CA1325610C - Recombinant baculovirus occlusion bodies in vaccines and biological insecticides - Google Patents

Abstract

ABSTRACT The present invention is directed to recombinant baculoviruses which encode fusion polyhedron proteins capable of forming occlusion bodies containing foreign peptides. The recombinant baculoviruses of the invention are formed by insertion into or replacement of regions of the polyhedrin gene that are not essential for occlusion body formation, with foreign DNA fragments by recombinant DNA techniques. The recombinant occlusion bodies produced in accordance with the present invention have uses in vaccine formulations, immunoassays, immobilized enzyme reactions, as biological insecticides, and as expression vectors.

Description

132~610 ~ECOM~INAN~ BACVLOVLRUS OCCLUSION BODIES
IN VACCINES AND BIOLOGIC_L INSECTICIDES

.
Paqe 1. Introduction................................... 6

2. Background o~ the Invenkion.................... 6 2.1 Insect Viruses and Occlus~on Bodies....
2.2 Polyhedrin................................. 8 - 2.3. Recombinant DNA Techniques and Baculovirus .......................... 9 2.4 Vaccines for Viral Infection6.............. 12 2.5 Vaccines for Parasitic and Bacterial Infection O~ 14

3. Summary of the Inven~ion....................... 14 3.1 Definition~................ O............... 15

4. Desaription of the Figures..................... 18

5. Detailed Description of the Invention.......... 23 5.1 The Generation of Recombinant Occlusion Bodies Containing Antigenic Determinants o~ Heterologous Proteins..... 24 5.1.1. Identification of Modi~iable Domains Encoded by the Polyhedrin Gene................... 24 5Ol.1.1. Hydrophilicity Analysis of Polyhedrin.............. 25 5.1.1.2. Se~uence Comparisons of Different Polyhedrin Genes...................... 27 5.1.1.3. Se~uence Analysis of Polyhedrin Genes Encoding Mutant or Truncated Polyhedrin Proteins... O.............................................. 29 5.1.1.4. Structural Analysis of Polyhedrin Amino ~cid Sequences................................................. 31 . . .

-2- ~32~

5.1Ol.5. ~reparatlon and Charac-terlzati~n ~ Monoclonal Antibodies to Occlusion Bodies....................... 33 5.1.1.5.~.
Tdent~ cation of Epitopes Recogniæed by Anti-Polyhedrin Monoclonal Antibodies.. 35 5.1.2. Identificati4n and Characteriza-tion of Immunodominant Peptides for Expression on or ~ithin Recom-binant Occlusion Bodies............. 38 5.1~3. Construction of Recombinant Polyhedrin Genes................... 39 5.1.4. Selection o~ Reco~binant Occlusion Bodie~.................... 45 5.1.5. Verification o~ Expression o~ -~oreign ~pitopes on or Within the Recombinant Occlusion Body.......... 47 5.2. Vector/Host Systems.... O~........................... 48 5.2.1. Hosts used in the Vector/Host System~.. ~.... ~.......................................... 53 5.2.1.1. In~ect Cell Lines............................... 53 5.2.1.2. Larva Hos~s..................................... 58 : : 25 5.2.2. Expression ln Other Microorganisms.. O........................................ 61 5.3. Detexmination of th~ Immunopotency of Foreign Epitopes Expressed on or Within Recombinant OccIusion Bodies........................ 63 5.4. Uses of Recombinant Occlusion Bodies................ 65 5-4-1. Vaccine O~ 65 5.4.1.1. Uses of Antibodies Generated by Immunization with Reco~binant Ocalusion Bodies....... O........................................... 68 - - ., -: --3- i~2~6~ ~

5.4.2. Bi~logical Insecticides............. 69 5.4.3. Express~on Vectors... ,.............. 71 5.4.4. Immunoassays........................ 7 5.4.5. Immobilized Enæymes............. ~ 73 5 6. Example: Construction of Trans~er Vectors Used for Introduc~ng Foreign Gene Sequences Into the Heliothis Poly-h~drin Gene to Produce ~z Recombinants..O... 73

6.1, Materials and Methods.................. 73 6.1.1. Restriction Mapping............. 73 5.1.2. Southern Blotting............... 74 6.1.3. DNA Seguencing.................. 74 6.2. Identification and 5equencing of the Polyhedrin Gene of Heliothis zea Virus.......... O.................. t ~ 76 6.3. Con~truction of Transfer Vectors........... 79 6.3.1. Parent Plasmids: pHH5 and pHX12.......................... 79 6.3.2. Construc~ion o~ Transfer vectors............................ 80 6.3.3. Trans~er Vectors Expressing Beta-Galactosidase................. ~1 6.3.4 Generation of Deletions of Heliothis Polyhedrin ~mino-Terminal Sequences................. 82

7. Example: Hellothi5 Viruse For Use In Generation of Recombinant Occlusion Bodies...... 83 7.1. Materials and ~ethods....... .O............. B4 7.1.1. In itro Propagation of HZSNPV............................. 84 7.1.2. Plaque Purification o~
HzSNPV Isolates.................... 85 7.1.3. Larval Propagation of Virus..,. 85 7.1.4. Isolation o~ Virions From Occlusion Bodies.... ~.............. 85 7.1.5. Isolation o~ Viral DNA.............. ~6 7.1.6. Restriction Endonuclease .

' . ~ " - : , -4- ~2~6~

Digestions.,....................... 86 7.l.7. SDS-Polyacrylamide Gel Electxophoresis.,.,................ 87 7.2. Character.ization o~ HæSNPV................ 87 7.2.l. In Vitro Propagatio~ and Plaque Puriflcation................ 8 7 7.2.2. Larval Infections with Occlusion Bodies................... ~8 7~2.3. Re triction Enzyme Dige~tion Patterns of Viral DNAs............. 92 7.2.4. Comparison o~ Virion 5tructural Pro~eins................ 95

8. Example: Cell Line ~o~ts For U~e In Generation of Reco~binant Occlusion Bodies.. 96 8.l. Materials and ~ethods.~................... ~7 8.l.l. Cloning o~ C~ll Strains............. 97 8.l.2. Cell Growth Curveæ.................. 97 8.l.3. Quantitation of Polyhedra and In~ectious Extrac~llular Virus Production................... 98 8.l.4. Isozym~ Analysis o~ Cell Isolate~....... ~................... 93 8.2. Characterization of the Cell Lines......... 99 8.2.l. Cell ~orphology...... ~............. 99 8.2.2. Cell Growth curYes........O........ l00 8.2.3. Susceptibility to HzSNPV............ 102 8.2.4. I~ozyme Analy~is of Cell Strain~ and Cell ~ines....... ,...... 102

9. Example: Larval Hosts or U~e in Generation of Recombinant Occlu~ion Bodies...... ~.... ~...... 104 9.l. Insect Diet Preparation........ O.............. ~...... 104 9.2. Colony Maintenance......................... 105 9.2.l. Rearing o~ T. ni or H. zea~.... l0 9.2.2. Rearing o G. melonella............. 107 9.3. Germ Free Colonie6............. ~ l07 lO. Example: Heliothis Polyhedrin Gene and Promoter in ~utographa Shuttl~ Vector... 108 , ' . ' ' ~5~ ~32~

10.1. Autographa ShuttlQ Vectors Encoding an Epitope of the Influenza Hemagglutinin within the Polyhedrin Gene.. ,... ~.......... ......~........ 1105 11. Example: Production of R~combinant Occlusion Bodies Exposing An Epitope of Influenza Hemagglutinin...... ~.............. 111

11.1. Construction of Shuttle Vectors.......... 112 11.2. Preparation of Recombinant Viruses... 116 11.3. Immunological Analyses of the Recombinant Occlusion Bodies............ 118 11.3.1. ELISA Analysis of Sur~ace Expresslon of Undenatured Influenza Epitope on Recom-binant O~clusion Bodies.... ~ 118 11.3.2. W~stern ~lot Analysis of Denatured Recombinant Occlusion Bodies..... ~......... 119 11.3.3. Immunoprecipitation Assays of Recombinant Influenza~
Polyhedrin Crystals.. ....~..... 119 11.3.4. I~munogenicity o~ R~com-binant Occlusion Bodies......... 121

12. Depo~it of Microorganisms......... ~............. 123 - - -:

, , :
. . . - ~ .:

-6~

1. INTRODUCTION
The present invention is directed to baculoviruses which encode recombinant polyhedrin proteins capable of forming occlusion bodies containing foreign peptides. The 5 recombinant baculoviruses of the invention are formed by replacing regions of the polyhedrin gene that are not essential for occlusion body formation with foreign DNA
fragments by recombinant DNA techniques. The recombinant occlusion bodies produced in accordance with the present 10 invention can be particularly useful as vaccines, biological insecticides, and expression vectors. The invention is demonstrated by way of examples in which recombinant baculoviruses ware engineered to express recombinant occlusion bodies that present an influenza hemagglutinin epitope. These 15 recomhinant occlusion bodies immunoreact with antibodies that define the authentic influenza hemagglutin epitope.

2. BACKGROUND OF THE INVENTION

2.1. INSECT VIRU5ES AND OCCLUSION BODIES
Baculoviruses are a group o~ viruses which are pathogenic for insects and some crustaceans. The virions of these viruses contain rod-shaped nucleocapsids enclosed by a lipoprotein membrane. Two ~orphologically distinct forms of 25 baculovirus are produced by infected cells: the nonoccluded virus and occluded virus. The nonoccluded virus is synthesized early after infection; nucleocapsids are assembled in the nucleus and acquire an envelope by budding through the plasma membrane to become extracellular virus. In occluded 30baculoviruses, the virions are embedded in the nucleus in large protein crystals, termed occlusion bodies.
Baculoviruses are members of the family Baculoviridae and the genus Baculovirus. This genus is composed of three subgroups of viruses: the nuclear polyhedrosis viruses (NPV), 35the granulosis viruses ~GV), and the non-occluded viruses.

-7~ 132~

NPV have ~cclusion bodies, termed polyhedra, which are polyhedral to cuboidal in shape, and 1-15 um in diameter. The lipoprotein membranes con~ain either single nucleocapsids (SNPV) or multiple (up to 39) nucleocapsids (MNPV) per 5 envelope. Up to 100 virions can be embedded in a single occlusion body ~Vlak, J. M. and Rohrmann, G. F., 1985, The Nature of Polyhedrin. In Viral Insecticides for Biological Control, Academic Press, pp. 489-542). Examples of this group of viruses include ~utographa californica NPV (AcNPV), 10 Heliothis æea NPV (HzNPV~, and Bombyx mori (B~NPV).
Comparison of DNA sequences of total viral genomes reveals a less than 2% homology between HzSNPV and AcMNPV, whereas a comparison among various MNPVs shows a greater degree of homology (Smith, G.E. and Summers, M.D., 1982, Virol.
15 123:393-406). HzSNPV is currently produced and sold in the United States for use as an insecticide under the trade name Elcar~.
The granulosis viruses have round to ellipsoidal occlusion bodies, termed granula, which are 0.1-1 um in size.
20 Each occlusion body contains one singly-enveloped nucleocapsid (Vlak, J. M. and Rohrmann, G. F., ~ ). (For review, see Tweeten, K.A., et al., 1981, Microbiol. Rev. 45O379-408~.
Baculoviruses contain double-stranded, circular D~A
molecules, which range from ~0-110 x 106 daltons. The 25 prototype of the Baculoviridae family is AcNPV, which has a genome of approximately 82-88 x 106 daltons (Miller, L. K., 1981, A Virus Vector for Genetic Engineering in Invertebrates.
In Genetic Engineering in the Plant Sciences. Praeger Publishers, New York, pp. 203-224). AcNPV replicates in the 30 nucleus of infected insect cells. Two forms o~ virus are produced as a result of wild-type AcNPV infection, occluded and non-occluded virions.
The apparent role of the occlusion body in the virus life cycle is to provide stability outside the host insect by 35protecting the virus from inactivating environmental factors.
Ingested occlusion bodies dissolve in the alkaline environment of the midgut, rel~asing virus particles for another round of , , - : , . . .

-8- ~ 3 2 ~

in~ection, late after viral replication. The occlusion body of NPV consists predominantly of a single, approximately 29,000 dalton molecular weight polypeptide, known as polyhedrin (Vlak, J. M. and Rohrmann, G. F., suPra). This 5 protein forms the paracrystalline lattice around the virions, and is present as a multimer. Polyhedrin is produced in enormous amounts during the course of viral infection, late after viral replication. As there is no evidence of gene amplification (T]ia, S. T., et al., 1979, Virology 99: 399-10 409)~ it is probable that the polyhedrin promoter is anextremely efficient one.

2.2. POLYHEDRIN
The occlusion body (OB) exists as a multimer of ~he approximately 30 kilodalton polyhedrin polypeptide which forms a paracrystalline lattice around the viral particle (Tinsley, T.W. and Harrap, K.A., 1978, Comprehensive Virology, Vol. 12, Fraenkel-Conrat, H. and R. Wagner (eds.), Plenum Press, New York, pp. 1-lOl). After alkali dissolutio~ of OBs, a 20 polyhedrin particle with a sedimentatîon coefficient of llS-13S (200-374 kilodaltons) can be isolated (Bergold, G.H. and Schramm, G., 1942, BiolO ZentralblattO 62:105; Bergold, G.H., 1947, Zeitschr. Naturforsch. 2b~122; Bergold, G~H. t 1948, Zeitschr. Naturforsch. 3b:338; Harrap, K.A., 1972, Virology 25 50:124; Eppstein, D.A. and Thoma, J.A., 1977~ Biochem. J.
1~7:321; Rohrmann, G.F., 1977, Biochem. 16 1631)o X-ray di~fraction studies determined that polyhedrin is crystallized in a body-centered cubic lattice (Engstrom, A., lg74, Biochem.
Exp. Biol. 11:7). Electron microscopic analysis of polyhedrin 30 crystals suggests the arrangement of subunits is consistent with six armed nodal units (Harrap, K.A., 1972, Virology 50:124). Crosslinking analysi of polyhedrin utilizing dimethyl suberimidate indicates a dodecameric structure.
There~or~, each arm of the nodal unit is composed of two 35subunits (Scharnhorst, D.W. and Weaver~ R.F., 1980, Virology 102:468). Alkali solubility of the crystal suggests that salt bridges are formed between the amino acid side chainsO This ' ~2~
indicates that the paracrystalline lattice is maintai~ed by noncovalent, ionic inter~olecular associations o~ the individual monomers~ Disulfide bond formation may also influence the quaternary structure of the multimeric form.
Baculovirus occlusion body protein has been termed polyhedrin for NPVs and granulin for GV~. .However, recent studies have shown that polyhedrins and granulins all belong to one group of related pro~eins (Rohrmann, G.F., et al., 1981, J. Mol. Evol. 17:329; Smith, G.E. and Summers, M.D., 10 1981, J. Virol. 39:125). Tryptic peptide analyses have shown that polyhedrins from MNPVs, SNPVs, and GVs have many common ~ragments (Summers, M.D. and Smi~h, G.E., 1975, Intervirology 6:168-180; Maiuniak, J.E. and Summers, M.D.,1978, J.

Invertebr. Pathol. 32-196). Such similarities in sequence 15 have been reported (Vlak, J.~. and Rohrmann,.G.F., 19~5, The Nature of Polyhedrin. In Viral Insecticides for Biological Control. Acad~mic Press, pp. 489-542.). Some polyhedrins have been found to be more closely related to granulins than to other polyhedrins (Rohrmann, ~.~., et al., su~ra). Thus, 20 hereinafter, the term polyhedrin will be used to re~er to the entire group of related proteins.
Comparison of the amino acid sequences of six lepidopteran NPV polyhedrins (Vlak, J.M. and Rohrmann, G.F., supra, pp. 506-508) reveals that 80-90% of amino acids are 25 conserved within these proteins. There are several regions which Can be distinguished on the basis of seguence conserVation. For example, amino acids 15-26 and 58-86 are highly conserved. The region between amino acids 38-55 is hydrophilic and highly variable. Other variable sites include 30 the N-terminal region, amino acids 120-127, 1~5-148, 165, 195, and 216 (Vlak, J.M. and Rohrmann, G.F., supra).

2.3 RECOMBINANT DNA TECHNIQUES AND BACULOVIRUS
The use of reco~binant DNA technology ~or the 35production of proteins involves the molecular cloning and expression in an appropriate vector of the genetic information encoding the desired proteins. Baculoviruses are useful as , ,: ' ~ , . ' .
- ~

, , .
" ~ , -lo- ~3256~
recombinant DNA vector systems since they are double-stranded DNA replicating units, into which can be inserted a large amount o~ foreiqn DNA (20 megadaltons or more), and which provide at least one strong pro~oter (polyhedrin) which 5 controls a gene with nonessential function for propagation in cell culture, which is available for replacement or insertion into by foreign DNA ~Miller, L. X., 1981, A Virus Vector for Genetic Engineering in Invertebrates, In Genetic Engineering in ~he Plant Sciences. Praeg~r Publishers, New York, pp.
10 203-224; Vlak, J. M. and Rohrmann, G. F., 1985, The Nature of Polyhedrin, In Viral Insecticides for Biological Control.
Academic Press, pp. 489-542). A method for the production of recombinant proteins using a baculovirus system has been described (Pennock et al., 1984, ~ol. Cell. Biol. 4:399; Smith et al., 1983, J. Virol. 46:584). Baculoviru~ vectors are constructed, which expre s foreign DNA which has been inserted into the viral genome. Upon introduction into an appropriate host~ the foreign protein is produced.
The expression of foreign DNA in recombinant 20 baculoviruses requires the ligation of baculovirus sequences to a DNA sequence encoding a foreign protein so that the protein-coding sequences are under the con~rol of a promoter.
Plasmid vectors, also called insertion vectors, ha~e been constructed to insert chimeric genes into AcNPV. One example of such an insertion vector is composed o~: (a~ an AcNPV
promoter with the transcriptional initiation site; (b~ several unique restriction endonuclease recognition sites located downstream from the transcriptional start site, which can be used for the insertion o~ foreign DNA fragments: (c) AcNPV DNA
30 sequences (such as the polyhedrin gene~, which flank the promoter and cloning sites, and which direct insertion of the chimeric gene into the homologous nonessential region of the virus ~enome and (d) a bacterial origin of replication and antibiotic resistance marker ~or replication and selection in 35E. coli. Examples of such vectors are described by ~iyamota et al. (1985, Mol. ~ell. Biol. 5:2860).

3 ~

Recombinant baculoviruses have been produced by cotran~fection o~ cells with recomb~nant bacterial plasmids cont~ining the foreign gene, together with baculovirus DN~.
The foreign gene is inserted in~o or replaces the nonessential 5 polyhedrin gene of the viral genome throuqh homologous recombination within the infect~d cell. The resulting recombinant plaques can be screened visually for lack of occlusion bodies resulting from the loss o~ the functional polyhedrin gene. The infected cells can also be screened 10 using immunological techniques, DNA plaque hybridization, or genetic selection for recombinant viruses which subæeguently can be isolated. These baculovirus recombinants retain their essential functions and infectivity.
Fnreign gene expression can be detected by enzymatic or 16 immunological assays (for example, immunoprecipitation, radioimmunoassay, or immunoblotting). High expression levels can be obtained by using strong promoters or by cloning multiple copies of a single gene.
Several foreign proteins have been successfully 20 expressed under control of the polyhedrin promoter in occlusion body-negative baculovirus systems. Human interleukin 2 (Smith et al., 1985, Proc. Natl. Acad. Sci.
U.S.A. 82: 8404-8408), human c-~y~ (Miyamoto et al., 1985, Mol. Cell. ~iol. 502860-2865), bacterial beta-galactosidase 25 (Pennock et al., 1984, Mol. Cell. Biol. 4~3~9-406), in~luenza virus haemagqlutinin (Kuroda et al., 1986, EMB0 5: 1359-1365), and human beta-interferon (Smith et al., 1983, Mol. Cell.
Biol. 3:2156-2165) have all been expressed in insect cells under the control of the polyhedrin promoter in recombinant 30 AcNPV expression vectors. Human alpha-interferon has been expr~ssed in sllkworms by ligation to the polyhedrin promoter of BmNPV (Maeda et al., 1985, Nature (London) 315: 592-594).
Smith and Summers (European P~tent ~pplication Publication No.
0 127 839, 12-12-84) propose a method for producing 35recombinant baculovirus expression vectors, and report the use . , ~ ~ ; .

, -12- ~32~

of recombinant ~cNPV vectors to express human beta-interferon and human interleuXin 2, under the control of the polyhedrin pr~moter.

2.4. VACCINES FOR VIR~L INFECTIONS
A number of methods are currently in use for the prevention and treatment of viral infections. These include vaccines which elicit an active immune response, treatment with chemotherapeutic agents and inter~eron treatment.
Traditional ways of preparing vaccines include the use o~ inactivated or attenuated viruses. Inactivation of the virus renders it harmless as a biological agent but does not destxoy its immunogenicity. Injection of these ~killedN virus particles into a hos~ will then elicit an immune response 15 capable of neutralizing a future infeckion with a live virus.
However, a major concern in the use of killed vaccines (using inactivated virus) is failure to`inactivate all the virus particles. Even when this is accomplished, since killed viruses do not multiply in their host, the immunity achieved 20 i5 often short lived and additional immunizations are usually required. Finally, the inactivation process may alter the viral proteins rendering them less e~fective as immunogens.
Attenuation refers to the production of virus strains which have essentially lost their disease producing ability.
25 One way to accomplish this is to subject the virus to unusual growth conditions and/or ~requent passage in cell culture.
Viral mutan~s are then selected which have lost virulance but yet are capable of eliciting an immune response. The attenuated viruses generally make good immunogens as they 30 actually replicate in the host cell and elicit long lasting immunity. However, several problems are encountered with the use of live vaccines, the most worrisome being insuf~icient attenuation.
An alternative to the above methods is the use of 35 subunit vaccines. This involves immuniza~ion only with those proteins which contain the relevant immunological material.
For many enveloped viruses, the virally encoded glycoprotein ' , .. ~ .:
,

-13- ~32~

contains those epitopes which are capable of eliciting neutralizing antibodies; these include the glycoproteins of La crosse Virus (Gonzalez-Scarano, F., Shope, R. E., Calisher, C.
E., and Nathanson, N., 1982, VirolOgy 120:42), Neonakal Calf 5 Diarrhea Virus (Matsuno, S~ and Inouye, S., 1983, Infection and Immunity 39-155), Venezuelan Equine Encephalomyelitis Vir~ls (Mathews, J. H. and Roehrig, J. T., 1982, J. Imm.
129:2763), Punta Toro Virus ~Dalrymple, J. M., Peters, C. J., Smith, J. F., and Gentry, ~ K., 1981, In Replication of 10 Negative Strand Viruses, D. H. L. Bishop and R. W. Compans, eds., p. 167. Else~ier, New York), Murine Leukemia Virus (Steeves, R. A., Strand, M., and August, J~ T., 1974, J.
Virol. 14:187), and Mouse Mammary Tumor Virus tMassey, R. J.
and Schochetman, G., 1981, Virology 115:~0). One advantage of subunit vaccines is that the irrelevant viral material is excluded.
Vaccines are oPten administered in conjunction with various adjuvants. The adjuvants aid in attaining a more durable and higher level of immunity using smaller amounts of 20 antigen in fewer doses than if the immunogen were administered alone. The mechanism of adjuvant action is complex and not completely understood. However, it may involve the stimulation of phagocytosis and other activities of the reticuloendothelial system as well as a delayed release and 25 degradation of the antigen. Examples of adjuvants include Freund's adjuvant (complete or incomplete), Adjuvant 65 (containing peanut oil, mannide monooleate and aluminum monostearate), the pluronic polyol ~-121, Avridine, and mineral gels such as aluminum hydroxide, aluminum phosphate, 30 or alum. Freund's adjuvant is no longer usPd in vaccine formulations for humans because it contains nonmetabolizable mineral oil and is a potential carcino~en.

.

.:
:
.-.-..

3 2 ~

2 . 5 . VACCINES FOR PARASITIC AND
BACTERIAL INF~CTIONS _ The de~elopment of vaccines for the prevention of parasitic or bacterial dis~ases is the focus of much research 5 effort. Vaccines are presently avai.lable for diphtheria, pertussis, and tetanus (Warren, K. S., 1985, In Vaccines85, Lerner, R. A., R. M. Chanockl and F. Brown (eds.3, Cold Spring Harbor Laboratory, New York, pp. 373-37~). In addition, a vaccine consisting of the polysaccharide capsule of Hemophilus 10 influenzae was recently licensed, although it is ineffective in pre~enting disease in certain su~groups of the population (Granoff, D. M. and Munson, R. S., Jr., 1986, J. Infect. Dis.
153:448-461). No vaccines currently exist ~or any of the many protozoan infections such as malaria or helminth infections such as schistosomiasis and ascariasis. The protective effects of anti era directed against epitopes of Escherichia coli toxins, cholera toxins, gonococcal pili, and malaria surface antigens (Vaccines85, 1985, Lerner, R. A., R. M.
Chanock, and F. Brown (eds.), Cold Spring Harbor Laboratory, 20 New York; Modern Approaches to Vaccines, 1~84, Chanock, R. ~., and R. A. Lerner (eds~), Cold Spring Harbor Laboratory, New York) are among the many systems presently under investigation.

3. SUMMARY OF THE INVENTION
The present invention is directed to recombinant baculoviruses whi~h encode ~usion polyhedrin proteins capable of forming occlusion bodies containing foreign peptides. The recombinant baculoviruses of the invention are formed by 30 insertion into or replacement of regions of the polyhedrin gene that are not essential for occlusion body formation with foreign DNA frayments by recombinant DNA techniques. The present invention also relates to vector/host systems which can direct the expression of the recombinant polyhedrin genes 35in different hosts, including but not limited to, cultured cells, larvae, or microorgani3ms.

,, :

; ~ . :
' ,' ~

-15- ~ 32~6~

The recombinant occlusion bodies ~OBs) of the present invention comprise crystallized polyhedrin fusion proteins which bear the heterologous gene product on the ~urface of or within the occlusi~n body. Where the heterolvgous gene 5 product comprises an epitope of a pathoge~ic microorganism, the re~ombinant OBs of the present inventio~ can be particularly useful in vaccine formulations. In another embodiment, the foreign sequence can encode a molecule with insecticidal activity/ thus increasing the lethality of the 1o baculovirus to the host agricultural pest. Recombinant OBs expressing foreign peptides comprising antigenic determinants have uses in immunoassays. In yet another embodiment of the invention, the foreign sequence can encode a molecule with enzymatic activity so that the recombi~ant OBs can be use~ as 15 a reaction surface. The recombinant viruses of the present invention can also be used as expression vectors for the production of the foreign peptide(s) contained on the recombinant OB. The production of recombinant OBs can also facilitate the isolation of the component heterologous gene 20 product in sub~tantially pure form.
The invention is demonstrated by way of examples in which recombinant baculoviruses were engineered to express recombinant occlusion bodies that present an influenza hemagglutinin epitope. These recombinant occlusion bodies 25 immunvreact with antibodies that define the authentic epitope.

3.1 DEFINITIONS
The following term~ and abbreviation~ have the meaninqs indicated:
Ac = Autographa californica Hz - Heliothis zea ECV = extracellular virus poly H = polyhedrin , ~ .
, . ' ' . ~' , .

Isozymes:
EST = esterase FUM = fumarate hydratase LD~ = lactate dehydrogenase MDH = malate dehydrogenase Buffers TE = 10 mM Tris-~ICl, lmM EDTA, pH 7.6 TBE = 81.2 mM Tris, 20 mM boric acid, 1.5 mM
EDTA, pH 8.9 TC = 9.7 mM Tris, 2.13 mM citric acid, pH 7.1 OB - Occlusion body, a paracrystalline protein matrix which occludes baculovirus virions. The paracrystalline protein matrix forms a refractile body which is polyhedral; cuboidal or spherical in shape.
The term OB will also be used hereinafter to refer to lattices formed in vitro by the recrystallization of soluble polyhedrin.
.

NPV= Nuclear Polyhedrosis Viruses, a subgroup of the baculovirus genus in which the nucleocapsid~ are enveloped by a lipoprotein membrane singly (SNPV) or in multiples (MNPV) per common envelope. Up to loO
of th~se virion packages are embedded in an occlu~ion body, polyhedral to cuboidal in shape and 1-15 um in diameter.

GY - Granulosis Virus, a subgroup of ~he baculovirus genus in which one singly-enveloped nucleocapsid is embedded per occlusion body, round to ellipsoidal in shape, and 0.1-1 um in size.

.

17 ~L32~

NOBV = Non-occluded baculoviruses.
MCS = Multiple cloning site. A region o~ DNA containing a series of unique restriction e~donuclease - cleavage sites.

