pst1 G (335) or 0.05 kb in Pst1

I (321). Many other deletions ranging in size from 0.1

to 2.3 kb have been reported and some are summarized

in Table 1. Some deletions inactivate viral genes such

as the UDPecdysteroid glucosyltransferase gene (egt;

O’Reilly & Miller, 1990) and the DA26 gene (O’Reilly

et al., 1990). Since these mutants survive, neither gene

is essential for replication in cell culture. The DA26

deletion occurred after passage of AcMNPV through

Manduca sexta larvae suggesting that deletions can

also occur by passage in insects. One consequence of

the passage effect is that some of the genotypic changes

which arise through passage involve sequential and

cumulative deletions of viral DNA (Kool et al., 1991;

Cusack & McCarthy, 1989; Wickham et al., 1991; Lee

& Krell, 1992; 1994; Croizier et al., 1985; Figure 1)

resulting in major losses of the standard genome. Such

highly deleted genomes are released from the infect-

ed cells as particles referred to as defective interfering

particles (DIP; Huang & Baltimore, 1977). A deletion

of 55 kb occurs during replication of M5, an AcM-

NPV PIB morphology mutant (Carstens, 1982; 1987)

and in the generation of AcMNPV DIP in a bioreac-

tor (Kool et al., 1991). Since DIP lack many essential

genes they cannot independently infect cells. Howev-

er, at HMP both DIP and helper virus can co-infect the

same cell allowing the DIP genome to replicate and

be expressed at the expense of the helper virus (Roux

et al., 1991; Bangham & Kirkwood, 1990). Moreover

the DIP concentration increases relative to the original

parental strain (wild type virus or recombinant virus)

and the virus titre and foreign gene expression would

decline with each passage (MacKinnon et al., 1974; De

Grooijer et al., 1992; Kompier et al., 1988; van Lier et

127

128

129

al., 1992). DIP can also mediate persistent infections

as observed for Heliothis zea NPV (Hz–1) DIP in IMC

Hz–1 and T. ni TN–368 cells (Wood & Burand, 1986).

Insertions of DNA into baculovirus genomes can

also occur at high frequency. Kumar & Miller (1987)

found 100 of 667 viral plaques with insertions of

0.08 to 2.5 kb of DNA, the majority of which (88)

were in Pst1

. The sizes of other insertions in bac-

ulovirus genomes range from 0.3 to 3.1 kb and up to

11.2 (Table 1) in both standard and recombinant bac-

uloviruses (Luckow, 1991). Insertion of DNA is usu-

ally mediated by homologous recombination, as for

the generation of recombinant baculoviruses (Luckow,

1991) and as demonstrated by the efficiency of marker

rescue (Carstens et al., 1987). However DNA inser-

tions can also be mediated by non-homologous recom-

bination as shown by Xiong et al. (1991) using aden-

ovirus DNA and, as is becoming increasingly obvious,

by transposons (Miller & Miller, 1982; Fraser et al.,

1983, 1985; Beames & Summers, 1990; Gombart et

al., 1989; Cary et al., 1989; Wang et al., 1989; O’Reil-

ly et al., 1990; Blissard & Rohrmann, 1990). DNA

insertions can be from the cell (e.g., transposons), co-

infecting viruses (Summers et al., 1980; Croizier et al.,

1994) or from transfected plasmids (Luckow, 1991;

Xiong et al., 1991). Other changes include reiterations

of viral DNA (Burand & Summers, 1982; Carstens &

Lu, 1990; Lee &Krell, 1994). These genetic exchanges

or rearrangements could dramatically change the virus

by altering its virulence or even host range as demon-

strated for an AcMNPV which exchanged as little as

79 bp of the Bombyx mori NPV (BmNPV) helicase

gene, allowing AcMNPV to replicate in B. mori cells

(Croizier et al., 1994).

