Heterologous promoter recognition leading to high-level expression of cloned foreign genes in Bombyx mori cell lines and larvae

GENE (jene190(1997)181-189

ELSEVIER

Heterologous promoter recognition leading to high-level expression of cloned foreign genes in Bombyx mori cell lines and larvae’ Satyanarayana

Sriram, V&as B. Palhan, Karumathil P. Gopinathan

*

Microbiology and Cell Biology Department, Indian Institute of Science, Bangalore 560012, India Received 4 February

1996; revised 11 June 1996; accepted

9 July 1996; Received

by P. Roy

Abstract The baculovirus expression system using the Autographa californica nuclear polyhedrosis virus (AcNPV) has been extensively utilized for high-level expression of cloned foreign genes, driven by the strong viral promoters of polyhedrin (polh) and p10 encoding genes. A parallel system using Bombyx mori nuclear polyhedrosis virus (BmNPV) is much less exploited because the choice and variety of BmNPV-based transfer vectors are limited. Using a transient expression assay, we have demonstrated here that the heterologous promoters of the very late genes polh and p10 from AcNPV function as efficiently in BmN cells as the BmNPV promoters. The location of the cloned foreign gene with respect to the promoter sequences was critical for achieving the highest levels of expression, following the order + 35 > + 1 > - 3 > - 8 nucleotides (nt) with respect to the poIh or ~10 start codons. We have successfully generated recombinant BmNPV harboring AcNPV promoters by homeologous recombination between AcNPV-based transfer vectors and BmNPV genomic DNA. Infection of BmN cell lines with recombinant BmNPV showed a

temporal expression pattern, reaching very high levels in 60-72 h post infection. The recombinant BmNPV harboring the firefly luciferase-encoding gene under the control of AcNPV polh or p10 promoters, on infection of the silkworm larvae led to the synthesis of large quantities of luciferase. Such larvae emanated significant luminiscence instantaneously on administration of the substrate luciferin resulting in ‘glowing silkworms’. The virus-infected larvae continued to glow for several hours and revealed the most abundant distribution of virus in the fat bodies. In larval expression also, the highest levels were achieved when the reporter gene was located at + 35 nt of the polh. Keywords: Baculovirus;

protection;

Homeologous recombination; Silkworm; Transient expression

Luciferase; Nuclear polyhedrosis virus; Polyhedrin; pI0 promoter; RNase

1. Introduction Baculoviruses for expression

been

exploited

of functionally

have

active,

as efficient

vectors

post-translationally

* Corresponding author. Tel. +91 80 3092410; Fax +91 80 3341683; e-mail: [email protected] 1Presented at the International Conference on ‘Eukaryotic Expression Vector Systems: Biology and Applications,’ National Institute of Immunology, New Delhi, India, 4-8 February 1996. Abbreviations: AcNPV, Autographa californica nuclear polyhedrosis virus; B, Bombyx; BmNPV, Bombyx mori nuclear polyhedrosis virus; bp, base pair(s); DOTAP, N-[ I-(2,3-dileoyloxy)propyl]-N,N,N-trimethyl ammonium methyl sulfate; HRP, horseradish peroxidase; kb, kiIobase(s) or 1000 bp; Luc, firefly luciferase; luc, gene encoding Luc; moi, multiplicity of infection; nt, nucleotide(s); ORF, open reading frame; ~10, gene encoding PIO; PAGE, polyacrylamide-gel electrophoresis; PBS, phosphate-buffered saline (8 mM Na,HPOJl.5 M KH,POJ136 mM NaC1/4 mM KCl); pfu, plaque-forming unit(s); polh, gene encoding polyhedrin; re-, recombinant; RLU, relative luminiscence unit(s); SDS, sodium dodecyl sulfate; wt, wild type. 0378-l 119/97/$17.00 0 1997 Elsevier Science B.V. All rights reserved PII SO378-1119(96)00678-6

processed re-proteins in insect cells (Luckow, 1991; O’Reilly et al., 1992). These expression systems are based on replacement of the nonessential viral genes such as polh or p10 with a foreign gene so that it is highly expressed, driven by the strong polh or pl0 promoters (Miller, 1993). Autographa californica nuclear polyhedrosis virus (AcNPV), employed most extensively for the expression of foreign genes, is considered as the prototype baculovirus system. An alternate baculovirusbased expression system utilizes Bombyx mori nuclear polyhedrosis virus (BmNPV), with a host range virtually limited to the mulberry silkworm, B. mori (Maeda, 1989a,b; Reis et al., 1992; Ishida et al., 1994; Palhan et al., 1995; Sumathy et al., 1996). This system offers economically viable high level expression of re-proteins in larval caterpillars of silkworm avoiding the high costs of cell cultures. The virus-infected B. mori larvae also provide better yields of re-protein because of their large size compared to Spodoptera litoralis and Trichoplusiu

S. Sriram et al. / Gene 190 (1997) 181-189

182

Plasmid Construct

Promoter

Context

Relative Luc activity (“?)