SDS-PAGE = Sodium dodecylsulfate polyacrylamide gel electrophoresis Cassette Trans~er Vector - A transfer vector 10 . comprising a polyhedrin promoter and a restriction enzyme recognition site into which a heterologous gene sequence can be inserted under the control of the polyhedrin promoter, in which a polyhedrin promoter and the restriction site are flanked by s~quences that are homologous to parent vector sequences.
Heterologous gene sequences can be insexted into the cassette transfer vectors which can then be used to construct recombinant expression v~ctors via homologous recombination ln ViVQ with a parent vector.

Transfer Vector = A transfer vector comprising a polyhedrin promoter and a heterologous ~ene sequence positioned under the control o~ th~
polyhedrin promoter, in which the polyhedrin promoter and the heterologous gene sequences are ~lanked by sequences that are homologous to parent vector sequencesO The transfer vector containing the heterologous gene equence can be used to construct recombinant expression vectors via homologous recombination in vivo with a parent vector.

Cassette Expresion Vector = An expression veckor comprising a polyhedrin promoter and a restriction enzyme recognition site into which .

- ; : ~ .
.' . . , ~ -' ' : . ~' .

-18- ~32~

a heterologous gene sequence can be inserted under the control o~ the polyhedrin promoter so that the gene is expressed in a suitable host.

Expression Vector = An expression vector comprisin~
a polyhedrin promoker and a heterologous gene sequence positioned under the control of the polyh~drin promoter so that the heterologous gene is expressed in a suitable host.
4. DESCRIPTION OF THE FIGURES
Figure 1. Nucleotide sequence of the polyhedrin gene of eliothis zea. The nuclectide sequence was determined using the dideoxy chain termination method o~ Sanger et al. ~1377, Proc. Natl. Acad. Sci. U.S.A. 74:5463~. The ded~ced amino acid sequence i5 presented below the nucleotide ~equence.
Restriction sites which were used in the cloning protocols are indicated. Solid bars indicate regions of the amino acid 20 sequence which have been identified as hydrophilic portions of polyhedrin.
Figure lA. Restriction map of the Heliothis polyhedrin yene. A restriction endonuclease digestion map of the Heliothis polyhedrin gene for the restriction endonu~leases 25 HlndIII, NruI, HincII, and AccI i6 presPnted. The map was derived from the nucleotide sequence shown in Figure 1.
Numbers in parentheses represe~t the number of the nucleotide in Figure 1.
~igure lB. Nucleotîde sequencing strategy of the 30 Heliothis polyhedrin gene. The sequencing strategy used in the dideoxy chain termination method (Sanger, F~ et al., 1977, Proc. Natl. Acad. Sci. U.S.A. 74:5463-5467) of sequencing of the Heliothis polyhedrin gene i~ depicted. The arrows indicate ~32~

the directi~n of sequencing on each fragment. The following abbreviations are used in the figure: Xho (XhoI), Sal (SalI), RI (EcoRI), H3 (Hind III~, HII (HindII), and Nr (NruI).
Figure 2. ~omology of amino acid sequences of 5 polyhedrin genes of Autographa cali~ornica MNPV ~Ac), Heliothis zea SNPV (Hz~ and Bom~yx mori NPV (Bm). The DNA and deduced amino acid sequence of the polyhedrin gene of Ac is presented (Hooft van Iddekinge, B.J.L,, et al., 1983, Virology 131:561).
The differences between the deduced amino acid sequence of the 10 Ac polyhedrin and that of Bm (Kozlov, E.A., et al., 1981, Xhim 7:1008) and Hz (see FIG. 1, supra), are indicated. Where an amino acid is present in one sequence and absent from another, the missing amino acid is indicated by the following symbol:
_ _ _ _ FigurP 3. Hydrophobicity profile of Hz and Ac polyhedrin. The hydrophobicity of each amino acid residue as measured by Hopp and Woods (1981, Proc. Natl. Acad. Sci. U.S.A.
78: 3824) was plotted against the amino acid residue number of the polyhedrin gene. The amino terminus and amino acid 20 residues 35-~0 were identified as hydrophilic regions of ~he Heliothis polyhedrin molecule.
Figure 4. Polyhedrin polylinker sequence. A synthetic polylinker gene sequence and its encoded amino acids are depicted. This gene segment encodes the Autographa californica 25 polyhedrin gene extending from the amino terminu~ to the BamHI
site corresponding to amino acid 58. Restriction endonuclease di~estion sites are indicated below the DNA sequence. The PvuI, ScaI, Bcll, and XbaI site~ correspond to amino acids 9, 19, 27, and 46, respectively.
Fiyure 5A. The construction of transfer vector PHE2.~, which can be used ~o insert ~oreign genes within the polyhedrin gene and to produce Hz viruses via in vivo recombination.
ThPse Hz recombinants, when placed in an appropriate host system, will express the foreign gene under the control of the 35 Heliothis polyhedrin promoter. The following abbreviations are used in the ~igure: polyH (~eliothis polyhedrin gene ': :

-20- ~32~ 0 sequence~, RI (EcoRI), Sst (SskI), Sma (SmaI), Bm (BamHI) Xb (XbaI), Sal (SalI), Pst ~PstI), H3 (HindIII), Nr (NruI), and Xho (XhoI).
Figure 5B. The construction of transfer vector 5 Phe2.61ac, whiCh contains a functional E. coli beta-galactosidase gene. Phe2.61ac can be used to insert other foreign genes or ~ragments thereof within the Heliothis polyhedrin yene, which can ~hen be transferred to the Helio~his virus genome by ln vivo recombination. The following 10 abbreviations are used in the figure: Xba ~XbaI), Bam (BamHI), Kpn (~nI~, Sma (SmaI), Oligo (Oligonucleotide), and MCS
(Multiple Cloning Site).
Figure 5c. The construction of transfer vectors which contain deletions in the amino terminal~coding region of the 15 Heliothis polyhedrin gene. These transfer vectors can be used to insert foreign genes within the polyhedrin gene and to produce Hz viruses via in vivo recombination. The following abbreviations are used in the ~igure: Xba (XbaI), bam (BamHI), kpn (~ , sma (SmaI~, sph (~I), bgl (B~lI), and pst (PstI).
Figure 6. HindIII restriction endonuclease analysis of HzSNPV Elcar~ wild-type and plaque purified isolates. ~iral DNA was purified from band isolated virions as described in ths methods. Viral DNAs were digested with HindIII and fractionated on a 0.75% agarose gel. Most of the differences 25 between the wild-type (W~) and plaque purified (1-25) strains occur between map units 23.5 and 43.3, which correspond to the region between HindIII bands H through G (Knell and Summers, 1984, J. Gen. Virol. 65:445-450).
Figure 7. Comparison of the wild~type Elcar~ isolate 30 and HzS-15 strain with enzymes BamHI, EcoRI, EcoRV, HindIII, KpnI, PstI, and SstI. Wild-type isolate (W+) and the plague-purified HzS-15 strain were digested with restriction enzymes and fractionated on a 0.75% agarose gel. The banding patterns for BamHI, ~I, and Ps~I are identical, while those of EcoRI, 35EcoRV, HindIII, and SstI are different. Many of the difference in the enzyme banding patterns of the two isolates could be localized to map units 23.5 through 43.3. The . ~ - .
;
.
':
-~32~

differences in the banding patterns for the enzyme EcoRV could not be positioned because no restriction maps exist for this enzyme.
Figure 8. A linear genomic map of the plague-purified 5 strain, ~zS-15. The genomic map was made with virion purified DNA digested either singly or with combina~ions of BamHI, PstI, and SstI. Any ambiguities in the map were resolved by double digestions o~ cloned Bam~I and PstI fragments. The genomic map of Knell and Summers (1984, J. Gen. Virol. 65:445-450) was used 10 as a reference since the restriction endonuclease banding pattern for BamHI was identical for both isolates~
Figure 9. Structural proteins of HzSNPV Elcar~ isolate and plaque purified strains. Sucrose gradient purified virions were elec~rophoresed for 4.5 hours on a 12% SDS-polyacrylamide 15 gel. The position and size of major wild-type proteins are labeled on the left, while unique proteins found in several of the plaque-purified strains are labeled on the riyht.
Figure 10. Cell growth curves for clonally i~olated cell strains derived from IP~B-HZ1075. Thr~e defined regions 20 of a tissue cultu~e flask (25 cm2) were counted at 48 hour intervals for a total of 8 days. Points on the graphs represent the average of the three counted areas with the error bars indi~ating 1 standard deviation. Letters in the lower left corner of each graph correspond to the nomenclature o~ the 25 specific cell strai~.
Figure 11. Comparison of isozyme banding patterns between all IPLB-HZ1075 derived cell strains and Heliothis zea larvae with isozymes FUM, L~H, and EST~ Cell and larval extracts were electrophoresed in a 5% polyacrylamide gel (95~
3~ acrylamide, 5% bis-acrylamide) in either TBE or TC buffer and stained for the appropriate enzyme~ Staining for FUM and for LDH confirm that the cell strains were derived from the parental IPLB-HZ1075 (W+) cell line and that they are ultimately derived from Heliothis zea larvae. The differences 35 in the EST gels suggest that all of the strains are not 22- ~2~

identical. Staining procedures for the isozymes are described infra. ~) EUM = Fumarate Hydratase B) LDH - Lactate Dehydrogenase C) EST = Esterase.
Figure 12. Comparison of isozyme banding patterns ~or 5 several insect cell lines. Cell lines prepared as described were electrophoresed in a 5% polyacrylamide gel (95%
acrylamide, 5% Bis-acrylamide) in TC buffer. LDH separates IPLB-HZ1075 from all other cell lines however, ATC-10 and IPLB-~Z1075 differ by an Rf value of only 0.03. MDH clearly 10 separates ~PLB-HZ1075 from ATC-10 and also BTI-EAA from IPLB-SF-2lAE which co-migrated when s~ained with LDH. Staining procedures are described infra.
Figure 13. The construction of cassette vector pAVl.5, which can be used to insert foreign genes within the Autographa 15 polyhedrin sequence which can then be transferred to the Autographa virus genome via in vivo recombination. pAVl.5 can also be used for further genetic manipulations such as insertion of the Heliothis polyhedrin gene, as shown in Figure

14. The following abbreviations are used in the figure: RI
20 ~EcoRI), Sst (SstI3, Bam (BamHI), Kpn (KpnI), Sal (SalI3, RV
(EcoRV~, Pst (PstI), H3 (HindIII), and Xho (XhoI).
Figure 14~ The construction of cassette vector pAVHp6, which can be used to insert foreign genes within the Heliothis and/or the Autographa polyhedrin genes, and which can be used 25 to transfer such foreign ~enes into Autographa virus by ln vivo recombination. The following abbreviations are used in the figure: 5st (SstI) 9 Bam (BamHI), RI ~EcoRI), Kpn (~I), Sal ~SalI), RV (EcoRV), Pst (PstI~, H3 (HindIII~, and Xho (XhoI).
Figure 15. The construction of a trans~er vector 30 containing a foreign DNA sequence encoding amino acids 98-106 of the influenza hemaqglutinin inserted at a specific HpaII
site within the Autographa polyhedrin gene. This transfer vector can be used to produce recombinant Autographa viruses containing the i~fluenza seguence via ln vivo recombination, 35 which express the foreign sequence under the control of the Autographa polyhedrin promoter.
.

- .
-:

-23- 132~6~

Figure 16A. The construction of vector pAV15InHem.
This vector contain~ an altered polyhedrin gene in which the polyhedrin sequenc~ between amino acid residue numbers 43 through 50 were replaced with an epitope of influenza 5 hemagglutinin.
Figure 16B. The construction of two transfer vectors, pAV15-InHem43 and pAV15InHem-50. These contain the epitope of influenza hemagglutinin at positions 43 and 50, respectively, of the polyhedrin sequ~nce.
Figure 17A and 17B. The construction o~ vector pAV17b InHem-l in which the epitope of in~luenza hemagglutinin is located after amino acid residue number 1 of polyhedrin. This plasmid encodes a unique SphI restriction site at the initiation codon of polyhedrin.
Fiyure 18. The construction of vector pAV17b InHem-2 in which the epitope of influenza hemagglutinin is located after amino acid resldue number 2 of polyhedrin.
Figure 19. The construction of pBRX13. This plasmid can be used to insert a coding sequence ~or any epitope into 20 the polyhedrin gene spanning the coding region for amino acid residues 36-SO. The resulting recombinant polyhedrin gene can be excised from pBRX13 and cloned into a transfer vector.

5. DETAILE~ DESCRIPTION OF THE INVENTION
The present invention relates to recombinant baculoviruses which encode fusion polyhedrin proteins capable of forming occlusion bodies containing foreign peptides. The recombinant OBs of the present invention comprise crystallized polyhedrin ~usion proteins which bear the heterologous gene 30 product on the surface of or within the occlusion body. ThP
re~ombinant OBs are formed by replacing regions of the baculovirus polyhedrin gene that are nonessPntial for occlusion body formation with sequences encoding foreign peptides. The present inv~ntion is also directed to vector/host systems which 35 can express the recombinant polyhPdrin gene in dif~erent hosts, including but not limited to, cultured cells, larvae, or microorganisms.

, ~ .

: ~ .

-24- ~32~

According to one embodiment of the invention, the recombinant OB which contains an immunogenic determinant of a pathogenic microorganism can be used in vaccine formulations.
In another emhodiment of the invention, recombinant OBs 5 comprising sequences with insecticidal activity can be used to increase lethality of the baculovirus to host agricultural pests.
According ko another embodiment of the invention, the recombinant OBs which expose an active site of an enzyme can be 10 u~ed as immobiliæed enzyme in appropriate procedures.
In other embodiments, the recombinant OBs of the present invention have uses in immunoassays and as expression vectors.
The production of recombinant polyhedrin crystals can also facilitate the isolation.of the component heterologous gene product in substantially pure form.
The method of the invention may be divided into the following general stages solely ~or the purpose of descriptionO
(a) identification of modifiable domains encoded by the polyhedrin gene, (b) identification and characterization of 20 immunodominant pepti~ss for expression on or within recombinant occlusion bodies, (c) construction of recombinant polyhedrin gene~, (d) selection of recombinant occlusion bodies, (e) verification of expression of foreign epitopes on or within the recombinant occlusion body, and (f) determination of 2~ immunopotency of foreign epitopes expressed on or within recombinant occlusion bodies.

5.l. THE GENERATION OF RECOMBINANT OBs CONTAINING ANTIGENIC DETERMINANTS OF
HETEROLOGOUS PROTEINS
5.1.1. IDENTIFICATION OF MODIFIABLE
DO~INS ENCODE:D BY THE
POLYHEDRIN GENE
Exposing new antiyenic d~terminants on the sur~ace of or within the OB requires identi~ying regions of the polyhedrin 35protein that can be modified without affecting the formation or stability of the lattice. Such segments can be altered by the insertion of new epitopes without interfering with the '' - ~

-25- 13~

integrity of the crystalline lattice. The identification of modifiable domains can be accomplished by comparing sequences of cloned polyhedrin genes, analyzing polyhedrin genes encoding truncated polyhedrin prot~in~, and by structural analysis of 5 polyhedrin a~i~o acid sequences. Exposing the heterologous gene product on the surface of the OB may be preferable (but not required) for formulation in vaccines. Surface domains of the OB may be more amenable to alteration without concomitant destruction of crystal integrity than internal domains. The 10 identification of polyhedrin surface domains can be accomplished by hydrophilicity analysis o~ polyhedrin, and by generating and characterizing monoclonal antibodies to OBs.
Segments of the sequence which are hydrophilic and hypervariable regions are prime candidates for insertion into

15 or replacement with the foreign sequence of interest. However, sequences of the polyhedrin ~ene which encode hydrophobic regions of the polyhedrin protein may also be insert~d into or replaced by heterologous sequences.
In the specific e~bodiments which exemplify the 20 invention described herein we have used the analyses described infra to identify the amino terminus and amino acid residue numbers 38-50 of the Autographa polyhedrin as modifiable domains. In particular, the data presented in the Examples, infra, demonstrate that foreign epitopes may be inserted in the 25 polyhedrin sequence at amino a~id residue number 1, 43 or 50 in order to produce recombinant OBs that expose the foreign epitope.

5.1~1.1. HYDROPHILICITY ANALYSIS OF POLYHEDRIN
In order to determine which portions of the polyhedrin gene sequence are more likely to be surface domains, hydrophobic and hydrophilic regions of the polyhedrin amino acid sequence and the corresponding regions of the gene sequence which encode the hydrophilic and hydrophobic regions 35 should be identified. Since the hydrophilic regions sf the amino acid sequence are likely to be external domains of the crystal, and, furthermore, are likely to be external domains of the polyhedrin monomer upon cry~tal dissolution, such regions -26- .~3 2 ~

may be especially useful in an embodiment of khe invention employing recombinant OBs in a vaccine formulation, since they would readily provide ~or presentation of the foreign epitope to the host immune system. Portions of the polyhedrin gene 5 which encode hydrophilic regions are prime candidates for insertion into or replacement by a heterologous yene sequence, since the foreign epitope inserted therein is thus likely to be immunogenic in a vaccine formulation. Thus, the portions of the polyhedrin gene sequence which encode regions of the 10 polyhedrin protein which are both hydrophilic and which are also determined to be highly variable (~ee Section 5.l.l.2, infra), are good candidates for replacement by heterologous gene sequence~, so that the resulting fusion polyhedrin proteins will crystallize and form recombinant OBs containing 15 immunogenic foreign epitopes.
Sequences of the polyhedrin gene which encode hydrophobic regions of the polyhedrin protein may also be inserted into or replaced by heterologous gene se~uences, and provide for a recombinant OB that is useful in a vaccine 20 formulation. In a particular embodiment, gene sequences which encode an amphipathic peptide (i.e. a peptide having one face which is hydrophobic, one ~ace which is hydrophilic) (see Section 5~4~lo~ infra) may be inserted into a region of the polyhedrin gene that is nonessential for crystallization and 25 which encodes a hydrophobic portion of the polyhedrin protein.
In a particular embodiment of the invention, we have identified the hydrophobic and hydrophilic regions of the Autographa, Heliothis and Bombyx morii polyhedrin amino acid sequence and the corresponding DNA fragments which encode them 30 (See FIG. 3 and FIG. l). We have identified the amino terminus and amino acid residue number~ 38 50 of Autographa polyhedrin (see FIG. 3) as hydrophilic regions of the polyhedrin protein which are proba~ly on the surface o~ the protein and are thus also likely to be on the surface o~ the OB. The heterologous 35 gene sequence can be inserted into the polyhedrin gene sequence so that the heterologou~ gene either interrupts or replaces all or a portion of nucleotide residue numbers 1 to 12, or 142 to : : : . : .

:- ~

-27- ~3~

180, which encode these regions of the amino acid sequence (see FIG. 13. It should be noted that these residue numbers are approximate and that any restriction site or sites which occur, or are genetically engineered to occur, within or in proximity 5 to these regions may be used to specifically cleave the polyhedrin gene seguence in order to inser~ the heterologous gene sequence. Some restriction sitPs which may be useful include, but are not limited to, the restriction ~ites indicatad in Figure 1. It is preferred to use restriction 10 sites that are unique, so that where no suitable sites exist, new sites may be obtained, for example, by in vitro mutagenesis (see S~ction 5.1.3, infra).

5.1.1.2. SEQUENCE COMPARISONS OF DIFFERENT

POLYHEDRIN GENES

Comparison of the sequences of different polyhedrin genes is with the aim of identi~ying regions of the gene that are hypervaxiable. The most hypervariable xegions probably constitute regions nonessential for lattice formation which may be genetically manipulated for use in exposing a new antigenic 20 determinant on the surface of or within the OB.
In a specific embodiment, regions of the polyhedrin gene which are variable and which encode alph~helical domains may be inserted i~to or replaced by a sequence encoding a foreign epitope that is also alpha-helical and which may be 25 amphipathic, so that a recombinant protein is produced which allows both the OB and the epitope within it to retain their structural integrity. ~An example of such a polyhedrin gene sequence, which i5 both variabl~ and ~ncodes an alpha-helical region, is that encoding amino acids 37-49; see the discussion 30 infra and section 5.1.1.4.).
In a specific embodiment, amino acid sequence comparisons of a number o~ baculovirus polyhedrin proteins reveal a high degree of sequence homology among these different proteins (FIG. 2). Overall, the Heliothis zea ~HzSNPV), Bombyx 35 mori (BmMNPV) and AutQgraPha californica (AcMNPV) polyhedrin proteins are 80-85% homologous. The sequence homology is '~ ' ''' ~ ;

' 132~10 expected given the similar functional and structural roles of th~se proteins in the different viruses. One might expect that the conservation of certain regions is essential for crystal formation. However, there are small regions where the amino 5 acid sequence is highly variable among the dif~erent polyhedrin proteins. In particular, the region between amino acids 37-~9 of the polyhedrin protein shows significant variability.
Interestingly, the procedures of Hopp and Woods (Hopp, T. and Woods, K., 1981, Proc. Natl. Acad. Sci. U.S~A. 78:3824) 10 indicatP that this se~ment coincides with the most hydrophilic region of the Heliothis zea polyhedrin protein. These results indicate that this variable hydrophilic region may de~ine a section of the pr~tein that is not essential for crystallization and is probably on the surface of the protein.
15 From this analysis the segment encoding amino acids 37-49 i5 a putative modifiable surface domain. Through a similar analysis, the amino terminus (roughly amino acids 1 through 4 is also identified as a hydrophilic, potentially variable region of the polyhedrin protein.
A comparison of the sequences from six lepidopteran NPV
polyhedrins (Vlak, J.M. and Rohrmann, GoF~ I lg85~ The Nature of Polyhedrin, In Viral Insecticides for Biological Control, Academic Press, pp. 489-452) reveals regions of homology which can be identified for modification; these include hut are not 25 limited to the amino terminus (~ , amino acid 1~; amino acids 37-49, 86-96, 120-127, 141-150, 193-198, and 208-216 (id.); as well as the carboxy terminus (e.g., the amino acid residues after amino acid number 245). In particular, amino acid residue numbers 58 and 211 are attractive candidates for 30 modification due to the location of unique BamHI and KpnI
sites, respectively, in the coding sequences for these regions.
As discussed previously, polyhedrins of NPVs and polyhedrins o~ GYs form a group o~ related molecules (Vlak, J.M. and Rohrmann, G.F., ~ , Tweeten, X.A., et al., 1981, 35 Microbiol. Rev. 45:379-408). Thus, polyhedrins of granulosis viruses (granulins) can also be altered in accordance with the : . , : : , : ., . . - , -: - t -29- ~32~0 present invention to express a foreign peptide on or within the occlusion ~ody. A comparison o~ some NPV and GV polyhedrin (Vlak, 3.M and Rohrmann; G.F, supra of pp. 489-542) reveals that the v~riable N-terminal amino acid sequences of some NPV
5 polyhedrins and a GV polyhedrin, may be manipulated to form recombinant OBs in particular embodiments of the present invention~

5 .1.1. 3 . SEQUENCE ANA~YSIS OF POLYHEDRIN
GENES ENCODING MUTANT OR TRUNCATED
POLYHE:DRIN PROTEINS
A second approach for identifying modifiable regions of polyhedrin is to analyze sequences of mutant or truncated polyhedrin proteins. Any mutant polyhed~in gene which is contained within the genome of a virus that still produces OBs, 15 can be analyzed to determi~e its mutation. The mutation contained in such a gene represents a point or region o~ the wild-type polyhedrin gene which can be altered without losing the ability to ~orm OBs.
For example, a baculovirus that produces a truncated 20 polyhedrin protein, ~hich still forms occlusion bodies, can be analyzed by cloning and sequencing its polyhedrin gene. Regions that are non essential for crystalliæation may be identified by comparison of the truncated sequence to that of oth~r known polyhedrins. The corresponding region of the full lenyth 25 p~lyhedrin gene could be replaced, or heterologous sequences could be inserted into the appropriate region of the truncated gene, in order to express a new antigenic determinant. One example of a mutant which has been isolated and may be analyzed in such a fashion, is a mutant AcMNPV which may differ from 30 other polyhedrins by a small deleti~n of 20-30 amino acids.
This AcMNPV produces tetrahedral rather than polygonal occlusion bodie~, which contain a 31 kD polyhedrin protein rather than the 33 kD wild-type polyhedrin.
Another AcMNPV mutant, termed ~5, has been described 35 (Carstens, E.B., 1982/ J. Virol. 43:~09-818 Carstens, E.B., et al., 1986, J. Virol. 58:684-688~. M5 has a single point ' 132~6~
mutation within the polyhedrin g~ne which results in a substitution of leucine ~or proline at amino acid 58 of the polyhedrin protein. However, this single alteration results in a drastic morphological change in the polyhedra, producing 5 cubic occlusion bodies. Thus amino acid 58 is possibly critical to the proper folding of the polyhedrin molecule (Carsten~, E.B., et al., supra).
Cloning and se~uencing the mutant polyhedrin gene can be accomplished by any technique known in the art (Maniatis, T., 10 et al., 1982, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, New York). For example, in order to generate DNA ~ragments encoding polyhedrin sequences (hereinafter referred to as polyhedrin DNA), the polyhedrin DNA
may be cleaved by restriction enzyme digestion, DNase 15 digestion, physical shearing, etc. Identification of the polyhedrin DNA can be accomplished in a number of way~, including, but not limited to, nucleic acid hybridiza~tion, comparison o restriction digestion patterns with known restriction maps, and mRN~ selection through nucleic acid 20 hybridization followed by in vitro translation. The polyhedrin DNA, or total baculovirus DNA, is inserted into the cloning vector which is used to transform appropriate host cells so that many copies of the polyhedrin sequences are generated.
This can be accomplished by ligating the polyhedrin DNA into a 25 cloning vector with comple~entary cohesive termini, with or without first ligating linkers onto DNA termini in order to generate desired restriction sites, or blunt-end ligation, homopolymeric tailing, etc. Any of a large number of vector-host systems may be used. Vector systems may be either 30 plasmids or modified viruses, but the vector system must be compatible with the host cell used. Recombinant molecules can be introduced into cells via trans~ormation, transfection, or infection. Identification of a cloned polyhedrin gene can be achievad by any technique known in the art. Such techniques 35 include, but are not limited to, screening for expression of the gene by colony blot analysis (Huynh, T.V., et al., 1985, In DNA Cloning: A Practical Approach, Vol. 2, Glover, D.M.

.

i ~ ' -31- 132~

(ed~), IRL Press, Oxford, pp. 49-78) using anti-polyhedrin antibody as a probe, or using a labelled polyhedrin gene fragment to screen colonies or plaques. DNA sequence analysis can be used to verify the identity of the polyhedrin gene.
In a specific embodiment of the present invention, plasmid or bacteriophage lambda libraries containing DNA
inserts derived from the mutant AcMNPV strain can be constructed. Clones containing the mutant polyhedrin gene can be identified by oolony or plaque hydridization to labelled 10 polyhedrin gene probes (Grunstein, M. and Hogness, D., 1975, Proc. Natl. Acad. Sci. U.S.A. 72:3961; Benton, W., and Davis, R., 1977, Science 196:180).
Once the polyhedrin DNA-containing clone has been identified, it may be grown, harvested and its DNA insert may 15 be characterized as to its restriction sites by various techniques known in the art (Maniatis, T., et al., Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, New York).
The sequence of the polyhedrin DNA insert can then be 20 determined. Methods by which this may be accomplished include the Maxam-Gilbert procedure (Maxam, A.M., and Gilbert, W., 1980, Meth. Enzymol. 65:499l or the Sanger dideoxy chain termination procedure (Sanger, ~., et al., 1977, Proc. Natl.
Acad. Sci. U.S.A. 74 5463). In a specific embodiment employing 25 the Sanger teshnique, appropriate ~egments of the polyhedrin DNA can be preferably subcloned into ~13 vectors ~Messing, J., 1983, Meth. Enzymol. 101:~0) for optimal sequencing efficiency.

5.1.1.4. STRUCTURAL ANALYSIS OF POLYHEDRIN
AMINO ACID SEQUENCES
While the hydrophilicity profile and sequence comparisons of the polyhedrin protein help identify modifiable domains, these analyses rely on the primary structure of the protein. There may be conformational restrictions that severely limit the ability to modify such domains. The region 35 may be either directly or indirectly involved in lattire formation such that modification of the region would '~ , . .

.