Passage, genotypic changes and the few polyhedra

(FP) phenotype

The serial passage of any virus results in the accumu-

lation of genotypic variations among any virus popu-

lation, including that of baculoviruses. One of the first

variations which arose during serial passage of bac-

uloviruses described at the genetic level, was a change

from the parental, many polyhedra per cell (MP) phe-

notype, to the few polyhedra per cell (FP) phenotype

reported for both AcMNPV and Galleria mellonella

NPV (GmNPV; Fraser, 1987; Fraser et al., 1983, 1985;

Wang et al., 1989) and TnNPV (Potter et al., 1976).

That this was the earliest and most common pheno-

typic change during serial passage (Kumar & Miller,

1987) reflects the high frequency of genetic alterations

in the FP-locus and the non lethal nature of this change

for virus grown in tissue culture.

Potter et al. (1976) showed for TnNPV that FP

plaques could be recovered as early as passage 9 of an

undiluted serial passage in TN–368 cells. The propor-

tion of FP plaques recovered increased during passage

so that by passage 14, no MP plaques were observed.

Also non occluded virus (NOV; i.e., extracellular virus)

from the FP isolates were released at a faster rate than

MP PIB. Although FP NOV were infectious to TN–368

cells and to insects when injected into the haemocoel,

the corresponding FP polyhedra were much less infec-

tious per os than the MP NOV. When TnNPV MP NOV

was passaged by injection into T. ni larvae, FP NOV

was observed by 4 passages and by passage 14 rep-

resented about 95% of plaques isolated from infected

haemolymph (Potter et al., 1978). FP virus was detect-

ed only sporadically for virus passaged as PIB and

per os infection of larvae, but was not maintained nor

was there a change in virulence in per os transmitted

virus. Thus one way to maintain the MP character of an

inoculum is to ensure that it is passaged as PIB through

insect larvae.

The generation of some AcMNPV FP mutants

occurs even during a single passage at low moi and

most are due to transposon insertion in the FP-locus

(Miller & Miller, 1982; Fraser, 1987). The very high

frequency of FP mutants from several insect cells and

insect species suggested that transposon mutagenesis

was common. From the first description of the FP

plaques by Hink & Vail (1973), the incidence of FP

and their molecular basis have been of major interest.

Miller & Miller (1982) in their analysis of a 7.3 kb

insertion in the genome of an FP mutant (FP-DL) later

established the role of transposon-mediated mutagen-

esis in the generation of some FP mutations. For AcM-

NPV, FP mutants were detected within only one or

two passages in TN–368 cells (Hink & Strauss, 1976).

FP virus was also detected in vivo from per os (PIB)

infected T. ni

Heliothis virescens

Helisidota caryae

and

. frugiperda larvae. For these, the LD

of the

parental MP PIB was 38 times higher than for FP PIB.

MP virus could be “rescued” (from 21 to 67%) if a

predominantly FP virus preparation was passed in vivo

(

per os

)

through T. ni larvae. Host DNA insertions of

0.8 to 2.8 kb were detected in the genomes of AcMN-

PV and GmNPV FP mutants (Fraser, 1987; Fraser et

al.,

1983)

Fraser & Hink (1982) first demonstrated how fast

the passage effect can manifest itself in the generation

130

of FP virus. The FP phenotype was detected within

two passages of GmNPV in TN–368 cells or in G. mel-

lonella larvae. The proportion of FP virus was virtually

100% by the 10th passage in TN–368 cells or by 5 ser-

ial injection passages in G. mellonella larvae. GmNPV

FP virus is a stable genotypic variant which was main-

tained through at least 10 serial injection passages in

G. mellonella larvae. The MP virus always generated

both MP and FP virus whether passaged through TN–

368 cells or by injection through G. mellonella larvae.

Thus the FP virus resulted from frequent spontaneous

mutations of a single MP virus rather than the FP virus

being present in the original MP virus preparation. For

GmNPV, the FP virus was up to 350 fold less infectious

than the parental MP virus.