1BmN cells I3 (1) pVL1392-luc

AcNPV polh

d=&

(3) pAc-luc yB%

(4) puw 1-luc X-

100

-3

15

14

AcNPVpolh

-8

6

6

AcNPV ~110

+ 1

70

78

polh

_

(B)

.G

c

100

-

X

IA

+ 35

BmNPV

(2) pBM030-Iuc

Sequencing

Sf 21 cells I

B

r

ladder G

‘I-’

tRNA

0

Fig. 1. Transient expression of luc from polh and p10 promoters. (A) To generate the plasmid constructs used in the study, the luc cDNA isolated as a 1892-bp BamHI fragment from plasmid pD0432 (Ow et al., 1986) was cloned into: (1) the BamHI site of pVL1392 (O’Reilly et al., 1992) under the control of AcNPV polh promoter, creating pVL1392-luc; (2) the BgflI site of pBm030 (Horiuchi et al., 1987) under the control of BmNPV polh promoter creating pBm030-luc; (3) the BarnHI site of pAc373 (Hasnain and Nakhai, 1990) under the control of AcNPV polh promoter creating pAc-luc; (4) the BgflI site of pUW1 @‘Reilly et al., 1992) under the control of AcNPV ~10 promoter, creating pUWl-lrrc. The promoter context refers to the site at which luc is cloned with respect to the ATG (nt + 1) of the viral polh or ~10 genes. The constructs harboring luc under the homologous or heterologous polh/plO promoters were transfected into BmN cells by lipofection (Palhan et al., 1995) followed by infection with BmNPV. For transient expression assays, 1 ug DNA of each of the above constructs (except pUWl-luc for which only 0.5 ug DNA was used to maintain the molar ratio of the transfecting plasmid because it was half the size of the other constructs) was lipofected separately into BmN cells (1 x lo6 cells/35 mm dish) using DOTAP (Boehringer-Mannheim) in serum-free TClOO medium (Gibco BRL). After 8 h, the cells were infected with BmNPV (moi = 10) and harvested at 48 h post infection. The cells were washed once with PBS and resuspended in 100 ul of Luc assay buffer (30 mM Tricine/3 mM ATPjl5 mM MgSO,/lO mM dithiothreitol, pH 7.8). The huniniscence emanated upon the addition of the substrate luciferin (final concentration 100 pm) was quantitated in a huninometer (Monolight 2010, Analytical Luminiscence Laboratory) as relative luminiscence units (RLU). Mock transfected cells were always included as controls to correct for background RLU, if any. An identical protocol was followed for transfection of Sf21 cells, followed by infection with AcNPV (moi= 10). The highest levels of Luc activity were from pVL1392luc, ranging from 5 to 7 x lo6 RLUs in both BmN or Sf9 cells. This activity was taken as 100% for comparison between the different plasmid constructs. A, the polyadenylation sites within the luc cDNA or from the polh downstream sequences as marked, B, BarnHI; Bs, BstYl; the box represents the luc ORF with orientation indicated by arrow; heavy arrowheads indicate promoters for polh orpl0. (B) Quantitation of luc transcripts by RNase protection. To monitor quantitatively the expression of luc mRNA driven by the different promoters, the transcripts arising from each of the plasmid constructs following transfection were quantitated by RNase protection assays. The total RNA was isolated by the guanidinium