~ 32 ~2~0 destabilize the crystal. Secondary structural analysis (Chou, P. and Fasman, G., ~974, Biochemi~try 13:222) can there~ore be done in order to identify regions that assume secondary structures which may he more conducive to modification without 5 concomitant lattice disruption. Similar analysis of the amino acid sequences ~ ~oreign epitopes to be expressed on the OB
may further help to identify compatible structures in the polyhedrin protein which may be altered by recombinant DNA
techniques~ As an example, secondary structural analysis of 10 the Autographa polyhedrin using the procedures of Chou and Fasman, supra, indicates that the region between amino acids 37-49 is primarily an alpha helical structure. Incorporating foreign peptides that are not alpha helical may alter the structure of surrounding regions and possibly disrupt 15 crystallization. Similarly, the conformation of the surrounding regions may interfere with the proper folding of the foreign peptide domain and block the proper presentation of the epitope.
Chou-Fasman analysis of the hydrophilic amino terminus 20 of the Autographa protein suggests that this region is involved primarily in the formation of a beta turn. This conformation provides some degree o~ structural flexibility for a new peptide lnserted in this region. A new determinant incorporated at thi~ site may not be forced into an in~ricate 25 secondary structure and could possibly assume its native structure without interfering with the stability of the polyhedrin crystal.
In a similar fashion, polyhedrin molecules and the foreign epitopes proposed for insertion can be analyzed to 30 determine secondary structures that are potentially compatible, resulting in recombinant OBs which express the foreign epitopes in their naturally antigenic conformations.
Other methods of structural analysis can also be employed to aid in identifying modifiable domains of the 3~ polyhedrin molecule. ~hese include but are not limited to X-ray crystallography and computer modelling. X-ray crystallography (Engstom, A., 1974, Biochem. Exp. Biol. 11:7-~33- 1~2~61~

13) can be used to analyze the domains of the polyhedrin protein which interact to form the paracrystalline lattice, and to confirm overall structures generated by computer modelling.
Computer modelling (Fletterick, R. and Zoller, M. (eds.), 1986, 5 Computer Graphics and Molecular Modeling, In Current communications in ~olecular siology~ Cold spring Harbor Laboratory, New York) can provide theoretical three-dimensional images representing sequences comprising foreign epitopes, in an attempt to identify higher order structures that are 10 compatible with the structure of the polyhedrin crystal.

5.1.1.5. PREPARATION AND CHARACTERIZATION OF
MONOCLONAL ANTIBODIES TO OCCLUSION BODIES
_ _ The isolation and characterization of monoclonal antibodies to intact 03s or to recrystallized polyhedrin will identify regions of the poly~edrin protein that are exposed on the surface of the crystal. Once these regions are identified, it will be possible to test whether these domains can be altered without interfering with the integrity of the crystalline lattice.
The occlusion bodies of NPVs are surrounded by a carbohydrate envelope (Minion, F.C., et al., 197~, J. Invert.
Pathol. 34:303) that may affect the immunogenicity of the OB.
Therefore, in order to avoid such inter~erence, monoclonal antibodies can be prepared to recrystallized polyhedrin, in 25 addition to using the purified OBs. An llS-13S polyhedrin aggregate can be purified from alkali solubilized occlusion bodies and recrystallized by standard techniques (Shigematsu, H., and Suzuki, S., 1971, J. Invert. Pathol. 17:375).
Recrystallization results in a polyhedrin particle free of 30 contaminating virions and carbohydrate.
OBs can be purified from a number of host cells by known techniques (for example, see Section 7.1.4. infra, and Tweeten, K.A., et al., 1981, Microbiol. Rev. 45:379-408). These host -cells include, but are not limited to, cell lines and larvae in 35 which the baculovirus can be propagated and produce OBs. For examplel such cell lines include, but are no~ limited to, .

-34- ~256~

Spodoptera frugiPerda IPLB-SF-21AE cells, Heliothis zea IPLB-HZ1075 cells, Estiqmene acrea BTI-EAA cell~, Trichoplusia ni TN-368 cells, Trichoplusia nl BTI-TN4BI, BTI-TN5F2, BTI-TN5F2P, and BTI-TN5F2A cells (Granados, R.R.I et al., 1986, Virology 5 152:472-476), Mamestra brassicae Mb 0503, and Mb 1203 cells (Miltenburger, HoG~ ~ et al., 1976, æ. Angew. Entomol.
82(3):306-323); Heliothis zea BCIRL-HZ-AMl,2, or 3 cells tMcIntosh, A.H. and Ignoffo, C.M., 1981, J. Invert. Pathol.
37:258~264); Heliothis virescens BCIRL-HV-AMl cells ~id.); and 10 their derivative cell lines. Infection of tissue culture cells can be accomplished by standard procedures known in the art (Smith, G., and Summers! M., 1979, J. Virol. 300828). In a specific embodiment of the invention, Spodoptera frugiperda cells can be infected with Autogra~ha californica MNPV at 1-2 pfu/cell. Polyhedrin protein can then be purified from OBs by techniques known in the art. For example, polyhedrin protein can be purified by incubating the purified occlusion bodies in 0.1 M Na2C03 (pH 11), 0.17 M NaCl, 1 mM EDTA, and spinning the dissolved prot~in at 24,000 rpm in an SW50.1 rotor ~or 30 20 minutes at 4C to remove virus particles and any insoluble material. The solubilized polyhedrin can be stor~d at -20C
(Huang, Y.S., et al., 1985, Virology 143:380), and the homogeneity of the preparation can be determined by SDS
polyacrylamide gel electrophoresis, among other methods.
Monoclonal antibodies to OBs or to recrystalliz~d polyhedrin can be prepared by using any technique which provides for the production of antibody molecules by continuous cells in culture. These include, but are not limited to, the hybridoma technique originally described by Kohler and Milstein 30 (1975, Nature ?56:495), and the more recent human B cell hybridoma technique (Xozbor et al., 1983, Immunology Today 4:72j and EBV-hybridoma technique lCole et al., 1985, Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc., pp. 77-96). In a specific embodiment of the invention, BALB/c 35 mouse monoclonal antibodies can be prepared by use of a fusion protocol utilizing the BALB/c myeloma cell line, NS-l, and the fusagent, polyethyleneglycol 1000, as described by Ploplis et .. . . .
. . . .
:

~ 3 2 ~ 0 al. (Ploplis, V.A., et al., 19~2, Biochemistry 21:5891).
Hybridomas producing antibodies to OBs can be identif ied preferably by use of a solid-phase immunoassay with a labelled ligand, such as an enzyme-linked immunosorbent assay (ELISA) 5 (as described, for example, by Ploplis, V.A.~ et al., supra).

Monoclonal antibodies generated to the OBs or to recrystallized polyhedrin can then be tested to ensure recognition of epitopes on the polyhedrin monomer. This can be accomplished by any immunoassay known in the art such as 10 immunoprecipitation or radioimmunoassay, although a Western blotting procedure is preferakle (Towbin, H., et al.O 197~l -Proc. Natl. ~cad. Sci. U.S.A. 76:4350). For instance, purified polyhedrin can be denatured and reduced with a sample buffer containing SDS and B-mercaptoethanol. The denatured sample can 15 then be electrophoresed through a gel, transferred to nitrocellulose, and incubated with the monoclonal antibody solution. The presence of an epitope on the monomeric polyhedrin molecule that can be recognized by a monoclonal antibody generated to the OB or to recrystallized polyhedrin, 20 can then be detected by use of a second antibody, directed against the monoclonal antibody, conjugated to a label such as an en2yme or radioisotope. For example, horseradish peroxidase can be used, in which case visualization of the antigen-antibody complex can be facilitated by using the enzyme 25 substrate 4-~hloro-1-naphthol.

5.1.1.5.1. IDENTIFICATION OF EPITCPES RECOGNIZED
BY ANTI-POLYHEDRIN MONOCLONAL ANTIBODIES

Characterization of the epitopes recognized by anti-polyhedrin monoclonal antibodies involves the study of 30 polyhedrin-encoded peptides' interactions with the monoclonal antibodies, Peptides representing putative epitopes can be tested ~or antibody binding. T~e peptides for use in such tests can be generated by any method known in the art, including, but not limited to, protease digestion or chemical 35 fragmentation of polyhedrin, chemical synthesis of peptides, or expression by recombinant DN~ vector-host systems.

'~, .
' -36- ~32~3 ~

In a preEerred embodiment of the invention, protease dige~tion of polyhedrin can be accomplished by use of vs protease (srown~ M., et al., 1980, J. Gen. Virol. 50:309), although any protease known in the art such as trypsin, 5 chymotrypsin, papain, or pepsin, among others, can be used.
Peptide~ generated by protease digestion are preferably isolated by HPLC utilizing reverse phase chromatography, although any standard techniques which result in the puri~ication o~ the peptides can be used. The peptides can 10 then be assayed for antibody binding. Peptides that e~fectively compete with the intact protein ~or antibody binding posses~ the antigenic determinant recognized by the antibody. Any type of competitive immunoassay may be used, such as radioimmunoassay~, and, preferably, ELISAs. Peptides 15 that are determined to comprise the antigenic determinant can be sequenced by techniques ~nown in the art, and their location in the polyhedrin protein determined by sequence analysis.
Chemical fragmentation of the polyhedrin protein can be accomplished, for example, by us~ of cyanogen bromide, partial 20 acid hydroly~is, BNPS~skatole, N-bromosuccinimide, hydroxylamine, or other methods of ~pecific cleavage.
In a~ alternative method for identification of the antigenic determinants, th~ identification of modiiable, hydrophilic (Hopp, T., and Woods, K., 1981, Proc. Natl. Acad.
25 Sci. U.S.A. 78:3824) and/or hypervariable regions (see sections 5.1.1.1 and 5.1.1.2, supra) can pinpoint potential antigenic sites. Small peptides that correspond ~o the putative epitopes can be ~hemically synthesized, for example, by the Merrifield solid phase method (Merrifield, R. B., 1963, J. Am. Chem. Soc.
30 85:2149). These peptides can then be analysed for crossreactivity with the monoclonal antibody as described supra.
Another way in which to pinpoint potentially modifiable regions o~ the polyhedrin protein which can then ~e synthesized 35 and tested for competitive antibody binding, is to identify species-specific epitopes. Monoclonal antibodies directed toward an OB antigenic determinant can be analyzed ~or .

.
' . ' " ~ ' ., ' .~ ~ .

132~6~ ~
crossreactivity to other polyhedra species. This can be most easily accomplished by an ELISA method although other immunoassays including, but not limited to, Western blotting, immunoprecipitation, and radioimmunoassays are within the scope 5 of the inventionO SpPcies-specific epitopes exposed on the surface o~ the crystal represent potentially modifiable regions of the polyhedrin protein. some such species-specific monoclonal antibodies to OBs have been described (Huang, Y. S., et al., 1985, Virology, 143:3~0), suggesting that there are 10 epitopes which can be varied on the occlusion body surface.
Another method for mapping epitopes on the polyhedrin protein is by comparing proteolytic digests of polyhedrin in the presence and absence of the monoclonal antibody. It has been shown that epitopes are protected from proteolysis in the 15 presence of their respective antibodies (Jemmerson, R., and Paterson, Y., 1986, BioTechniques 4:18). Protease-digested fragments can be generated by known methods in the ar~, including, but not limited to, the use of proteases such as trypsin, chymotrypsin, V8 protease, papain, pepsin, etc.
2~ Protease-generated peptides can be identified by various techniques, including, but not limited to, reverse-phase chromatography and two-dimensional gel electrophoresis. In a preferred embodiment of the invention, tryptic peptides of polyhedrin can be identified using reverse-phase 25 chromatography. The digestion patterns of a nonspecific immunoglobulin molecule, digested in the presence o~
polyhedrin, can also be determined, in order to identify those peptides which are derived from the immunoglobulin molecule that have retention times, electrophoretic migrations, or other 30 characteristics that are similar to those of the polyhedrin-derived peptides. The thus identified polyhedrin-speci~ic and immunoglobulin-specific digestion patterns can be compared to the digestion patterns obtained ~rom the polyhedrin-antipolyhedrin complex. Presumably, the antibody will protect 35 some proteolytic cleavage site~ and reduce the recovery of peptides containing the epitope. These peptides, putatively comprising the epitope, can be isolated and characterized by , , ., . ..... : . .. . ~ .

~ 3 ~
sequence analysis. Due to probable steric hindrance of the antibody on overall proteolysis of the protein, thi~ procedur~
could potentially identify some peptides which are not involved in antibody binding. Ther~fore, a variety of other proteases 5 can be used. Such proteases include, but are not limited to, trypsin, chymotrypsin, V8 protease, papain, and pepsin.
Comparing the results generated by several proteases will identify the region containing the epitope. This method has proven to be effective in mapping conformational dependent 10 epitopes that cannot be easily mimicked by synthetic peptides (Jemmerson, R., and Paterson, Y., 1986, BioTechniques 4:18).

5.1.2. IDENTIFICATION AND CHA~ACTERIZATIO~
OF IMMUNODOMINANT PEPTIDES FOR
EXPRESSION ON OR WITHIN RECOMBINANT OBs In one embodiment of the invention, the generation of recombinant OBs which contain one or more foreign antigenic determinants, for use in vaccine formulations or immunoassays, requires the identification and characterization of specific antigenic determinants which may bP used in constructing 20 recombinants. For use in vaccine formulations, a peptide or protein should be identified which encodes an immunopotent sequence of a pathogenic microorganism. In other words, the peptide should be capable o~ eliciting an immune response against a pathogen. In addition, molecules which are haptens 25 (i-e- antigenic, but not immunogenic) may also be used, since the polyhedra functions as a carrier molecule in conferring immunogenicity on the hapten. (For a further discussion of peptides which may be exposed on or within recombinant OBs, see Section 5.4.1., infra.) Peptides containing epitopes which are 30 reactive with antibody although incapable of eliciting immune responses, even when exposed on recombinant OBs, still have potential uses in immunoassays (see Section 5.4.4., infra)O
Peptides or proteins which are known to encode antigenic determinants can be incorporated into recombinant polyhedra.
35 If specific antigens are unknown, identification and characterization o~ immunoreactive sequences shou}d be carried .

-39- ~32~0 out. One way in which to accomplish this is through the use of monoclonal antibodies generated to the surface molecules of the pathogen. Such a technique has been used to help identify and characterize the major epitopes of myoglobin (Berzo~sky, J.A., 5 et al., 1982, J. Biol. Chem. 257 3189), lysozyme (Smith-Gill, S.J., et al., 1982, J. Immunol. 128:314), and in~luenza hemagglutinin (Wilson, I.A,, et al., 1984, Cell 37:767). The peptide sequences capable of being recognized by the antibodies are defined epitopes. These peptide sequences can be 10 identified, for example, by virtue o~ the ability of small synthetic peptides containing such sequences, to compete with the intact protein for binding of monospecific antibodies.
Alternatively, small synthetic peptides conjugated to carrier molecules can be tested for generation of monoclonal antibodies that bind to these sites, encoded by the peptide, on the intact mole~ule. Such an approach has been used for the recognition of an immunodominant peptide determinant in the influenza hemagglutin protein (Wilson, IoA~ ~ et al~ 84, Cell 37:767).
Other method~ which may be employed for the identification and 20 characterization of antigenic determinants are also within the scope of the invention. These include, but are not limited to, protease protection experiments such as described in Section 5.1.1.5.1, su~ra. In this technique, epitopes are identified by their protection Prom proteolysis in the presence of their 26 respective antib~dies.

5.1.3. CONSTRUCTION OF RECOMBINANT POLYHEDRIN GENES
Once modifiable, and preferably surface, regions of the polyhedrin protein have been identified, all or part of the 30 corresponding segments of the gene can be replaced with sequences encoding one or more foreign epitopes. Many strategies known in the art can be used for this purpose, provided the antigenicity of the heterologous sequence and the ability of the polyhedrin to form an occlusion body are not 35 destroyed. The relevant s~quences of the polyhedrin gene and of the heterologous gene can, by techniques known in the art, be cleaved at appropriate sites with restriction ':

:

~40- ~32~
endonuclease(s), isolated, and ligated in vitro. If cohesive termini are generated by restriction endonuclease digestion, no further modification of DNA before ligation may be n~eded. If, however, cohesive termini of the polyhedrin DNA are not 5 available ~or generation by restriction endonuclease digestion, or different sites other than those available are preferred, any of numerous techni~ues known in the art may be used to accomplish ligation of the heterologous DNA at the desired sites. For example, cleavage with a restriction enzyme can be 10 followed by modification to create blunt ends by digesting back or filling in single-stranded DNA termini before ligation.
Alternatively, the cleaved ends of the polyhedrin or heterologous DNA can be ~chewed backn using a nuclease such as nuclease Bal 31, exonuclease III, lambda exonuclease, mung bean nuclease, or T4 DNA polymerase exonuclease activity, to name but a few, in order to remove portions of the sequence. An oligonucleotide sequence which encodes one or more restriction sites that are unique to the polyhedrin gene sequence and/or to the baculoviral genome itsel~ can ~e inserted in a region of 20 the polyhedrin gene that is nonessential for crystallization (hereinafter this oligonucleotide linker will be referred to as a polylinker)O The polylinker can be inserted into the polyhedrin sequence by in vitro techniques such as those discussed supra. The resulting recombinant gene is akin to a 25 ncassette vectorn into which any heterologous gene can be inserted using appropriate restriction enzymes. In this embodiment, it is beneficial to insert a polylinker sequence within the polyhedrin gene so that the interrupted polyhedrin sequence is no longer in the correct translational reading 30 frame, in which case the recombinant virus containing the cloning sites will be OB-. The subsequent ligation of a heterologous gene into the cloning site located within the region of the polyhedrin gene sequence that is non-essential for crystallization, so that both sequences are in the correct 35 translational reading ~rame uninterrupted by translational stop signals, will result in a construct that directs the production of a fusion polyhedrin protein that will crystallize and form .

~ .

..... ~ -.

-41 ~32~

recombinant occlusion bodies. A polylinker may also ~e used to generate suitable sites in the heterologous gene sequence.
Additionally, polyhedrin or heterologous gene sequence~ can be mutated in vitro or in Yivo in order to ~orm new restriction __ 5 endonuclease sites or destroy preexisting ones, to ~acilitate in vitro ligation procedures. Any technique for mutagenesis known in the art can be used, including but not limited to, ln vitro site-directed mutagenesis (Xunkel, 1985, Proc. Natl.
Acad. Scio 82:488 492; Hutchi~son, C., et al., 1978, J. Biol.
10 Chem. 253:6551), use of TAB~ linkers (Pharmacia), etc.
The particular strategy for constructing gene fusions will depend on the specific polyhedrin sequence to be replaced or inserted into, as well as the heterologous gene to be inserted. The discussion infra relates several strategies by 15 which manipulation of restriction sites of the polyhedrin gene f~r in vitro recombination purposes may ~e accomplished, and is intended for descriptive purposes only. Many other recombinatiGn ~trategies are within the scope of the invention.
One specific embodiment o~ the invention is a strategy 20 for replacing the region of the AcMNPV polyhedrin gene encoding amino acids 37-49 with oligonucleotides encoding an epitope of influenza hemagglutinin. Thers is a BamHI site within the polyhedrin structural gene at the sequence encoding amino acid 58. A deletion strain can be constructed by cleavage with 25 BamHI ~ollowed by digestion with exonucleasP Bal 31, and ligation to a synthetic BamHI polylinker. In this way, we constructed a d~letion strain in which DNA sequences encoding amino acids 35 through the BamHI site were replaced with a synthetic BamHI linker. The plasmid containing this gene can 30 then be cut at the BamHI site, ~blunt-ended~ with either Sl or mung bean nuclease, and ligated to ~he termini of the following synthetic oligonucleotide:

Lys His Phe Ala Leu Asp Asn Tyr Leu Val Ala Glu Asp 5' AAG CAC TTC GCG AGA TCTA GAC AAC TAC CTA GTG GCT GAG GATC5 3' TTC GTG AAG CGC TCT AGAT CTG TTG ATG GAT CAC CGA CTC CTAG
Nru I BglII Xba I

:
.
.
.: ' .

, . , ~ ' , ~ . :

-~2- ~325~

The synthetic oligonucleotide is thus inserted into the polyhedrin gene, and, by virtue of its encoded uni~ue restriction sites, m~kes more cleavage sites available for recombina~ion purposes. The clone thus generated contains 5 the AcMNPV polyhedrin gene ~rom the amino terminus to amino acid 37, the oligonucleotidP with unique NruI, BglII, and XbaI sites, and the polyhedrin coding sequence from amino acid 50 through the BamHI site at amino acid 58. Once this clone is generat~d, additional sequences can be inserted 10 into the unique NruI, ~II, or XbaI sites, that would restore the translational reading frame an~ encode antigenic determinants of foreign proteins, such as an epitope of a pathogenic microorganism. As one example, cloning the following oligonucleotide:

Tyr Pro Tyr Asp Val Pro Glu Tyr Ala 5' - CG TAT CCG TAC GAT GTA CCG GAT TAC ~CT
3' - GC ATA GGC ATG CTA CAT GGC CTA ATG CGA GATC

into the NruI and XbaI sites would generate a gene fusion in 20 which the influenza hemagglutinin epitope extending from amino acid 98-106 would be inserted in frame between amino acids 3~ and 50 o the Autographa polyhedrin gene.
Alternative strategies for inserting the foreign oligonucleotide include, but ar~ not limited to, insertion 25 Of a BqlII site at amino acid 43 by changing nucleotide 127 o~ the Autographa polyhedrin gene from a G to a T by in vitro mutagenesis ~Xunkel, 1985, Proc. Natl. Acad. Sci.
82:488-492; Hutchinson, C., et al., 1978, J. Biol. Chem.
253:6551). Synthetic oligonucleotides can then be inserted 30 between the unique BglII and BamHI sites. Thus, many gene fusions can be generated, adding new antigenic determinants and deleting as much of the region between the BglII and the BamHI sites as desired. A DNA synthesizer (e.g., Applied Biosystems Model 380A) may be used to gen~rat~ a large 35 variety of constructs. A methsd for preparing this type of ~onstruct is described in detail in Section 11 inra.

. : .

: . !
. ' ' ., -43- 1 3 ~

Similar strategies may be used to recombine other regions of the polyhedrin gene. In vitro mutagenesis follow~d by insertion of synthetic polylinkers can enable manipulation of most, if not all, of the regions of the 5 polyhedrin gene.
In another particular embodiment of the invention, a strategy can be employed to in~ert foreig~ DNA within the polyhedrin gene at a particular restriction site, which may or may not be a unique restriction site. In this 10 embodiment, single-stranded DNA of the polyhedrin gene of a recombinant vector is manipulated to producs ~ite-cpecific cleavage at a speciflc restriction site. (For an example using thi~ strategy, see Section lO.l., infra3. Single-stranded DNA from the polyhedrin gene i~ isolated. This can 15 be accomplished by many standard techniques such as heat-denaturation of the double-stranded ~orm Pollowed by fractionation, or pr~ferably, by i501ating the single-stranded DNA o~ a vector ~uch as a bacteriophage derivative ~e.g. an Ml3 phage, a phagemid) which contains the 20 polyhedrin DNA inserted within its genome. Specific cleavage at a par~icular restriction ~ite within the DNA is accomplished by annealing a complementary synthetic oligonucleotide (oligo-l) to the single-stranded DNA, before restriction digestion. This annealing creates the requisite 25 double-~tranded region for recognition and cleav~ge by the restriction endonuclease. A~ter cleavage, the single-stranded linear DNA can then be isolated by known techniques (e.g. heat denaturation and column chromatography). An oligonucleotide with a seguence encoding a Por~i~n epitope 30 can also be ~ynthesized (termed hereinafter oligo 2~.
Another oliqonucleotide can then be ~ynthesized (termed hereina~ter oligo 3) which i~ complementary to oligo 2 and which, in addition, has 5' and 3' termini which extend beyo~d oligo 2 which are co~ple~entary to the single-35 stranded termini of the polyhedrin DNA. Oligo 2 and oligo 3can then be annealed together, followed by ligation of the duplex to the ~ingle-~tranded polyhedr~n DNA.

, . .

1 3 2 ~
Transformation of a suitable vec~or host 6uch as E. coli will produce a recombinant transfer vector which contains the DNA encoding a foreign peptide inserted at a specific restriction site within the polyhedrin gene.
Irrespective of the manner of constructing the recombinant polyhedrin gene, transfer of the gene fusion into the baculovirus can be accomplished by homologous recombination ln vivo between viral DNA and DNA sequPnces containinq the fusion. In a pre~erred embodiment, such DNA
10 sequences are contained in a transfer vector such as a plasmid (Pennock, G., et al., 1984~ Mol. Cell. Biol. 4:399;
Smith, G., et al., 1983 , MG1 ~ Cell. Biol. 3:2156). The transf~r vector can be constructed to contain the heterologous gene inserted within the polyhedrin gene 15 sequence and flanked by baculoviral sequences adjacent to the viral polyhedrin gene. This can be accomplished by DNA
recombination involving the use of standard techniques in molecular biology (Maniatis, T., et al., 1982, Nolecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, 20 New York). Parental baculoviral DNA plu5 transfer vector DNA can be cotransfected into cells susceptible to infection, where in vivo recombination will then take place, producing the recombinant virus of the invention. The transfections can be accomplished by any procedures known in 25 the art, including, but not limited to, the calcium-phosphate precipitation method (see, for example, Smith, G., et al., 1983, J. Virol. 46:~8~), treatment with polybrene an~ dimethyl sul~oxide (Kawai, S. and Nishi~awa, M., 1984, Mol. Cell. Biol. 4:1172), or electroporation (e.g., Kuta, 30 A.E., Rhine, R.S. and Hebner, G.M., 1985, ~Electrofusion-A
New Tool for Biotechnology~, Amer. Biotech. Lab. 3:31-37).
In another embodiment, a cassette vector can be constructed which comprises a polylinker sequence inserted within a region of the polyhedrin gene that is nonessential ~or 35 crystallization. This recombinant polyhedrin gene can then be transferred to a baculovirus by in vivo recombination within baculovirus-in~ected cells, producing a ncassette-~45~ ~32~
expression" virus. The genome of this cassette-expressic?n virus can be isolated for in vitro recombination purposes, in which insertion or replacement o~ polyhedrin regions which are nonessential for crystallization, by a 5 heterologous sequence, is ~acilitated by virtue of the polylinker.

5 .1. 4 . SELECTION OF RECOMBINANT
OCCLUSION BODIES
As explained _upra, baculoviruses that produce 10 recombinant occlusion bodies may be constructed via recombination by replacing or interrupting regions of the polyhedrin gene sequence, that are noness~ntial for crystallization, with the heterologous gene sequence so that the sequences are not interrupted by translational stop signals. Since the gene products of these recombinants will be expressed as recombinant occlusion bodies, it is preferred to use an OB- parent baculovirus strain, in order to select OB~ virus plaques against an OB- background.
Viruses generating OBs can be detected in plaque assays 20 among the large number of parental viruses which ~ail to make OBs, since OB+ viruses form more refractile plagues than OB- viruses. Alternatively, an OB+ background would be preferred where the recombinant OBS form plaques that are less refractile than those formed by wild type viruses. For 25 instanc~, the recombinant OBs expressed by InHem-43 and InHem-50 detailed in the Examples, infra (see Section 11 et seq) are morphologically very different from wild type occlusion bodies. These recombinants were found to produce cuboidal occlusion bodies that express the foreign epitope.
30 The cuboidal recombinant OBs, ~ormed placgues which were less refractile than those produced by wild type virus.
Therefore, the recombinant OBs were easily selected against an OB~ background.
Selection can also be done on the ba~is o~ physical, 35 immunological, or functional properties of the inserted heterologous gene product. For example, an enzyme-linked , .
? ~ ` : . ' .

, ' . '' ' ' ' ' ~

-~6- ~32~

immunosorbent assay (ELISA~ can be used to detect expression of a foreign antigenic determinant. I~ a heterologous gene has been incorporated that encodes an enzyme, selection may be done on the basis of enzymatic activity. Staining 5 technigues based on chemical reactivity of the foreign peptide may be used. Many other techniques known in the art can be used, depending on the foreign sequence expressed, and are within the scope of the invention.
While attempting to select recombinant viruses that 10 produce recombinant occlusion bodies, whether the seleckion i5 against an os- or o~+ background, it is possible to employ other markers which aid in selection. For example, a second gene, encoding a selectable marker, can also bP
introduced into a region of the baculovirus genome which is not essential fo~ crystallization. Prior to transfer to the baculovirus genome, the selectable marker may exist as a totally distinct DNA fra~ment or, preferably may be contained in adjacent sequences of the transfer vector containing the recombinant polyhedrin gene. The selectable 20 marker should be cotransfected with the recombinant polyhedrin gene into the baculovirus where in vivo recombination will occur. OB+ recombinants which also express the selectable marker can $hen be selected. Many cloned genes known in the art can be used as the selectable 25 marker, including, but not limited to, genes encoding enzymes such as beta-galactosidase, which have standard procedures for selection.
Another method for selection is to screen for presence of the heterologous DNA sequence inserted into the 30 polyhedrin gene. This can be accomplished by techniques known in the art, such as nucleic acid hybridization to replica plaques (Ben~sn, W.D. and Davis, R.W., 1977, Science 196:180), and variations thereof Another technique known in the art which may be used 35 for selection is loss of a marker gene activity through inactivation or replacement of the marker gene. In this embodiment, parental baculoviruses can be constructed which ' . .. , " ' ' ' 1 3 2 ~
contain a selectable marker flanked by sequences homologous to those surrounding the racombinant polyhedrin. For example, a parental baculovirus can be constructed containing a beta-galact~sidase gene downstream of the 5 polyhedrin promoter. In vivo recombination between the recombinant polyhedrin gene and the construicted parental strain will result in insertion of the recombinant polyhedrin gene by virtue of its homologies with the parental polyhedrin sequences ~urrounding the beta-10 galactosidase qene. The recombination which results ininsertion of the rec~mbinant polyhedrin gene and inactivation or replacement of the b~ta-galactosidase gene may be selected for by the lack of beta-galactosidase activity by known methods (Messing, J., et al., 1977, Proc.
Natl. Acad. Sci. U.S.A. 74:3642). This selection can be accomplished against an OB or OB+ background as previously described.
A control selection experiment which may be done is to cotransfect with wild-type ~iral DNA (OB+) in order to 20 detect, among the wild-type progeny, recombinants failing to make OBs. Although the identification and characterization of the recombinant generated in this type of cotrans~ection represents a negative result, it will provide valuable information regarding what modifications of the polyhedrin 25 gene interfere with lattice formation and are therefore unsuitabl~ for the practice of the present invention.