Genomic changes and serial passage

Wickham et al. (1991) followed the effect of serial

passage and moi on genomic changes in a recombi-

nant, AcMNPV in Sf21 cells. Virus production

decreased from 814 pfu/cell for the original isolate to

35 pfu/cell for a late passage isolate. DIP at late pas-

sage were detected by electron microscopy showing

up to 60% short (200 nm) particles. The DIP genome

had a deletion of contiguous Eco

1 A, I, J, K, N, and

O fragments which includes the gene and corre-

sponded to the region that Kool et al. (1991) reported

was missing in their DIP (Figure 1E). The DIP prepa-

ration also interfered with replication of the original

AcMNPV. When DIP-containing virus (moi of

100) was added, the amount of AcMNPV pro-

duced per cell decreased from 814 pfu/cell (without

DIP) to 40 pfu/cell. However the percent of standard

virus in the DIP preparation increased from 15 to 65%

after a single passage at a much reduced moi of 0.001.

While the production of per cell decreases due

to the DIP, this effect is less obvious at low moi. It is

unfortunate that Wickham et al. (1991) did not give

sufficient details (e.g., initial moi, frequency of pas-

saging, passage number analyzed) so that their results

could be compared to those of Kool et al. (1991) and

others.

In a more systematic study, Kool et al. (1993) fol-

lowed HMP of AcMNPV in Sf21 cells through 40

passages in culture flasks. As for studies of virus in

bioreactors (Kompier et al., 1988) most cells infected

with P40 virus no longer produced polyhedra and the

yield of NOV was about 100 fold lower than that of the

inoculum. Much of the genome was deleted in late pas-

sage DIP and seven novel and hypermolar AcMNPV-

specific Eco

1 fragments were detected within cells

infected with P40 virus and in P41 NOV preparations.

Since these fragments supported infection-dependent

DNA replication, Kool et al. (1993) suggested that

they contain cis sequences important for DNA repli-

cation and would have a selective advantage over the

replication of the larger standard virus genome.

In a more extended study, Lee & Krell (1992, 1994)

followed the generation of defective genomes during

HMP up to passage 81 (Figure 1). Like Kool et al.

(1993) the region from 1.7 to 45.0 m.u. was deleted

within 40 passages in the defective genomes. The size

of the defective genomes became smaller during pas-

sage down to an average size of about 50 kb. Up to

passage 81 even more of the genomic DNA genome

was deleted although the overall size of the genome

was not reduced any further. By passage 80 only about

2.8 kb of the standard genome (m.u. 85.0 to 87.2; Fig-

ure 1) was maintained in the defective genomes isolated

from P80 infected cells and from P81 NOV. This short

sequence was reiterated and originated from the AcM-

NPV Hin

III K region which Lee & Krell (1994) sug-

gest contains an origin of DNA replication. Because of

the smaller size and presence of multiple origins such

genomes could compete for essential trans-acting fac-

tors such as DNA polymerase and helicase and would

replicate at the expense of the standard genome.

Passage effect and bioreactors

For large scale production of foreign proteins at eco-

nomical levels cells have to be grown and infected in

bioreactors where cell growth, virus infection and for-

eign protein synthesis occur simultaneously within the

same chamber. Under a continuous system the virus

undergoes many passages over a short period of time.

Virus grown in flasks undergoes only a single replica-

tion cycle and virus is normally passaged about every

3 or more days. However, in a bioreactor where unin-

fected cells are being continuously supplied, virus is

“passaged” after each replication cycle time (e.g., 8–

24 hr). In the bioreactor setting, since the virus is being

passaged at a high frequency and at high moi the con-

sequences of the passage effect are noted within about

two weeks (Figure 2).

In one of the first systematic evaluations of the

passage effect on wild type AcMNPV in a bioreac-

tor, Kompier et al. (1988) followed the production

of a standard virus (AcMNPV-E2 in Sf21 cells) in a

131