S. Sriram et al. / Gene 190 (1997) 181-189

ni larvae, susceptible to AcNPV infection. However, the major handicaps of the BmNPV system have been the limited variety of transfer vectors available for optimized expression and the lower transfection efficiencies of BmN cell cultures. In the prototype AcNPV system, on the other hand, an extensive collection of transfer vectors with multiple cloning sites placed optimally under the polh promoter for highest levels of expression of the cloned foreign gene, and harboring protein secretory signals and purification tags as well, are now available (Wang et al., 1991, 1994; Belyaev and Roy, 1993; Davies et al., 1993). It is generally believed that AcNPV does not multiply in the B. mori derived BmN cells and BmNPV does not multiply in the Spodopteru derived Sf9 or Sf21 cells, although some evidences to the contrary are emerging, albeit at low efficiency (Mukherjee et al., 1995). There is considerable sequence homology between polh and p10 sequences of both the viruses (Ayres et al., 1994; Hu et al., 1994; Palhan, 1995) and therefore it is likely that the two viruses are utilizing similar truns-acting factors to regulate these promoters. Virally encoded very late gene expression factors are necessary for the promoter recognition. We wanted to examine whether the BmNPV and AcNPV very late gene (polh and ~20) promoters are recognized in Sf21 and BmN cells following infection with the corresponding viruses. Here we report that they are functional in BmN cells and in Sf21 cells to similar extents. We have also been successful in generating recombinant BmNPVs at both polh and p10 loci using the AcNPV polh- and plO-based transfer vectors.

2. Results and discussion 2.1. Comparison of luc expression from AcNPV and BmNPV very late gene promoters The BmNPV and AcNPV polyhedrin genes share 100% homology at the transcription initiation region up to 72 nt upstream from the ATG (nt + 1) and are nearly identical in the +35 nt region. These promoters share the highly conserved, very late gene motif ‘TAAGTATT’ (Rankin et al., 1988; Ooi et al., 1989). Besides they also share the ‘TTTGTA’ motif with the AcNPV pZ0 gene

183

(Zanotto et al., 1992). In a transient expression assay, we examined the functioning of AcNPV polh and ~10 promoters in BmN cells, using firefly luc as a reporter gene. Towards this goal, a series of plasmid constructs harboring luc located at different positions under the control of AcNPV polh or pI0 promoters and BmNPV polh were generated (Fig. 1A). The reporter gene constructs were transfected into BmN cells followed by infection with BmNPV to provide the trans-acting factors needed for the expression from these promoters. Very high levels of Luc activity were seen following transfection with all these constructs whether the promoters originated from BmNPV polh or AcNPV polh or p10 genes. The highest level of Luc activity (ranging from 5 to 7 x lo6 RLU) was obtained from the construct where the luc gene was placed at + 35 nt (with respect to nt + 1 of ATG) of AcNPV polh (Fig. 1A). The p10 promoter from AcNPV was also functional (70% activity of that obtained from AcNPV polh promoter) in BmN cells infected with BmNPV. The expression of luc was also compared in Sf21 cells infected with AcNPV and the levels as well as the patterns of expression were nearly identical. The luc expression levels were influenced by the promoter context, viz., the position at which the reporter gene was placed with respect to the polh start point. The relative Luc levels in both cell lines followed the order + 35 > - 3 > - 8 nt with respect to + 1 nt of polyhedrin ATG. This observation is in agreement with the previous report (Luckow and Summers, 1989) that cis elements up to 1-35 nt are required for optimal expression from polh promoter. The higher level of luc expression from AcNPV ~10 promoter than from the homologous BmNPV polh promoter in BmN cells was due to the more optimal placement of the reporter gene at + 1 nt in pUWlluc as compared to -3 nt in pBM0301uc. The luc mRNA levels following transfection in transient expression assays in BmN cells were also quantitated using an RNase protection assay. A highly transcribed copy of tRNA, G1yfrom B. mori was used as an internal control in transfections to normalize the efficiency. The tRNAy’Y copy was modified by oligotagging to give rise to a 10 nt longer transcript in order to facilitate the discrimination of the exogenous transcripts from the endogenous tRNAs (S.S. and K.P.G., unpublished results). The tRNA gene was preferred as

isothiocyanate method (Sambrook et al., 1989) from BmN cells (lipofected with plasmid DNA followed by viral infection as specified for (A) 48 h after viral infection. The luc-specific transcripts were quantitated by RNase protection assay (Sambrook et al., 1989) using “P-labeled antisense luc transcript generated in vitro by T7 RNA polymerase from an appropriate clone (a 154~nt fragment, antisense to the 5’ end of the /UC RNA) as probe for hybridization. The excess probe and the unhybridized RNA were removed by digestion with a combination of RNase Tl and RNase A, and the RNA samples were separated on a 6 M urea/8% polyacrylamide sequencing gel. The transcripts arising from an oligo-tagged copy of fRNAylY gene from B. mari included as an internal control for normalizing the transfection efficiencies were visualized by RNase protection assay using the corresponding antisense probe. For precise size determination of the transcripts, a DNA sequencing ladder was also included (lanes 5-S). The digitized computer imaging of the autoradiograms is presented. The protected fragments, luc RNA (125 nt) and tRNA (90-94 nt) are indicated by arrows. The right panel presents the quantitation of the transcripts by laser beam densitometer scanning.