5.1.5. VERIFICATION OF EXPRESSION OF
FOREIGN EPITOPES ON OR WITHIN THE
RECOMBINANT OCCWSION BODY
After selection for recombinant OBs, the OBs should be isolated (see, for example, Section 7.1.4., infra and Tweeten, K.S., et al. 1981, Microbiol. Rev. 45:379-408) and analysed for the expression of the foreign epitope.
OBs can be purified by any standard technigue (for 35 example, see Section 7.1.4., infra, and Tweeten, K.A., et al., 1981, Microbiol. Rev. 45:379-408). In one embodiment, . .' . ` ' ~32~a monoclonal antibodies directed against the recombinant polyhedrin protein can be used as an effective means of purifying the polyhedrin proteinO As an example, isolated preparations of recombinant OBs can be solubilized, the 5 recombinant polyhedrin protein purified by immunoaf~inity chromatography ~Goding, J.W., 1983, Monoclonal Antibodies:
Principles and Practice, Ch. 6, nAffinity Chromatography Using Monoclonal Antibodiesn, Academic Press, Inc., London, pp. 188-207) and then recrystallized to form a purified 10 preparation of recombinant OBs. This method requires an antigen that remains capable of binding to antibody e~en after the stringent condition~ necessary for crystal dissolution.
The foreign gene product can be analyzed by assays 15 based on physical, immunological, or functional properties of the product. Immunological analysis is especially important where the ultimate goal is to use the recombinant OBs that expres6 the product in vaccine formulations and/or as antigens in diagnostic immunoassays. Antibodies to the 20 peptide, preferably monoclonal, can be tested for their ability to interact with the crystalline recombinan~
polyhedrin. ~his can be a~complished by various tPchnigues known in the ark including, but not limited to, an enzyme-linked immunosorbent assay (ELISA) method (for example, a 25 solid-phase binding assay on polyvinyl chloride plates), or a radioimmunoassay. Methods known in the art such as western blottinq or immunoprecipitation procedures can be used to determine the presence of the peptide on the polyhedrin monomer.

5.2. VECTOR/HOST SYSTEMS
Any baculovirus may be used as the parent for construction of the recombinant baculoviruses of the present invention. These vectors include but are not limited to 35 NPVs and GVs. For example, NPVs which may be used in accordance with the present invention include but are not limited to AcMNPV, HzSNPV, Heliothis virescens NPV, S.
littoralis NPV, Rachoplusia ou MNPV, Galleria mellonella . . .,.. ~ . ~ , .

4~ 5~

MNPV, Lymantria dis~ar MNPV, Bombyx mori SNPV, Orygia pseudotsugata SNPV and MNPV, Orygia leucosti~a NPV, Choristoneura fumiferana MNPV, Pseudohazis eglanterina SNPV, N. sertifer SNPV, To paludosa SNPV, Trichoplusia ni MNPV, 5 and Spodoptera frugiperda MNPV (Vlak, J.M. and Rohrmann, G.F., supra). GVs which may be used in accordance with the present invention include but are not limited to P.
brassicae GV, Esti~nene acrea GV, Choristoneura vindis GV, Plodia interpunctella GV, Choristoneura vindis GV, Plodia 10 interpunctella GV, T. ni GV, Choristoneura murinana GV, Cirphis unipuncta GV, L. pomonella GV, Cydia ~omonella GV, Mamestra oleracea GY, Pseudaletia unipuncta GV, Pygera anastomosis GV, S. frugiperda GV, Zeiraphera diniana GV, and Choristoneura fumiferana GV (Vlak, J.M. and Rohrmann, G.F., 15 su~ra; Tweeten, K.A., et al., 1981, Microbiol. Rev. 45:379-408).
In a preferred embodiment, a pla~e-purified isolate with a homogeneous genotype should be used as the parent baculovirus. Moreover, a recombinant baculovirus can be 20 constructed from parent viruses which possess particularly advantageous properties with respect to the host systems used in accordance with the present invention. For example, viruses which demonstrate high infectivity and hiqh virus titers in the host system are preferred.
When using larvae host systems, viruses which do not cause melanization are preferred. Melanization is a normal response to viral infection which comprises the production of melanin, a pigment which is incorporated into the insect's cuticle, and appears to involve the polymerization 30 of indol rin~ compounds derived by oxidation of tyrosine (Wigglesworth, V.B~, 1974, in The Principles of Insect Physiology, Chapman and Hall, London, p. 610). The tyrosinase which is involved in the melanization process appears to be abundant in the hemolymph of the insect and 35 can react fairly non-specifically with available proteins.
Thus, the tyrosinase activity in an insect carrying the recombinant baculoviruses o~ the invention may non-: . , ~32~

specifically metabolize the recombinant polyhedrin protein, inter~ering with and decreasing the yield and purity of the recombinant product. Melanization o~ occlusion bodies can cause subsequent chemical alkeration of virion proteins and 5 nucleic acids. Melanization can al o severely reduce infectious extracellular virus titers in collected hemolymph, as well as poison cultured c~lls followin~
inoculation. Thus non-melanizing or slow-melanizing host strains are preferred in order to avoid these problems.
In the discussion that follows, parent vectors and cell lines are discussed in terms of Heliothis zea SNPV and Heliothis zea cell lines. It should be noted that this discussion is for descriptive purposes only and the scope of the invention includes many other baculoviruses, such as GVs 15 and other NPVs.
In a specific embodiment, Heliothis zea SNPV may be used as the parent virus strain. Restriction digestion patterns of eight different geographi~ isolates of HzSNPV
suggest each is a separate population of viruses having a 20 slightly different predominant genotyp~, but none represents a totally unique virus species (Gettig and McCarthy, 1982, Virology 117:245-252). Seven of the eight geographical isolates examined have similar major occluded virus structural protein profile in SD~-polyacrylamide gel 25 electrophoresis (SDS-PAGE) (Monroe and ~cCarthy, 1984, J.
Invert. Path. 43O32-40). Even though the eight Heliothis SNPV isolates are genetically and biochemically similar, several isolates exhibit significant differences in virulence towards H. zea larvae (Gettig and McCarthy, 1982, 30 Virology 117:245-252).
We have thus far analyzed plaque purified strains derived from the Elcar~ isolate of HzSNPV (originally isolated by Dr. J.J. Hamm, U.S.D.A., Tifton, GA). We characterized the genotypic and phenotypic heterogeneity of 35 the Elkar~ isolate by comparing restriction enzyme -51- i32~61 0 digesti~ns of viral genomes, SDS-PAGE pr~files of occluded virus structural protein~, and difference~ in larval pathology among the plaque-puri~ied stra~s.
After purifying and analyzing 20 ~trains from the 5 Elcar~ isolate we found no single predominant genotype.
Each strain could be distinguished using D~e or more restriction enzymes, and none was identical to the molar restriction pattern of the wild-type isol~te. The inability to identify a predominant genotype indica~ed that this virus 10 is highly variable.
We have localized a major region ~ variability to between 23.4 and 43.4 map units. ~This r~yion includes HindIII fragments G, H, M, and N ~see FIG~ 6). At least 15 of the twenty plaque-purified strains di~rge from the 15 wild-type strain in this region, with alt~rations in onP or more of these HindIII fragments.
When working with HzSNPV i~ larva~ host systems it is advantageous to time the in~ction carefu~ly to avoid melanization of either virion containing hemolymph or 20 occlusion bodies, and to use a s10w or n~~melanizing strain of virus. Since melanization is a const~n~ problem when studying the biochemistry of HzSN~V, the -~n-melanizing strains described herein are preferr~d f~r use in larval host systems in this specific emb~iment ~ the invention.
25 In particular, strains 5, 7, 8, ~" ~1, 2~, 24 an~ 25 described infra (See Section 7, i~ partls~ar, Table II) cause ~low melanization. In ~ pr~rred ~mbodiment, strain HzS-15, (also described in ~ti~ 7 in-~r~3 which causes extremely slow melanization ~an b~ ~sed ~ the parent strain 30 for constructing recombinant Hel.i~s ~3~ses for use in larval host systems~
The genotype map ~f the H~3-~5 stx~in ~see Figs. 7 and 8) differs from the ~ild-typ~ m~p (~ell and Su~mers, 1984, J. Gen. Virol. 65:4A~-450) ~ the ~pervariable 35 region. The divergence .3 ~ t~s ~in ~ ~vident with the enzymes EcoRI, HindIlI, and ~stI ~ompa~lson o~ the maps of HzS-15 with those ~ the wil~-~y~ ~sol~ shows that the . .

-52- 13 2~

divergence is caused by changes in the relative positions of several restriction sites, rather than in ov~rall genome size or apparent organization.
The region of hypervariability in HzSNPV is not 5 restricted to the Elcar~ strain. Rsexamination of ~ettig and McCarthy's (1982) study of geographic variation reveals that this window of hypervariability exists in other Heliothis species SNPVs. In addition, many Hi~dIII
fragments that are consarved among our plaque-purified 10 isolates are also conserved among the previously analyzed geographic variants. This is especially true of HindIII
fragments A, B, C, L, M, and 0, all of which appear in 7 out of 8 geographic variants and in 13 out of 20 of the present isolates. Whether or not thP high degree of variability 1~ found in the Heliothis spp. SNPVs confers advantages to the virus population under different geographic conditions remains to be determined (Gettig and McCarthy, 1982, Virology 117:245-252).
The restriction enzyme map of HzS-15 (Figs. 7 and 8) 20 substantially confirms the previous map of Knell and Summers (1984, J. Gen. Virol. 64:445-450). While we found no evidence of major errors in the position of SstI or BamHI
restriction sites, we did find that o~e alteration of band position was necessary in the SstI map of the wild-type 25 isolate. Band hybridization was not employed in this analysis, but we were able to confirm the relative position of most restriction ~iite~ with a high degree of certainty using cloned PstI or BamHI fragments.
Working with the HzS-15 plaque purified isolate, we 30 obtained a slightly different estimation of the overall genomic size. Our estimate o~ genome size based upon double digests and analysis of individual cloned fragments is approximately 131 kb, rather than the 120 kb previously reported (Rnell and Summers, 1984, supra). Sinc~ we found 35 little difference in the restriction maps outside of -53~ ~32~

alterations in posi~ion o~ restriction sites, we feel that our estimate is a closer approximation of HzSNPV genome size.
The variability between isol~tes is not limited to 5 the genotype, but is also refle~ted in the structural proteins of the virions (see FIG. 9). we observed differences between isolates in several of the occluded virus proteins. In fact, there were more differences between these plaque-puri~ied isolates from the single 10 Elcar~ strain than were observed between several different geographical isolates of HzSNPV (Monroe and McCarthy, 1984, J. Invert. Path. 43:32-40).
The exact reason for the di~ferences in rate of melanization may be related to the relative ability o~
15 individual strains to lyse cells, or in some tissue tropism.
Evidence that cell lysis may be responsible for the differences in larval melanization response comes from freeze-thaw experiments with larvae infected with the non-melanizing strain, HzS-15. Unmelanized HzS-15 in~ected 20 larvae will quickly melanize following freezing and thawing.
Preliminary cell culture data also supports this hypothesis, but further work is reguired to con~irm that cell lysis i5 the predominant factor.

5.2.l. HOSTS USED IN THE VECTOR/HOST SYSTEMS
The recombinant baculoviruses of the present invention can be used to direct the expression of the heterologous gene product in a number of host systems including but not limited to cell lines and larvae in which 30 the virus can be propagated. Some useful cell lines and larval systems which can be used in accordance with the invention are described in the subsections below.

5~2.l.l. INSECT CELL LINES
Any insect cell line in which the baculovirus can be ' propagated can be used in the procedures of the present invention. Such cell lines inclu~e but are not limited to , .' : ~ . .

, -54- ~32~

IPLB-SF-21AE (SpodoPtera frugiperda cells): TN-368, BTI-TN~BI, BTI-TN5F2, BTI TN5F2P, BTI-~N5F2A (Trichoplusia nl cells: Granados, R.R., et al., 1986, Virology 152:472~476);
ILPB-HZ1075 (Heliothis zea cells; Goodwin, R.H., et al., 5 1982, In Vitro Cell. Dev. Biol. 18:843 850); BCIRL HZ-AMl,2, or 3 (Heliothis zea cells; McIntosh, A.H. and Ignoffo, C.M., 1981, J. Invert. Pathol. 37:258-264); BCIRL-~V-AM1 (Heliothis virescens cells; id.); BTI-EAA (Estigmene acrea cells); Mb 0503, Mb 1203 (Mamestra brassicae cells) 10 (Miltenburger, ~.G., et al., 1976, Z. Angew. Entomol.
82(3):30~-323); and cell lines derived from these lines, etc.
For an informative discussion of the in vitro replication of baculoviruses, see Volkman, L.E. and Knudson, 15 D.L., 1986, ~In Vitro Replication of Baculoviruses, in The Biology of Baculoviruses, Vol. I, Biological Properties and Molecular Biology, Granados, R.R. and B.A. ~ederici, eds., CRC Press, Florida_~k~ or~to~ by re~rcncc ~.
In the discussion which follows, cell lines for use in a specific embodiment of the invention involving ~zSNPV
and Heliothis cell lines is described, which is intended for descriptive purposes only, and in no way limits the scope of the invention.
The ln vitro propagation of must SNPVs has been difficult to achieve. The IPLB-HZ1075 cell line, which was originally established by Goodwin (Goodwin, R.H., et al., 1982, In Vitro Cell. Dev. Bio. 18:843-850) from larval ovaries and fat body of the cotton bollworm, Heliothis zea, 30 can support the growth of HzSNPV. However, numerous reports have demonstrated that 100% infection is difficult to routinely achieve with this system (Granados, R.R., et al., 1981, Intervirology 16:71~79; Yamada, K., et al., 1982, J.
Invert. Path. 39:185-I91). It is more common to obtain only 35 50 to 70% inf~ction as measured by the presence of intranuclear OBs; thus, the usefulness of this system for analysis of HzSNPV replication procasses is severely -55- 132~

limited. In an e~fort to increase the productivity of HzSNPV in vitro, some investigators have established new cell lines from Helio~his zea (Goodwin et al., supra;
McIntosh, A.H., et al., lg85, Intervirology 23:150-156).
5 McIntosh and Ignoffo (1981, J. Invert. Pathol. 37:258-264) have demonstrated replication of HzNPV, with production of OBs, in cell lines derived from ~eliothis zea or Heliothis virescens.
During routine subculturing of the HZ1075 cell line (Goodwin et al., 19~2, In Vitro Cell. Dev. Bio. 18: 843 850), we fre~uently observed clonal outgrowth of cells having similar morphologies. This observation led us to consider the possibility that this cell line is highly heterogeneous, and that perhaps not all cells of the population were equally susceptible to in~ection with HzSNPV. We reasoned that isolation and characterization of subclones might provide a cell strain which would be more susceptible and produce more occlusion bodies upon infection with HzSNPV.
In the specific embodiments of the examples herein, we describe the isolation and characterizakion of clonal cell strains derived from the IPLB-HZ1075 insect cell line which can be used in the practice of the present invention.
These strains exhibited different growth characteristics, 25 morphologies, and productivities of HzSNPV, which are defined herein. We also utilized isozyme markers to characterize the cell strains and demonstrate that they are all derived from Heliothis zea.
The IPLB-HZ1075 cell line consists of a heterogeneous 30 population of cells, and this heterogeneity seems to account in part for the inability to obtain 100% infection upon inoculation with HzSNPV. To obtain a more homogeneous response to infection through cloning of individual c~ll strains, we subcloned and characterized twelve strains from 35 IPLB-HZ1075 using dilution plating.

, ~ ~ .

. 5~ 132~
All of the subcloned cell lines were similar to the parental cell line in that none of them were capable o~ 100%
infection upon inoculation with the HzSNPV isolate HzS-15 under our culture conditions. Each cell strain exhibited 5 slightly different growth kinetics ~see FIG. 8), predominant cell morphology, and ability to replicate virus (see Table VI infra) While predominant cell morphologies were different for most of the cell lines, only two (UND-B and UND-G) 10 exhibited a fairly homogeneous cell morphology. The others were composed of several different morphologies even though they originated from clonal populations. All of the cell strains were fibroblasticl sîmilar to the parental cell line (Goodwin, R.H., et al., 1982, In Vitro 18:843-8503.
Population doubling times for the cell strain~ varied widely between 37.33 and 65.48 hours, and the parental cell line had a doubling time near the maximum (63.15 hours).
Routine subculturing of the parental cell line does not appear to select for cells with similar growkh kinetics.
20 Perhaps the mixed cell population regulates the growth of individual cell strains in some way, possibly by suppressing the growth of faster growing strains.
There was an equally wide variability in OB and ECV
production upon infection, and no apparent correlation 25 between any of the measured parameters. OB production could not be linked to either ECV titers or relative cell growth rates. If the results were influenoed by the low multiplicity of infection, we might expect some correlation between OB or ECV production and cell growth rates.
30 Analysis of the fast growing strains did not reveal any consistent pattern of either higher or lower OB or ECV
production relative to the slower growing strains. In addition, we did not observe significant cell growth in any of the strains during the 10 day infection period.
35 Consequently, we feel that the estimates do reflect the ~57~ ~32~10 relative productivities of the cloned cell strains for either OBs or ECVs, but may not accurately reflect the overall relative susceptibility of the cell strains for HzSNPV.
SeYeral of the subcloned populations seemed to replicate HzSNPV better than the parental IPLB-HZ1075 cell line with respect to both OB (UND-B,C,G,K,M,R) and ECV
production tUND-B,C,F,G,K,L,M,o,R,V). This result differed from earlier stud.ies with Trichoplusia ni cell line TN-368 10 (Faulkner, P., et alO/ 197S, In Invertebrate Tissue Culture Applications in MedicinP, Biology, and Agriculture, Kurstak, E. and K. Maramorosch, eds., Academic Press, New York, pp.347-360; Volkman, L.E. and Summers, M.D., 1976, In Invertebrate Tissue Culture Applications in Medicine, 15 Biology, and Agriculture, supra, pp. 289-2g6~. Subcloned cell strains of T~-368 were distinguishable from the parental cell line with respect to c~ll doubling time (Faulkner, P., et al., supra) and ability to plaque Autographa californica MNPV (AcMNPV) (Volkman, L.E. and _ 20 Summers, M.D., supra), but none of the TN-368 subcloned cell strains replicated Ac~NPV better than the uncloned parental cell line.
Confirmation that all cell strains were derived from the IPLB-HZ1075 cell line came from comparisons of isozyme 25 patterns (see FIG. 9). We also compared isozyme patterns of several invertebrate cell lines (see FIG. 10) and found that all could be separated unambiguously using the enz~mes MDH
and LDH in tandem. This is in contrast to earlier reports that were unable to separate IPLB-SF21AE and IPLB-HZ1075 30 using a large number of enzymes and two different gel systems (Brown, S.E. and Knudson, D.L., 1980, In Vitro

16:829-832; Taba~hnik, W.J. and Knudson, D.L., 1980, In Vitro 16: 389-392).

~ ` ~.' ~' ' ,' ,' ' ' ' ` ' ' , -58- 1325~

In a specifi.c embodiment, the IPLB-HZ1075 UND-K cell line may be pr~ferred for use in the expression vector/host systems of the present invention because o~ its ability to grow Heliothis virus quickly and at high titers which 5 plaque, thus enabling identification.
In a preferred embodiment, the growth medium of any IPLB-HZ1075 cell line used as a host in accoxdance with the invention should contain 1% bovine serum albumin and 2 g/liter L-glutamine in order to improve infectivity to about 10 100%. When using cultured cells as hosts in the expression/host syst~ms of the present invention, it is preferred to infect the cell cultures with ECVs rather than OBs. The infectious oell culture supernatant can be stabilized by the addition of liquid agarose to a final concentration of 0.1%. Alternatively, virions can be isolated from the OBs according to procedures known in the art such as the technique described by Smith and Summers, 1978, Yirology 84: 390-402, and the modified procedure described herein in Section 7.1~4. 'nfra.

5.2.1.2 hARYA HOSTS
Baculoviruses expressing the recombinant polyhedrin genes of the present invention can be propagated and/or mass-produced by infection of various host insect larvae.
25 The propagation and isolation of baculoviruses using laboratory larval populations has been previously described (e.q., Wood, H.A., et al., 1981, J. Invertebr. Pathol.
38:236-241; Ignoffo, C.M. and Garcia, CO~ 1979, EnvironO
Entomol. 8:1102-1104). Larva hosts which may be used in the 30 propa~ation and production of viruses expessing recombinant polyhedrin genes include but are not limited to those species listed in Table I, infra.

~ ~ .

-59~-~32~
-TABLE I
INSECT L~RVA SPECIES WHICH CAN ~E USED FOR
THE PROPAGATION AND PRODUCTION OF VIRUSES EXPRESSING
RECOMBINANT POLYHEDRIN GENES OF THE PRESENT INVENTION

Heliothis zea (Boddie) Trichoplusia ni (Huber) Galleria mellonella Spodoptera frugiperda Esti~mene acrea -Aedes aegypti Chori~toneura fumiferana Heliothis virescens Autographa californica S. littoralis Rachoplusia _ Lymantria dispar Bom yx m _ ory~ia pseudot~ ata Pseudohazi~ eglanterina N. sertifer _ paludo6a P. brassicae Oryg'a leuco~tigma Choristoneura indis Plodia interpunctella Choristoneura murinana Cirphis unipuncta L. ~omonella Cydia pomonella Mamestra oleracea Pseudaletia unipuncta Pygera anastomosis Zeiraphera diniana .

~.
,, .
.:
. .

.- . - ~ . : .

132~6~
In particular embodiments, T. ni or G mellonella larvae can be used for the propagation and production of recombinan~ AcMNPV, G. mellonella MNPV, or T. ni MNPV, while H. zea can be used to support the growth of recombinant 5 HzSNPV.
Any rearing conditions and diet formulations can be used which support the growth and maintenance of the larvae.
One example of a diet mix which can be used to support the growth of T. ni or H. zea larvae is described in Section 10 9.1. in~ra. Examples of rearing conditio~s which can be used for H. zea, T. nl, or G. melonella are described infra, in Sections 9.2.1., 9.2.2., and 9.3. It is possible that the larval strains are cannibalistic ~ , Heliothis zea) and, therefore, cannot be grown all together. It would 15 therefore be preferable to separate the larvae so that only one or two insects are dispensed into each container ~or growth .
It is preferable, but not reguired, to maintain the larval cultures in a germ free environment. The cultures 20 thus maintained would be free ~rom the presence of exogenous microorganisms which can potentially produce substances toxic or allergenic for humansO
In a futher preferred method of the invention, insect larvae can be cultured free of both exogenous and endogenous 25 microorganisms. 5ince Lepidopterans (~ , Heliothis zea, Trichoplusia ni) contain no endogenous symbiotic microorganisms, they can be maintained in the absence of both endogenous and exogenous microorganisms, thus eliminating the danger of contamination by microorganisms 3~ pathogenic for humans. In accomplishing and maintaining these germ-free conditions, the insect eqgs can be sterilized (e.g. by treatment with peracetic acid; see Section 9.3., infra). The insect diet mix can be sterilized, e.~. by the use of radlation .

~32~
In addition, as dis~ussed in section 5.2~, supra, a non-melanizing strain of virus is preferred for use, in order to optimize yield and purity of the recombinant polyhedrin obtained from the infected larvae.
In another embodiment of this aspect of the invention, it is pos~ible to produce and use giant larvae for the propagation of the recombinant baculoviruses of the invention. Selective inhibition of juvenile hormone (JH) esterase has been shown to result in the maintenance of JH
10 titers and in the production of giant larvae (Sparks, T.C., et al., 1983, Insect Biochem. 13:529; Hammock, B.D. and Roe, R.M., 1985, Meth. Enzymol. lllB:487). The use of such larvae in the mass production of the recombinant polyhedrins of the invention can greatly increase the obtained yields.

5.2.2. EXPRESSION IN OTHER MICROORGANISMS
The recombinant polyhedrin genes of the present invention can also be expressed in vector/host systems involving other microorganisms including but not limited to 20 other viruses such as vaccinia viruses, adenoviruses, retroviruses, etc., yeast; and bacteria.
As ~ne embodime~t, the production of recombinant polyhedrin crystals in bacterial cells has a number o~
attractive advantages. It would eliminats the need to 25 transfer ~ene fusions into a baculovirus and to identify and characterize the resulting recombinant virus. In addition, it is cheaper and easier to grow large quantities of bacterial cells than to culture insect cells. Many different strains of bacteria and types o~ plasmids known in 30 the art can be used in this embodiment of the present invention, as long as the host allows for appropriate expression of the recombinant polyhedrin gene of the vector.
Production in bacterial expres~ion systems can be accomplished by use of the same type of genetic 35 manipulations as described in Section 5~1.3., supra. For example, by taking advantage of the degeneracy of the genPtic code, a polyhedrin gene segment can be synthesized ~ .: . . . : , . .
; . . - . . .
.

:, ~ . . ', ' ~, ': ' -62- ~32~

which contains new and unique re~triction sites, yet encodes the same amino acids as the wild~type polyhedrin gene.
Thus, a ~polyhedrin polylinker~ seguence is created, a type of cassette vector, which can be ligated to the remainder of 5 the parental polyhedrin genet and which can be utilized to insert sequences encoding foreign epitopes at its unique restriction sites. Such a gene construction provides the potential to easily enyineer a large number of changes into the polyhedrin gene. In addition, the gene construction can 10 be designed with flanking restriction sites suitable for insertion into E. coli expression vectors. Insertion of the polyhedrin polylinker sequence into an E. coli expression vector will produce a cassette-expression vector which can greatly facilitate construction and expression, in bacteria, of a recombinant polyhedrin gene of the present invention.
Such a construction is not restricted to use in ~. coli; it can also be engineered for use in the baculovirus or other systems. One example of a polyhedrin polylinker is shown in Figure 4. Figure 4 shows a gene segment encoding the amino 20 terminus to amino acid 58 (at the BamHI site) of the Autographa polyhedrin protein, which also contains new PvuI, ScaI, BclI, and XbaI sites at positions corresponding to amino acids 9, l9, 27, and 46, respectively. Ligation of a two kilobase pair BamHI fragment containing the 3' end of 25 the AcMNPV polyhedrin gene will reconstruct the entire gene.
The unigue restriction sites can facilitate the replacement oP small regions in the 5' ~ection of the gene with synthetic oligonucleotides encoding new antigenic determinants. In particular, replacing the segment between the XbaI site and eith2r the BamHI or BclI site enables the - _ insertion of new determinants in a putatively modifiable region between amino acids 37-49O The ~I site can be used to add determinants to the amino terminus of the protein.
~he unique EcoRI site immediately 5' to the synthetic gene 35 permits cloning the fusion gene into an EcoRI site of an Eo coli expression vector. For example, in an embodiment involving the synthetic gene o~ Figure ~, a vector which may - .

':

-63- ~32~

be used is the E. coli expression vector PK223-3 (Pharmacia). This plasmid contains the tac promoter (de Boer, H.A., et al., 1983, Proc. Natl. Acad. Sci. U.S.A., 78:21) and Shine-Delgarno sequences upstream oP unique 5 cloning sites. In addition, there are strong ribosomal termination sequences downstream of the cl~ning sites.
Thus, the PK223-3 plasmid construction would permit efficient regulated expression of genes inserted at its cloning site. Numerous other plasmids with suitable cloning 10 sites and signals for expression may also be used.
The preceding discussion is intended only as an example of the types of manipulations and constructions which may be employed in the cloning and expression of recombinant polyhedrins. Other vectors, hosts, and 15 synthetic gene sequences may be manipulated in ~imilar fashions to express the recombinant polyhedrins of the present invention~ Appropriate cassette vectors, transfer vectors, and/or cassette-expression vectors can be constructed and used to facilitate the appropriate 20 recombinations.
Construction and expression in a suitable vector/host system will determine whether the recombinant polyhedrin expressed in such a system will crystallize. If the protein will not crystallize ln vivo, the solubilized polyhedrin 25 protein can be purified and crystallized in vitro (Shigematsu, H. and Suzuki, S., 1971, J. Invert. Pathol.

17:375-382). Thus by expressing the recombinant polyhedrin genes, polyhedrin crystals exposing new epitopes can be generated~

5.3. DETERNINATION OF THE IMMUNOPOTENCY OF
FOREIGN EPITOPES EXPRESSED ON OR WITHIN
RECOMBINANI OCCLUSION BODIES
Demonstration of i~munopotency ~f the epit~pe of a pathogenic microorganism, carried on or within a recombinant 35 occlusion body in accordance with the present invention, is a necessary step prior to vaccine formulation.

,. ~ , : . ~ ..

- .

~, . .. .

64~ ~32~10 Immunopotency of the foreign epitope expressed on ox within recombinant occlusion bodies can be determined by monitoring the immune response of test animals following immunization with the recombinant os. Occlusion bodies for immunizati~n 5 purposes ~an be obtained by purification from insects or insect cell cultures (for example, by the procedures of Section 7~1.4. lnfra, and Tweeten, K.A., et al., 1981, Microbiol. Rev. 45:379-408), or by in vitro recrystallization of polyhedrin (Shigematsu, H. and Suzuki, 10 S., 1971, J. Invert. Pathol. 17:375 382). TPst animals may include but are not limited to mice, rabbits, chimpanzees, and eventually human subje¢ts. Methods o~ introduction of the immunogen may include oral, intradermal, intramuscular, in$raperitoneal~ intravenous, subcutaneous, intranasal, or 15 any other standard route of immunization. The immune response o~ the test subjects can be analyzed by three approaches: (a) the reactivity of the resultant immune`
serum to the authentic pathogenic molecule or a fragment thereof containing the desired epitope, or to the isolated 20 naturally occurring pathogenic microorganism, as assayed by known techniques, e.gO enzyme linked immunosorbant assay (ELISA), immunoblots, radioimmunoprecipitations, etc., (b) the ability o~ the immune serum to neutralize infectivity of the pathogen in vitro, and (c) protection from infection 25 and~or attenuation of in~ectious symptoms in immunized animals.
In a specific embodiment of the invention, rabbits can be inoculated by a variety of protocols with recombinant AcMNPV OBs expressing amino acids 98-106 of the influenza A
30 virus hemagglutinin. Rabbit antisera reacting to tha recombinant OBs can be examined for cross-reactivity to the influenza hemagglutin monomer as well as to the homologous strain of influen7a virus, by he~agglutin A inhibition and by an~ibady titer determination. In protection exp~riments, 35 mice can be inoculated intraperitoneally with the , -65- ~32~

recombinant OBs, ~ubsequently challenged by intranasal inoculation of virulent virus, and monitored for the onset o~ disea~e symptoms.

5.4. USES OF RECOMBINANT OCCLUSION BODIES

5.4.l. VACCINES
Recombinant OBs expre~sing epitopes of pathogenic microorganisms are particularly useful in the formulation of 10 vaccines. In a preferred embodiment, the foreign epitope is exposed on the surface o~ the crystal. Since the crystalline lattice of the occlusion body is composed predominantly of the polvhedrin molecule, foreign epitopes within this molecule are presented a large number of times on the surface of the OB~ Recombinant OBs can be used in vaccine ~ormulations evan if the foreign epitope is not presented on the surface of the crystal but is internal, sinc~ alterations in crystallization proper~ies (e.g.
dissociation in vivo) can allow slow release of protein 20 (comprising the ~oreign epitope) from the crys~al, allowing presentation o~ the epitope to the host's immune system. In addition, the occlusion bodies are produced in large ~uantity, are stable ~tructures, and are easy to purify.
They can be generated in insect cell cultures that do not 25 produce known human pathogens. ~n a particular embodiment, recombinant OBs for vaccine us~ can be obtained ~rom infected insest larvae grown in a germ-~ree environment.
The use o~ recombinant OBs may be especially advantageous when the heterologous peptide or protein to be 3~ used in a vaccine formulation i5 a hapten (i.e., a molecule that is antigenic but not immunogenic) which ordinarily must be coupled to a carrier molecule that confers immunogeni-city. The production of recombinant OBs carrying the heterologous hapten on their surface using the expression 35 vector/host systems o~ the present invention would render the molecule immunogenic and ~liminat~ coupling reaction~.
Furthermore, the reco~binant OBs can be dissociated and , - ~ : - :

- -66- 13 2 ~ ~

recrystallized so that (a~ the enveloped recombinant virions can be removed from the solubilized recombinant OBs, which can then be recrystallized without the recombinant virus;
and/or (b) a mixture of recombinant OBs, each o~ which bears 5 a different heterologous protein, can be solubilized and recrystalliæed. The resulting OBs would bear each of the heterologous proteins and would be particularly useful as a multivalent vaccine. Alternatively, multivalent vaccines can be produced by engineering multiple epitopes into the 10 polyhedrin gene so that multiple epitopes are expressed on each recombinant polyhedrin molecule which ~orm the recombinant OBs.
In another specific embodiment of this aspect of the invention, the foreign peptide or protein to be expressed on 15 or within a recombinant OB may be amphipathic, that is, having one face hydrophilic and one face hydrophobic. Such a foreign peptide may be especially useful in the induction of T cell-mediated immunity. An amphipathic epitope may provoke T cell stimulation by providing for the interaction 20 of its hydrophobic face with the presenting cell membrane or Ia, and the interaction of its hydrophilic face with the T cell receptor (Allen, P.M., et al., 1984, Proc. Natl.
Acad. Sci. U.S.A. 81:248~; Berzofsky, J.A~, et al., 1985, ln Immune Recognition of Protein Antigens, Laver, W.G. and G.M.
25 Air, eds., Current Communications in Molecular Biology, Cold Spring Harbor Laboratory, New York, pp~ 156-160). In a particular embodiment, a heterologous sequence encoding an amphipathic alpha-helical structure can be insPrted into or replace portions of the polyhedrin gene which are 30 nonessential for crystallization and which encode alpha-helical regions, which may be hydrophobic.
Any protein epitope o~ a pathogenic microorganism which is capable of inducing an immune response ~pecific to the microorganism can potentially be used in a recombinant 35 OB vaccine formulation. Demonstration o~ the production of recombinant OBs which express the foreign epitope in an immunopotent state, as provided for by the present invention, is necessary prior to formulation as a vaccine.

':

~ ~ , '' ~

` -67~ ~325~1~

Potentially useful antigens for recombinant OB
vaccine formulations can be identified by various criteria, such as the antigen~s involvement in neutralization of the pathogen's infectivity (Norrby, E., 1985, Summary, In 5 Vaccines85, Lerner, R.A., R.M. Chanock, and F. Brown (eds.), Cold Spring ~arbor Laboratory, New York, pp. 388-389), type or group speci~icity, recognition by patients' antisera, and/or the demonstration of protective effects of antisera generated to the antigen. In addition, the antigen's 10 encoded epitope should preferably display a small or no degree of antigenic variation in time. The gene sequence encoding the epitope to be expressed on or within recombinant OBs may be obtained by techniques known in the art includinq but not limited to purification from genomic 15 DNA of the microorganism, by cDNA synthesis from RNA of the microorganism, by recombinant DNA techniques, or by chemical synthesis.
Recombinant OBs have potential uses as vaccines for diseases and disorders of viral, parasitic, and bacterial 20 origins. Many viral-specific antlgens are known and can potentially be incorporated into the recombinant OB vaccine formulations of the invention. For example, such antigens, and/or portions thereof which encode the epitope(s~, which may be used include but are not limited to in~luenza ~
25 hemagglutinin; Hepatitis A virus VPl; Hepatitis B surface, core, or e antigens; retroviral envelope glycoproteins or capsid proteins; poliovirus capsid protein VPl; rabies virus glycoprotein; foot and mouth disease virus VPl; Herpes simplex virus glycoprotein D; Epstein-Barr virus 30 glycoprotein; pseudorabies virus glycoprotein; vesicular stomatitis virus glycoprotein, etc. In a particular embodiment, the recombinant OBs of the present invention can comprise an epitope o~ the AIDS virus (HTLV-III/LAV/HIV) glycoprotein and/or capsid proteinsO Such an embodiment may 35 be particularly useful in vaccinating against AIDS without : . ..
:, :

-68- ~32~

concomitant induction of detrimental effects caused by the presence of the active AIDS virus glycoprotein such as the induction of T lymphocyte cell fusion and death.
Recent research has identified many potential 5 antigens of bacteria or parasites which may be forml7lated in vaccines in accordance with the present invention. For example, such antigens, or fragments thereof which encode the epitope(s), which may be formulated in vaccines in accordance with the present invention include but are not 10 limited to malaria antigens (Millér, L.H., 1985, In Vaccines85, Lerner, R.A., R.M. Chanock, and F. Brown (eds.), Cold Spring Har~or Laboratory, New York, pp. 1-5), cholera toxin, diptheria toxin, and gonococci antigens. As more specific examples, microbial genes which have been 15 successfully clonPd and may be used in recombinant OB
vaccine formulations include but are not limited to, enterotoxin genes o~ E. coli, the toxin and filamentous hemagglutinin genes of Bordetella pertussis, and the circumsporozoite (CS) antigen of the malaria parasite 20 Plasmodium falciparum (Norrby, E., 1985, In Vaccines85, supra, pp. 387-3g~; Dame, J.B., et al., 1985, In Vaccines85, supra, pp. 7-11).

5.4,1.1. USES OF ANTIBODIES GENERATED BY
IMMUNIZATION WITH RECOMBIN~NT
OCCLUSION BODIES
The antibodies generated against pathogenic microorganisms by immunization with the recombinant OBs of the present invention also have potential uses in diagnostic immunoassays, passive immunotherapy, and generation of 30 antiidiotypic antibodies.
The generated antibodies may be isolated by standard techniques known in the art (~. immunoaffinity chromatography, centri~ugation, precipitation, etc.), and used in diagnostic immunoassays to detect the presence of 35 viruses, bacteria, or parasites of medical or veterinary importance in human or animal tissues, blood, serum, etc.

.

~69- 132~9 The antibodies may also be used to monitor treatment and/or disease progression. Any immunoassay system known in the art may be used for this purpose including but not limited to competitive and noncompetitive assay systems using 5 technique~ such as radioimmunoassays, ELISA (enzyme linked immunosorbent assays), ~sandwichn immunoassays, precipitin xeactions, gel diffusion precipitin reactions, immunodiffusion assays, agglutination assays~ complement-fixation assays, immunoradiometric assays, fluorescent 10 immunoassays, protein A immunoassays and immunoelectrophoresi~ assays, to name but a few.
The vaccine formulations of the present invention can also be used to produce antibodies ~or use in passive immunotherapy, in which short-term protection of a host is 15 achieved by the administration of pre-formed antibody directed against a pathogenic microorganism. Passive immunization could be used on an emergency basis for immediate protection of unimmunized individuals who have been exposed to a pathogenic mi~roorganism, for instance, in 20 hospitals and other health-care facilities. Human immunoglobulin is preferred for use in humans since a heterologous immunoglobulin will i~duce an immune response directed against its foreign immunogenic components.
The antibodies yenerated by the vaccine formulations 25 of the present invention can also be used in the production of antiidiotypic ~ntibody. The antiidiotypic antibody can then in turn be used for i~munization, in order to produce a subpopulation o~ antibodies that bind the initial antigen of the pathogenic microorganism (Jerne, N.K.~ 1974, Ann.
30 Immunol. (Paris) 125c:373; Jerne, N.K., et al., 1982, EMBO
1:~34).

5.4.2. BIOLOGICAL INSECTICIDES
Baculoviruses are major pathogens of a large number 35 of agricultural pest~ (Vlak, J.M. and Rohrmann, G.F., supra;
Tweeten, K.A., et al., 1981, Microbiol. Rev. 45:379-408).
For example, one baculovirus host, the corn earworm .
t .
. . ~ , .
' . ' -70- 132~

Heliothis zea, in many areas routinely damages 90-100% of the ears of sweet corn (Kirk-Othmer, Encyclopedia of Chemical Technology, 1981l 3rd Ed., Vol. 13, John Wiley &
Sons, New YorX, p~ 415). HzSNPV has been approved as a 5 viral insecticide and is used as a pathogen for the cotton b~llworm and the corn earworm. The OB is the infectious particle responsible for transmission of the virus from organism to organism in the wild. The production of recombinant occlusion bodies in accordance with the present 10 invention thus provides for horizontal transmission of infection with concomitant expression o the foreign gene.
Manipulation of the polyhedrin protein to incorporate enzymatic activities, toxic peptides, or any molecule with insecticidal activity can increase the lethality of the OB
to host agricultural pests. Thus, the recombinant OBs of the prqsent invention have valuable applications as biological insecticides.
Genes which may be reco~bined into OBs in accordance with this embodiment of the invention include any genes 20 which encode molecules that effectively increase the desired insecticidal activity of the baculovirus without impairing the viability or infectivity of the virus itsel~. Such molecules include but are not limited to those which encode enzymes, enzyme inhibitors, insect hormone antagonists, 25 neurotoxins, metabolic inhibitors, insect chemattractants, endotoxins of other insect pathogens, etc. For example, molecules which interfere with physiological and/or developmental processes unique to arthropods that are susceptible to baculoviral i~fection, may be expressed on or 30 within recombinan~ OBs. Such molecule~ include but are not limited to insect growth regulators such as hormone antagonists (e.g. neotenin antagonists), and chitin synthesi~ inhibitors. Neuropeptides which are toxic or which induce detrimental behavioral modification~ loss 35 of appetite or mating behavior) may be encoded within the polyhedrin gene. Sex pheromones which act as chemattractants may be used to increase spread of the .. , ' . . .
,; . . .
. ..... .
" ' ' `. ,' , " ~
, : `
32~6 baculovirus infection throughout the insect population. A
chitinase incorporated into the OB may increase the infectivity of the virus. ~n endotoxin of another insect pathogen, such as the Bacillus thuringiensis endotoxin, may 5 be expressed in order to increase pathogenicity. Many specific embodiments of the invention are possible, provided that the recombinant form of the insecticidal molecule is functionally active within the physiological environment of the infected insect. Metabolic precursors to insecticidal 10 molecules may also be encoded by the recombinant polyhedrin gene, provided that the metabolic machinery to convert the peptide to a biologically active form is available and functional at the site of infection within the host insect.
Any standard method can b2 used to assay lethality of 15 the recombinant baculovirus. Such methods include but are not limited to the diet-surface technique and container, to bioassay OB activity ~Ignoffo, C.M., 1966, J~ Invert. Pathol.
8:531-536; Ignoffo, C.M. and Boening, O.P., 1970, J. Econ.
Entomol. 63:1696-1697).

5.4.3. EXPR~SSION VECTORS
The recombinant viruses which form occlusion bodies expressing heterologous peptides, the production of which is provided for by the present invention, can be used generally 25 as expre~sion vector systems for the production of the fore~gn peptide(s) which they ~ncode. In this embodiment of the inventionO the recombinant baculoviruses which express the foreign peptide under control of the polyhedrin promoter are used to infect an appropriate host cell in order to 30 obtain the desired quantities of the heterologous peptide.
To this end, the foreign peptide (which is a fusion polyhedrin protein~ may be purified from the occluded virions, isolated occlusion bodies, cell culture media, infected larvae, etc., by standard techniques known in the 35 art for the purification of proteins, including ~ut not limited to chromatography (e.~. ion exchange, affinity, and siæing column chroma~ography), entrifugation, differential : '' ' .

. -72- ~3~ a solubility, isoelectric focusing, and preparative electrophoresis. Expression as part of a (polyhedrin) crystal can greatly facilitate isolation o~ the heterologous peptide or protein in substantially pure form. In addition, 5 standaxd procedures such as solu~iliza~ion o~ the crystal, followed by immunoaffinity chromatography, and, if desired, recrystallization (as described in Section 5.l.5., supra) can be used to increase purity of the final product.

5.4.4 IMMUNOASSAYS
The recombinant OBs of the pres~nt invention, expressing foreign epitope(s), may be used as antigens in immUnGaSsayS for the detection of antibodies to the epitope(s). The recombinant OBs may also be used to detect 15 the same or related epitope(~) by competition assays. The recombinant OBs, or the foreign epitope(s~ expressed by them, may be used in any immunoassay system known in the art including but not limited to competitive and noncompetitive assay systems using techniques such as radioimmunoassays, 20 ELISA (enzyme-linked immunosorbent assay), ~sandwich~ -immunoassays, precipitin reactions, gel diffusion precipitin reactions, immunodiffusion assays, agglutination assays, complement-fixation assays, immunoradiometric assays, fluore~cent immunoassays, protein A immunoassay~ and 25 immunoelectrophoresis assays, to name but a few.
As demonstra-~ed in the Examples detailed herein, the recombinant OBs of the invention are capable of capturing and precipitating antibodies specific for the foreign epitope(s) presented on the recombinant OBs. This is an attrac~ive 30feature of the recombinant OBs which makes them particularly useful for the detection o~ antibodies in sample fluids, especially where the antibodies ara presented at low concentrations. ~or example, the recombinant OBs bound to their captured antibodies could be immobilized by anti-35polyhedrin antibodies. The presence of the captured antibodycan be deterted using appropriate anti-immunoglobulin antibodies. Thus, a ~sandwich~ type of immunoassay for .

-- ~;

, -73- 1325~

detecting dif~erent antibodies in sample fluids can be accomplished using ~universal~ capture (e.g., antipolyhedrin) and detection antibodies (e.g., anti human Ig).
Alternatively, recombinant OBs which present the Fc 5 binding region o~ protein G or A could be produced by cloning appropriate regions of those genes into the polyhedrin gene within the modifiable domains described herein. These recombinant OBs can b~ used to bind any antibody. The resulting recombinant OB/antibody complexes can then be used 10 in immunoassays to bind and capture antigens in sample fluids. ~hese could similarly be used in the nsandwich" type of assay system described above to detect antigens in ~ample fluids.
5 r 4.5. IMMOBILIZED ENZYMES
The recombinant OBs of the present invention which express the active site of an enzyme on their sur~ace can be used in a variety of procedures which r~quire immobiliæed enzymes. For example, the recombinant enzymatic OBs may be packed into a column on whlch reactions catalyzed by the 20 enzyme can be carried out. The resulting products can easily be separated from the mixture of reactants an~ enzyme.

6. EXAMPLE: CONSTRUCTION OF TRANSFER VECTORS
USED FOR INTRODUCING FOREIGN GENE SEQUENCES
INTO THE HELIOTHIS POLYHEDRIN &ENE TO PRODUCE
Hz RECOMBINANTS
The subsections below describe the sequencing o~ th~
Heliothis polyhedrin gene and the construction o~ a family of plasmid transfer vectors which allow for the production of Heliothis zea virus recombinants which contain foreign genes 0within the polyhedrin gene se~uence.

6.l. MATERIALS AND METHODS

6.l.l. RESTRICTION MAPPING
Plasmid DNAs were digested with restriction endonucleases HindIII, Eco~I, PstI, XbaI, BamHI, SalII XhoI, NruI, ClaI, H cII, BclI, or KpnI under conditions specified by the manufacturer (Bethesda Research Laboratories or : ''- . ' - :

., ~.

`` _74_ 132~

Promega Biotec). Digested DNAs were size fractionated on O.7~ to 1.2% agaros~ or 8% acrylamide gels containing 90 mM
Tris-borate, 90 mM boric acid, 2 mM EDTA (pH 8.0), and 0.1 ug/ml of ethidium hromide. DNA bands were visualized with an 5 ultraviolet transilluminator and photographed. Analyses of single and multiple diqestions were used to construct the restriction maps.

6.1.2. _UTHERN BLOTTING
Gels were soaked in denaturing solution (0.5 M NaOH, 1.5 M NaCl) for 30 minutes. The gels were neutralized by soaking in 1.0 M Tris-HC1 pH 8.0, 1.5 M NaCl. DNA was transferred to nitrocellulose by a modification of the method of Southern (JO Mol. Biol. 98:503-517). DNA was blotted 16 using 1.0 M NH4 acetate. Filters were baXed under vacuum for two hours and soaked in prehybridization solution (0.12 M
NaP04 pH 6.3, 2 X SSC, 50% formamide, 10 mM EDTA, 1% sarcosyl and 3X Denhardts) for more than three hours. Filters were rinsed with distilled water and incubated at 37C overnight 20 with fresh prehybridization solution plus denatured labelled radioactive probes. Filters were rinsed in 0.2 X SSC and washed for one h~ur i~ prehybridization solution without Denhardts. Rinses and washes were repeated four times.
Filters were dried and autoradiographed with Xodak X-Omat AR5 25 film.
SSC = 150 mM NaCl, 15 mM sodium citrate pH 7.0 ~enhardt's Solution = 0.02% ficoll, 0.02~
polyvinylpyrrolidone, 0.02% BSA

6.1.3. DNA SEQUENCING
DNA sequences were determined using the dideoxy chain termination method with M13 subclones (Sanger, F., Nicklen, S., and Coulson, A.R., 1977, Proc. Natl. Acad. Sci. U.S.A.
74:5463-5467; Messing, J~, Crea, R., and Seeburg, R.H., 1981, 35Nucl. Acid Res. 9:309-321) of the Heliothis polyhedrin gene.
Supernatants of M13 infected cells were centrifuged two times at 5000 rpm for 20 minutes to remove c~lls and cell debris.

.

.

.l'~
_75_ 1325~

Phage were precipitated by adding 1/5 volume 20% PEG6000, 2.5 M NaCl. Pellets were r~suspended in G, 6, .Z (6 mM Tris-HCl p~ 8.o, 6 mM NaCl o.2 mM EDTA) and virus reprecipitated with 1/5 volume 20% PEG, 2.5 M NaCl. Care was taken to remove as 5 much liquid from the pellet as possible. Pellets were resuspended in 6, 6, .2 and DNA was extracted with phenol saturated with 0.1 M Tris pH 8Ø DNA was reextracted with phenol:chloroform (1:1), then with chloroform, and finally with ether. DNA was precipitated with e~hanol twice and 1O rinsed one time.
Single stranded DNA templates were annealed to sequencing primers (Bethesda Research Laboratories or Pharmacia) in 10 ul reactions containing 5 ul of single stranded template, 2 ul of primer, 2 ul of HB buffer (70 mM
15 Tris pH 7.5, 70 m~ MgC12, 500 mM NaCl). Reactions were heated to 95C for 5 minutes and allowed to cool to room temperature ~or 45 minutes. Aft~r annealing, 2 ul of alpha-32P-dATP, 1 ul oP 25 uM dATP and 2 Units of DNA Polymerase Large Fragment (Bethesda Research Laboratories, Pharmacia, or 20 Promega Biotec) were added. Primer extensions in the presence o~ the dideoxy nucleotides were initiated by adding 3 ul of the annealing mix to tubes containin~ the appropriate mix o~ dideoxy (dd) and deoxy nucleotides. A reaction: 1 ul of 0.5 ~M ddATP and 1 ul 125 uM dCTP, dGTP and dTTP. G
25 reaction: 1 ul o~ 0.625 mM ddGTP and 1 ul 8 uM dGTP, 170 uM
dCTP and 170 uM dTTP. C reaction: 1 ul 0.5 mM ddCTP and 1 ul 8 uM dCTP, 170 uM dGTP and 17U uM dTTP. T reaction: 1 ul O.84 mM ddTTP and 1 ul o~ 8 uM dTTP, 170 uM dCTP and 170 uM
dGTP. Reactions were incubated at 45C for 15 minutes. 1 ul 30of 0.5 mM dATP, 0.5 mM dGTP, 0.5 mM dCTP and 0.5 mM dTTP were added and the reactions were incubated for an additional 15 minutes. The reactions were stopped by adding 12 ul of 95%
formamide and 10 mM EDTA pH 8~0. The samples were heated to g5C and loaded on denaturing ~crylamide gels containing 8 M
35urea, 90 mM Tris pH 8.3, 90 mM boric acid and 2 mM EDTA.
Gels were fixed in 10% acetic acid, 10% methanol, dried and autoradiographed.

~. . . , : ~
'~

~76 1~2~ 9 Alternatively, DNA was sequenced as described by Chen, E., and Seeburg, P., 1985, DNA 4:165-170 using the Sequenase~
system (U.S. Biochemicals, OH) 6.2. IDENTIFICATION ~ND SEQUENCING OF
THE POLYHEDRIN GENE OF HELIOTHIS
ZEA VIRUS
Heliothis zea DNA was obtained from virus isolated from Heliothis zea infected larvae. Viral HindIII and XhoI
fra~ments wer~ cloned into the HindIII and SalI site, 10 respectively, of pUC12. Two plasmids were characterized:
pHH5 which contains a 3.1 kb RindIII virus fragment and pHX12 which contains a 6.5 kb XhoI virus fragment. Both inserts cross-hybridized to a DN~ fragment encoding part of the Autographa polyhedrin gene.
Restriction maps of pHH5 and pHX12 demonstrated that the pHH5 insert is contained within p~X12. Southern blots of pHH5 using the Autographa polyhedrin gene as a probe indicated that a 1.7 kb NruI fragment cross-hybridized to the Autographa polyhedrin gene. HincII and HindIII/SalI
20 ~ragments of pHH5 and the HindIII/EcoRI ~ragments of pHX12 were subcloned into M13mpl8 (~anisch-Perron, C., et al., 1985, Gene 33:103-119). The EcoRI/NruI fragment of pRE2.5 (described in~ra in Section 6.~.2.) and the EcoRI/HindIII
fragment of transfer vector 1 (described infra in Section 25 6.3.4. were also subcloned. The DNA seguence of selected subclones was determined using the dideoxy chain termination method (Sanger, F., et al., 1977, Proc~ Natl. Acad. Sci.
U.S.A. 74:5463-5467). The sequencing strategy used is diagrammed in Figure 1~. The sequence of the Heliothis 30 polyhedrin gene is shown in Figure 1. A restriction enzyme map ~or the restriction endonucleases HindIII, NruI, HincII, and AccI was derived from the nucleotide sequence and is shown in Figure 1~, The digest fragments and sizes are shown in Table II.

.

~77~ ~3 2~ ~

TABLE II

HindIII/ NruI/HincII/AccI RESTRICTION ENZYME ~IG~STION
FRAGMENTS OF THE HELIOTHIS POLHED~IN GENE

List of Fragments from 5' to 3' ~nd Fra~ment # _ Fragment sizeStarts atEnds at ~ * +2 2 250 ~3 +252 3 9 +253 +~61 4 14 +262 +275 12 ~276 ~2~7 6 ___ +288 --*

~___________ 1 Fragments shown are those expected from the DNA sequence 20 of th~ Heliothi~ polyhedrin gene shown in Figure 1, after digestion with HindIII, NruI, HincII, and AccI.

* Not within the sequenced region.

The DNA sequence of the subclones was compared with .
that of the Autographa polyhedrin gene in order to identify the coding sequence of the Heliothis polyhedrin gene. The DNA sequence shown in FIG. 1 reveals an open reading frame of 30 753 nucleotides. The 7th codon of the open reading frame encodes a methionine which is ~ollowed by a sequence encoding 244 amino acids which have 84~ amino acid sequence homology to the Autographa polyhedrin seguence. This sequence terminates with a TAA codon as found in the cvrresponding '` ` ' ' ' , ' ~ ~

-78- ~32~

position of the Autographa ~ene. We defined the first methionine in this open reading frame as the initiating codon of the Heliothis polyhedrin gene.
If the Autographa (MNPV) and Heliothis ~SNPV) 5 polyhedrin amino acid sequences are aligned to maximize sequence homology, there is 84~ sequence homology between the two proteins. This compares with 77% seguence homology between Heliothis (SNPV) and Bombyx mori (MNPV) proteins (FIG. 2). The Autographa and Bombyx proteins also share 84%
10 sequence homology. If the tyr-ser-tyr sequence at amino acid residues 5-7 of the Heliothis sequence mark the beginning of the homology between the two proteins, the Heliothis sequence contains an insertion of an additional amino acid residue at the amino terminus as compared with the Autographa and Bombyx 15 proteins. There are 36 amino acid substitutions between the Autographa and Heliothis proteins as well as a deletion of an amino acid between positions 226 and 227 o~ the Heliothis sequenoe.
The Heliothis and Bombyx sequences are somewhat more 20 divergent, sharing only 77% sequence homology. In addition to the 52 amino acid substitutions, there are two single amino acid insertions as well as two d~letions in the Heliothis sequence. Interestingly, of the four deletions and insertions between these two sequences, one insertion and one deletion are found in the Heliothis-Autographa comparison and the other insertion and d~letion are fou~d in the Autographa-Bombyx comparison. This suggests that the evolutionary divergence between Autographa and either Heliothis or Bombyx is approximately the same and that 30Heliothis and Bombyx are more highly divergent. Similar conclusions can be reached by comparing the overall sequence homology of thP polyhedrin proteins ~f the three species.
There does not appear to be a relationship between the degree of seguence divergence and whether the virus is an SNPV or 35MNPV. The degree of sequence divergence between Autographa (MNPV) and either the Heliothis (SNPV) or Bombyx (MNPV) is similar.