S. Sriram

184

et al. 1 Gene 190 (1997)

181-189

wt BmNPV

(3.9 kb) HindlII-K* (5.7 kb)

vBml392iuc polh

123456

1234S6

23 9.4 4.6

Hind III-K” (5.7 kb)

4.4 2.3

sulr*

2.0

(2.6 kb)

W 4500

1 kDa

vBml392luc

4000 c : 3soo” A 53000:

u

wt

vBm 030

vBm 1392

he

lnc

vBm 030

u

vBm 1392

wt

vBm0301uc

Luc

&O;

Lot

P .$2000r 3 IS00 ; ‘3 looO3 WO; 0

I-

12

24

36

48

Hours Pnst Infection

60

72

Polh 27

Fig. 2. Isolation of re-vBm1392 luc. (A) To generate the re-BmNPV harboring luc under the control of AcNPV polh promoter, homeologous recombination between AcNPV polh-based transfer vector, pVL1392 luc and BmNPV genomic DNA was carried out. BmN cells (2 x lo6 cells in serum-free TClOO medium) were cotransfected with 1 ug of pVLl3921ucand 5 ug of BmNPV DNAs by lipofection. The re-viruses in the cotransfection supematants were screened by two rounds of 96-well cultures (at 10e3 and 10m4 dilutions) using Luc assay and further purified by Luc+ phenotype, and the re-virus, vBml3921uc, was isolated. two rounds of plaque assays. The individual plaques were screened for occlusion-, DNA isolated from vBml392luc-infected BmN cells (moi = 10, 72 h post infection) was digested with restriction enzymes, as indicated below, and subjected to electrophoresis on 0.8% agarose gels. The DNA was visualized by ethidium-bromide staining (see A, left panel) and in Southern blots using 32P-labeled /UC DNA fragment as probe (right panel). The HindIII-K restriction pattern and the location of the 3.9-kb K fragment harboringpolh are presented on top. Lanes: l-3, Hind111 digests of DNAs from uninfected BmN cells, BmNPV genomic DNA and vBml392luc-infected

S. S&am et al. J Gene I90 (1997) 181-189

the internal control because its transcription levels do not change during viral infection unlike most other cellular promoters which get repressed during viral infection. Maximum levels of Zuc transcripts were obtained from the pVL1392 luc followed by pUWlZuc (Fig. 1B). The faint signal of luc transcript arising from pBm0301uc in the autoradiograms become discernible only after prolonged exposure and no RNA signals could be detected in pACluc transfection. After normalizing for the tRNA ply transcripts, the Zuc mRNA levels from pVL13921uc and pUWlluc were found to be at least five- and fourfold higher than from pBm0301uc. These results agreed with the Luc activity levels monitored by the luminometer. Our observations that the heterologous AcNPV polh and ~10 promoters are functional in BmN cells infected with BmNPV implied that the viral trans-acting factors which modulate the expression from these promoters are conserved between the two baculoviruses. This opened up the possibility of utilizing the optimized array of AcNPV promoter-based transfer vectors for foreign gene expression in the BmNPV system. Likewise, the exploitation of the pZ0 promoter-based expression vectors so far not utilized in the BmNPV system could also be contemplated making use of the AcNPV pZ0 promoter. 2.2. Generation of re-BmNP V harboring the AcNP V polh promoter A major difficulty encountered in generating re-BmNPV has been the low transfection efficiency of BmN cells, We have previously standardized a lipofection procedure which provided a 50-lOO-fold improvement in transfection efficiency over the calcium phosphate method in both BmN and Sf9 or Sf21 cells (Palhan et al., 1995). From the nt sequences of flanking regions of polh and pZ0 genes of BmNPV (Palhan, 1995; S. Maeda, personal communication), a high degree of homology to the flanking regions of their counterparts from AcNPV was evident. We therefore attempted to generate a re-BmNPV by homologous recombination between the AcNPV polh flanking sequences present in the transfer vector and BmNPV genomic DNA (Fig. 2).

185

Such recombinations between homologous nt sequences from heterologous systems are referred to as homeologous recombination. The transfer vector pVL1392Zuc and the BmNPV genomic DNA were transfected into BmN cells and the cotransfection supernatants were screened for the presence of re-virus based on the appearance of occlusionphenotype as well as the expression of luc in infected cells (Fig. 2). A modified, 96-well screening procedure based on the sensitive and noninvasive Luc assay enabled the identification of low populations of re-virus. The Luc-based screening was especially useful in the early rounds of screening of recombinants and proved to be simpler and more sensitive than DNA dot blot analysis using 32P-labeled Zuc probes. After two rounds of enrichment by 96-well screening based on Luc activity followed by two rounds of plaque purification a re-virus, vBm1392luc, was isolated. In order to determine the site of recombination, the re-virus DNA was analyzed by restriction digestion (Fig. 2A). The wt viral genomic DNA fragment (HindIII-K, 3.9 kb), harboring the polh region (Maeda and Majima, 1990) was replaced by a 5.7-kb fragment containing the Zuc gene as well as the ORF 603 present in the polh 5’flanking region of AcNPV. Likewise, SalI, the 1.2-kb (wt) genomic fragment, was replaced by a 2.6-kb fragment containing the luc gene insertion. The presence of luc sequences in these fragments was confirmed from Southern hybridizations. The homeologous recombination approach would facilitate the exploitation of the varied collection of AcNPV polh-based vectors, optimized for expression and harboring protein secretory signals and purification tags currently available, for manipulating the re-BmNPV system. The expression of Zuc in vBml392luc-infected BmN cells followed a temporal pattern, detectable from 24 h and reaching very high levels around 72 h post infection (Fig. 2B). At all time points tested, Luc activity was significantly higher in vBm13921uc (harboring luc at +35 nt with respect to polh ATG) as compared to vBm0301uc (harboring luc at - 3 nt). The latter accounted for 21% of the homeologous re-virus (at 48 h post infection) consistent with the transient expression patterns seen following transfection.

BmN cells, respectively; 4, size markers, h DNA digested with HindIII; 5 and 6, Sa[I digests of BmNPV genomic DNA and vBml392luc-infected BmN DNA. (B) Temporal expression of luc from vBm13921uc. The expression pattern of luc following infection of BmN cells with purified re-vBml3921uc was examined upto 72 h post viral infection. For comparison, the re-virus vBm0301uc (harboring luc under the control of BmNPV polh and located at - 3 nt with respect to the polh start codon ATG) (Palhan et al., 1995) was included. BmN cells (1 x 106) were infected with the re-viruses (moi = 10) and harvested at various times (12, 24, 36, 48, 60 and 72 h) following infection. The Luc activity was assayed as given in Fig. 1A. (C) Western blotting of re-luciferase. BmN cells (2 x 106) were infected with the re-virus (moi= lo), and at 48 h post infection, the cell pellet was collected and lysed using 1 x SDS gel loading buffer. The total proteins (10 pg) were subjected to electrophoresis on 0.1% SDS-IO% polyacrylamide gels in duplicate sets. The proteins were visualized by silver staining (Sambrook et al., 1989) in one set. The other set was transferred to nitrocellulose membranes by electroblotting, and probed using a 1:500 dilution of rabbit polyclonal antiserum against Luc (gift from S.E. Hasnain) followed by treatment with a 1:lOOO dilution of HRP-conjugated, goat antirabbit immunoglobulin. The immunoreactive bands were visualized by staining with diaminobenzidine (Sambrook et al., 1989). Left panel, silver staining pattern of total proteins from uninfected cells (U), wt BmNPV-infected cells (wt), homologous re-vBm0301uc- (Palhan et al., 1995) and heterologous re-vBm1392luc-infected cells. Right panel, Western blots of identical samples, stained for peroxidase activity.