. .
.

..
: ~: . ' : .
, ~79~ 132~6~

The pattern of hydrophilicity is very similar for the Autographa and Heliothis proteins (FIG. 3). Interestingly, the region of highest hydrophillcity of the polyhedrin proteins is the region o greatest sequence divergence.

5 There is only 54~ sequence homology between the Autographa and Heliothis polyhedrins in the region between amino acids 3~ and 50 of the sequence. The Autographa and Bombyx sequences share only 31% sequence homology, while the Heliothis and ~ombyx sequences are 39% homologous in this 10 region. These values compare with approximately 80% se~uence homology for the entire protein. Conceivably, these hydrophilic reqions identify a site involved in some species specific interaction with other viral or cellular components.
Small peptides generated from this region perhaps may be used 15 to rai~e monoclonal antibodies that could discriminate among different baculoviruses.

6.3. CONSTRUCTION OF TRANSFER VECTORS
The plasmids pHH5 and pHX12 were used to construct a 20 transfer vector, termed pHE2.6, which allows for the insertion of foreign genes within the polyhedrin gene sequence so that recombinant Hz viruses containing the foreign genes can be produced via in vivo recombination.
The construction of this transfer vector is outlined 25 in FI~. 5A, which should be referred to in order to simplify the description that ~ollows.

6.3.1. P~RENT PLASMIDS: PHH5 AND pHXl2 The preparation of pHH5 and pHXl2 is described above 30in Section 6.2. The pHH5 plasmid contains a 3.1 kh HindIII
fragm0nt of the Hz virus DNA (starting from nucleotide residue number 281 of the polyhedrin gene sequence depicted in FIG. l) in the HindIII site of pUCl2 (Vieira, J. and Messing, J., ~982, Gene 19:259) (FIG. 5A). Th~ HindIII Hz 35DNA insert of pH~5 encodes approximately two-thirds of the polyhedrin gene comprising the carboxy-coding re~ion ~iOe., approximately one-third of the polyhedrin coding sequence -80- i 3 ~ 0 comprising the amino-coding region is missing). The polyhedrin gene sequence is oriented so that the polylinker of the pUC12 parent plasmid is located upstream or 5' to the polyhedrin gene sequence (FIG. 5A).
The XhoI H~ DNA insert of the pHX12 plasmid contains the entire polyhedrin gene sequence inserted into the SalI
site of pUC12. The Hz polyhedrin gene se~uence in pHX12 is oriented in the opposite direction with respect to the pUC12 polylinker as compared with the Hz polyhedrin coding sequence 10 contained in pHH5; that is, the EcoRI site of the polylinker of pUC12 is located 3' to the polyhedrin gene sequence in pHX12 ~FIG. 5A).

6. 3 . 2 . CONSTRUCTION OF TRANSFER VECTORS
A restriction fragment of pHX12, containing portions of the amino-coding terminus of ~he Hz polyhedrin gene sequence, was used to reconstruct the polyhedrin gene in pHH5 so that a transer vector containing the polyhedrin gene sequence interrupted at its amino-coding terminus by a 20 multiple cloning site (MCS), would be produced.
The pHX12 plasmid was cleaved with EcoRI and NruI, and an approximately 1125 bp EcoRI-NruI fragment, containing the promoter and amino-terminal portion o~ the ~eliothis polyhedrin gene, was isolated. This 1125 bp fragment was 25 cloned into the EcoRI and SmaI sites o~ the pUC1~ polylinker in pHH5. The BamHI to PstI sequence of the polylinker of this clone was then replaced with a synthetic oligonucleotide containing various restriction endonuclease recognition sites (a multiple cloning site/ MCS~. The 30sequences at the cloning junctions ~f the oligonucleotide have been confirmed by DNA sequence analysis.
The resulting plasmid, termed pH~2.6, contains the polyhedrin 5' flanking region including promoter sequences, . 5~ polyhedrin coding sequences, an MCS, 3' polyhedrin coding 35sequences, 3' polyhedrin flanking region, and pUC12 sequences. Foreign gene ~equence~ can be inserted into the polylinkar and the resulting plasmid vector can be used to - .

~:' - , .
:

81- ~32~Q

transfect host cells infected with Heliothis zea virus. Hz virus recombinants will be ~ormed which contain the ~oreign gene and direct its expression using the polyhedrin promoter.

6.3~3~ TRANSFER VECTORS EXPRESSING
BETA-GALACTOSIDASE
A transfer vector, termed pHE2.61ac, was constructed to contain the E. coli beta-galactosida~e (B-gal) gene inserted within an MCS flanked by Heliothis polyhedrin 1 sequences (FIG. 5B). A 3 kb fragment of plasmid pMC1871 (Pharmacia), containing the E. coli B-gal gene, was isolated by cleavage with BamHI followed by gel puri~ication.
Plasmid pHE2.6 was cleaved with BglII, treated with bacterial alkaline phosphatase, and ligated (T4 DNA ligase) 15 to the pMC1871-derived ~ragment in order to insert the B-gal gene into the ~II site of pHE~.6. The resulting plasmids contained the B-gal gene in both orientations (5' to 3', and 3' to 5') with respect to the polyhedrin promoter and coding sequences. E. coli strain DHl was transformed with the 20 resulting plasmids, and the identity of transformants was confirmed by DNA fra~ment size determinations upon restriction digestion of their plasmid DNA. E. coli strain DH5 alpha (Bethesda Research Laboratories) was also transformed with the B-gal-containing plasmids, and the 25 resulting transformants were tested according to the ~blue-white~ screening ~echnique of Messing et al. (1977, Proc.
Natl. Acad. Sci. U.S.A. 74:3642-3646). Briefly, the transformed bacterial cells were mixed with the chromogenic substrate ~X-gal~ (5-bro~o 4-chloro-3-indolyl-B-D-30galactopyranoside) before plating. Blue bacterial coloniesarise from bacteria containing plasmids that express functional B-gal activity. Such plasmids contain a beta-galactosidase gene which encodes the enzyme responsible for hydroly~is of nX-gal~ and re~ultant production of a blue 5-~ ~ ~ f ~ ~

~82- ~32~6~

bromo-4-chloro-indigo. Bacteria harboring plasmids that do not express functional B-gal activity give rise to white colonies since there is no ability to hydrolyze the chromogenic substrate.
The plasmid construction which contained B-gal in the appropriate orientation produced blue colonies indicative of beta-galactosidase activity. This plasmid was termed pHE2.61ac. pHE2.61ac was determined by appropriate restriction dige~tions to contain the B-gal gene in the 10 proper orientation (i.e. the same 5' to 3' direction) to the polyhedrin promoter. Those transformants containing the parental plasmid pHE2.~ or the B-gal gene in the wrong orientation produced only white colonies.
Plasmid pHE2.61ac has been used for transfections into Heliothis virus-infected cells in order to transfer the B-~al gene fusion into Heliothis virus. Once a Heliothis virus expressing beta-galactosidase is obtained, the virus can be used as the parental virus for further manipulations involving insertions and deletions of the polyhedrin gene, 20 through transfection o~ parental virus-infected cells with trans~er vectors such as plasmid pHE2.6. Selection of the appropriate recombinant viruses would be greatly fa~ilitated by detection of white plaques amidst a background of blue plaques.

6.3.4. GENERATION OF DELETIONS OF
HELIOTHIS POLYHEDRIN AMINO-TERMINAL SEQUENCES
The strategy we have used to create deletions in the amino-terminus of the polyhedrin gene in Heliothis transfer 30vectors is diagrammed in Figure 5C.
Plasmid pHE2~6 was digested with EcoRI and PstI. A
1.1 kb EcoRI-PstI fragment containing the Heliothis polyhedrin gene was generated, which was subcloned into M13mpl~. The double-stranded replicative fo~m of the , -83- ~32~

resulting Ml3 derivative was digested with XbaI and KpnI, which ~ut within the MCS, to generate a single-stranded 5' overhang at the XbaI cleavage site and a single-~tranded 3' overhang at the KpnI cleavage site. The resulting DNA was 5 treated with Exonuclease III (exo III) which digested a single strand of the double-stranded ~NA for a variable le~gth in the 3' to 5' direction starting ~rom thz XbaI-cleaved end. The XpnI-cleaved end which has a 3' overhang is resistant to exo III digestion. The DNA was then 10 digested with Mung Bean nuclease, which digests single-stranded DNA, to generate blunt ends by removing the single-stranded DN~ left after exo III digestion. The blunt ends were ligated together with DNA ligase, resulting in transfer vectors that contain deletions of various length within the N-terminal portions of the Heliothis polyhedrin gene. The transfer vectors thus derived contain the Heliothis polyhedrin promot~r, 5' polyhedrin regions of various length, an abbreviated MCS, and 3' polyhedrin sequences.
Thus far, several N-terminal deletion transfer vector~ have been obtained in this fashion. Transfer vector l has a 28~ base pair (bp) deletion spanning nucleotide number 63 of Figure 1 through the XpnI site.
Transfer vector 2 has a 274 bp deletion spanning nucleotide 25 number 71 of Figure l through the KpnI site. Additional deletion mutations are being generated. Similar manipulations can be done at other suitable restriction sites in order to obtain deletions of regions that are nonessential for OB formation.
7. EXAMPLE: HELIOTHIS VIRUSES FOR USE IN
GENERATION OF RECOMBINANT OCCLUSION BODIES
Twenty plaque-purified str~ins of HzSNPV Elcar) were characterized based on their restriction endonuclease diye~tion patterns of viral DNA and structural protein ~ ~ 2 ~
profiles. Each of the twenty strains had a unique genotype which was distinguishable by digestion with restriction endonucleases samHI~ EcoRI, HindIII, or PstI. Most of th~
genomic heterogeneity between strains was located between 5 map units 23.5 and 43.3. Differences were evident in the o~cluded virus structural protein profiles of all the plague-purified strains relative to the wild-type isolate.
We noted differences in pathology of the plaque-purified strains upon inoculation of ~. zea larvae, and were able to 10 segregate the strains into three categories based upon the relative rates of death (as measured by immobility) and of melanization (as measured by darkened appearance and disruption of cuticle) from the time of inoculation. The genotype of the weakly melanizing strain, HzS-15, was 15 extensively characterized relative to the wild population genotype, using numerous restriction enzymes. A genomic map was constructed for HzS-15 using the enzym~s BamHI, PstI, and SstI.

7~ lo MATERIAL AND METHODS
7.1.1. IN VITRO PROPAGATION OF HzSNPV
Infectious extracellular virus (ECV) was obtained from larvae five days post infection and before 25 melanization. Hemolymph wa~ collected by clipping a proleg and bleeding 10 larvae into 5 ml of TNM~FH medium (Hink, W.F. 9 1970, Nature ~66:4~6-467) containing 5 x 10 3 M 1-cysteine-HCl and 5 x 10 3 M dithiothreitol. The diluted hemolymph was filter-sterilized and used as inoculum for 30 IPLB-HZ1075 cells (Goodwin, R.H., et al., 1982, In Vitro

18:843-850) adapted for growth in TNM-FH medium.
Inoculation of cell cultures was accomplished by adding 5 ml of filtered inoculum to a 24 hour old monolayer of 1.0 x 106 IP~B-HZ1075 cells in a tissue culture flask (25 cm2)O After 35 one hour at 29~C, the inoculum was removed and the monolayer . " ~, ~ .

.
.

-85- 1 3 ~

washed once with fresh media. Five ml of TNM-FH was added and the cultures were incubated at 29C and monitored at 24 hour intervals ~or the presence of occlusion bodies.

7.1.~. PLAQUE PURIFICATION OF HZSNPV ISOLATES
Plague assays were performed on supernatants of cell cultures in~ectad with the larval-isolated HzSNPV as previously described ~Fraser, 1982, J. Tis. Cult. Meth.
7:43-46). Twenty-four wild-type (~P type) plaques were 10 picked and used as ino~ulum for 1 x 105 ~ells in each well of a 24 well plate. These once-plaque-purified isolates were re-assayed and individual pla~ues from the second assay were amplified first in 24-well plate cultures, and then in 25 cm~ flask cultures o~ IPLB-HZ1075 cells.

7.1.3. LARYAL PROPAGATION OF VIRUS
Individual strains were amplified by oral inoculation of one to two inch long Heliothis ea larvae (third to fourth instar) with occlusion bodies isolated from the 20 second cell culture passage of each plaque-purified strain.
Infections were allowed ~o progress for ~ to 7 days before collecting the larvae. LarYae infected with each isolate were ~eparated into either melanized or non-melanized pools upon collection, and both pools were frozen at -20~C until 25 useO

7.1.4. ISOLATION OF VIRIONS FROM OCCLUSION BODIES
Infectious occlusion bodies (OBs) were harvested from pools of infected, non-melanized larvae by homogenization in 30TE bu~fer (10 mM Tris-HCl, lmM EDTA, pH 7.6) containing 0.1%
SDS. The larval homogenate was filtered through two layers of cheesecloth and centrifug~d at 1,800 x g for 15 minutes.
The supernatants were discarded and the OB psllet washed .

:; .

-8~ ~32~

twice by resuspension in 20 ml of TE buffer and centrifugation at 1,800 x g. The washed OBs were resuspended in a final volume of 10 ml TE bu~fer.
Virions were isolat~d from the washed, partially 5 purified OBs according to the procedure of Smith and Summers (1978, Virology 84:390-402) with slight modifications. Five ml of dis~olution buffer (0.3 M Na2C03, 0.03 M EDTA, 0.51 M
NaCl, pH 10.9) was added to 10 ml of washed OBs (approximately 15 mg/ml), and the oss were dissolved by 10 incubation ~or 10 minutes at room temperature. The mixture was layered on 20-60% (w/w) sucrose gradien~s (in TE buffer) and centrifuged at 75,000 x g for 60 minutes at 4C. The single visible virion band was collected, diluted with an equal volume of TE buffer, and the virions pelleted at 15 55,000 x g for 30 minutes (4 C)~ The pelleted virions were r~suspended in distilled water and stored at -20~C until use.

7 .1. 5 . ISOI~TION OF VI~L DNA
Gradient-puri~ied virions were incubated in T~ buffer containing 0.1% KCl, 0.1~ SDS, and 0.1 mg/ml proteinase X
(Sigma) for 3 hours at 65C. Following two extractions with phenol and two extractions with chloroform:isoamyl alcohol (24:1), the DNA was precipitated by the addition of 1/10 25 volume of 2 M sodium acetate, and 2 volum~s of 95% ethanol.
The precipitated DNA was pelleted at 1,800 x g for 15 minutes and resuspended in st~rile distilled water by heating at 65-C for 30 minutes. The concentration of DNA
was determined by absorbance at 260 nm, and the DNA was stored at 4-C until use.

7.1.6. RESTRICTION ENDONUCLEASE DIGESTIONS
Viral DNAs were digested with BamHI, EcoRI, EcoRV, HindIII, ~I, PstI, and SstI restriction endonucleases 35 (Bethesda Research Laboratories~ under conditions specified ~ .

~ 3 ~
by the supplier. Restriction enzyme ~ragments were separated by electrophoresi6 in 0.75% agarose gels (20 x 20 cm) in Tris-acetate buffer (0.04 M Tris-acetate, 0.1 mM
EDTA, pH 8.0) sontaining 0.25 ug/ml ethidium bromide. Gels 5 were electrophoresed for 17 hours at 70 volt~ and DNA
fragments detected with UV light (306 nm~. Gels were photographed using a Kodak Wratten 23A filter and Polaroid type 55 positive/negative ~ilm.

1o 7.1.7. SDS-POLYACRYLAMIDE GEL ELECTROPHORESIS
Structural proteins of virions released from occlusion bodies by alkali treatment were compared by electrophoresis in discontinuous polyacrylamid~ slab gels accordîng to the method of Laemmli (1970, Nature 227:680).
15 Virion proteins were solubilized by boiling for 3 minutes in denaturation buffer (62.2 mM Tris-HCl, 2.0% SDS, 20%
glycerol, 2.5% dithiothreitol, pH 6.8~ at a concentration of 1 mg protein/ml. Electrophore~is was carried out at 30 milliamp~ for 4.5 ~ours in a 12% separating gel (10 cm long 20 x 12 cm wide x 1.5 ~m thick~. Gels were stained with 0.125%
Coomassie brilliant blue R 250 ~ollowing standard protocols (Summers, M.D. and Smith, ~.E~, 1978, Virology 84:390-402;
Monroe, J.E. and McCarthy, W.J., 1984, J. Invert. Path.
43:32-40).

7~2. ~HARACTERIZATION OF HzSNPV

7.2.1. IN VITRO PROPAGATION AND PLAQUE PURIFICATION
The IPLB-HZ1075 insect cell line grew well in TNM FH
30medium supplemented with 8% fetal calf serum. Cells remained susceptible to infection by HzSNPV, but infectivity was not 100~ under these conditions. ~he highest levels observed were between 50 and 70% infected cells with maximal titers of S x 106 plaque forming units per ml. The be~t 35infections were achieved when cells were allowed to grow at -88 ~ 3 ~

least 24 hours before inoculation with virus. We ha~e since discovered that the addition of 1% bovine serum albumin (BSA) and 2 g/l L-glutamine to the growth medium improv~s infectivity to about 100%.
Plaques were produced on monolaye.rs of ~P~B-HZ1075 cells using the procedures described previously ~Fraser, 1982, J. Tis. Cult. Meth. 7:43~46; Fraser and McCarthy, 1984, J. Invert. Path. 43:~27-~2~). No FP-like plaques (few polyhedra) were observed in this study. All plaques picked 10 for isolation exhibited the wild type morphology and produced many occlusion hodies per infected cell.

7~2.2. LARVAL INFECTIONS WITH OCCLUSION BODIES
The plaque-purified strains were amplified in third to fourth instar H. zea larvae. Larval propagation was necessary to rapidly expand the virus and reduce the probability of selecting in vitro passage mutants.
Mutant selectîon is a phenomenon which oc~ur~ readily during in vitro propagation of baculoviruses (Potter, K.N., 20 et al., 1976, J. Virol. 18~1040-1050: Hink and Strauss, 1976, J. Invert. Path. 27:49-55; Fraser and Hink, 1982, Virology 117:366-378; Fraser and McCarthy, 1984, J. Invert.
Path. 43:427-429), but i5 not observed during short term in vivo propagations of HzSNPV (McIntosh, A.~. and Ignoffo, 25 C.M., 1986, Intervirol. 25:172-176).
To amplify the virus in larvae, the inoculations were performed by placiny a drop of a 1 x 106 OBJml suspension directly on the head capsule of each larvae. Larval derived OBs were used for subsequent inoculations to characterize 30the relative virulence and degree of pathogenicity of each strain and the wild-type i~olate.
During these in vivo amplifications, we noted di~ferences in the gross pathology of several plaque-purified strains relative to the pathology of the wild-type 35isolate. Nany o~ the plaque-purified strains caused rapid ~ .
., , ~ . .

.:, . ~ ,- \ , -89- ~32~

melanization and instability of the cuticle upon death of the larvae, a pathology normally ~een following infection with HzSNPV. In contrast, several strains caused mortality without the usual attendant rapid melanization and cuticular 5 breakdown.
The plaque-purified virus strains could be separated into three groups based on their relative ability to cause melanization in in~ected third instar larvae (Table III)~

TABLE III
Separation of HzSNPV Elcar) Strains on the Basis of Ability to Cause ~elanization Melanization Ability Isolate __ Rapid ~elanization and Death W+, l, 2, 4, ll, 12, 13, l~, 17, 18, 20, 23 Slow Melanization and Death 5, 7, B, 9, 21, 22, 24, ~5 No Melanizationa 15 25 ___~__ ____________ a nNO Melanization~ is.defined as less than 30%
melanization by nine days after larval death.

-The wild-type vlrus lsolate (W+) and several of the plaque-purified ~trains (l, 2, 4, ll, 12, 13, 14, 17, 18, 20, 233 caused larval death within four to five days post 3~

. :

:

~ .

- - 9o -:132~0 inorulation. The dead larvae rapidly melanized over a period of l to 3 hours, turning a dark brown overall, and the cuticle was easily disrupted.
Larvae infected with several other strains (5, 7, 8, 5 9, 15, 21, 22, 24, 25~ also reached apparently complete infection by 4-5 days post inoculation as evidenced by the abundance of occlusion bodies in infected tissues, but the larvae did not die or melanize rapidly. The larvae became soft and in~apable of motion in the posterior two-thirds of 10 the body after 4-5 days, but actual death ( e., unresponsiveness to probing) and subsequent melaniza~ion required several more days, and in some cases even weeks, e.g. HzS-15.
Strain HzS-15 caused a similar pathology to the other 15 slow-melanizing strains. However, HzS-15 was remarkable in that most larvae in~ected with this strain did not begin melanizing until greater than 7 days post inoculation, with many taking several wee~s to completely melanize.
Furthermore, HzS-15 is highly virulent.
To further characterize these apparent differences in pathology between plague-purified s~rains, we standardized the inoculations using a surface treatment bioassay with 20 neonate (24 hour old) larvae per isolate. Two larvae were added to individual 1 ounce plastic portion cups containing 25 an agar based diet (Ignoffo, C.M., 1963, Ann. Entom. Assoc.
Am. 56:178-182) surface treated with 100 ul of a 1 x 10 OB/ml suspension (1250 OBmm2). All infected larvae died within 4 days post in~ection at this dosage. Larvae were monitored daily ~or mortality (as measured by 30 unresponsiveness to probing) and melanization ~as measured by coloration and cuticular di~ruption upon prodding~.
Three groupings were generated based upon relative percentages of larvae melanizing within a given time period (Table IV), essentially confirming the earlier observations 35 (Table III).

, ' ., ~. ~ '2 ~

_ TABLE IV
PERCENT OF LARVAE MELANIZING OVER TIME

5 Melanization Days After Larval Death Rate Strain 0-1 2-3 4-9 9 , W 75 ~5 - -rapid a 12 78 17 5 melanization 13 94 6 14 94 6 - ~

~ 0 26 67 7 lo 20 20 50 slow b 19 6 31 5G
20 melanization 21 69 12 13 6 2~ 44 19 19 1 24 ~ 7 27 67 ~5 5 1~ 35 50 non-melanizingC 15 5 11 11 73 25 a Greater than 90% mortality within 3 days, and at least 75%
of the larvae melanized within one day of death.
b Greater than 30% of the larvae melanized within 9 days of death.
c Less than 30% of the larvae melanized by 9 days after death.

At least 75% of the larvae infected with the rapidly melanizing strains completely melanized within 24 hours of death. The slow melanizing strains produced greater than ~5 .

~ . :
..
. .
: :
, '~

92- 13~

30% melanization response within nine days following larval death. Once again, ~zS-15 was remarkable, causing less than 30% melanization by nine days post mortality.

7.2.3. RESTRICTION ENZYME DIGESTION PATTERNS OF VIRAL DNAs The genomes of all twenty strains wre compared following digestions with BamHI, EcoRI, HindIII, and PstI.
Each strain could be distin~uished from all others on the basis of the combined restriction digestion patterns. No 10 single genotype was predominant. For example, in the HindIII digests there were only three strains with identical fragment patterns (FIG. 4). These three strains could be distinguished from the others upon digestion with BamHI.
Comparisons of the several individual digests suggested that 15 there is a hypervariable region of the HzSNPV genome between map units 23.5 and 43.3 (Knell and Summers, 1984, J. Gen.
Virol. 65:445-450).
Strain HzS-15 was singled out as unique due to its remarkably long melanization period and co~plex occluded 20 virivn structural protein profile. We compared this strain with the wild-type isolate using several additional enzymes (FIG. 5). The wild-type vixus and HzS-15 exhibited similar restriction patterns with enz~mes BamHI, KpnI, and PstI.
Different restriction patterns were observed with the 25 enzymes EcoRI, EsoRV~ ~indIII, and SstI. The differences in EcoRI and SstI banding patterns can be attributed to the hypervariable region identified in t~e HindIII digests. No conclusions can be drawn about the region of EcoRV
variability since no mapping data is available for this 30 enzyme.
Since HzS-15 and the wild-type isolate produce similar Bam~I restriction patterns, we used the restriction enzyme map of Knell and Summers supra as a reference in constructing a physical map of HzS-15 (FIG. 6~. The map of ~ . - .
, .

.

~93- 132~$10 HzS-15 was constructed from ~naly~e~ o~ single restri~tion enzyme digests and double restriction enzyme digests with BamHI, PstI, and SstI (~ables V, VI).

__ _ TABLE V
SIZES OF llzS-15 RESTRICTIO~ F~AGMENTSa Enzymes (Sinqle Digestion) 10 Fra~ment BamHI EcoRV HindIlI PstI SstI
.

A 36.8214.08 25.70 39.33 28.97 B 33.6112.93 16.27 36.72 25.66 C 15.4~10.92 15.15 33.47 23.22 D 13.Ç69.53 14.25 11.50 19.22 E 12.628.94 11.35 6.26 11.59 F 7.628.94 10.72 3.54 9.76 4.008.32 10.11 9.66 9.67 3.968.3~ 7.82 4.06 I 1.877.68 7.55 J 1.836.~3 3~69 K 1.296.01 2.75 20L 3.90 2.60 M 3.38 1.72 N 3.01 0 3.01 P 2.84 Q 2.66 R 1.70 2 S 1.60 6 T 1.60 U 1.~7 V 0.99 W 0.92 X 0.54 Y 0.~3 Z 0.33 30AA 0.28 Total 132.74131.25 1290 .......... ~........................ 0 ~ Sizes are given in kilobase pairs.

- , ~

: ~

_~ -94~

_ _ _ . _ TABLE VI
. SIZES OF ~zS-15 RESTRICTION F}~AGMENTSa Enz~me~ ~Double Digestion) 5 Fragment BamHI/PstI BamHI/SstI PstI/SstI
, A 28 . 64 22 . 65 28 . 97 B 25 . 78 15 . 46 25 . 66 C~ 15.~6 ~5.20 21.76 D 8 . 59 11. 50 lï . 50 E 7 . 26 11.19 11. 50 F 6.56 10.36 8.00 G 6.26 7.83 6.14 H 5.22 7.83 5.52 I 5.12 7.62 5.42 J 4.33 5.53 3.35 K 4.00 4.06 2.69 L 3.95 4.06 2.09 M 3.54 4.û0 1.40 N 3.08 3.96 1.03 O 1.~7 3.4~ 0.66 P 1.83 3.31 Q 1.29 1.83 R 0.99 1.45 S 1.29 T 0.45 V

W
X
3~ ~
z ~A

a Sizes are given in kilobase pairs.
- ~

: -, : -; . .
t ~95~ 132~9 Ambiguities in double digestion analyses of the completeviral genome w~re resolved using double and triple diyestions of individual cloned PstI and BamHI fragments.
Several differences were evident between HzS-15 and 5 the wild-type map of Knell and Summers (1984, supra). An additional SstI restriction site present in HzS-15 at map unit 43.9 produced two fragments, B and G, related to the wild-type SstI fragment A. The HzS-15 SstI-~ fragment is related to wild-type ~rayment B. The loss of an SstI
10 restriction site at 93.6 map units of the wild-type map generates a fused frayment, H, in ~zS-15 that is related to both SstI-G and -H of the wild-type genome. In addition, the location of wild-type SstI fragments ~ and F of Knell and Summers (1984, supra) had to be interchanged to correspond with our mapping data.

7.2.4. COMPARISON OF VIRION STRUCTURAL PROTEINS
The varied larval pathology prompted an inves igation of potential similarities in structural 20 proteins between strains exhibiting similar pathology.
Yirions were liberated from larval-derived occlusion bodies by alkali treatment and purified by banding in linear sucrose gradientsu Electrophoresis of occluded virion structural proteins of the wild-type isolate revealed 13 25 major polypeptides following staining with Coomassie blue ~-250. These proteins ranged in size from 69.9 to 17.8 kilodaltons (FIG. 7). Five of these polypeptides (VP 32.1, VP 37.2, VP 41.1, VP 49.2, and VP 62.9) were found in the occluded virion protein profiles of all the plaque-purified 30 strains. The remaining eight wild-type polypeptides varied in occurrence among the plague-purified strains.
The total number of major polypeptides in the plaque-puri~ied strains varied ~rom a low o~ 13 to a high of

19, and ranged in size from 17.8 to ~4.1 ~ilkdaltons. VP
3546.6 was easily visible in profiles of all the plaque-- ' . : ' ~

~96- 132~

puri~ied strains, but was not apparent in the profile of the wild-type isolate. Other unusual polypeptides were VP 62.0, found only in HzS-21, and several proteins between 21.1 and 25.7 kilodaltons which were ~vident only in strains 18 5 through 25.
The HzS-15 strain exhibited most of the wild-type polypeptides except VP 33.~ and VP 66.1, and also exhibited many of the additional polypeptides found individually in several of the other strains. Several protein bands were 10 apparently unique to HzS-15 including VP 51.0, and three bands above VP 69Ø
There was no apparent correlation between any of the occluded virion structural proteins and the observed dif~erences in rate of melanization of infected larvae.