S. Sriram et al. 1 Gene I90 (1997) 181-189

186

(4 A

B

I

wtBmNF’V vBmUW

C

1/UC z lUC

12345

KpnI-D* (3.65 kb)

PlO

12345

23 kb 9.4 kb 6.6 kb

Kpn I-D* (3.5 kb)

2.3 kb 2.0 kb

Eco RI (12 kb) Eco HI (0.43 kb)

3

140-

vBmLW1 luc

L ._ = E 120g

NO;

; * ‘3

80:

2t ._ g

60: 40200

I 12

24

36

48

Hours post infection

60

1 12

Fig. 3. Generation of pI0 re-BmNPV. The re-BmNPV, vBmUW 1 luc containing the AcNPV pI0 promoter and flanking sequences and luc reporter gene was generated by homeologous recombination between AcNPV plO-based transfer vector pUWlluc and the BmNPV genomic DNA following cotransfection (panel a). The methodology was identical to that described in Fig. 2 legend. In this instance, the re-virus screening was based only on Luc activity since the plaques continue to be occlusion+ (due to the presence of intact polh in the re-virus). The re-vBmUWlluc was purified by two rounds of 96-well plate enrichment at 10m3 and 10e4 dilutions, followed by three rounds of plaque purification. The plaque-purified virus was scaled up by a low moi infection of BmN cells (2 x lo6 cells in 60 mm dish). The total cellular DNA was isolated 4 days post infection, digested with restriction enzymes as indicated and electrophoresed on 0.8% agarose gels. The DNA was visualized by staining with ethidium bromide (left panel) or probed using 3ZP-labeled luc fragment following Southern transfer (right panel). Lanes: 1 and 2, @I digest of wt BmNPV DNA and vBmUWlluc DNA; 3 and 4, EcoRI digest of wt BmNPV DNA and vBmUWlluc DNA; 5, size markers. The top panel presents the KpnI restriction pattern of BmNPV DNA and the 1.9-kb D fragment harboring thepl0 sequences. Temporal expression of luc in BmN cells infected with vBmUWlluc was analyzed (panel b), exactly as described under Fig. 2C for the re-vBml392luc. Luc activity in the virus-infected cells was monitored at 12, 24, 36, 48, 60 and 72 h post infection.

S. Sriram

et al. / Gene 190 (1997)

181-189

187

188

S. Sriram

et al. / Gene 190 (1997)

The proteins from re-BmNPV-infected cells showed the presence of a 62-kDa band corresponding to Luc. The immunological authenticity of the re-Luc expressed from vBml3921uc was confirmed by Western blots using polyclonal antibodies against Luc (Fig. 2C). 2.3. Generation of p10 re-BmNPVby

homeologous

recombination

We also attempted to generate a re-BmNPV harboring the AcNPVpZO promoter by a similar approach (Fig. 3). Cotransfections into BmN cells were carried out using the plasmid transfer vector pUWlluc harboring the AcNPV ~10 flanking regions (-250 nt upstream and 1 kb downstream sequences) along with BmNPV genomic DNA. Since the occlusion- phenotype selection procedure was not applicable to the p10 re-s, once again the Luc activity-based screening procedure was adopted. The re-virus, vBmUWlluc, was isolated by two rounds of 96-well plates screening followed by three rounds of plaque purification. The site of recombination was determined by restriction analysis with KpnI and EcoRI (Fig. 3). The wt BmNPV DNA, Kpnl-D fragment (1.9 kb) harboring the pZ0 region increased by 1.6 kb due to the insertion of Zuc gene replacing the pZ0 sequences. The homeologous recombination at the p10 region also results in the deletion of an internal BamHI site at the C-terminal end of the ~7.5 ORF of BmNPV pl0 (Hu et al., 1994; Palhan, 1995). Correspondingly, the 8.9-kb BamHI fragment from wt BmNPV DNA carrying the pI0 region disappeared to give rise to a much higher molecular weight DNA fragment comigrating with the not wellresolved, BamHI-A,B,C fragments as evidenced in Southern blots (data not shown). The near 80% homology between AcNPV and BmNPV at the ~10 region (V.B.P. and K.P.G., unpublished results) was sufficient to drive homeologous recombination between the viral sequences at this locus. The expression of luc in BmN cells following vBmUWlZuc infection (luc placed under the control of AcNPVplO promoter) also followed a temporal pattern, the enzyme activities appearing from 24 h and reaching the peak around 60 h post infection. The maximal levels of expression, however, corresponded to only lo-15% of that obtained from polh promoter. Nevertheless, foreign gene expression driven by p10 promoter has an advantage over polh promoter in that the former expresses maximally 12-24 h earlier, providing greater scope for post-translational modifications, secretion and proper folding of the re-protein, before completion of cell lysis. 2.4. Luc synthesis from re-BmNPVs in B. mori larvae A major advantage of the re-BmNPV over the AcNPV system is the possible utilization of the B. mori larvae

181-189

in place of cell lines for large scale production of biomolecules. The large size of B. mori larvae, coupled with the ease in their commercial scale rearing processes make re-protein production through virus-infected caterpillars an economically more viable option than cell lines in culture. We examined the synthesis of Luc in B. mori larvae following infection with luc re-BmNPVs. The infection of larvae with either vBml3921uc or vBmUWlluc resulted in high levels of Luc expression. On administration of the substrate luciferin, these re-virus-infected larvae emanated significant levels of luminiscence (Fig. 4A and B). Clearly, the extent of ‘glow’ emanated was much higher in larvae infected with vBml392Zuc (larvae 4 and 5, extreme right) as compared to vBmUWlZuc (larvae 2 and 3). On dissection of the re-virus-infected larvae, luc expression was observed mainly in the fat bodies but not in the alimentary canal, silk glands or trachea (Fig. 4C and D) implicating that the fat bodies are the main target of BmNPV infection when the virus is directly injected into the hemolymph. This is in contrast to the infection of the gut when the viral entry is through the oral route by ingestion of polyhedra. The glow emanated from such virus-infected larvae is so high that the path of virus infection and the distribution of virus in the tissues could be traced readily using the re-luc viruses.

3. Conclusions ( 1) The very late baculoviral promoters (polh and ~10) from both AcNPV and BmNPV are functional in Sf21 as well as BmN cells to similar extents. (2) The levels of expression from the strong polh and p10 promoters depended on the location of the reporter gene with respect to the + 1 ATG codon of either genes, and followed the order +35>+1>-3>-8nt. (3) Homeologous recombination (see Section 2.3) between polh or p10 flanking regions of AcNPV and BmNPV was achieved resulting in the generation of re-BmNPVs harboring AcNPV promoter/ flanking sequences. (4) The homeologous recombination approach facilitates direct exploitation of the currently available, varied collection of second generation AcNPV polhbased transfer vectors without further modifications in the BmNPV system. These vectors have been already optimized for high levels of expression and secretion of the synthesized recombinant proteins, and are tagged for easy protein purification. The AcNPV plO-based vectors can also be used in the BmNPV system.

S. Sriram et al. / Gene 190 (1997) 181-189

(5) The BmNPV system offers the additional advantage (6)

of economic, large scale production of biomolecules through larval caterpillars. The re-virus harboring the Zuc reporter gene generated here can also serve in pathological investigations on viral infection process.

Acknowledgement

We thank Dr. Seyed Hasnain for plasmids pD0432 and pAclue, and the Luc antibody, S. Maeda for plasmids pBm030 and pBE283, and M.D. Summers for plasmid pVL1392. S.S. is supported by a research fellowship from the Council of Scientific and Industrial Research, India. This research was supported by grants from the Department of Biotechnology, Govt. of India, and by a collaborative scientific project (CII*CT94-0092) under Indo-European Economic Community cooperation.

References Ayres, M.D., Howard, SC., Kuzio, J., Lopez-Ferber, M. and Possee, R.D. (1994) The complete sequence of Autographa californica nuclear polyhedrosis virus. Virology 202, 586-605. Belyaev, AS. and Roy, P. (1993) Development of baculovirus triple and quadruple expression vectors: co-expression of three or four bluetongue virus proteins and the synthesis of bluetongue virus-like particles in insect cells. Nucleic Acids Res. 21, 1219-1223. Davies, A.H., Jowett, J.B.M. and Jones, I.M. (1993) Recombinant baculovirus vectors expressing glutathione-4transferase fusion proteins. Bio/Technology 11, 933-936. Hasnain, S.E. and Nakhai, B. (1990) Expression of the gene encoding firefly luciferase in insect cells using a baculovirus vector. Gene 91, 135-138. Horiuchi, T., Marumoto, Y., Seiki, Y., Sato, Y., Furusawa, M., Kondo, A. and Maeda, S. (1987) High level expression of the human-cc-interferon gene through the use of an improved baculovirus vector in the silkworm, Bombyx mori. Agric. Biol. Chem. 51, 1573-1580. Hu, N.-T., Lu, Y.-F., Hashimoto, Y., Maeda, S. and Hou, R.F. ( 1994) The p10 gene of natural isolates of Bombyx mori nuclear polyhedrosis virus encodes a truncated protein with an M, of 7700. J. Gen. Virol. 75, 2085-2088. Ishida, N., Tsujimoto, M., Kanaya, T., Shimamura, A., Tsuruoka, N., Kodama, S., Katayama, T., Oikawa, S., Matsui, M., Nakanishi, T., Kobayashi, J. and Nakazato, H. (1994) Expression and characterization of human bone morphogenetic protein-2 in silkworm larvae infected with recombinant Bombyx mori nuclear polyhedrosis virus. J. Biochem. 115, 279-285. Luckow, V.E. (1991) Cloning and expression of heterologous genes in insect cells with baculovirus vectors. In: Prokop, A., Bajpai, R.K.