8. EXAMPLE: CELL LINE HOSTS FOR USE IN
GENERATION OF RECOMBINANT OCCLUSION BODIES
A Heliothis zea derived cell line, IPLB-HZ1075, was subcloned by dilution plating. Twenty-four isolates, (HZ1075/UND A through X), were originally identified. Many

20 of the isolates with highly vacuolated cells eventually died during subcloning and amplification. Surviving isolates were characterized as to predominant morphology, cell doubling time, and ability to produce both ECV and OBs when infected with HzSNPV. Confirmation of the cell isolates' 25 origin was made by iso~yme analysis of the isolates, parental IPLB-HZ1075 cell line and H. zea larvae using the en~ymes FUM, LDH and MDH. A comparison of invertebrate cell lines in our laboratory by isozyme analysis proved that all were separable using the enzymes LDH and MDH.

'' : .~ ' .
, ,~ ' ~ ~ ''.
- .
- . :
-~97~ ~32~
8.1~ MATERIALS AND METHODS

8.1.1. C~ONING OF CELL STRAINS
The IPLB-HZ1075 cell line was obtained from Dr. J.
5 Vaughn (USDA Invertebrate Pathology Laboratory, Beltsville) and adapted to growth in TN~-FH medium over several passages. Cloning of cell strains was accomplished by diluting cells to an averaye density of 1 cell per 100 ul, and plating 100 ul in each well of a 96 well culture plate.
10 The wells ware examined after 12 hours and those containing only one cell each were marked. The growth medium for cell clones was composed of an equal mixture of filter-sterilized, conditioned TNM-FH medium and fresh TNM-FH
me.dium (50% conditioned medium). The conditioned medium was 15 obtained from 24 hour old, actively growing cultures of IPLB-HZ1075 cells, and was filter st~rilized to lnsure no carryover of cells. The stxain6 were maintained in the 96 well plates by replenishing the 50% conditioned medium every 5 days until crowding forced subculturing into 24 well 20 plates.
The cell strains were designated HZ1075/UND-A
through X. A total of 24 strains were originally isolated but many eventually died during amplification and subculturing. Of those that survived, one was lost to 25 contamination after only partial characterization.

8~1.2. CELL GROW~H CURVES
-.
Individual tissue culture flasks (25 cm ) were seeded with 1 x 106 cells of each strain in 3 ml of TNM-FH.
The cells were allowed to attach and enter log phase growth for 24 hours after which an initial cell count was made.
Three defined regions on the flask were counted at 48 hour intervals for up to 8 days following the initial count.
Cell clumping was not a probiem for most cell strains, and when it did occur~ counts were taken on several focal planes .

- ~ : . ,. . .:
. : .: . : . . . .
. . - , i,, .
: . . ' ' ': , `
"' . : .
~. ' ' ' ' " '', '' " ,' ':' , ~
' ` : ., -98 ~325~

8.1.3. QUANTITATION OF POLYHEDRA AND INFECTIOUS
EXTRACELL~LAR VIRUS PRODUCTION
Each strain was inoculated at a density of 1.25 x 105 cells per well in separate wells of a 24 well cluster plate and the cells were allowed to attach for 24 hours.
After the attachment~period the medium was replaced with 100 ul of virus inoculum containing approximately 0.5 x 104 plaque forming units of a plaque-puriXied strain of HzSNPV
(HzS-15). One hour was allowed for adsorption of the virus, after which the inoculum was replaced with 1 ml of fresh TNM-FH medium. The cells were monitored for 10 days, after which both the media and cells were collected for quantitation of ECV and OBs.
At 10 days post-inoculation, the cultures were collected and the cells and OBs were pelleted by centrifugation at 15,000 x g for 2 minutes. The ECV-con~aining supernatants were decanted and titered using the 50% tissue culture infective dose (TCID50) method tYamada et al., 1982, J. Invert. Path. 39:185-191) in Tarasaki microtiter plates (Lux)~o Tenfold serial dilutions of 20 in~ectious cell culture supernatants were co~bined with an equal volume of HZ1075/UND-K cells at a density of 5 x 105 cells per ml, and 100 ul of each dilution mix was aliquoted to each of ten wells in the Tarasaki plates. Wells were scored five days post infection for the presence of OBs in 25 cell nuclei, and the TCID5~ was calculated according to Reed and Meunch (1938, AmerO J. Hyg. 27:493-497).
Th~ OBs were released from the infected cells by resuspanding the pellets from each well in 1 ml of TE buffer (0.01 M Tris-HCl, 0.001 M EDTA, pH 7.5) containing 0.1% SDS.
30 The OBs were pelleted from the cell lysate at 15,000 x g for 5 minutes~ washed once with TE buffer, and resuspend~d in a final volume of 250 ul TE. The average number of OBs per ml of the starting cell culture ~as calculated from three independent direct hemocytometer counts.

.
' : ::

- -99- l 3 ~ a 8.1.4. ISOZYME ANALYSIS OF CELL ISOI~TES
,.
Monolayers of cells (25 cmG) were collected and pelleted at 1800 x g for 10 minute~. The media was decanted and the cells resuspended in lysis buffer (ly~is buf~er =
5 0.0152 M Tris, 0.046 M citric acid, 10% sucrose, 1% Triton X-100, 0.02 m~ bromophenol blue). The cells were broken by ~reezing ~at -70-C) and thawing (at 37~C) three times, and the cell lysates were cleared by centrifugation at 15,000 x g for 3 minutes. Cleared supernatants could be stored at -70C for prolonged periods without notic~able alteration of enzymatic activity.
Isozymes were detected following electrophoresi~ of the cleared cell lysates in 5% polyacrylamide gels in either TBE buffer (81.~ mM ~ris, 20 mM boric acid, 1.5 mM EDT~, pH
8.9) for enzymes esterase (EST~ and fumarate dehydratase (FUM~, or 2X TC buffer (19.4 m~ ~ris, 4.25 m~5 citric acid, pH 7.1) for enzymes lactate dehydrogenase (LDH) and malate dehydrogenase ~MDH). Vertical sla~ gels (20 x 20 cm~ were run at 350 volts for 2 hours in either TBE bu~fer or TC
20 bu~fer, and stained for the respective enzymes ~ollowing the protocols of Harris and Hopkinson, 1977, Handbook of Enzyme Electrophoresis in Human Genetics. ~msterdam: North Holland Publishing Co., p. 297.

8.2. C~ARACTERIZA~ION OF THE CELL LINES
8.2.1. CELL MORPHOLOGY
Twenty-~our cell strains labeled HZ1075/UND-A
through X were originally isolated by limited dilution 30 plating in 96-well plates. Many of the strains were composed of c~lls with extensive vacuolation. Most of these highly vacuolated strains eventually died, leaving a total of 13 strains, one o~ which was eventually lost to contamination.
The twelve surviving strains were fibroblastic in character, and each could ba distinguished basad upon a predominant cell morphology. Overall morphologies were . . . :
: ' - - ' ';
~, .

-100~ O

characterized as predominantly ellipsoidal with 2 or more extensions (uND-s~c~F~H~M~o~R~u) or irregular with several protoplasmic extensions (UND-G,H,L,K,U,V) (Note that the UND-H and UND-U cell population~ consisted largely of cells 5 of both morphologies.) All cell strains exhibit~d mixed morphologies even though each had arisen from a single cell.
Strain UND-B had the most uniform morphology with predominantly ellipsoid-shzped cells. Strain UND-G was characterized by extensive cytoplasmic vacuolation.

8.2.2. CELL GROWTH CURVES
Cell doubling times for the 12 surviving strains and the parental IPLB-HZ1075 cell line were determined by counting three defined areas of each cell monolayer in 25 15 cm2 tissue culture flasks at 48 hour intervals. All but two of the cell strain~ reached stationary growth phase after 96 hours. Strain UND-C entered stationary growth phase by 144 hours, while UND-K exhibited a biphasic growth curve with an apparent primary stationary phase from 96 to 144 hours, and 20 a second growth period between 144 and 1~6 hours (FIG. 8).
The population doubling times were calculated for each strain (Table VII) and ranged from 37.33 hours to 65.48 hours. The majority of cell strains had calculated doubling times between 45 and 60 hours.

. .

-101- ~. 3 2 ~

TA~LE VII
CELL DOU~LING TIMES AND RELATIVE VIRAL PR ~UCTION

AVERA~E
. DOUBLING POLYHEDRAL
STRAIN TIME(HRS.) COUNT ~/ML~ TCID50 -B 50.90 4.23 x 10~ c,d 6.02 x 104 10 ~ 48.~0 5.09 x 106 c 2.86 x 10~
F 59.19 1.37 x 1o6 f,~ 4.47 x 103 G 52~16 9.29 x 106 a 1.09 x 104 H 37.33 1.36 x 1o6 f~g 1.00 x 103 K 46.~5 9.74 x 106 k 1069 x 104 15 L 65.48 3.76 x 1o6 d,e 7.36 x 103 M 39.02 6.61 x 10 8.48 x 103 O 64.57 2.42 x 1o6 e,f 1.82 x 104 R 41.08 5.08 x 10 4.47 x 104 ~ 51.94 1.0~ x 106 g 1.09 x 103 59.82 2.62 x 106 e 6.02 x 103 HZ1075 63.15 3.42 x 10 ' 1.00 x 10 1 Cell doubling times and relative viral production of cloned IPLB-~Z1075 cell strains. Doubling times for each cell strain were calculated using the cell growth curves of Figure 8. Doubling times varied ~rom a low of 37.33 hours in strain UND-~ to a high of 65.48 hours in strain UND-L. The average doubling time was 51.37 hours. Duncan's multiple range analysis was used to determine significant differences in OB production. The average number of OBs presented for each cell strain represents three hemocytometer counts. Cell strains with the sam~ superscript letter are not significantly di~ferent.

:
~ ' : ' ` :

', :, : .
' -102- ~32~0 8.2.3. SUSCEPTIBILITY TO HzSNPV
The relative susceptibility of each cell strain to a plaque-purified HzS-15 strain of HzSNPV was gauged by determining the total number of OBs produced and the total 5 infectious ECV released by 10 days post infection. The relative number of OBs produced by 10 days post infection in each strain varied from l.OS x 106 per ml (UND-U) to 9.74 x 106 per ml (UND-K). Duncan's multiple range analy~is showed that the average OB counts can be separated into seven statistically 10 related groups (Table VII). The TCID50 values ranged from 1.0 x 10 per ml (UND-H~ to 6.02 x 104 per ml (UND-B) (Table VII).
There was no apparent correlation between the ability to produce OBs and high ~CV titers. The population doubling times were also unrelated to either ECV or OB production. For example, UND-B, with a popula~ion doubling time of ~0.9 hours, produced relatively moderate numbers of O~s (4.23 x 10 per ml) but released a relatively high titer of ECV (6.02 x 104 TCID50 per ml). Strain UND-U with a doubling time of 51.94 hours (similar to that of UND-B~ produced significantly fewer OBs 20 (1.06 x 106 per ml) and released relatively few ECV (1.09 x 103 TCID50/ml). Finally, UND~G had a doubling time of 52.16 hours (not significantly different from UND-B or ~U~ and had the second highest level of OB production (9.29 x 106 per ml) but only a moderate level of ECV release (1.09 x 104 TCID50/ml).

8.2.4. ISOZYME ANALYSIS OF CELL
STRAINS AND CEL~ LINES
To confirm the origin of the cloned cell strains, we compared their staining patterns for the isozymes fumarate hydratase (FUM), lactate dehydrogenase (LDH), esterase (EST) 30 (FIG. 9), and malate dehydrogenase (MDH, not shown) with those of both the IPLB-HZlG75 parental cell line and larval tissues from the host of origin, H. zea. The FUM, LDH, and MDH
patterns of all the cloned cell strains were identical to those of H. zea larval tissues and the parental IPLB-HZ1075 cell 35line.

11 3 2 ~

The pa tern obtained with esterase staining was particularly complex. While all the cloned strain patterns were similar to the larval and parental cell line patterns, individual differences were apparent between cloned cell 5 strains. ~his Purther subs~an~iates the clonal character of these cell strains.
The IPLB-HZ1075 cell line was compared to okher lepidopteran and one dipteran cell lines maintained in our laboratory. Cell homogenates were prepared and electrophoresed 10 as described above, and stained ~or either LDH or MDH (FIG.
10). The Rf values for each of the cell lina isozyme bands were calculated using the IPLB-HZ1075 bands as reference (Rf=1.0) and are presented in Table VIII.

TABLE VIII
Rf VALUES OF SEVERAL INSECT CELL LINES FOR LDH AND MDHa CELL LINE LDHb MDHc INSECT OF ORIGIN

ACT-10 1.03 10.0 AEDES AEGYPTI
BTI-EAA 0.77 4.0 ESTIGMENE ACREA
IPLB-HZ1075 1.00 1.0 HELIOTHIS ZEA
IPLB-SF-21AE 0.77 1~0 SPODOPTERA FURGIPERDA
25 TN-368 0.58 1.9 TRICHOPLUSIA NI
a Cell extracts were electrophoresed in a 5% polyacrylamide gel (95% acrylamide, 5% bis-acrylamide) in TC buffer and stained for L~H or MDH. Rf values were calculated relative to the migration of the IPLB-HZ1075 enzyme~
Lactate Dehydrogenase c Malate Dehydrogenase _ . ~ . .-. .
' - . ' ~ .: .

r-~ 104~

The pattern obtained for MDH distinguished the IPLB-HZ1075 cell line from all but one (IPLB-SF21AE) of the lepidopteran cell lines, and from the single dipteran cell line originated from Aedes aeqypti. Although the Spodoptera ~gi~ IPLB-SF21AE
5 cell line wa~ indistinquishable from IPLB-HZ1075 by ~taining for MDH, it was distinguishable by staining for LDH.

9. EXAMPLE: LARV~L HOSTS FOR USE IN GENERATION OF
RECOMBINANr OCOWSIC~ ODIES
The subsections below describe a method for growing Heliothis zea larvae in order to mass-Gulture recombinant viruses made in accordance with the invention. The non-melanizing ~rains of HzSNPV, such as those described in Section 6, supra, ara pr~ferred for use in the larval expression system~ of the present inv~ntion.

9.1. INSECT DIET PREP~RATION
This procedure descri~es the preparatio~ o~ diet or Trichoplu~ia ni and Heliothi~ zea. Th~ diet ~or Estigmene acrea require3 twics the ~ount oP vita~in ~ix.

FORMUL~

Per Liter Per 50~ ml Pinto bean~ in 90 ml water 18 g 9 g 25 A5ar (Sigma) 25 g12.5 g Vita~in diet forti~ic~tion 3.3 g 1.66 g mix~ur~ (ICN) Casein (Sig~a) 42 g21 g Sucros~ (ICN~ 42 g21 g 30 Wheat Germ (NIC) 36 g18 g *Wesson's Salt Mix (ICN) 12 g 6 g *Al~acel ~non-nutritive bulk) ~g 3 g Methyl Parabe~ in ~8 ~1 95% e~anol~.~6 g 0.83 g Sorbic Acid (Sig~a) 1.66 gO.83 g *Trade-mark :,4 .~., , - ~ '-" : ;. ~ ` ~

:-'- ,' . :~

-105 132~

Ascorbic Acid (Sigma) 5 g 2.5 g Streptomycin Sulfate (Sigma) 0.16 g 0.08 g The diet mix is prepared as follows:
1. The methyl paraben is diss~lved in alcohol before the procedure is started.
2. Bring 100 ml of water to a boil. While rapidly stirring, slowly add the agarO Rapid stirring is necessary to prevent the agar from clumping. I~ the agar clumps, ik 10 must be broken up with a spatula.
3. Reheat the mix until small bubbles appear on the side of the beaker.
4. Mix the melted agar and the pinto beans using a blender.
5. Mix on high for 1.5 minutes.
6. Add the rest of the components and mix on high for another 1.5 minutes.
7. Dispense the media into the appropriate containers.
Allow to cool at room temperature (15-30 minutes).
8. Store the media in the refrigerator in a sealed container until needed.

9.2. COLONY MAINTENANCE
The diet prepared as described is dispensed so that each larva receives a minimum of approximately 10 ml.

9.2.1 REARING OF T. NI OR H. ZEA
The rearing conditions ~or T. ni or H. zea are: 28-30~C, relative humidity of 65-70%, and photoperiod of 12 hours 30 (each 24 hours).
Upon pupation, the pupae are combined into a ~light cage of one cubic foot size, at 40 pupae/cage. The adult moths are allowed to emerge and feed on a mixture of 250 g sucrose, 25 ml honey, 5 g ascorbic acid, 5 g methyl paraben (dissolved 35 initially in 5 ml 95~ ethanol before addition), plus 500 ml distilled H20 ~components are dissolved with moderate heat).
The feeding mix is presented to ~he adult moths in a 1~ ml . ~

~ 3 '~
conical centrifuge tube equipped with a cap through which a two inch long dental wick extends, so that the adults can feed on the dental wick. The walls o~ the flight cage are lined with sterilized paper towels, on which the adult moths 5 lay their eggs. The paper towels containing the eggs are removed from the cage with aseptic precautions and transferred to a plastic box, termed a crisper, in which some of the agar-based diet mixture is present. The larvae emerge from the eggs in the crisper. Since the larvae are 10 positively phototropic, a light is shone at the end of the crisper where the diet mix is located/ so that the larvae move toward the diet mix and feed upon it. The feeding of the larvae on the diet mix increases yield of larvae, since the larvae are cannibalistic and would otherwise eat other 16 larva~ and unhatched eggs. Because the larvae are cannibalistic, they are segregated within a day after they emerge. This segregation is accomplished by hand, using an artist's brush which has been sterilized in 0.25% C~orox~, and rinsed with double distilled ~2 The brush is used to 20 gently lift the larvae and place each individual larva alone inko a cup. The larvae are allowed to grow in the cups until pupation, at which time the pupae are placed into the flight cages~
Since T. ni are not as cannibalistic as H. zea, a 25 slightly dif~erent protocol is alternatively used. After T.
ni eggs are laid upon the paper towels, a one inch square piece o~ the paper towel is placed on the lid o~ a quart cup (J. Cupt Dart Container Corp., Mason, Michigan, Cat. No.
8SJ20) containing 20-30 ml of liquid agar-based diet mixture 30 that has solidi~ied. The cup is inverted on top of the paper. The larvae hatch and migrate up toward the diet surfa~e. The paper is then removed ~rom the bottom and the ¢ups are turned right-side up. The larvae are allowed to grow in the cups until pupation, at which time the pupae are 35 placed into the flight cayes.

-107- 132~

9.2.2. REARING OF G~ MELONELLA
The reari~g of G. melonella largely follows the same protocol as described in Ssction 9.2.1. ~upra for T. ni or H.
zea, with the following dif~erences:
No photoperiod is necessary ~or G. melonella growth.
The temperature can range from 25 to 30C. The relative humidity of ambient room aondition~ is suitable.
The larva diet for G. melonella is composed of a mixture of 200 ml honey, 100 ml glycerine, and 1 box Gerber~'s mixed 10 cereal~ The diet mixture is put into a quart Mason jar fitted with a wire~mesh screen at the top, into whi~h the G.
melonella eggs are placed. After the larvae emerge, more diet mix is added as necessary until the larvae form cocoons.
Since G. melonella larvae are not as cannibalistic as H. zea 15 larvae, it is not necessary to segregate the larvae. When the larvae are in the last instar stage and start forming cocoons, they are taken out of the jar by hand (with gloves) ~nd placed together in a crisper. The insects pupate and the adults emerge withi~ the crisper. The adult moths lay eggs 2~ in the cracks or crevices of the crisper without feeding.
Eggs are thus laid at the interface of the lid and the box, so that when the lid is removed, the eggs adhere to the lid.
The eggs are then stripped off the lid by using a razor blade, and placed in a Mason jar containing the diet mixture.

9.3. GERM-FREE COLONIES
We are engaged i~ the establishment of insect colonies which are totally germ-free. We have been able to sterilize H. zea and T. ni. eggs, with egg survival of the 30 sterilization process. The eggs are placed on toweling paper and exposed to peracetia acid for 30 minutes. The eggs are then placed in a sterile environmen~ (an isolater) and rinsed off with sterile water while wi~hin the isolater. The eggs thus sterilized give rise to ge~m-free larvae.

.
: .. .
.
' , : : -`` -108- ~ 3 ~

10. EXAMPLE: HELIOTHIS POLYHEDRIN GENE AND
PROMOTER IN AUTOGRAPHA SHUTTLE VECTOR
Plasmid pEcoRI-I (Adang, M.J. and Miller, L.K., 1982, J.
Virol. 4~:782-7~3: Rohel, D~Zo~ et al., 1983, Virology 124:357~365; Smith, G.~., et al., 1982, J. Virol. 440199-208~, containin~ the polyhedrin gene of AcNPV, was used as starting material for the construction of an Autographa shuttle vector containing the ~eliothis polyhedrin gene and promoter (Figs.
13, 14). A 2 kb XhoI to BamHI ~ragment was isolated and subcloned into the SalI and BamHI site~ of ~13mpl9, generating 10 clone mpl9pEcoIXB. A DNA fragment was synthesized, corresponding to the Autographa polyhedrin sequence extending from the EcoRV site in the promoter region to the transcription inîtiation site, followed by a multiple cloning site (MCS) containing BamHI, EcoRI, SalI, and ~I restric-tion enzyme 15 recognition sites. Synthesis of the oligonucleotide was by use of` an Applied Biosystems ~odel 380A DNA synthesizer (automated phosphoramidite chemistry). The sy~thetic fragment was cloned between the E RV and KpnI ~ites of mpl9pEcolXB, resulting in clone mpl9Ald.
2~ The HindIII to ~I fragment of ~pl9Ald was isolated.
(The XhoI site was lost in the cloning into the Sal~ site of mpl9, and the HindIII site of the mpl9 MCS i5 a convenient nearby site). The KPnI to BamHI fragment of pEcoXI-I was also isolated. These two fragments were cloned into the HindIII and 25 SstI sites of the MCS o~ pUC12 (FIG. 13). The ligation included a synthetic oligonucleotide, 5'-GATCAGCT-3', in order to permit the ligation of the BamHI end of the pEcoRI-I
fragment into an SstI end of pUC12, and to remove the BamHI
site probably by the mechanism shown below.

--log- ~32~6~ ~

---G + 5'-GATCAGCT-3' ~ c~---- ~CCT~G 5~ oligo 3-tcgag~

BamHI end SstI end ---GGATCAGCTc~
---CCTAGtcgag----0 The resulting clone, pAVl.5, included Autographa sequences extending frvm the XhoI ~ite 5~ of the polyhedrin gene to the transcription initiation site, a MCS, and Autographa se~uences extending from the KPnI site in the carboxy-coding end of the polyhedrin gens to a ~amHI site 3 t of the gene. The XhoI and BamHI sites were lost.
Plasmid pAVl.5 and plasmid pHXl2 were used as the parental plasmids for the construction shown in Figure 14. A 2 kb PstI-EcoRI ~ragment of pAVl.5 ~contalning the Autographa polyhedrin promoter) and a 4.2 kb SalI-PstI fragment of pAVl.5 20 (containing pUCl2 sequences) were isolated. A ~.2 kb EcoRI-SalI fragment of pHXl~ (containing the Heliothis polyhedrin promoter and coding sequences) was isolated, and ligated to the PstI-EcoRI and SalI-PstI pAVl 5-derived fragm~nts. The resulting plasmid, termed pAVHp6, contains the Heliothi~
25 polyhedrin promater and coding sequences, flanked by Autographa polyhedrin ~equences includin~ the Autographa polyhedrin promoter. pAVHp6 can thu~ be used to transfer the Heliothis polyhedrin gene into AcNPV through ln vivo recombination, resulting in a recombinant virus that can comprise an 30 expression system in accordance with the present invention.
pAVHp6 can also be used to create a recom~inant AcNPV with two polyhedrin promoters~ One thus has the potential to exprPss two different heterologou genes within the same virus. In additionl if foreign DNA is inserted and expressed under the ,, - - - ,, ,. ~ .: , , : .
~: : , : --llO- ~2~6~0 control of the Autographa promoter in such a recombinant virus, ~he parental Heliothis polyhedrin promoter and gene can presumably ensure the retention o~ occlusion body formation.

1 0 .1. AUTOGRAPHA SHUTTLE VECTORS
ENCODING AN EPITOPE OF THE
INFLUENZA HEMAGGLUTININ WITHIN
THE POLYHEDRIN GENE
The strate~y b~ing used to construct an Autographa shuttlP
vector containing sequences which encode an epitope of 10 in~luenza hemagglutinin within a portion of the polyhedrin coding ~equences is diagrammed in Figure 1~.
Figure 15 depicts a ~trategy for cloning amino acids 98-106 of the influenza hemagglutinin lnto the amino-terminal coding ~equence o~ the Autographa polyhedrin gene. This 15 strategy can be u~2d to attempt to insert the influenza se~uence into the Autographa polyhedrin sequence contain~d in the M13 derivative mpl9EcoIXB (described in Section 10., su~ra) within the sequence encoding the second amino acid of the polyhedrin protein. An oligonucleotide (termed Rol-l) can be 20 synthesized (Applied Biosystems Model 380A), which is homologous to the region containing the ~II cleavage site within the codon ~or amino acid 2. Rol--l is annealed to mpl9EcoIXB single-stranded DNA, which is then cut with HpaII.
Annealing of the oligonucleotide creates the requisite double-25 stranded region for restriction endonuclease cleavage. Thelinear single-stranded DNA with paII-derived ~nd~ is isolated by heat denaturation and gel purification. An oligonucleotide corresponding to ami~o acid 98-106 of influenza hemagglutinin (termed Rol-2) is synthesizedO Rol-2 is then ann~aled tn a 30third synthetic oligon~cleotide tRol-3) which i5 co~plementary to Rol-2. In addition, Rol-3 ha~ 5' and 3' termini which extend beyond Rol-2 which are complementary to the ~
derived ends of the i~olated single-stranded phage DNA. Thus the ann~aled Rv1-2/Rol-3 DNA can be ligat~d to the i~vlated 35single-stranded phage DNA, forming a circular DNA moleculeO

.
'` ' ' ; ' ' ', .

3 2 ~
After transformation of bacterial cells with the ligated complex, the desired transformant can be selected by hydridization to radiolabeled Rol-3 according to the procedure of Benton and Davis (1977, Science 196:180-182). In addition, 5 Rol-3 encodes two restriction sites, MluI and NsiI, which are not found in the parental mpl9EcoIXB DNA. Thus, the identity of selected transformants can be confirmed by the presence of MluI and NsiI restriction Bites in the phage DNA isolated from transformants.
As an alternative, a similar strategy to that described supra may be used in order to cut the polyhedrin sequence contained within mpl9EcoIXB at the BamHI site within the sequence encoding amino acid 58.

11. EXAMPLE: PRODUCTION OF RECOMBINANT
OCCLUSION BODIES EXPOSING ~N EPITOPE
OF INFLUENZA HEMAGG~UTININ _ The subsections below describe manipulations of the polyhedrin gene of Autograæha californica to form recombinant occlusion bodies that expose antigenic determ.inants of 20 foreign organisms. ~he construction o~ 5 different recombinant polyhedrin genes containing a short DNA sequence encoding an influen~a hemaglutinin epitope are described.
The five recombinants are named In~em-1, InHem-~, InHem-43, InHem-50, and InHem-43/50, in which nInHem~ signifies the 25 influenza hemagglutinin epitope and the numbered suffix indicates the amino acid residue of the baculovirus polyh~drin sequence into which the hemagglutinin epitope was inserted. Three of these genes encode proteins that form recombinant OBs (InHem-l, InHem-43 and InHem-50) while the 30 other two do not form lattices. Interestingly, insertion of the hemagglutinin epitope into the polyhedrin variable region around amino acids 38 50 results in cuboidal OBs that do not e~bed virions.