189

and Ho, C. (Eds.), Recombinant DNA Technology and Applications. McGraw Hill, New York, NY, pp. 97-152. Luckow, V.E. and Summers, M.D. (1989) High level expression of nonfused foreign genes with AcNPV expression vectors. Virology 170, 31-39. Maeda, S. (1989a) Expression of foreign genes in insects using baculovirus expression vectors. Annu. Rev. Entomol. 34. 351-372. Bomhyx Maeda, S. (1989b) Gene transfer vectors of a baculovirus, mori nuclear polyhedrosis virus, and their use for expression of foreign genes in insect cells. In: Mitsuhashi, J. (Ed.), Invertebrate Cell Systems and Applications, Vol. I. CRC Press, Boca Raton, FL. pp. 167-182. Maeda, S. and Majima, K. (1990) Molecular cloning and physical mapping of the genome of Bombyx mori nuclear polyhedrosis virus. J. Gen. Virol. 71, 1851-1855. Miller, L.K. (1993) Baculoviruses: high level expression in insect cells. Curr. Opin. Genet. Dev. 3, 977101. Mukherjee, B., Burma, S., Talwar, G.P. and Hasnain, S.E. (1995) Transcriptional regulation of cell line-dependent, baculovirus-mediated expression of foreign genes. DNA Cell Biol. 14, 7-14. Ooi, B.G., Rankin, C. and Miller, L.K. (1989) Downstream sequences augment transcription from the essential initiation site of a baculovirus polyhedrin gene. J. Mol. Biol. 210, 721-736. O’Reilly, D.R., Miller, L.K. and Luckow, V.E. (1992) Baculovirus Expression Vectors. A Laboratory Manual. W.H. Freeman. New York, NY. Ow, D.W., Wood, K.V., DeLuca, M., de Wet, J.R., Helinski, D.R. and Howell, S. (1986) Transient and stable expression of firefly luciferase gene in plant cells and transgenic plants. Science 234, 856-859. Palhan, V.B. (1995) Bombyx mori Nuclear Polyhedrosis Virus - Molecular Biology and Biotechnological Applications, Ph.D. Thesis, Indian Institute of Science, Bangalore, India. Palhan, V.B., Sumathy, S. and Gopinathan, K.P. (1995) Baculovirus mediated high level expression of luciferase in silkworm cells and larvae. BioTechniques 19, 97-104. Rankin, C., Ooi, B.G. and Miller, L.K. ( 1988) Eight base pairs encompassing the transcriptional start point are the major determinant for baculovirus polyhedrin gene expression. Gene 70, 39-49. Reis, U., Blum, B., von Specht, B.-U., Domdey, H. and Collins, J. (1992) Antibody production in silkworm cells and silkworm larvae infected with a dual recombinant Bomb_vx mori nuclear polyhedrosis virus. Bio/Technology 10, 9 10-9 12. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sumathy, S., Palhan, V.B. and Gopinathan, K.P. (1996) Expression of human growth hormone in silkworm larvae through recombinant Bombyx mori nuclear polyhedrosis virus. Protein Expr. Purif. 7, 262-268. Wang, X., Ooi, B.C. and Miller, L.K. (1991) Baculovirus vectors for multiple gene expression and for occluded virus production. Gene 100, 131-137. Wang, M.-Y., Bentley, W.E. and Vakharia, V. ( 1994) Puritication of a recombinant protein produced in a baculovirus expression system by immobilized metal affinity chromatography. Biotechnol. Bioeng. 43, 3499356. Zanotto, P.M. de A., Sampaio, M.J.A., Johnson, D.W., Rocha, T.L. and Maruniak, J.E. (1992) The Anticarsia gemmatalis nuclear polyhedrosis virus polyhedrin gene region sequence analysis-gene product and structural comparisons. J. Virol. 73. 104991056.