: ~ . :
. - , ~. ~ -,.. -,................ :;

~2~6~
The immunological data generated demonstrate that the recombinant OBs are antigenic and immunoreact with antibodies specific for the foreign epitope. For example, monoclonal antibodies raised to the authentic influenza hemagglutinin 5 epitope bind to the dena~ured recombinant polyhedrin proteins in Western blots. Furthermore, these antibodies also interact with non-denatured purified recombinant OBs. The ability of the recombinant OBs to precipitate or capture antibodies to the influenza hemagglutinin epitope suggests 10 that the recombinant structures may be valuable as diagnostic reagents. Additionally, preliminary results with a limited number of animal~ indicates that one o~ the recombinants induces an immunogenic response to the hemagglutinin epitope.

11.1. CONSTRUCTION OF SHUTTLE VECTORS
Alternations within the polyhedrin gene were introduced into the baculovirus genome by homologous recombinations in vivo following cotrans~ection of susceptible cells with both viral DNA and transfer plasmids 20 containing the altered gene. The transfer plasmids are bacterial plasmids containing the viral segment surrounding the polyhedrin gene. In particular, a series of transfer vectors that contain 2kb of baculovirus sequences 5' of the polyhedrin gene, the sequence of the altered polyhedrin gene, and approximately 1.5 kb of 3' flanking seguences were used.
The long flanking sequences facilitated the transfer of the polyhedrin gene in the transfer plasmid into the viral genome by homologous recombination in vlvo.
Since the initial regions chosen for manipulation were 30 contained on a 2 kb fragm2nt extending from an XhoI site upstream of the polyhedrin gene to a BamHI site corresponding to amino acid residue number 58 of the polyhedrin gene, this fragment was subcloned into mpl9 (see FIG. 16), As shown in ~;
':
': , - !

-113~

FIG. l~A, the ml9 subclone of the Autographa polyhedrin gene, mpl9Xho/Bam, described supra, contains a 2kb insert extending from an XhoI site 5' of the polyhedrin gene.
New restriction sites were introduc~d into the gene by 5 in vitro mutagenesis u~ing the procedures developed by ~unkel, 1985, ProcO Natl. Acad. Sci. 82:488-492. By propagating the mpl9 Xho/Bam subclone in a dut ung strain in the presence of uridine monophosphate, we isolated uracil-containing plus-strand DNA. The minus strand was synthesized 10 in vitro in the presence of deoxyribonucleotides and primed with a synthetic oligonucl00tide that hybridized to the region to be mutated. The primer contained the appropriate mismatches to introduc~ the desired mutation. When the double strand was used to tran~fect dut+ung~ E. coli, progeny 15 derived from the minus strand were preferentially recovered.
The uracil containing plus strand is not efficiently used as a template in a ung+dut~ strain.
The following procedures were used to introduce the influenza epitope in the modifi~ble region betwèen amino 20 acids 43 and 50 of the polyhedrin sequence (See FIG. 16A).
Using the oligonucleotid~ Crec5:
5'GGTAGCCTCTTAGATCTCATGTTCGGCG-3' a GC base pair at nucleotide 127 was changed to a TA base pair. This change introduced a ~glII site into the wild type 25 polyhedrin gene ~equence. ~he alteration in the mpl9 subclone was tr~nsferred into the transfer vector by replacing the Pst/Bam fragment of the transfer vector with the corresponding fragment of the mutated mpl9 subclon~
(Crec5mpl9Xho/Bam). The resulting transfer vector, pAV15, 30 contained a new unique B~lII site at a position corresponding to a~ino acid residue number 43 and a naturally occurring BamHI site at a position corresponding to amino acid residue number 58. A synthetic oligonucleotide encodinq the influenza epitope followed by the polyhedrin gequence from 3 amino acid residue 50 to 58 was cloned into the BglII/BamHI

:

, :, - ~
. ~: . : ~

-114~ Z~

si~e of pAV15 (see FIG. 16A). Taking advantage of the degeneracy of the genetic code the oligonucleotide introduced an XbaI site at a position corresponding to amino acid residue number 50 of the polyhedrin se~uence. The resulting 5 trans~er vector, pAVlSInhem, contained an altered polyhedrin gen~ coding for a polyhedrin in which amino acid residues between 43 and 50 were replaced with the influenza epitope.
By cloning synthetic oli~onucleotides between the BglII and XbaI site~ of pAV15Inhem, new trans~er vectors were 10 constructed coding for polyhedrin pxoducts in which amino acids between 43 and 50 (those that were deleted in pAV15Inhem) were reinserte~ either before or after the influenæa epitope (see FIG. 16B). Thus, three transfer vectors were constructed. In the ~irst, pAVl5Inhem, the 15 influenza epitope replaced the ~equence between amino acids, 43 and 50. In the second, pAVInhem-43, the epitope was inserted at amino acid residue 43 of the polyhedrin sequence.
In the third, pAVInhem-50, the epitope was placed at amino acid residue 50. No polyhedrin sequences were deleted from the latter two constructs.
Using similar strategies, a SphI site was introduced at the initiating methionine~ The influenza epitope wa~ then inserted after amino acid xesidue number 1 (p~V17b Inhem-l) as shown in FIG. 17A and FIG. 17B. As shown in FIG. 17A and 17B, pAVl.5 (FIG. 13) was used to construct three intermediate plasmids, pAV11, pAV12 and pAV13; these, in turn, were used to construct pAV17 which has a unique SphI
site located at the ATG of the polyhedrin gene (FIG. 17A).
Using a synthetic oli~onucleotide and the in vitro mutagenesis technique, pAV17 was converted to pAV17b (FIG.
17B) which is characterized by the unique SphI site at the initiating ATG of the polyhedrin gene and a unique ~amHI si~e within the polyhedrin gene (i.e., the BamHI ~ite located at amino acid -8 in pAV17 was elim~nated). A synthetia -115- ~32~
oligonucleotide encoding the infIllenza hemagglutinin epitope was then cloned into the ~hI site o~ pAV17b resulting in pAV17b-InHem-1 (FIG. 17B).
In another set of constructs (see FIG. 18) a unlque ~I site was introduced within the polyhedrin gene enabling the insertion of the influenza epitope after amino acid resi-due number 2 (pAV17bInHem-2 as shown in FIG. 18).
A cassette vector was constructed using pBR322 as the plasmid backbone. This vector, called pBRX13, allows for the insertion of the coding se~lence for any epitope into the poly-hedrin gene spanning the coding region for amino acid residue numbers 36 through 50. Once the recombinant polyhedrin is con-structed, the entire polyhedrin gene sequence can be cut out of the pBRX13 vector and cloned into a transfer vector where it is flanked by baculoviral sequences that allow for n YlYQ recom-bination wlth virus.
The construction of the pBRX13 vector, illustrated in FIG. 19, was accomplished as follows: pBR322 was cut with ~çQRI and ~I and the following oligonucleotide (HE50/HE51) was cloned into the pBR322 backbone so that the E~oRI, PstI and one ~m~I site of pBR322 is eliminated and replaced with a unlque ~I and ~hQI site:
PstI BqlII
Not I ~I EcoRII ~oI
5'-AATTGCGGCCGCTGCAGGTACCGAATTCTCGAGATCTTGCA-3' 3'-~GCCGGCGACGTCCATGGCTTAAGAGCTCTAGA-5' The resulting plasmid, which is now ~m~S~Q~r, was cut with XmnI
and PvuII and religated in order to eliminate a second ~m~I
site (and ~II site~ located in the pBR32~ backbone. The re-sulting plasmid was named pBRX. Then the ~ I fragment of pAVlSInHem (which encodes the entire polyhedrin gene contain-ing the influenza hemagglutinin epitope) was cloned into the ~hQI/EanI site of pBRX. The resulting plasmid, pBRX13 contains the first 213 a.a. of the polyhedrin sequence except that ~'~3'i~
-- i ' , , : ' " ,: `

~3~6~ ~
the coding region for amino acid residues 43-50 is replaced by the influenza epitop~. The polyhedrin sequence of pBRX13 contains a uniqu~ XmnI, BqlII and XbaI site spanning the amino acid 36-50 re~ion so that the coding sequence for any 5 epitope can be cloned into this region. Once a foreign epitope is cloned into the unique sites, the entire recombinant polyhedrin gene can be excis~d from pBRX-13 using KpnI and EcoRV. This ~ragment can then be cloned into a transfer vector so that the recombinant polyhedrin sequence 10 is flanked by baculovirus sequencee to allow for in vivo recombination with virus.

11.2. PREPARATION OF RECOMBINANT VIRUSES
The transfer of the altered polyhedrin genes into the baculovirus genome was accomplished by homologous recombination in vivo between viral DNA and the transfer plasmids. The viral and plasmid DNAs were introduced into susceptible cells by cotransfection. Transfections involving calcium phosphate precipitation of DNA yielded the most 20 consistent results. Cells were preseeded on 60 mm culture dishes in growth mediaO Calcium phosphate precipitated ~NA
was added to the media and the cells incubated for 12-18 hours. The media was then removed and fresh media added.
The cells were then incubated for 4 or 5 days at which time 25 most cells were infected with virus. Although only a small perc~ntage of the cells are initially transfected, the rest of the cells are infected by the progeny of later rounds of infectio~.
The progeny of the transfection were plaqued and the 30 recombinant viruses were id~nti~ied from the parental virus on the basis of plaque morphology7 Two types of cotransfections were set up to identify recombinant occlusion body formation. In the first cotransfection, viral DNA was derived from a strain in which the polyhedrin g~ne had bPen 35 replaced with the bacterial CAT gene. Since this virus has -117 1 ~ 2 ~

no polyhedrin gene it fails to make OBs. If the recombinant polyhedrin gene encodes a protein that will form an occlusion body, the recombinant virus is detected in plaque assays among the large number of parental types which ~ail to make 5 OBs. Since viruses producing occlusion bodies form refractile plaques, these recombinants are easily detected against the OB negative background. The second cotransfection involved the use of wild type viral DNA. In this case, a recombinant failing to make OBs could be 10 detected among the wild type progeny. In this case the rare recombinant forms a non-refractile plaque.
We identified five recombinant viruses: Inhem-1, Inhem-2, Inhem-43 and Inhem-50 containing the altered polyhedrin gene in which the influenza hemaglutinin epitope is inserted after amino acids 1, 2, 43, and 50, respectively.
In the polyhedrin gene o~ the recombinant virus ~nhem-43/SO, the influenza epitope replaces tha polyhedrin sequence between amino acids 43 and 50.
Three of the recombinant viruses encode polyhedrin 20 proteins that form occlusion bodies. Inhem-1 forms OBs that are indistinguishable from wild type by light microscopy.
Unlike the irregular occlusions formed by wild type, Inhem-43 and Inhem-50 form large cuboidal occlusions. A clue indicating how the Inhem-43 and Inhem-50 alterations result 25 in the formakion of a regular cuboidal lattice W2S provided by alectron micrographs of the recombinant occlusions.
Apparently Inhem-43 and Inhem-50 do not embed virions in the OBs. Conceivably the virions embedded in the wild type OB
act as impurities and interfere with regular lattice 30 formation. By failing to embed virions these mutants may form large, regular lattices. The other two recombinants, Inhem-2 and Inhem-43/50 did not fo~m occlusion bodies.

.. .
, .. . . .

, 11. 3 . IMMUNOLO(::ICAL ANALYSES OF THE
RECOMBINANT CCLUSION BODIES
The data discussed below indicate that the recombinant OBs described above expose the influenza hemagglutinin epitope as analyzed by ELISA immunoassay, immunoprecipitation and Western immunoblotting. In addition, preliminary results indicate that the recombinant OBs are immunogenic and capable of eliciting an immune response specific for the hemagglutinin epitope.
11. 3 .1. ELISA ANALYSIS OF SURFACE EXPRESSION OF
UNDENATURED INFLUENZA EPITOPE ON
P~ECOMBINANT OCCLUSION BODIES
Plates were coated with lOO ~g/ml ACTR (Autographa wild type virus) or lOO ~l of recombinant OBs isolated from 15 one T75 flask in 2 ml of TE buffer (43-2Bl, 43-2BlA, 50-llAl, 50-l lBl). After overnight incubation at 4~C, the plates were washed three times with PBS and coated with 1% BSA for 45 minutes at 37C. lOO ~l of anti-influenza hemagglutinin in 2% BSA was added to each well and incu~ated for 90 minutes at 37C. After three washes with PBS lOO ~l an anti-mouse IgG conjugated to alkaline phosphata~e (103 dilution) in 1%

~SA was added to each well. The wells were washed 3 times with PBS after which lOO ~l of 0.4 mg/ml p-nitrophenol phosphats in diethanolamine buffer was added to each well and incubated at room temperature for 30 minutes. The plates were read at a wavelength of 405 nm after 30 minutes. The results are shown in Table IX below.

-, -llg ~32~

. _ .
TABLE IX

ELISA ASSAY FOR PRESENTATION OF INFLUENZA
EPITOPE BY RECOMBINANT OCCLUSION BODIES
_ Recombinant Virus ~405 nm AC~R 0.168 43-2B1 0.864 43 2BlA 0.77~
50-1 lAl 0.895 50-1 lBl 0.787 TE Buffer 0.141 .

11.3.2. WESTERN BL0T ANALYSES OF DENATU~ED
RECOMBINANT OCCLUSION BODIES
Western blot analyses indicated that monoclonal antibodies raised to the peptide sequence of the influenza hemagglutinin epitope cross react with the denatured 20 recombinant polyhedrin protein. Proteins from lysates of cells infected with the recombinant viruses were electrophoretically separat~d in acrylamide gels and blotted onto nitocellulose. Incubation of the blots with a monoclonal antibody (courtesy of Dr. Ian Wilson) raised to 25 the peptide sequense of the epitope indicated that the monoclonal antibody recognized the recombinant polyhedrin but did not recognize wild type polyhedrin.

llo 3.3. IMMUNOPRECIPITATION ASSAYS OF ~ECOMBINANT
INFLUENZA~POLYHEDRIN CRYSTALS _ _ Immunoprecipitation data i.ndicate that the anti-influenza hemagglutinin monoclonal antibody ~M~b) al50 ~nteracts with non-denatur~d recombinant polyhedrin. In these experiments purified occlusion bodies from InH~m-43, , .
.~ ; ... :
, - , .

, :: . . ...

InHem-50 or ACTR (wild type) infected cells were incubated with either BSA, mouse anti-influenza Mabl or mouse anti-plasminogen Mab (as the negative control). The occlusion bodies were pelleted and washed repeatedly. The OBs were 5 then incubated with alkaline phosphatase conjugated rabbit anti-mouse antibody. The OBs were pelleted, washed several times and then incubated with a chromogenic substrate.
The results indicate that the OBs from InHem-43 and InHem-50, but not from ACTR, bound to and precipitated the 10 anti-influenza Mab. These recombinant OBs did not precipitate the anti-plasminogen Mab, indicating that the binding to the anti-influenza epitope represents a specific rather than non-specific interaction. These results indicate that the influenza epitope is exposed on non-denatured recombinant OBs and can be recognized by the antibodies raised to the authentic epitope.
Details of the method and results presented in Table X
are described in more d~tail: 1.3 ml polypropylene tubes were coated for about 1 hour at 37C with 1% BSA in PBS. 100 20 ~1 of purified polyhedra (1 T75 flask prep in 2 mls TE
Buffer) was added to each coated ~ube and washed 3 times with 1% BSA ~centrifuge between washes for 5 minutes). 100 ~1 anti-influenz2 hemagglutinin (at about 0.1 mg/ml) in 2% BSA
was added to the polyhedra pellet and made up to 0.5 ml with 1% BSA. The samples were incubated at room temperature for about 90 minutes with mixing. The ~amples were pelleted and washed 3 times with 1% BS~. Anti-mouse IgG conjugated to alkaline phosphatase (100 ~1~ at 103 dilution in 1% BSA was added and made up to 500 ~1 with 1% BSA. The samples were incubated at room temperature for 90 minutes. They were then centrifuged and washed 3 times with 1% BSA. 0.5 ml of 0-nitrophenol P04 substrate (0.4 mg/ml) in diethanolamine buffer was added and the samples mixed for 30 minutes at room temperature. They were then centri~uged and the supernatant 3 read at a wavelength of 405 nm.

. ~. . .

132~6~ ~

TABLE X
PRECIPITATION OF INFLUENZA HEMAGGLUTININ MONOCLONAL
ANTIBODY WITH RECOMBINANT OCCLUSION BODIES
A
Sample Antibody 405 nm 43-2Bl l~ BSA 0.00l 43-2BlA BSA 0.004 50-llAl BSA 0.003 50-llAl BSA 0.002 10 ACTR BSA 0.003 43-2~1 anti-inf. he~ag. l.291 43-2BlA anti-inf. hemag. l.219 50-llAl anti-inf. hemag. 0.9l5 15 50-l lBl anti-inf . hemag. 0.596 ACTR anti-in~. hemag. 0.0l7 43-2Bl 28G tPg antibody) N.D.
43-2BlA 28G (Pg antibody) 0.055 20 ~0-l lAl ~8G (Pg antibody) 0.073 50-l lBl 28G (Pg antibody) 0.063 ACTR 28G (Pg antibody) 0.064 _ 25 Similar result~ were obtained for InHem-l.

11.3.4. IMMUNOGENICITY OF THE RECOMBINANT
OCCLUSION BODIES
The data in Table XI show results of preliminary 30 studies testing the immunogenicity of the recombinant OBs.

In the~e tests, mice were inoculated with purified OBs from Inhem-43 infected cells. At various times the animals were bled and the ~era tested for antlbodies specific for the influenza hemagglutinin epitope in ~ISA assays. In these .
: . . ; . . .

~2~

assays we looked for the presence of mouse antibodies that would bind to the 9 amino acid peptide ~ixe~ to plastic wells. The preliminary results indicate that the recombinant OBs induced an immunogenic response to the influenza peptide.
5 The measurements represent a single animal per time point.

TA~LE XI
INDUCTION OF ANTIBODIES SPECIFIC FOR INFLUENZA
HEMAGGLUTININ EPITOPE USING RECOMBINANT_OBs AS IMMUNOGEN

A405 nm~ _ Serum Dilution Time lo2 2 x 102 5 x Io2 5 0 weeks 0.280 0.296 0.269 1 0.361 0.319 0.308 2 0.45~ 0.377 0.339 3 0.813 0.636 0.384 4 0.548 0.492 0.366 *Mab to peptide 1.480 ~-.
. . . . . .
.. , ~, . ~, , .

~: . . . . .
. . . .. .

123 ~32~

12. DEPOSIT OF MICROORGANISMS
The following E. coli strains carrying the listed plasmids have been deposited with the Agricultural Research Culture Collection (NRRL), Peoria, IL, and have been a~signed the 5 following accession numbers:

E. coli Strain Plasmid Accession Number K12 (DH5) pHX12 NRRL B-18172 X12 (DH5) pHH5 NRRL B-18173 K12 (DH5) pHE2.6 NRRL B-18174 K12 (DH5) pHE2.61ac NRRL B-18175 K12 (DH5) pAVHp6 NRRL B-18176 K12 (Dh5 alpha) pAV15InHem-43 B-1&308 15 K12 (Dh5 aIpha) pAV15InHem-50 B-18309 K12 (Dh5 alpha) pAV17bInHem-1 B-18310 K12 (Dh5 alpha) pAV17bInHem-2 B-18311 K12 (Dh5 alpha) pBRX-13 B-18312 The following eliothis zea cell line and Heliothis zea NPV i~olate have ~een depo~ited with the American Type Culture Collection, Rockville~ MD, and have been assigned the listed accession numbers:
Accession Number Heliothis zea cell line: IPLB-HZ1075/UND-K ATCC CRL 9281 Heliothis zea NPV isolate: HzS-15 ATCC VR 2156 The present invention is not to be limited in scope by the microorganisms and cell line deposited since the deposited 30embodiment is intended as a singl~ illustration of one aspect oX the invention and any microorganisms or viruses which are functionally equivalent are within the scope of this invention.
Indeed, various modifications of the invention in addition to those shown and described hereln will become apparent to those ,: "
, ' ; , ,, . ' - . ' ' - .

1 3 2 ~
skilled in the art from the ~oregoing description and accompanying drawings. Such modifications are intended to fall within the scope oE the appended claims.
It is also to be under~tood that all ba~e pair sizes given 5 for nucleotides are approximate and are used for purposes of description, and figures which diagrammatically depict DNA
sequences are not necessarily drawn to scale.

' :' : : ' ,' , .;, . .

'' ~ ~' - '' - .
,.

-' ~; . -:

Claims (51)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A recombinant occlusion body comprising repeating subunits in which each subunit comprises a polyhedrin fusion protein comprising a portion of the polyhedrin protein which participates in crystallization fused to a foreign amino acid sequence.

2. The recombinant occlusion body according to claim 1 in which the foreign amino acid sequence is related to an epitope of a pathogenic microorganism.

3. The recombinant occlusion body according to claim 2 in which the pathogenic microorganism comprises a virus.

4. The recombinant occlusion body according to claim 3 in which the epitope comprises influenza hemagglutinin.

5. The recombinant occlusion body according to claim in which the epitope comprises amino acids 98-106 of influenza hemagglutinin.

6. The recombinant occlusion body according to claim 3 in which the virus comprises Hepatitis A virus.

7. The recombinant occlusion body according to claim 2 in which the foreign amino acid sequence is exposed on the surface of the occlusion body.

8. The recombinant occlusion body according to claim 1 in which the foreign amino acid sequence comprises an antigenic determinant of a foreign protein.

9. The recombinant occlusion body according to claim 1 in which the foreign amino acid sequence replaces all or a portion of the amino terminus of the polyhedrin protein amino acid sequence.

10. The recombinant occlusion body according to claim 1 in which the foreign amino acid sequence replaces all or a portion of a region homologous to the amino acid sequence substantially as depicted in FIG. 1 from amino acid residue number 37 to 49.

11. The recombinant occlusion body according to claim 5 in which the foreign amino acid sequence replaces all or a portion of a region homologous to the amino acid sequence substantially as depicted in FIG. 1 from amino acid residue number 37 to 49.

12. The recombinant occlusion body according to claim 1 in which the foreign amino acid sequence is inserted after amino acid residue number 1 of the Autographa polyhedrin sequence substantially depicted in FIG. 2.

13. The recombinant occlusion body according to claim 1 in which the foreign amino acid sequence replaces amino acid residue number 43 of the Autographa polyhedrin sequence substantially depicted in FIG. 2.

14. The recombinant occlusion body according to claim 1 in which the foreign amino acid sequence replaces amino acid residue number 50 of the Autographa polyhedrin sequence substantially depicted in FIG. 2.

15. A polyhedrin fusion protein which is capable of crystallizing with other polyhedrin proteins to form recombinant occlusion bodies, comprising: a portion of the polyhedrin protein which participates in crystallization fused to a foreign amino acid sequence.

16. The recombinant polyhedrin protein of claim 15 in which the foreign amino acid sequence comprises an epitope of a pathogenic microorganism.

17. The recombinant polyhedrin protein of claim 16 in which the pathogenic microorganism comprises a virus.

18. The recombinant polyhedrin protein of claim 17 in which the pathogenic microorganism comprises influenza virus.

19. The recombinant polyhedrin protein of claim 18 in which the foreign amino acid sequence comprises amino acids 98-106 of the influenza hemagglutinin.

20. The recombinant polyhedrin protein of claim 17 in which the pathogenic microorganism comprises Hepatitis A virus.

21. The recombinant polyhedrin protein of claim 15 in which the foreign amino acid sequence replaces all or a portion of the amino terminus of the polyhedrin protein.

22. The recombinant polyhedrin protein of claim 15 in which the foreign amino acid sequence replaces all or a portion of a region homologous to the amino acid sequence substantially as depicted in FIG. 1 from amino acid residue number 37 to 49.

23. The recombinant polyhedrin protein of claim 15 in which the foreign amino acid sequence comprises an antigenic determinant of a foreign protein.

24. The recombinant polyhedrin protein of claim 15 in which the foreign amino acid sequence is inserted after amino acid residue number 1 of the Autographa polyhedrin sequence sub-stantially as depicted in FIG. 2.

25. The recombinant polyhedrin protein of claim 15 in which the foreign amino acid sequence replaces amino acid residue number 43 of the Autographa polyhedrin sequence substantially as depicted in FIG. 2.

26. The recombinant polyhedrin protein of claim 15 in which the foreign amino acid sequence replaces amino acid residue number 50 of the Autographa polyhedrin sequence substantially as depicted in FIG. 2.

27. A recombinant virus which directs the expression of polyhedrin fusion proteins that crystallize to form recombinant occlusion bodies, comprising:
(a) a polyhedrin promoter; and (b) a nucleotide sequence encoding a polyhedrin fusion protein comprising (i) a first nucleotide sequence encoding a portion of the polyhedrin structural protein that participates in crystallization and (ii) a second nucleotide sequence encoding a foreign protein, in which the first and second nucleotide sequences are in the same translational reading frame uninterrupted by translation termination signals: and in which the nucleotide sequence encoding the polyhedrin fusion protein is under the control of the polyhedrin promoter so that polyhedrin fusion proteins which crystallize to form recombinant occlusion bodies are produced in a suitable host infected with recombinant virus.

28. The recombinant virus according to claim 27 comprising a baculovirus.

29. The recombinant virus according to claim 28 comprising a nuclear polyhedrosis virus.

30. The recombinant virus according to claim 29 comprising Autographa californica nuclear polyhedrosis virus.

31. The recombinant virus according to claim 29 comprising Heliothis zea nuclear polyhedrosis virus.

32. The recombinant virus according to claim 28 comprising a granulosis virus.

33. The recombinant virus according to claim 27 in which the foreign peptide comprises an epitope of a pathogenic microorganism.

34. The recombinant virus according to claim 33 in which the pathogenic microorganism comprises a virus.

35. The recombinant virus according to claim 34 in which the epitope is related to an epitope of influenza hemagglutinin.

36. The recombinant virus according to claim 27 in which the nucleotide sequence encoding the foreign peptide replaces all or part of the polyhedrin gene that encodes the amino terminus of the polyhedrin protein.

37. The recombinant virus according to claim 27 in which the nucleotide sequence encoding the foreign peptide replaces all or part of a region of the polyhedrin gene that is homologous to the nucleotide sequence substantially as depicted in FIG. 1 from nucleotide number 142 to 180.

38. The recombinant virus according to claim 27 in which the nucleotide sequence encoding the foreign peptide is inserted after amino acid number 1 of the Autographa polyhedrin gene substantially as depicted in FIG. 2.

39. The recombinant virus according to claim 27 in which the nucleotide sequence encoding the foreign peptide replaces the part of the Autographa polyhedrin gene that encodes amino acid residue number 43 substantially as depicted in FIG. 2.

40. The recombinant virus according to claim 27 in which the nucleotide sequence encoding the foreign peptide replaces the part of the Autographa polyhedrin gene that encodes amino acid residue number 50 substantially as depicted in FIG. 2.

41. A transfer vector encoding a polyhedrin fusion protein that crystallizes to form recombinant occlusion bodies comprising:
(a) a first nucleotide sequence encoding a portion of the polyhedrin structural protein that participates in crystallization; and (b) a second nucleotide sequence encoding a foreign peptide in which the first and second nucleotide sequences are in the same translational reading frame uninterrupted by translation termination signals: and (c) baculovirus flanking sequences surrounding the first and second nucleotide sequence so that recombination with baculovirus can occur in vivo.

42. The transfer vector according to claim 41, in which the nucleotide sequence encoding the foreign peptide replaces all or part of the polyhedrin gene that encodes the amino terminus of the polyhedrin protein.

43. The transfer vector according to claim 41, in which the nucleotide sequence encoding the foreign peptide replaces all or part of a region of the polyhedrin gene that is homologous to the nucleotide sequence substantially as depicted in FIG. 1 from nucleotide number 142 to 180.

44. The transfer vector according to claim 41 in which the nucleotide sequence encoding the foreign peptide is inserted after the coding sequence for amino acid number 1 of the Autographa polyhedrin gene substantially as depicted in FIG. 2.

45. The transfer vector according to claim 41 in which the nucleotide sequence encoding the foreign peptide replaces the part of the Autographa polyhedrin gene that encodes amino acid residue number 43 substantially as depicted in FIG. 2.

46. The transfer vector according to claim 41 in which the nucleotide sequence encoding the foreign peptide replaces the part of the Autographa polyhedrin gene that encodes amino acid residue number 50 substantially as depicted in FIG. 2.

47. The transfer vector according to claim 41, comprising pAV15InHem-43, substantially as deposited with the NRRL and assigned accession number B18308.

48. The transfer vector according to claim 41, comprising pAV15InHem-50, substantially as deposited with the NRRL and assigned accession number B18309.

49. The transfer vector according to claim 41, comprising pAV17bInHem-1, substantially as deposited with the NRRL and assigned accession number B18310.

50. The transfer vector according to claim 41, comprising pAV17bInHem-2, substantially as deposited with the NRRL and assigned accession number B18311.

51. A recombinant vector, pBRX13, substantially as deposited with the NRRL and assigned accession number B18312.